Practical asymmetric synthesis of a non-peptidic αvβ 3 antagonist

Article abstract of DOI:10.1021/jo048082n

(Chemical Equation Presented) The development of a practical and highly convergent synthesis of an α_vβ₃ antagonist is described. The two key fragments present in this compound, a tetrahydropyrido[2,3-b]azepine ring system and a chiral 3-aryl-5-oxopentanoic acid, were constructed independently and then coupled at a late stage using a Wittig reaction. The pyridoazepine moiety was prepared from N-Boc 6-chloro-2-aminopyridine via directed ortho-metalation/alkylation followed by in situ cyclization. A Suzuki reaction was then used to attach the propionaldehyde side-chain required for Wittig coupling. The coupling partner was prepared from asymmetric methanolysis of a 3-substituted glutaric anhydride followed by elaboration of the acid moiety to the requisite β-keto phosphorane. Using this route, kilogram quantities of the desired drug candidate were prepared.

Full text of DOI:10.1021/jo048082n

Practical Asymmetric Synthesis of a Non-Peptidic rvâ3 Antagonist
Stephen P. Keen,* Cameron J. Cowden,* Brian C. Bishop, Karel M. J. Brands,
Antony J. Davies, Ulf H. Dolling, David R. Lieberman, and Gavin W. Stewart
Department of Process Research, Merck Sharp & Dohme Research Laboratories, Hertford Road,
Hoddesdon, Hertfordshire EN11 9BU, United Kingdom
stephen_keen@merck.com; cameron_cowden@merck.com
Received October 28, 2004
The development of a practical and highly convergent synthesis of an Rvâ3 antagonist is described.
The two key fragments present in this compound, a tetrahydropyrido[2,3-b]azepine ring system
and a chiral 3-aryl-5-oxopentanoic acid, were constructed independently and then coupled at a
late stage using a Wittig reaction. The pyridoazepine moiety was prepared from N-Boc 6-chloro-
2-aminopyridine via directed ortho-metalation/alkylation followed by in situ cyclization. A Suzuki
reaction was then used to attach the propionaldehyde side-chain required for Wittig coupling. The
coupling partner was prepared from asymmetric methanolysis of a 3-substituted glutaric anhydride
followed by elaboration of the acid moiety to the requisite â-keto phosphorane. Using this route,
kilogram quantities of the desired drug candidate were prepared.
Introduction
Structurally, 1 consists of a tetrahydropyrido[2,3-b]-
azepine attached via a 4-carbon tether to a 3-aryl-5-
oxopentanoic acid moiety containing the sole chiral
center. At the outset, one of the main objectives was to
The vitronectin receptor Rvâ3 is a member of the
integrin family, a class of adhesion receptors that mediate
cell-cell and cell-matrix interactions.1 In particular, Rvâ3
is highly expressed in osteoclasts, multinucleated cells
responsible for bone resorption, and is believed to play a
critical role in the adhesion and migration of osteoclasts
on the bone surface.2 Furthermore, it has been demon-
strated that ligands which bind with high affinity to Rvâ3,
such as RGD-containing peptides or small molecule RGD
(Arg-Gly-Asp) mimetics, inhibit bone resorption in vivo.3
Therefore, Rvâ3 antagonists are of considerable interest
for the treatment of osteoporosis,4 a disease characterized
by a decrease in bone mass resulting from increased bone
resorption relative to formation. As part of a program to
develop an orally active antagonist of the Rvâ3 receptor
for the treatment of osteoporosis,5 compound 1 was
identified as a candidate for pre-clinical development.
(3) For example, see: (a) Fisher, J. E.; Caulfield, M. P.; Sato, M.;
Quartuccio, H. A.; Gould, R. J.; Garsky, V. M.; Rodan, G. A.; Rosenblatt,
M. Endocrinology 1993, 132, 1411-1413. (b) King, K. L.; D’anza, J.
J.; Bodary, S.; Pitti, R.; Siegel, M.; Lazarus, R. A.; Dennis, M. S.;
Hammonds, R. G.; Kukreja, S. C. J. Bone Miner. Res. 1994, 9, 381-
387. (c) Engleman, V. W.; Nickols, G. A.; Ross, F. P.; Horton, M. A.;
Griggs, D. W.; Settle, S. L.; Ruminski, P. G.; Teitelbaum, S. L. J. Clin.
Invest. 1997, 99, 2284-2292. (d) Yamamoto, M.; Fisher, J. E.; Gentile,
M.; Seedor, J. G.; Leu, C.-T.; Rodan, S. B.; Rodan, G. A. Endocrinology
1998, 139, 1411-1419. (e) Lark, M. W.; Stroup, G. B.; Hwang, S. M.;
James, I. E.; Rieman, D. J.; Drake, F. H.; Bradbeer, J. N.; Mathur, A.;
Erhard, K. F.; Newlander, K. A.; Ross, S. T.; Salyers, K. L.; Smith, B.
R.; Miller, W. H.; Huffman, W. F.; Gowen, M. J. Pharmacol. Exp. Ther.
1999, 291, 612-617. (f) Miller, W. H.; Alberts, D. P.; Bhatnagar, P.
K.; Bondinell, W. E.; Callahan, J. F.; Calvo, R. R.; Cousins, R. D.;
Erhard, K. F.; Heerding, D. A.; Keenan, R. M.; Kwon, C.; Manley, P.
J.; Newlander, K. A.; Ross, S. T.; Samanen, J. M.; Uzinskas, I. N.;
Venslavsky, J. W.; Yuan, C. C.-K.; Haltiwanger, R. C.; Gowen, M.;
Hwang, S.-M.; James, I. E.; Lark, M. W.; Rieman, D. J.; Stroup, G. B.;
Azzarano, L. M.; Salyers, K. L.; Smith, B. R.; Ward, K. W.; Johanson,
K. O.; Huffman, W. F. J. Med. Chem. 2000, 43, 22-26. (g) Duggan, M.
E.; Duong, L. T.; Fisher, J. E.; Hamill, T. G.; Hoffman, W. F.; Huff, J.
R.; Ihle, N. C.; Leu, C.-T.; Nagy, R. M.; Perkins, J. J.; Rodan, S. B.;
Wesolowski, G.; Whitman, D. B.; Zartman, A. E.; Rodan, G. A.;
Hartman, G. D. J. Med. Chem. 2000, 43, 3736-3745.
(1) (a) Hynes, R. O. Cell 1992, 69, 11-25. (b) Horton, M. A. Int. J.
Biochem. Cell Biol. 1997, 29, 721-725. (c) van der Flier, A.; Sonnen-
berg, A. Cell Tissue Res. 2001, 305, 285-298.
(2) (a) Rodan, S. B.; Rodan, G. A. J. Endocrinol. 1997, 154, S47-
S56. (b) Duong, L. T.; Rodan, G. A. J. Bone Miner. Metab. 1999, 17,
1-6. (c) Teitelbaum, S. L. Science 2000, 289, 1504-1508.
10.1021/jo048082n CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/05/2005
J. Org. Chem. 2005, 70, 1771-1779
1771

Keen et al.
SCHEME 1
SCHEME 2
SCHEME 3
make the synthesis of 1 as convergent as possible. With
this in mind, it was envisaged that the carbon-carbon
bond R-â to the keto group could be formed late in the
synthesis via a Wittig or Horner-Wadsworth-Emmons
reaction of aldehyde 2 with the requisite phosphorane/
phosphonate 3 (Scheme 1). The success of this proposed
route therefore depended upon developing efficient syn-
theses of these two key fragments. This paper describes
the successful accomplishment of these goals.
Results and Discussion
Synthesis of Aldehyde 2. It was anticipated that
aldehyde 2 would be available from elaboration of a
pyridoazepine containing a suitable substituent at C-2.
However, reported syntheses of the tetrahydropyrido[2,3-
b]azepine ring system are limited to several specific
examples,6,7 only a couple of which are compatible with
a single C-2 substituent. Of these, we were particularly
interested in the intramolecular Diels-Alder reaction
of N-hex-5-yn-1-yl-5-(trifluoromethyl)-1,2,4-triazin-3-
amine, reported by Seitz and co-workers.7 We quickly
discovered, however, that under similar conditions the
5-methyl analogue (5), prepared via reaction of triazine
48 with hex-5-ynylamine,9,10 does not undergo this cy-
cloaddition (Scheme 2). This is perhaps not surprising
given the inverse electron demand nature of these Diels-
Alder reactions.11
The focus, therefore, switched to alternate intramo-
lecular Diels-Alder substrates. Based on precedent from
the Padwa group, for forming the analogous 6,6-fused
tetrahydro-1,8-naphthyridine system,12 the intra-
molecular Diels-Alder reactivity of oxazole 7, obtained
via alkylation of 613,14 with 6-bromo-1-hexene, was ex-
amined (Scheme 3). Disappointingly, under thermal
conditions, 7 failed to provide either the desired pyri-
doazepine or the intermediate Diels-Alder adduct. At-
tempts to cyclize the analogous oxazoles 9 or 10 proved
equally unsuccessful.15
(4) For recent reviews on the development of Rvâ3 antagonists, see:
(a) Miller, W. H.; Keenan, R. M.; Willette, R. N.; Lark, M. W. Drug
Discovery Today 2000, 5, 397-408. (b) Duggan, M. E.; Hutchinson, J.
