Published on Web 09/11/2003
`
`Asymmetric, Stereocontrolled Total Synthesis of
`Paraherquamide A
`
`Robert M. Williams,* Jianhua Cao, Hidekazu Tsujishima, and Rhona J. Cox
`Contribution from the Department of Chemistry, Colorado State UniVersity,
`Fort Collins, Colorado 80523
`
`Received June 16, 2003; E-mail: rmw@chem.colostate.edu
`
`Abstract:The first total synthesis of paraherquamide A, a potent anthelmintic agent isolated from various
`Penicilliumsp. with promising activity against drug-resistant intestinal parasites, is reported. Key steps in
`this asymmetric, stereocontrolled total synthesis include a new enantioselective synthesis of R-alkylated-
`(cid:226)-hydroxyproline derivatives to access the substituted proline nucleus and a highly diastereoselective
`intramolecular SN2¢ cyclization to generate the core bicyclo[2.2.2]diazaoctane ring system.
`
`Introduction
`The paraherquamides1-4 are an unusual family of fungal
`natural products which contain a bicyclo[2.2.2]diazaoctane core
`structure, a spiro-oxindole, and a substituted proline moiety.
`The parent member, paraherquamide A (1), was first isolated
`from cultures of Penicillium paraherquei by Yamazaki and co-
`workers in 1981.1 Since then, paraherquamides B-G,2 VM55595,
`VM55596, and VM55597,3 SB203105 and SB200437,4 and
`sclerotamide5 have been isolated from various Penicillium and
`Aspergillus species. Marcfortines A-C are structurally similar,
`containing a pipecolic acid unit in place of proline.6 Also closely
`related are VM55599,3 aspergamides A and B,7 avrainvillamide
`(CJ-17,665),8 and the most recently isolated members of this
`family, stephacidins A and B .9 These last six compounds
`contain a 2,3-disubstituted indole in place of the spiro-oxindole.
`Brevianamides A and B,10 which contain a spiro-indoxyl rather
`
`(1) Yamazaki, M.; Okuyama E.; Kobayashi, M.; Inoue, H. Tetrahedron Lett.
`1981, 22, 135-136.
`(2) (a) Ondeyka, J. G.; Goegelman, R. T.; Schaeffer, J. M.; Kelemen, L.; Zitano,
`L. J. Antibiot. 1990, 43, 1375-1379. (b) Liesch, J. M.; Wichmann, C. F.
`J. Antibiot. 1990, 43, 1380-1386. (c) Blanchflower, S. E.; Banks, R. M.;
`Everett, J. R.; Manger, B. R.; Reading, C. J. Antibiot. 1991, 44, 492-497.
`(3) Blanchflower, S. E.; Banks, R. M.; Everett, J. R.; Reading, C. J. Antibiot.
`1993, 46, 1355-1363.
`(4) Banks, R. M.; Blanchflower, S. E.; Everett, J. R.; Manger, B. R.; Reading,
`C. J. Antibiot. 1997, 50, 840-846.
`(5) Whyte, A. C.; Gloer, J. B.; Wicklow, D. T.; Dowd, P. F. J. Nat. Prod.
`1996, 59, 1093-1095.
`(6) (a) Polonsky, J.; Merrien, M.-A.; Prange´, T.; Pascard, C.; Moreau, S. J.
`Chem. Soc., Chem. Commun. 1980, 601-602. (b) Prange´, T.; Billion, M.-
`A.; Vuilhorgne, M.; Pascard, C.; Polonsky, J.; Moreau, S. Tetrahedron Lett.
`1981, 22, 1977-1980.
`(7) Fuchser, Jens. Beeinflussung der Sekundarstoffbildung bei Aspergillus
`ochraceus durch Variation der Kulturbedingungen sowie Isolierung,
`Strukturaufklarung und Biosynthese der neuen Naturstoffe. Ph.D. Thesis,
`University of Go¨ttingen, Germany, 1995. K. Bielefeld Verlag: Friedland,
`1996 (Prof. A. Zeeck).
`(8) (a) Fenical, W.; Jensen, P. R.; Cheng, X. C. U.S. Patent 6,066,635, 2000.
`(b) Sugie, Y.; Hirai, H.; Inagaki, T.; Ishiguro, M.; Kim, Y.-J.; Kojima, Y.;
`Sakakibara, T.; Sakemi, S.; Sugiura, A.; Suzuki, Y.; Brennan, L.; Duignan,
`J.; Huang, L. H.; Sutcliffe, J.; Kojima, N. J. Antibiot. 2001, 54, 911-916.