H. Expert Opin. Ther. Pat. 2000, 10, 1367-1383. (c) Hartman, G. D.;
Duggan, M. E. Expert Opin. Invest. Drugs 2000, 9, 1281-1291. (d)
Coleman, P. J.; Duong, L. T. Expert Opin. Ther. Pat. 2002, 12, 1009-
1021.
(5) For the most recent publications in a series documenting the
discovery of non-peptide Rvâ3 antagonists, see: (a) Breslin, M. J.;
Duggan, M. E.; Halczenko, W.; Hartman, G. D.; Duong, L. T.;
Fernandez-Metzler, C.; Gentile, M. A.; Kimmel, D. B.; Leu, C.-T.;
Merkle, K.; Prueksaritanont, T.; Rodan, G. A.; Rodan, S. B.; Hutch-
inson, J. H. Bioorg. Med. Chem. Lett. 2004, 14, 4515-4518. (b)
Coleman, P. J.; Brashear, K. M.; Askew, B. C.; Hutchinson, J. H.;
McVean, C. A.; Duong, L. T.; Feuston, B. P.; Fernandez-Metzler, C.;
Gentile, M. A.; Hartman, G. D.; Kimmel, D. B.; Leu, C.-T.; Lipfert, L.;
Merkle, K.; Pennypacker, B.; Prueksaritanont, T.; Rodan, G. A.;
Wesolowski, G. A.; Rodan, S. B.; Duggan, M. E. J. Med. Chem. 2004,
47, 4829-4837.
(6) For example, see: (a) Hawes, E. M.; Davis, H. L. J. Heterocycl.
Chem. 1973, 10, 39-42. (b) Jo¨ssang-Yanagida, A.; Gansser, C. J.
Heterocycl. Chem. 1978, 15, 249-251. (c) Seitz, G.; Dietrich, S.; Go¨rge,
L.; Richter, J. Tetrahedron Lett. 1986, 27, 2747-2750. (d) Marcelis,
A. T. M.; van der Plas, H. C. Tetrahedron 1989, 45, 2693-2702. (e)
Pa¨tzel, M.; Ushmajev, A.; Liebscher, J.; Granik, V.; Grisik, S.;
Polievktov, M. J. Heterocycl. Chem. 1992, 29, 1067-1068.
(7) (a) John, R.; Seitz, G. Arch. Pharm. (Weinheim, Ger.) 1989, 322,
561-564. (b) John, R.; Seitz, G. Chem. Ber. 1990, 123, 133-136.
(8) Paudler, W. W.; Chen, T.-K. J. Heterocycl. Chem. 1970, 7, 767-
771.
(9) Mu¨ller, T. E.; Pleier, A.-K. J. Chem. Soc., Dalton Trans. 1999,
583-587.
(10) For an example of an analogous reaction between an alkylamine
and triazine 4, see: Jacobsen, N. W.; de Jonge, I. Aust. J. Chem. 1987,
40, 1979-1988.
(11) Intramolecular Diels-Alder reactions of 1,2,4-triazines have
been used to prepare 2,3-dihydropyrrolo[2,3-b]pyridines without electron-
withdrawing substituents on the triazine moiety: Taylor, E. C.; Pont,
J. L. Tetrahedron Lett. 1987, 28, 379-382.
(12) Padwa, A.; Brodney, M. A.; Liu, B.; Satake, K.; Wu, T. J. Org.
Chem. 1999, 64, 3595-3607.
(13) Cockerill, A. F.; Deacon, A.; Harrison, R. G.; Osborne, D. J.;
Prime, D. M.; Ross, W. J.; Todd, A.; Verge, J. P. Synthesis 1976, 591-
593.
(14) Crank, G.; Foulis, M. J. J. Med. Chem. 1971, 14, 1075-1077.
1772 J. Org. Chem., Vol. 70, No. 5, 2005

Practical Synthesis of an Rvâ3 Antagonist
SCHEME 4
SCHEME 6
SCHEME 5
aldehyde 2, is not trivial. Because 2-amino-3-picolines
containing suitable functionality at this position are not
readily available, our focus shifted to 2-aminopyridine
substrates. Therefore, 2-amino-6-chloropyridine (16), pre-
pared from commercially available 2,6-dichloropyridine,20
was initially converted into the N-Piv derivative 17 using
a slightly modified literature procedure (Scheme 6).17
Dilithiation of 17 with BuLi/TMEDA and subsequent
reaction with 1-chloro-4-iodobutane afforded a 1:1 mix-
ture of alkylation product 18 and starting pyridine.
Surprisingly, cyclization of 18 was not observed upon
deprotonation with KHMDS under the previously de-
scribed conditions. Similarly, Reed and co-workers re-
ported that an analogous N-Piv 2-amino-3-(3-chloropropyl)-
pyridine failed to cyclize to the desired tetrahydro-1,8-
naphthyridine under comparable conditions.16c
Having already demonstrated that the analogous
chloride 14, containing the N-Boc group, could be readily
cyclized, we reexamined this chloropyridine route using
N-Boc derivative 19. Therefore, N-Boc aminopyridine 19
was prepared by deprotonation of aminopyridine 16 with
LiHMDS followed by addition of (Boc)2O (Scheme 6).
Dilithiation of 19 with BuLi/TMEDA at -78 °C, followed
by addition of 1-chloro-4-iodobutane and warming to
ambient temperature, resulted in clean conversion (90-
95%) to the chlorobutyl intermediate. Upon heating the
resulting reaction mixture to reflux, we were delighted
to find that this lithiated intermediate cyclized to afford
the desired pyridoazepine 20 in 78% yield. It was
subsequently found that this yield can be improved to
86% by transmetalation of dilithiated 19 with CuI prior
to addition of the alkylating agent.21
Following the failure of these Diels-Alder approaches,
we then considered the possibility of building the tet-
rahydroazepine ring onto a suitably functionalized pyri-
dine core. We envisaged that directed ortho-metalation
of an N-acyl 2-aminopyridine (11) followed by alkylation
with a suitable dihaloalkane and in situ cyclization would
provide the desired bicycle 12 (Scheme 4).
Several examples of dilithiation and subsequent C-
alkylation/acylation of N-Boc and N-Piv 2-aminopyridines
(11a)16,17 and 2-amino-3-picolines (11b)18,19 are docu-
mented in the literature. In particular, a previous report
from these laboratories had demonstrated that picoline
derivative 13 could be converted into the desired cycliza-
tion substrate 14.19 In that case, however, no attempt at
cyclization was made. Using this procedure, chloride 14
was readily obtained in high yield, and we were pleased
to find that treating this compound with KHMDS re-
sulted in clean conversion to the desired pyridoazepine
15 (Scheme 5). It was subsequently shown that these two
steps could be combined into a single operation to afford
pyridoazepine 15 directly from picoline 13. Thus, after
dilithiation of 13 and subsequent alkylation with 1-chloro-
3-iodopropane, the resulting reaction mixture, containing
lithiated 14, was simply heated to reflux to afford
pyridoazepine 15.
Although this is a direct and highly efficient route to
the tetrahydropyrido[2,3-b]azepine ring system, func-
tionalization of 15 at C-2, as required for elaboration to
(15) N-Benzyl oxazole 8 was prepared from benzylcyanamide using
the procedure described in ref 13.
(16) (a) Turner, J. A. J. Org. Chem. 1983, 48, 3401-3408. (b) Venuti,
M. C.; Stephenson, R. A.; Alvarez, R.; Bruno, J. J.; Strosberg, A. M. J.
Med. Chem. 1988, 31, 2136-2145. (c) Reed, J. N.; Rotchford, J.;
Strickland, D. Tetrahedron Lett. 1988, 29, 5725-5728. (d) Estel, L.;
Linard, F.; Marsais, F.; Godard, A.; Que´guiner, G. J. Heterocycl. Chem.
1989, 26, 105-112. (e) Turner, J. A. J. Org. Chem. 1990, 55, 4744-
4750. (f) Murray, T. J.; Zimmerman, S. C.; Kolotuchin, S. V. Tetrahe-
dron 1995, 51, 635-648. (g) Zimmerman, S. C.; Kwan, W.-S. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2404-2406. (h) Takayama, K.; Iwata,
M.; Hisamichi, H.; Okamoto, Y.; Aoki, M.; Niwa, A. Chem. Pharm. Bull.
2002, 50, 1050-1059. (i) Eldrup, A. B.; Christensen, C.; Haaima, G.;
Nielsen, P. E. J. Am. Chem. Soc. 2002, 124, 3254-3262.
With chloropyridoazepine 20 in hand, we envisaged
that elaboration into the required aldehyde could be
achieved via palladium-catalyzed coupling with a suitable
3-carbon unit. In particular, a Heck reaction of 20 with
allyl alcohol would provide aldehyde 24 in a single step.