`(9) Qian-Cutrone, J.; Huang, S.; Shu, Y.-Z.; Vyas, D.; Fairchild, C.; Menendez,
`A.; Krampitz, K.; Dalterio, R.; Klohr, S. E.; Gao, Q. J. Am. Chem. Soc.
`2002, 124, 14556-14557.
`(10) (a) Birch, A. J.; Wright, J. J. J. Chem. Soc., Chem. Commun. 1969, 644-
`645. (b) Birch, A. J.; Wright, J. J. Tetrahedron 1970, 26, 2329-2344. (c)
`Birch, A. J.; Russell, R. A. Tetrahedron 1972, 28, 2999-3008.
`12172 9 J. AM. CHEM. SOC. 2003, 125, 12172-12178
`
`than a spiro-oxindole, and the asperparalines, which contain a
`spiro-succinimide,11 are also structurally comparable (Figure 1).
`The paraherquamides have attracted considerable attention
`due to their molecular complexity, intriguing biogenesis,12,13 and
`biological activity. Some members, most notably paraherqua-
`mide A, display potent anthelmintic activity and antinematodal
`properties.14 Due to the appearance of drug resistance developed
`by helminths, broad spectrum anthelmintic agents such as the
`macrolide endectocides, benzimidazoles, tetrahydropyrimidines,
`and imidazothiazoles are beginning to lose efficacy and there
`has arisen an urgent need to discover new families of antipara-
`sitic agents. The paraherquamides represent an entirely new
`structural class of anthelmintic compounds, and as such, they
`hold great potential as drugs for the treatment of intestinal para-
`sites in animals.15 The mode of action of the paraherquamides
`is, as yet, incompletely characterized, but recent work suggests
`that they are selective competitive cholinergic antagonists.16
`
`(11) (a) Hayashi, H.; Nishimoto, Y.; Nozaki, H. Tetrahedron Lett. 1997, 38,
`5655-5658. (b) Hayashi, H.; Nishimoto, Y.; Akiyama, K.; Nozaki, H.
`Biosci. Biotechnol. Biochem. 2000, 64, 111-115.
`(12) (a) Porter, A. E. A.; Sammes, P. G. J. Chem. Soc., Chem. Commun. 1970,
`1103. (b) Baldas, J.; Birch, A. J.; Russell, R. A. J. Chem. Soc., Perkin
`Trans. 1 1974, 50-52. (c) Birch, A. J. J. Agric. Food Chem. 1971, 19,
`1088-1092. (d) Kuo, M. S.; Wiley, V. H.; Cialdella, J. I.; Yurek, D. A.;
`Whaley, H. A.; Marshall, V. P. J. Antibiot. 1996, 49, 1006-1013.
`(13) (a) Stocking, E. M.; Sanz-Cervera, J. F.; Williams, R. M.; Unkefer, C. J.
`J. Am. Chem. Soc. 1996, 118, 7008-7009. (b) Williams, R. M.; Sanz-
`Cervera, J. F.; Sanceno´n, F.; Marco, J. A.; Halligan, K. M. Bioorg. Med.
`Chem. 1998, 6, 1233-1241. (c) Stocking, E. M.; Williams, R. M.; Sanz-
`Cervera, J. F. J. Am. Chem. Soc. 2000, 122, 9089-9098. (d) Stocking, E.
`M.; Martinez, R. A.; Silks, L. A.; Sanz-Cervera, J. F.; Williams, R. M. J.
`Am. Chem. Soc. 2001, 123, 3391-3392. (e) Stocking, E. M.; Sanz-Cervera,
`J. F.; Unkefer, C. J.; Williams, R. M. Tetrahedron 2001, 57, 5303-5320.
`(f) Stocking, E. M.; Sanz-Cervera, J. F.; Williams, R. M. Angew. Chem.,
`Int. Ed. 2001, 40, 1296-1298.
`(14) (a) Ostlind, D. A.; Mickle, W. G.; Ewanciw, D. V.; Andriuli, F. J.;
`Campbell, W. C.; Hernandez, S.; Mochales, S.; Munguira, E. Res. Vet.
`Sci. 1990, 48, 260-261. (b) Shoop, W. L.; Egerton, J. R.; Eary, C. H.;
`Suhayda, D. J. Parasitol. 1990, 76, 349-351. (c) Shoop, W. L.; Eary, C.