Disappointingly, using either standard Heck conditions22a
or Jeffery modifications,22b none of the desired product
was observed. Although iodides are typically used for
Heck reactions between aryl/vinyl halides and allylic
(17) Thompson, A. M.; Rewcastle, G. W.; Boushelle, S. L.; Hartl, B.
G.; Kraker, A. J.; Lu, G. H.; Batley, B. L.; Panek, R. L.; Showalter, H.
D. H.; Denny, W. A. J. Med. Chem. 2000, 43, 3134-3147.
(18) (a) Clark, R. D.; Muchowski, J. M.; Fisher, L. E.; Flippin, L.
A.; Repke, D. B.; Souchet, M. Synthesis 1991, 871-878. (b) Curtis, N.
R.; Kulagowski, J. J.; Leeson, P. D.; Ridgill, M. P.; Emms, F.;
Freedman, S. B.; Patel, S.; Patel, S. Bioorg. Med. Chem. Lett. 1999, 9,
585-588.
(20) Kaminski, J. J.; Perkins, D. G.; Frantz, J. D.; Solomon, D. M.;
Elliott, A. J.; Chiu, P. J. S.; Long, J. F. J. Med. Chem. 1987, 30, 2047-
2051.
(21) For a more detailed discussion on the development and scope
of this methodology, see: Davies, A. J.; Brands, K. M. J.; Cowden, C.
J.; Dolling, U. H.; Lieberman, D. R. Tetrahedron Lett. 2004, 45, 1721-
1724.
(19) Hands, D.; Bishop, B.; Cameron, M.; Edwards, J. S.; Cottrell,
I. F.; Wright, S. H. B. Synthesis 1996, 877-882.
J. Org. Chem, Vol. 70, No. 5, 2005 1773

Keen et al.
SCHEME 7
SCHEME 8
alcohols,22,23 2-chloropyridines have been successfully
utilized in Heck reactions with other alkenes,24 as well
as in palladium-catalyzed Stille,25 Suzuki,26 Negishi,27
and Sonogashira28 couplings. We therefore tested the
reactivity of chloride 20 in a Heck reaction with ethyl
acrylate and were pleased to find that the expected
adduct 21 was obtained in high yield (Scheme 7).
Although ester 21 could theoretically be converted into
desired aldehyde 24, we continued to pursue a more
concise method of installing the propionaldehyde chain.
Thus, Suzuki coupling of chloride 20 with trialkylborane
22, prepared from commercially available acrolein diethyl
acetal and 9-BBN,29 was examined (Scheme 8). The only
precedent we could find for Suzuki reactions of similar
trialkylborane reagents derived from acrolein acetals
were a couple of examples which used vinyl triflates as
the coupling partners.30,31 We were, therefore, gratified
to find that a Pd(OAc)2/dppf-catalyzed reaction of chloride
20 with borane 22 cleanly afforded the desired acetal 23
SCHEME 9
in near quantitative yield (98% after chromatography).32
Alternatively, the crude Suzuki reaction mixture contain-
ing 23 could be treated with hydrochloric acid to afford
aldehyde 24 directly.
Synthesis of Phosphorane/Phosphonate 3. With
a route to aldehyde 24 in place, we next required a
synthesis of its coupling partner, phosphorane/phospho-
nate 3. It was hypothesized that this could be obtained
via enantioselective desymmetrization of cyclic anhydride
25, which in turn would be available from dehydration
of the corresponding diacid 26 (Scheme 9).
Preparation of diacid 26 started from the known
pyrimidine aldehyde 28,33 which was obtained from
reaction of O-methylisourea with vinamidinium salt 27
(Scheme 10).34 After examining a variety of bases and
solvents for this reaction, it was found that biphasic
conditions using aqueous KHCO3, with isopropyl acetate
as the cosolvent, gave the highest yields of aldehyde 28.
This was then converted into the desired diacid 26 using
an adaptation of a procedure reported by Smith and co-
workers,35 which was itself based on original Knoevena-
gel chemistry. Thus, piperidine-catalyzed condensation
of aldehyde 28 with 2 equiv of ethyl acetoacetate, followed
(22) For example, see: (a) Melpolder, J. B.; Heck, R. F. J. Org. Chem.
1976, 41, 265-272. (b) Jeffery, T. J. Chem. Soc., Chem. Commun. 1984,
1287-1289. (c) Jeffery, T. Tetrahedron Lett. 1990, 31, 6641-6644. (d)
Zhao, H.; Cai, M.-Z.; Hu, R.-H.; Song, C.-S. Synth. Commun. 2001,
31, 3665-3669.
(23) A handful of examples using aryl/vinyl bromides are also
known: (a) Groen, M. B.; Zeelen, F. J. Recl. Trav. Chim. Pays-Bas 1978,
97, 301-304. (b) Kao, L.-C.; Stakem, F. G.; Patel, B. A.; Heck, R. F. J.
Org. Chem. 1982, 47, 1267-1277. (c) Kirby, A. J.; Walwyn, D. R. Gazz.
Chim. Ital. 1987, 117, 667-680. In addition, since completing this work
an example utilizing an aryl chloride has been reported: (d) Littke,
A. F.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 6989-7000.
(24) (a) Li, J.; Chen, S.-H.; Li, X.; Niu, C.; Doyle, T. W. Tetrahedron
1998, 54, 393-400. (b) Li, J.; Luo, X.; Wang, Q.; Zheng, L.-M.; King,
I.; Doyle, T. W.; Chen, S.-H. Bioorg. Med. Chem. Lett. 1998, 8, 3159-
3164. (c) Niu, C.; Li, J.; Doyle, T. W.; Chen, S.-H. Tetrahedron 1998,
54, 6311-6318.
(25) For example, see: (a) Pomel, V.; Rovera, J. C.; Godard, A.;
Marsais, F.; Que´guiner, G. J. Heterocycl. Chem. 1996, 33, 1995-2005.
(b) Holladay, M. W.; Bai, H.; Li, Y.; Lin, N.-H.; Daanen, J. F.; Ryther,
K. B.; Wasicak, J. T.; Kincaid, J. F.; He, Y.; Hettinger, A.-M.; Huang,
P.; Anderson, D. J.; Bannon, A. W.; Buckley, M. J.; Campbell, J. E.;
Donnelly-Roberts, D. L.; Gunther, K. L.; Kim, D. J. B.; Kuntzweiler,
T. A.; Sullivan, J. P.; Decker, M. W.; Arneric, S. P. Bioorg. Med. Chem.
Lett. 1998, 8, 2797-2802. (c) Bates, R. W.; Boonsombat, J. J. Chem.
Soc., Perkin Trans. 1 2001, 654-656.
(26) For example, see: (a) Ohta, A.; Takahashi, N.; Shirokoma, Y.;
Yuasa, K.; Kurihara, T.; Miyamae, H. Heterocycles 1990, 30, 875-884.
(b) Molander, G. A.; Yun, C.-S. Tetrahedron 2002, 58, 1465-1470. (c)
Molander, G. A.; Katona, B. W.; Machrouhi, F. J. Org. Chem. 2002,
67, 8416-8423.
(27) For example, see: (a) Amat, M.; Hadida, S.; Pshenichnyi, G.;
Bosch, J. J. Org. Chem. 1997, 62, 3158-3175. (b) Tre´court, F.; Gervais,
B.; Mongin, O.; Le Gal, C.; Mongin, F.; Que´guiner, G. J. Org. Chem.
1998, 63, 2892-2897. (c) Gauthier, D. R.; Szumigala, R. H.; Dormer,
P. G.; Armstrong, J. D.; Volante, R. P.; Reider, P. J. Org. Lett. 2002, 4,
375-378.
(28) For example, see: (a) Sakamoto, T.; Shiraiwa, M.; Kondo, Y.;
Yamanaka, H. Synthesis 1983, 312-314. (b) Sakamoto, T.; An-naka,
M.; Kondo, Y.; Araki, T.; Yamanaka, H. Chem. Pharm. Bull. 1988, 36,
1890-1894. (c) Kim, T.-S.; White, J. D. Tetrahedron Lett. 1993, 34,
5535-5536.
(29) Hydroboration of acrolein diethyl acetal with 9-BBN has been
reported to give a 98:2 mixture of regioisomers favoring the terminal
position: Brown, H. C.; Chen, J. C. J. Org. Chem. 1981, 46, 3978-
3988.
(30) (a) Oh-e, T.; Miyaura, N.; Suzuki, A. J. Org. Chem. 1993, 58,
2201-2208. (b) Grigg, R.; Kennewell, P.; Savic, V. Tetrahedron 1994,
50, 5489-5494. (c) Grigg, R.; Savic, V.; Thornton-Pett, M. Tetrahedron
1997, 53, 10633-10642.
(31) Since completing this work, a couple of examples which use
aryl halides as the coupling partners have also been reported: (a)
Young, J. R.; Huang, S. X.; Walsh, T. F.; Wyvratt, M. J.; Yang, Y. T.;
Yudkovitz, J. B.; Cui, J.; Mount, G. R.; Ren, R. N.; Wu, T.-J.; Shen, X.;
Lyons, K. A.; Mao, A.-H.; Carlin, J. R.; Karanam, B. V.; Vincent, S.
H.; Cheng, K.; Goulet, M. T. Bioorg. Med. Chem. Lett. 2002, 12, 827-
832. (b) Beinhoff, M.; Karakaya, B.; Schlu¨ter, A. D. Synthesis 2003,
79-90.
(32) A more detailed account of this Suzuki coupling has been
reported elsewhere: Cowden, C. J.; Hammond, D. C.; Bishop, B. C.;
Brands, K. M. J.; Davies, A. J.; Dolling, U. H.; Brewer, S. E.
Tetrahedron Lett. 2004, 45, 6125-6128.
(33) (a) Gupton, J. T.; Gall, J. E.; Reisinger, S. W.; Smith, S. Q.;
Bevirt, K. M.; Sikorski, J. A.; Dahl, M. L.; Arnold, Z. J. Heterocycl.
Chem. 1991, 28, 1281-1285. (b) Ragan, J. A.; McDermott, R. E.; Jones,
B. P.; am Ende, D. J.; Clifford, P. J.; McHardy, S. J.; Heck, S. D.; Liras,
S.; Segelstein, B. E. Synlett 2000, 1172-1174.
(34) McWilliams, J. C.; Buck, E.; Eng, K. K.; Maligres, P. E.; Sager,
J. W.; Waters, M. S.; Humphrey, G. R. WO Patent 2002028840.
1774 J. Org. Chem., Vol. 70, No. 5, 2005

Practical Synthesis of an Rvâ3 Antagonist
SCHEME 10
the more challenging 3-substituted glutaric anhydrides
was reported to occur with up to 91% ee.
Therefore, these commercially available catalysts were
screened for their ability to catalyze methanolysis of
anhydride 25. Using 10 mol % of catalyst in THF at room
temperature, the enantiomeric excesses obtained ranged
from very poor to moderate (3-68% ee) with quinine-
derived catalysts tending to give products enriched in the
(R)-enantiomer (ent-29) and catalysts in the quinidine
series favoring the (S)-enantiomer (29).42 In the absence
of a catalyst, the reaction was very slow and only a trace
amount (<5%) of the acid-ester was formed after aging
for 48 h. In agreement with Deng’s findings, the most
promising results were achieved using anthraquinone
catalysts (DHQD)2AQN and (DHQ)2AQN (68% and 54%
ee, respectively),43 and reaction conditions were further
optimized therefore using these catalysts. Changes to
concentration, solvent, methanol charge, and catalyst
loading were all examined. However, only by lowering
the reaction temperature was a significant improvement
in enantioselectivity observed [82% ee at -40 °C using
(DHQD)2AQN]. Unfortunately, the reaction rate slowed
dramatically with decreasing temperature, and, at -40
°C, conversion of anhydride 25 to acid-ester 29 was
unacceptably slow (∼50% after 40 h). Moreover, these
catalysts are expensive and not readily available in bulk
quantities. Desymmetrization of anhydride 25 using
naturally occurring cinchona alkaloids was therefore
investigated.44
by deacylation/hydrolysis with aqueous NaOH and finally
acidification, afforded diacid 26 as a crystalline solid in
81% yield.
Dehydration of diacid 26 using conventional proce-
dures, such as treating with DCC36 or heating with excess
acetic anhydride,37 resulted in poor yields (50-60%) of
anhydride 25. However, using trifluoroacetic anhydride
(TFAA) as the dehydrating agent led to significant
improvements.38 Thus, treating diacid 26 with TFAA
resulted in rapid and clean conversion to anhydride 25,
which crystallized directly from the reaction mixture in
94% yield.
Literature precedent indicated that desymmetrization
of prochiral anhydride 25, to set the absolute stereo-
chemistry of our sole stereogenic center, would be best
accomplished by enantioselective alcoholysis.39 The re-
sulting enantiomerically pure half-ester could then be
readily converted into the desired phosphorane/phospho-
nate. There are numerous examples in the literature of
asymmetric alcoholysis of cyclic anhydrides;40 however,
we were particularly interested in a publication by Deng
and co-workers who reported excellent selectivity using
cinchona-based catalysts.41 Notably, desymmetrization of
(35) (a) Smith, W. T.; Kort, P. G. J. Am. Chem. Soc. 1950, 72, 1877-
1878. (b) Smith, W. T.; Shelton, R. W. J. Am. Chem. Soc. 1954, 76,
2731-2732.
(36) de Laszlo, S. E.; Bush, B. L.; Doyle, J. J.; Greenlee, W. J.;
Hangauer, D. G.; Halgren, T. A.; Lynch, R. J.; Schorn, T. W.; Siegl, P.
K. S. J. Med. Chem. 1992, 35, 833-846.
(37) For a typical procedure, see: Matsuda, F.; Kawasaki, M.;
Ohsaki, M.; Yamada, K.; Terashima, S. Tetrahedron 1988, 44, 5745-
5759.
(38) Using TFAA for dehydration of diacids to five-, six-, and seven-
membered cyclic anhydrides has been reported before. For example,
see: (a) Boger, D. L.; Hong, J.; Hikota, M.; Ishida, M. J. Am. Chem.
Soc. 1999, 121, 2471-2477. (b) Kim, H. O.; Ji, X.; Melman, N.; Olah,
M. E.; Stiles, G. L.; Jacobson, K. A. J. Med. Chem. 1994, 37, 3373-
3382. (c) Boger, D. L.; Baldino, C. M. J. Am. Chem. Soc. 1993, 115,
11418-11425.
Methanolysis of anhydride 25 using either catalytic or
stoichiometric amounts of the natural alkaloids, in THF
at 20 °C, gave disappointingly low enantioselectivities
(3-34% ee). However, by utilizing conditions (toluene,
-40 °C) similar to those reported by Bolm and co-
workers, for quinine/quinidine-mediated methanolysis of
succinic anhydrides,45 significantly improved enantiose-
lectivities were obtained. Thus, reaction of anhydride 25
(39) To the best of our knowledge, at the start of this project the
only reported asymmetric desymmetrization of cyclic anhydrides using
carbon nucleophiles was a single example using chiral Grignard
reagents: (a) Real, S. D.; Kronenthal, D. R.; Wu, H. Y. Tetrahedron
Lett. 1993, 34, 8063-8066. More recently, enantioselective desymme-
trizations of cyclic anhydrides using either aryl Grignard reagents or
diethylzinc in the presence of chiral ligands have been reported: (b)
Shintani, R.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 1057-1059.
(c) Bercot, E. A.; Rovis, T. J. Am. Chem. Soc. 2002, 124, 174-175.
(40) For recent reviews, see: (a) Spivey, A. C.; Andrews, B. I. Angew.
Chem., Int. Ed. 2001, 40, 3131-3134. (b) Chen, Y.; McDaid, P.; Deng,
L. Chem. Rev. 2003, 103, 2965-2984.
(42) The absolute stereochemistry was assigned by analogy with
Deng’s results (see ref 41). In addition, completing the synthesis of 1
using the acid-ester obtained from desymmetrization with quinidine-
based catalysts provided final material with the correct absolute
stereochemistry for the biologically active Rvâ3 antagonist.
(43) The following catalysts were also screened: (DHQ)2PYR, 3%
ee; cinchonidine, 3% ee; hydroquinidine, 5% ee; quinidine, 11% ee;
(DHQ)2PHAL, 12% ee; DHQ-CLB, 16% ee; DHQ-MEQ, 16% ee; quinine,
20% ee; DHQ-PHN, 44% ee; DHQD-PHN, 65% ee.
(41) Chen, Y.; Tian, S.-K.; Deng, L. J. Am. Chem. Soc. 2000, 122,
9542-9543.
J. Org. Chem, Vol. 70, No. 5, 2005 1775

Keen et al.
SCHEME 11
SCHEME 12
ee.47 Using this procedure, (S)-acid-ester 29 was obtained
in typical yields of 60-64%.
With a route to both enantiomers of the acid-ester in
place, we wished to compare the formation and subse-
quent coupling performance of â-ketophosphonate 3148
and phosphorane 34. The desired phosphonate was
prepared by reaction of acid-ester ent-29 with 4 equiv of
lithiated diisopropyl methylphosphonate, followed by
esterification of the resulting acid (Scheme 11).49 The
fairly disappointing yield (46%) initially obtained for the
first step was significantly improved by transmetalation
of the lithiated phosphonate with MgBr2 prior to reaction
with ent-29. Phosphorane 34 was in turn prepared from
(S)-acid-ester 29 via activation of the acid moiety and
subsequent reaction with methylenetriphenylphos-
phorane. After examining several methods for activating
acid-ester 29,50 it was found that the best yields of
phosphorane 34 were obtained via either isobutoxy
anhydride 32 or tert-butyl mixed anhydride 33. Although
comparable yields (65-73%) of 34 were obtained using
either of these methods, mixed anhydride 33 was found
to be more stable and easier to handle.
Completing the Synthesis of 1. With both phos-
phonate 31 and phosphorane 34 in hand, a comparison
of the Wittig and Horner-Wadsworth-Emmons cou-
plings revealed that the reaction of phosphorane 34 with
aldehyde 24 occurred in significantly higher yield than
the corresponding coupling using phosphonate 31 (Scheme
12). Thus, heating a solution of aldehyde 24 and phos-
phorane 34 to 80 °C afforded a 90% yield of desired enone
with methanol in the presence of 1.0 equiv of quinidine
afforded 29 with 62% ee (Scheme 11). Similarly, the
analogous reaction with quinine provided ent-29 in 58%
ee.46 Attempts to further improve the enantioselectivity
of these reactions by optimization of temperature, sol-
vent, methanol charge, and concentration proved unsuc-
cessful.
We therefore required an upgrade procedure to obtain
the desired acid-ester in enantiomerically pure form. It
was discovered that for quinine-mediated methanolysis
of anhydride 25, the enantiopurity of the resulting (R)-
acid-ester could be increased, to 87% ee, by simply
filtering off the quinine salt of ent-29 at the end of the
reaction (65% yield). Unfortunately, a similar upgrade
of (S)-acid-ester 29, via isolation of the quinidine salt
from the reaction mixture, was not possible. A break-
through was achieved, however, when it was discovered
that crystallization of acid-ester 29 from aqueous HCl,
during a salt-break of the quinidine salt (62% de), was
accompanied by an upgrade in enantiopurity to >96%
(44) Use of cinchona alkaloids for desymmetrization of meso and
prochiral cyclic anhydrides has been documented in the following
publications: (a) Hiratake, J.; Yamamoto, Y.; Oda, J. J. Chem. Soc.,
Chem. Commun. 1985, 1717-1719. (b) Hiratake, J.; Inagaki, M.;
Yamamoto, Y.; Oda, J. J. Chem. Soc., Perkin Trans. 1 1987, 1053-
1058. (c) Aitken, R. A.; Gopal, J.; Hirst, J. A. J. Chem. Soc., Chem.
Commun. 1988, 632-634. (d) Aitken, R. A.; Gopal, J. Tetrahedron:
Asymmetry 1990, 1, 517-520. (e) Shimizu, M.; Matsukawa, K.;
Fujisawa, T. Bull. Chem. Soc. Jpn. 1993, 66, 2128-2130. (f) Starr, J.
T.; Koch, G.; Carreira, E. M. J. Am. Chem. Soc. 2000, 122, 8793-8794.
(g) Mittendorf, J.; Benet-Buchholz, J.; Fey, P.; Mohrs, K.-H. Synthesis
2003, 136-140.
(45) (a) Bolm, C.; Gerlach, A.; Dinter, C. L. Synlett 1999, 195-196.
(b) Bolm, C.; Schiffers, I.; Dinter, C. L.; Gerlach, A. J. Org. Chem. 2000,
65, 6984-6991.
(46) Methanolysis of 25 in the presence of cinchonine or cinchonidine
occurred with poor enantioselectivity (<23% ee) under all conditions
examined.
(47) A more detailed study of this crystallization, and the different
crystal morphologies of acid-ester 29, revealed that the crystal form
of 29 isolated from this ee upgrade was not the most thermodynami-
cally stable form. The observation that extended aging (>6 h) occasion-
ally led to isolation of material with much lower ee (<70%) also
supports this conclusion.
(48) The diisopropyl phosphonate was prepared in place of the more
commonly used dimethyl analogue due to isolation problems associated
with the high water solubility of the intermediate acid in the latter
case.
(49) For analogous preparations of â-ketophosphonates from 3-sub-
stituted glutaric acid monomethyl esters, see: (a) Karanewsky, D. S.;
Malley, M. F.; Gougoutas, J. Z. J. Org. Chem. 1991, 56, 3744-3747.
(b) Konoike, T.; Araki, Y. J. Org. Chem. 1994, 59, 7849-7854.
(50) Preparation of phosphorane 34 via the acid chloride, imida-
zolide, or Weinreb amide of 29 resulted in significantly lower yields.
1776 J. Org. Chem., Vol. 70, No. 5, 2005

Practical Synthesis of an Rvâ3 Antagonist
red-brown dianion solution was aged for 1 h at -70 °C, and a
solution of 1-chloro-4-iodobutane (7.57 kg, 34.7 mol) in THF
(5 L) was then added over 35 min, once again keeping the
temperature below -65 °C. Upon completion of the addition,
the reaction was left to gradually warm to room temperature
overnight and then heated to reflux for 9 h. The resulting
solution was cooled to 60 °C and washed with water (54.5 L),
while maintaining the internal temperature at >40 °C. The
35, whereas deprotonation of phosphonate 31, using
KHMDS, and reaction with aldehyde 24 provided the
same product in only 53% yield. A further advantage of
the Wittig route is that, unlike phosphonate 31, which
is an oil, phosphorane 34 can be conveniently purified
by crystallization. Typically, the Wittig reaction was run
in 2-propanol, and the resulting crude reaction mixture,
containing enone 35, was hydrogenated directly to afford
keto-ester 36, in overall assay yields of 81-88%. Crude
36 was then hydrolyzed to keto-acid 37, which was
isolated, in 70% overall yield, via crystallization. Finally,
removal of the Boc group provided enantiomerically pure
(>99% ee) zwitterionic 1 in typical yields of 74-81%.
i
aqueous layer was back-extracted with PrOAc (54.5 L), and
the combined organic layers were then washed with water (27
L). The resulting organic solution was concentrated in vacuo
to a volume of 26 L and then solvent switched to heptane (26
L). Product crystallized from solution during this solvent
switch, and the slurry obtained was cooled to 5 °C and aged
for 1 h. The solid was then isolated by filtration, washing the
wet-cake with cold heptane (10 L). Drying under vacuum at
40 °C afforded pyridoazepine 20 (5.05 kg, 78%) as a white
solid: mp 164.5-167.5 °C; 1H NMR (400 MHz, CD2Cl2) δ 7.52
(d, J ) 7.9 Hz, 1H), 7.13 (d, J ) 7.9 Hz, 1H), 3.9-3.1 (br, 2H),
2.70 (dd, J ) 6.4, 5.0 Hz, 2H), 1.84-1.78 (m, 2H), 1.71-1.58
(m, 2H), 1.41 (s, 9H); 13C NMR (101 MHz, CD2Cl2) δ 155.9,
154.1, 147.5, 141.8, 134.1, 123.0, 80.9, 47.3, 33.3, 29.9, 28.6,
26.2. Anal. Calcd for C14H19ClN2O2: C, 59.47; H, 6.77; N, 9.91.
Found: C, 59.44; H, 6.78; N, 9.86.
Conclusions
In summary, Rvâ3 antagonist 1 has been synthesized
in 17% yield, over the longest linear sequence, via a
scalable and highly convergent route. Highlights of this
synthesis include a one-pot alkylation/cyclization to
construct the seven-membered ring of the pyrido[2,3-b]-
azepine moiety, highly efficient Suzuki coupling to attach
the propionaldehyde side-chain, and enantioselective
desymmetrization of a 3-substituted glutaric anhydride
to set the sole chiral center.
tert-Butyl 2-(3-Oxopropyl)-5,6,7,8-tetrahydro-9H-pyrido-
[2,3-b]azepine-9-carboxylate (24). A stirred slurry of pyri-
doazepine 20 (3.32 kg, 11.7 mol), potassium carbonate (3.25
kg, 23.5 mol), palladium(II) acetate (132 g, 0.59 mol), and 1,1′-
bis(diphenylphosphino)ferrocene (326 g, 0.59 mol) in THF (16.5
L) was degassed and then put under an atmosphere of
nitrogen. A solution of borane 22 in THF [prepared by adding
acrolein diethyl acetal (3.51 kg, 27.0 mol) to 9-BBN (57.4 L,
0.41 M in THF, 23.5 mol) at 0 °C and then stirring at room
temperature for 5 h] was then added, and the resulting
reaction mixture was heated to reflux for 26 h. After cooling
to 20 °C, water (66 L) was added and the mixture was stirred
for 30 min. The two layers were allowed to settle, and the lower
aqueous phase was discarded. iPrOAc (10 L) was then added,
and, after the mixture was stirred for 5 min and allowed to
settle, the aqueous phase was again discarded. The resulting
organic layer was concentrated, under reduced pressure, to
Experimental Section
tert-Butyl N-(6-Chloropyridin-2-yl)carbamate (19). Hex-
yllithium (34.6 L, 2.5 M in hexanes, 86.5 mol) was slowly
added, over 2 h, to a stirred solution of 1,1,1,3,3,3-hexameth-
yldisilazane (13.9 kg, 86.1 mol) in THF (50 L) at -5 °C. A
solution of 2-amino-6-chloropyridine (16) (5.00 kg, 38.9 mol)
in THF (10 L), followed by a solution of di-tert-butyl dicarbon-
ate (8.63 kg, 39.5 mol) in THF (6.7 L), were then added,
ensuring the internal temperature remained below 0 °C. The
resulting reaction mixture was aged for 30 min at room
temperature and then carefully acidified to pH 3 by addition
of 1 M hydrochloric acid (168 L, 168 mol). The two layers were
separated, and the aqueous phase was extracted with iPrOAc
(50 L). The combined organic layers were then washed
sequentially with 4% aqueous NaHCO3 (50 L) and water (50
L). The resulting organic layer was concentrated to 20 L under
reduced pressure and then solvent switched to 2-propanol (20
L). The solvent switch was carried out by adding 65 L of
2-propanol to the batch, concentrating in vacuo to 24 L, adding
a second portion of 2-propanol (50 L), and then concentrating
to a final volume of 20 L. Product had begun to crystallize from
solution by this point, and water (10 L) was then slowly added
to the stirred slurry. Once the addition was complete, the
slurry was cooled to 5 °C and the solid was isolated by
filtration, washing the wet-cake first with cold 2-propanol/
water (9.8 L/7.3 L) and then water (10 L). Drying under
vacuum at 40 °C furnished N-Boc aminopyridine 19 (8.09 kg,
i
minimum volume and then diluted with PrOAc (33 L). Once
again, the lower aqueous phase was discarded and the
remaining organic layer was concentrated to 10 L and then
i
diluted with PrOAc (23 L). The solution obtained was cooled
to 0 °C, and pre-cooled (0 °C) 2 M hydrochloric acid (23.3 L,
46.6 mol) was then added. This biphasic mixture was stirred
at 0 °C for 4 h, and the two layers were then separated. The
i
aqueous phase was filtered, cooled to 5 °C, and PrOAc (16.5
L) was added. The resulting mixture was adjusted to pH 8 by
addition of 10% aqueous K2CO3 (55 L), and once again the two
layers were separated. The aqueous phase was extracted twice
i
with PrOAc (2 × 16.5 L), and the combined organic layers
i
were then washed with water (8.3 L). The resulting PrOAc
solution was solvent switched, under reduced pressure, to
2-propanol (15 L), and this crude solution of aldehyde 24 (3.08
kg, based on HPLC assay) was used in the Wittig reaction
directly. HPLC conditions: 250 × 4.6 mm Zorbax RX-C8
column; UV detection at 210 nm; gradient elution with 0.1%
phosphoric acid/acetonitrile from 90:10 to 10:90 over 20 min
then 10:90 isocratic elution for 5 min; 1.0 mL/min; 35 °C.
Retention time for aldehyde 24 ) 9.1 min.
1
91%) as a white solid: mp 87.5-89 °C (lit.17 88-89.5 °C); H
NMR (400 MHz, CDCl3) δ 7.86 (d, J ) 8.2 Hz, 1H), 7.61 (dd,
J ) 8.2, 7.7 Hz, 1H), 7.25 (br s, 1H), 6.98 (d, J ) 7.7 Hz, 1H),
1.52 (s, 9H); 13C NMR (101 MHz, CDCl3) δ 152.1, 152.0, 149.2,
140.8, 118.5, 110.4, 81.7, 28.4. Anal. Calcd for C10H13ClN2O2:
C, 52.52; H, 5.73; N, 12.25. Found: C, 52.44; H, 5.70; N, 12.21.
tert-Butyl 2-Chloro-5,6,7,8-tetrahydro-9H-pyrido[2,3-
b]azepine-9-carboxylate (20). Hexyllithium (20.0 L, 2.5 M
solution in hexanes, 50.0 mol) was added, over a period of 35
min, to a stirred solution of N,N,N′,N′-tetramethylethylene-
diamine (5.85 kg, 50.3 mol) in THF (54.5 L) at -20 °C. The
resulting yellow solution was aged for 30 min at this temper-
ature, cooled to -78 °C, and a solution of N-Boc aminopyridine
19 (5.23 kg, 22.9 mol) in THF (24 L) was then added over 45
min, maintaining the temperature below -65 °C. The resulting
Characterization data for an isolated sample of 24: 1H NMR
(250 MHz, CD2Cl2) δ 9.82 (t, J ) 1.3 Hz, 1H), 7.46 (d, J ) 7.6
Hz, 1H), 7.00 (d, J ) 7.6 Hz, 1H), 3.9-3.1 (br, 2H), 3.08-3.00
(m, 2H), 2.92-2.83 (m, 2H), 2.67 (dd, J ) 6.3, 5.0 Hz, 2H),
1.86-1.74 (m, 2H), 1.74-1.54 (m, 2H), 1.38 (s, 9H); 13C NMR
(63 MHz, CD2Cl2) δ 202.1, 157.4, 155.4, 154.1, 139.3, 132.4,
121.7, 80.0, 47.0, 43.1, 33.5, 30.1, 29.9, 28.4, 26.4.
2-Methoxypyrimidine-5-carbaldehyde (28). To a stirred
slurry of O-methylisourea sulfate (30.4 g, 247 mmol) and
i
vinamidinium salt 27 (90 g, 70 wt % pure, 248 mmol) in -
J. Org. Chem, Vol. 70, No. 5, 2005 1777

Keen et al.
PrOAc (475 mL) was added a solution of KHCO3 (35.0 g, 350
mmol) in water (140 mL) over a 10 min period. Once the
addition was complete, the resulting reaction mixture was
stirred at room temperature for 40 h and then diluted with
water (195 mL). The biphasic mixture was allowed to settle,
the two layers were separated, and the aqueous phase was
extracted twice with PrOAc (2 × 500 mL). The combined -
PrOAc layers were concentrated under reduced pressure
(keeping temperature below 40 °C) to a volume of 500 mL and
then solvent switched to heptane (500 mL). The resulting
slurry was cooled to -8 °C, aged for 1 h, and the solid was
then isolated by filtration, washing the wet-cake with pre-
cooled heptane (360 mL). Drying under vacuum at 30 °C
afforded aldehyde 28 (31.8 g, 93%) as an off-white solid: mp
94-96 °C (lit.33a 89-90 °C); 1H NMR (250 MHz, CD2Cl2) δ 9.99
(s, 1H), 8.96 (s, 2H), 4.09 (s, 3H); 13C NMR (101 MHz, CD2Cl2)
δ 188.3, 168.2, 161.9, 124.8, 56.2. Anal. Calcd for C6H6N2O2:
C, 52.17; H, 4.38; N, 20.28. Found: C, 51.94; H, 4.32; N, 20.10.
allowing it to gradually warm to 20 °C overnight. The resulting
clear yellow solution was extracted twice with water (2 × 60
L), and the combined aqueous extracts (pH ) 5.8) were
returned to the empty vessel, rinsing it with water (12 L). To
this aqueous solution, concentrated hydrochloric acid (3.39 L,
39.5 mol) was added (pH after addition ) 3.8). After the
mixture aged for 5 min, (S)-acid-ester seed (45 g, 0.18 mol)
was added and stirring was continued for a further 10 min. A
second portion of concentrated hydrochloric acid (3.39 L, 39.5
mol) was then added, and acid-ester began to crystallize from
solution (pH of solution ) 1.9). The resulting stirred slurry,
which had warmed to 30 °C during acidification, was slowly
cooled to 20 °C over 1.5 h and then aged at this temperature
for 2.5 h. The solid was collected by filtration, washing the
wet-cake with water (24 L). Drying under vacuum at 35 °C
afforded the title compound (6.54 kg, 63%) as a white solid.
The enantiomeric excess of this product was determined as
97.8% by chiral stationary phase HPLC (250 × 4.6 mm
Chirobiotic T column; UV detection at 274 nm; isocratic elution
with 80:20 aqueous Et3N (0.036 M)-AcOH (0.044 M)/methanol
for 20 min; 0.5 mL/min; 25 °C. Retention times: (R)-acid ester
i
i
3-(2-Methoxypyrimidin-5-yl)pentanedioic Acid (26). To
a stirred slurry of aldehyde 28 (9.00 kg, 65.2 mol) in 2-propanol
(80 L) was added ethyl acetoacetate (17.8 kg, 136.8 mol). A
solution of piperidine (0.56 kg, 6.6 mol) in 2-propanol (10 L)
was then added, and the resulting solution was warmed to 50
°C. After being stirred at this temperature for 3.5 h, the
reaction mixture was cooled to 0 °C and aqueous NaOH (24.2
kg of 46% w/w aqueous NaOH in 30 L of water, 278.3 mol)
was slowly added over a period of 1 h. Once the addition was
complete, the reaction mixture was warmed to 23 °C and
stirred for 3 h. The resulting biphasic mixture was allowed to
settle, the two layers were separated, and the lower aqueous
phase was then diluted with saturated aqueous sodium
chloride (10.6 kg NaCl in 30 L of water). This aqueous solution
was cooled to 0 °C and then acidified, to a pH of 2-3, by careful
addition of concentrated hydrochloric acid (19.6 L, 228.3 mol).
The resulting slurry was stirred overnight at ambient tem-
perature and then cooled to 5 °C. After aging at this temper-
ature for 1 h, the solid was collected by filtration, washing the
wet-cake with water (23 L). Drying under vacuum at 40 °C
afforded diacid 26 (12.7 kg, 81%) as a white solid: mp 177-
ent-29 ) 10.5 min; (S)-acid ester 29 ) 11.4 min): mp 143.5-
1
145.5 °C; [R]20 +12.9 (c 2.0 in MeOH); H NMR (250 MHz,
D
CD3OD) δ 8.51 (s, 2H), 3.98 (s, 3H), 3.59 (s, 3H), 3.55 (tt, J )
9.0, 6.2 Hz, 1H), 2.86 (dd, J ) 16.2, 6.1 Hz, 1H), 2.81 (dd, J )
16.3, 6.2 Hz, 1H), 2.73 (dd, J ) 16.2, 9.0 Hz, 1H), 2.67 (dd, J
) 16.3, 9.0 Hz, 1H); 13C NMR (63 MHz, CD3OD) δ 174.8, 173.5,
165.9, 160.2, 131.3, 55.6, 52.3, 40.4, 40.4, 34.6. Anal. Calcd
for C11H14N2O5: C, 51.97; H, 5.55; N, 11.02. Found: C, 51.99;
H, 5.49; N, 10.81.
Methyl (3S)-3-(2-Methoxypyrimidin-5-yl)-5-oxo-6-(tri-
phenylphosphoranylidene)hexanoate (34). A stirred sus-
pension of methyltriphenylphosphonium bromide (18.2 kg, 51.0
mol) in THF (82 L) was cooled to -60 °C. Hexyllithium (20.6
L, 2.4 M in hexanes, 49.4 mol) was then slowly added over a
period of 30 min, keeping the internal temperature below -10
°C. Once the addition was complete, the batch was aged at 0
°C for 90 min and then cooled to -80 °C.
In a second vessel, triethylamine (2.37 L, 17.0 mol) was
added, over a period of 30 min, to a stirred slurry of acid-ester
29 (4.38 kg, 16.7 mol) and trimethylacetyl chloride (2.06 kg,
17.1 mol) in THF (34 L) at -5 °C. The resulting reaction
mixture was aged at -5 to 0 °C for 30 min and then added,
over a period of 40 min, to the cooled (-80 °C) ylide mixture
prepared above. The resulting batch was aged for 40 min at
-70 °C and then quenched into aqueous potassium dihydro-
genphosphate (1.20 kg of KH2PO4 in 64 L of water, 8.8 mol),
keeping the temperature of the quenched mixture between 0
and 10 °C. Once the quench was complete, the batch was
1
177.5 °C; H NMR (250 MHz, CD3OD) δ 8.52 (s, 2H), 3.98 (s,
3H), 3.54 (tt, J ) 9.1, 6.1 Hz, 1H), 2.82 (dd, J ) 16.3, 6.1 Hz,
2H), 2.68 (dd, J ) 16.2, 9.1 Hz, 2H); 13C NMR (63 MHz, CD3-
OD) δ 174.9, 165.9, 160.2, 131.4, 55.6, 40.5, 34.6. Anal. Calcd
for C10H12N2O5: C, 50.00; H, 5.04; N, 11.66. Found: C, 49.94;
H, 4.95; N, 11.64.
4-(2-Methoxypyrimidin-5-yl)dihydropyran-2,6-dione
(25). To a stirred slurry of diacid 26 (11.5 kg, 47.9 mol) in THF
(58 L) was slowly added trifluoroacetic anhydride (12.1 kg, 57.6
mol) over a period of 40 min. The resulting reaction mixture
was heated to 55 °C, stirred at this temperature for 80 min,
and then cooled to 50 °C. Heptane (195 L) was then slowly
added over a period of 90 min, during which the temperature
of the batch dropped to 30 °C. The resulting stirred slurry was
allowed to cool to 23 °C overnight, and the solid was then
collected by filtration, washing the wet-cake with 2:1 heptane/
THF (51 L). Drying under vacuum at 35 °C afforded anhydride
25 (9.95 kg, 94%) as a beige solid: mp 150.5-152.5 °C; 1H
NMR (250 MHz, CD2Cl2) δ 8.42 (s, 2H), 3.99 (s, 3H), 3.44 (tt,
J ) 11.9, 4.4 Hz, 1H), 3.13 (dd, J ) 17.3, 4.4 Hz, 2H), 2.87
(dd, J ) 17.3, 11.9 Hz, 2H); 13C NMR (63 MHz, CD2Cl2) δ 166.0,
165.6, 158.2, 126.3, 55.6, 37.0, 30.0. Anal. Calcd for C10H10-
N2O4: C, 54.05; H, 4.54; N, 12.61. Found: C, 53.85; H, 4.55;
N, 12.43.
(3S)-5-Methoxy-3-(2-methoxypyrimidin-5-yl)-5-oxo-
pentanoic Acid (29). Toluene (180 L) was charged to a vessel
containing anhydride 25 (9.0 kg, 40.5 mol) and quinidine (13.15
kg, 40.5 mol) under a nitrogen atmosphere. The resulting
slurry was cooled, with stirring, to -40 °C. Methanol (16.4 L,
406 mol), which had been pre-cooled to approximately 5 °C,
was then added over 15 min. The internal temperature of the
reaction mixture had risen to -35 °C by this point, and the
slurry was then stirred at this temperature for 8 h before
i
extracted twice with PrOAc (2 × 85 L) and the combined
extracts were washed with half-saturated aqueous NaCl (2 ×
38 L). The resulting organic layer was concentrated under
reduced pressure to a volume of 25 L, diluted with iPrOAc (34
L), and then concentrated again to 25 L. Crystallization of the
product had occurred during this distillation, and, after cooling
to 0 °C, the solid was collected by filtration, washing the wet-
cake with tert-butyl methyl ether (10.5 L). Drying under
vacuum at 30 °C afforded phosphorane 34 (6.28 kg, 71%) as a
cream-colored solid: mp 149.0-151.5 °C; [R]20 -85.9 (c 2.0
D
1
in MeOH); H NMR (250 MHz, CD2Cl2) δ 8.42 (s, 2H), 7.61-
7.38 (m, 15H), 3.97 (s, 3H), 3.72-3.55 (m, 2H), 3.56 (s, 3H),
2.79 (dd, J ) 15.7, 5.9 Hz, 1H), 2.65-2.54 (m, 3H); 13C NMR
(63 MHz, CD2Cl2) δ 189.3 (d, J ) 2 Hz), 172.4, 165.0, 159.1,
133.3 (d, J ) 10 Hz), 132.5 (d, J ) 3 Hz), 130.6, 129.1 (d, J )
12 Hz), 127.3 (d, J ) 91 Hz), 55.0, 53.2 (d, J ) 107 Hz), 51.8,
46.8 (d, J ) 16 Hz), 40.7, 34.8. Anal. Calcd for C30H29N2O4P:
C, 70.30; H, 5.70; N, 5.47. Found: C, 70.06; H, 5.71; N, 5.23.
tert-Butyl 2-[(7S)-9-Methoxy-7-(2-methoxypyrimidin-
5-yl)-5,9-dioxononyl]-5,6,7,8-tetrahydro-9H-pyrido[2,3-b]-
azepine-9-carboxylate (36). To the previously prepared
solution of aldehyde 24 (3.02 kg, 9.92 mol, based on HPLC
assay) in 2-propanol (∼15 L) was added phosphorane 34 (4.84
1778 J. Org. Chem., Vol. 70, No. 5, 2005

Practical Synthesis of an Rvâ3 Antagonist
kg, 9.44 mol). The slurry obtained was degassed, put under
an atmosphere of nitrogen, and then heated to reflux. After
being stirred at reflux for 12 h, the resulting solution of crude
enone 35 was allowed to cool to room temperature and then
used in the following hydrogenation step directly.
[R]20 -2.7 (c 2.0 in MeOH); 1H NMR (250 MHz, CD2Cl2) δ
D
8.42 (s, 2H), 8.2-7.4 (br, 1H), 7.49 (d, J ) 7.6 Hz, 1H), 6.97
(d, J ) 7.6 Hz, 1H), 4.3-2.5 (br, 2H), 3.94 (s, 3H), 3.66 (app
quintet, J ) 7.2 Hz, 1H), 2.96 (dd, J ) 17.3, 6.8 Hz, 1H), 2.78-
2.56 (m, 7H), 2.55-2.33 (m, 2H), 1.85-1.47 (m, 8H), 1.35 (s,
9H); 13C NMR (63 MHz, CD2Cl2) δ 208.5, 173.8, 164.9, 159.2,
159.0, 154.9, 154.0, 140.1, 132.7, 130.7, 122.1, 80.4, 55.3, 48.0,
47.2, 43.1, 40.5, 36.9, 33.5, 32.2, 29.9, 29.8, 28.4, 26.4, 23.5.
Anal. Calcd for C28H38N4O6: C, 63.86; H, 7.27; N, 10.64.
Found: C, 63.75; H, 7.30; N, 10.46.
To the crude solution of enone 35 in 2-propanol, prepared
above, was added a slurry of wet (58% H2O) 10% palladium
on carbon (1.27 kg) in 2-propanol (25 L). The resulting reaction
mixture was degassed, and the vessel was filled with hydrogen
to a pressure of 2.8 bar gauge.51 The reaction mixture was then
stirred under this pressure of hydrogen for 2 h. Upon comple-
tion of the reaction, the vessel was degassed and put under
an atmosphere of nitrogen. The reaction mixture was then
filtered, washing the catalyst with 2-propanol (4 × 15 L), and
the resulting solution was concentrated under reduced pres-
sure to ∼20 L. This crude solution of keto-ester 36 (4.50 kg,
based on HPLC assay) was used directly in the subsequent
hydrolysis step. HPLC conditions: 250 × 4.6 mm Phenomenex
Luna C8(2) 5 µm column; UV detection at 210 nm; gradient
elution with aqueous CH3(CH2)5SO3Na (0.01 M)-KH2PO4
(0.025 M)/acetonitrile from 60:40 to 35:65 over 30 min; 1.0 mL/
min; 25 °C. Retention time for keto-ester 36 ) 19.0 min.
An analytical sample of 36 was obtained, as a colorless oil,
by flash column chromatography (EtOAc): 1H NMR (250 MHz,
CD2Cl2) δ 8.37 (s, 2H), 7.44 (d, J ) 7.6 Hz, 1H), 6.92 (d, J )
7.6 Hz, 1H), 3.93 (s, 3H), 3.7-3.0 (br, 2H), 3.66-3.52 (m, 1H),
3.57 (s, 3H), 2.91-2.62 (m, 7H), 2.56 (dd, J ) 15.9, 8.7 Hz,
1H), 2.48-2.26 (m, 2H), 1.86-1.48 (m, 8H), 1.37 (s, 9H); 13C
NMR (101 MHz, CD2Cl2) δ 208.3, 172.0, 165.4, 159.5, 159.0,
155.6, 154.3, 139.3, 132.2, 130.2, 121.6, 80.1, 55.2, 52.1, 48.1,
47.3, 43.5, 40.2, 37.8, 33.7, 32.2, 30.1, 29.7, 28.6, 26.7, 23.7.
(3S)-9-[9-(tert-Butoxycarbonyl)-6,7,8,9-tetrahydro-5H-
pyrido[2,3-b]azepin-2-yl]-3-(2-methoxypyrimidin-5-yl)-5-
oxononanoic Acid (37). The crude solution of keto-ester 36
(4.50 kg, 8.32 mol, based on HPLC assay) in 2-propanol (∼20
L), prepared above, was cooled to 0 °C. 2 M aqueous sodium
hydroxide (5.6 L, 11.2 mol) was added to the stirred solution
over a period of 30 min, and the resulting reaction mixture
was then stirred at 0 °C for 2 h. The thin slurry obtained was
diluted with water (43 L), warmed to 20 °C, and then washed
once with tert-butyl methyl ether (43 L) and twice with iPrOAc
(2 × 43 L). To the resulting aqueous layer was added 2 M
hydrochloric acid (0.56 L, 1.12 mol) followed by iPrOAc (43 L).
This biphasic mixture was then acidified by careful addition
of a second portion of 2 M hydrochloric acid (5.04 L, 10.1 mol).
Once addition was complete, the two layers were separated
and the aqueous phase (pH ) 3.8) was extracted with iPrOAc
(43 L). The two organic layers were combined, washed with
water (21 L), and then treated with Ecosorb C-941 activated
carbon (0.43 kg). After being stirred for 1 h, at room temper-
ature, the mixture was filtered, washing the carbon with
iPrOAc (2 × 12 L). The filtrate was concentrated, under
reduced pressure, to ∼20 L and then diluted with iPrOAc (40
L). Solid began to crystallize from solution at this point, and
the slurry was concentrated to 20 L. Heptane (10 L) was then
added over a period of 30 min at room temperature, and the
slurry was left to stir over the weekend (∼65 h). The resulting
slurry was cooled to 0 °C and aged for 5 h at this temperature.
The solid was then collected by filtration, washing the wet-
(3S)-3-(2-Methoxypyrimidin-5-yl)-5-oxo-9-(6,7,8,9-
tetrahydro-5H-pyrido[2,3-b]azepin-2-yl)nonanoic Acid
(1). Trifluoroacetic acid (7.36 L, 95.6 mol) was added over a
period of 10 min to a solution of keto-acid 37 (3.35 kg, 6.37
mol) in DCM (17 L), maintaining the internal temperature
below 30 °C. Once the addition was complete, the resulting
reaction mixture was heated to 30 °C. After stirring at this
temperature for 2 h, the solution was cooled to 20 °C, and pre-
cooled (5 °C) 2 M aqueous NaOH (56.5 L, 113 mol) was
carefully added over a period of 20 min. The two layers were
then separated, and the aqueous phase was treated with
Ecosorb C-941 activated carbon (168 g). After being stirred for
1 h at room temperature, the mixture was filtered, washing
the carbon with water (2.4 L). The resulting filtrate was then
acidified to pH 6.0 by careful addition of concentrated hydro-
chloric acid (1.47 L, 17.1 mol), keeping the internal temper-
ature below 20 °C. Once the addition was complete, the
mixture was extracted twice with DCM (2 × 33.5 L). The
combined organic layers were washed with water (11.3 L) and
then solvent switched, using partial vacuum at 40 °C, to
2-propanol (15 L). With the resulting 2-propanol solution still
at 40 °C, a small amount of 1 (14.5 g) was added as seed. The
batch was then left to stir at this temperature for 12 h. By
this point, solid had begun to crystallize from solution and the
slurry was allowed to gradually cool to 20 °C overnight. The
solid was then collected by filtration, washing the wet-cake
with 2-propanol (3 × 3.3 L). Drying under vacuum at 40 °C
afforded 1 (2.07 kg, 76%) as a white solid. The enantiomeric
excess of this product was determined as 99.4% by chiral
stationary phase HPLC (150 × 4.0 mm Chromtech Chiral AGP
column; UV detection at 245 nm; gradient elution with 1%
triethylamine phosphate(aq)/10% IPA in 1% triethylamine
phosphate(aq) from 100:0 to 90:10 over 5 min and then to
50:50 over the next 10 min; 0.9 mL/min; 25 °C. Retention
times: (3S) enantiomer ) 6.5 min; (3R) enantiomer ) 10.0
1
min): mp 105.5-108.5 °C; [R]20 -17.1 (c 2.0 in MeOH); H
D
NMR (250 MHz, CD2Cl2) δ 8.41 (s, 2H), 7.25 (d, J ) 7.4 Hz,
1H), 6.34 (d, J ) 7.3 Hz, 1H), 3.91 (s, 3H), 3.72-3.59 (m, 1H),
3.35-3.27 (m, 2H), 2.97 (dd, J ) 16.5, 7.0 Hz, 1H), 2.78-2.41
(m, 9H), 1.91-1.77 (m, 4H), 1.66-1.46 (m, 4H); 13C NMR (63
MHz, CD2Cl2) δ 208.4, 177.9, 164.9, 158.5, 158.3, 151.9, 143.0,
132.6, 124.5, 111.5, 55.0, 49.5, 43.5, 42.9, 40.5, 33.1, 32.0, 31.4,
29.0, 28.6, 25.5, 22.3. Anal. Calcd for C23H30N4O4: C, 64.77;
H, 7.09; N, 13.14. Found: C, 64.50; H, 7.07; N, 13.04.
Acknowledgment. We thank Derek Kennedy for
NMR support, Andy Kirtley for determining specific
rotations, and Bill Leonard and Tony Houck for sup-
plying kilogram quantities of 2-amino-6-chloropyridine.
We also thank Andy Gibson, John Edwards, Ray
Alabaster, and Simon Johnson for their assistance in
running this chemistry on multi-kilogram scale.
i
cake with 2:1 PrOAc/heptane (4.5 L). Drying under vacuum
at 45 °C afforded keto-acid 37 (3.45 kg, 70% overall yield from
phosphorane 34) as a cream-colored solid. The enantiomeric
excess of this product was determined as 96.6% by chiral
stationary phase HPLC (150 × 4.0 mm Chromtech Chiral AGP
column; UV detection at 210 nm; gradient elution with
aqueous KH2PO4 (0.025 M)/methanol from 75:25 to 50:50 over
15 min; 0.9 mL/min; 25 °C. Retention times: (3R) enantiomer
) 8.6 min; (3S) enantiomer ) 10.0 min): mp 113.5-117.5 °C;
Supporting Information Available: Experimental pro-
cedures and characterization data for compounds 5, 7, 9, 10,
15, 17, 21, ent-29, 30, 31, and 35. Alternative procedures for
the preparation of 34. 1H NMR spectra of compounds 5 and 9.
1H and 13C NMR spectra of compounds 7, 10, 24, 30, 31, 35,
and 36. This material is available free of charge via the
Internet at http://pubs.acs.org.
(51) Gauge denotes that the pressure is measured with respect to
ambient pressure.
JO048082N
J. Org. Chem, Vol. 70, No. 5, 2005 1779