`H.; Michael, H. W.; Haines, H. W.; Seward, R. L. Vet. Parasitol. 1991,
`40, 339-341. (d) Shoop, W. L.; Michael, B. F.; Haines, H. W.; Eary, C.
`H. Vet. Parasitol. 1992, 43, 259-263. (e) Shoop, W. L.; Haines, H. W.;
`Eary, C. H.; Michael, B. F. Am. J. Vet. Res. 1992, 53, 2032-2034. (f)
`Schaeffer, J. M.; Blizzard, T. A.; Ondeyka, J.; Goegelman, R.; Sinclair, P.
`J.; Mrozik, H. Biochem. Pharmacol. 1992, 43, 679-684. For a review,
`see: (g) Geary, T. G.; Sangster, N. C.; Thompson, D. P. Vet. Parasitol.
`1999, 84, 275-295.
`
`10.1021/ja036713+ CCC: $25.00 © 2003 American Chemical Society
`
` P. 1
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`Total Synthesis of Paraherquamide A
`
`A R T I C L E S
`
`Scheme1. Retrosynthetic Plan for Paraherquamide A
`
`Figure1. Structures of some paraherquamides and related compounds.
`
`The small quantities of paraherquamide A that can be isolated
`from cultures for biological study have slowed the development
`of these agents. Recently, Lee and Clothier reported the
`interesting semisynthetic conversion of marcfortine A (3), a
`metabolite more readily available by fermentation, into para-
`herquamide A via paraherquamide B (2).17 Following synthetic
`studies on brevianamide B (12),18 our laboratory reported the
`first total synthesis of a member of the paraherquamide family,
`ent-paraherquamide B, in 1993, in which a diastereoselective
`intramolecular SN2¢ cyclization reaction was used to construct
`the core bicyclo[2.2.2]diazaoctane ring system.19 We have
`further exploited this reaction strategy, and we described the
`first total synthesis of paraherquamide A in 2000.20 Herein, we
`detail a full account of this work.
`
`(15) (a) Blizzard, T. A.; Marino, G.; Mrozik, H.; Fisher, M. H.; Hoogsteen, K.;
`Springer, J. P. J. Org. Chem. 1989, 54, 2657-2663. (b) Blizzard, T. A.;
`Mrozik, H.; Fisher, M. H.; Schaeffer, J. M. J. Org. Chem. 1990, 55, 2256-
`2259. (c) Blizzard, T. A.; Margiatto, G.; Mrozik, H.; Schaeffer, J. M.; Fisher,
`M. H. Tetrahedron Lett. 1991, 32, 2437-2440. (d) Blizzard, T. A.;
`Margiatto, G.; Mrozik, H.; Schaeffer, J. M.; Fisher, M. H. Tetrahedron
`Lett. 1991, 32, 2441-2444. (e) Lee, B. H.; Clothier, M. F.; Johnson,
`S. S. Bioorg. Med. Chem. Lett. 2001, 11, 553-554. (f) Lee, B. H.; Clothier,
`M. F.; Dutton, F. E.; Nelson, S. J.; Johnson, S. S.; Thompson, D. P.;
`Geary, T. G.; Whaley, H. D.; Haber, C. L.; Marshall, V. P.; Kornis, G. I.;
`McNally, P. L.; Ciadella, J. I.; Martin, D. G.; Bowman, J. W.; Baker, C.
`A.; Coscarelli, E. M.; Alexander-Bowman, S. J.; Davis, J. P.; Zinser, E.
`W.; Wiley, V.; Lipton, M. F.; Mauragis, M. A. Curr. Top. Med. Chem.
`2002, 2, 779-793. (g) Lee, B. H.; Clothier, M. F. U.S. Patent 5,750,695,
`1998.
`(16) (a) Robertson, A. P.; Clark, C. L.; Burns, T. A.; Thompson, D. P.; Geary,
`T. G.; Trailovic, S. M.; Martin, R. J. J. Pharmacol. Exp. Ther. 2002, 302,
`853-860. (b) Zinser, E. W.; Wolfe, M. L.; Alexander-Bowman, S. J.;
`Thomas, E. M.; Davis, J. P.; Groppi, V. E.; Lee, B. H.; Thompson, D. P.;
`Geary, T. G. J. Vet. Pharmacol. Ther. 2002, 25, 241-250.
`(17) (a) Lee, B. H.; Clothier, M. F. J. Org. Chem. 1997, 62, 1795-1798. (b)
`Lee, B. H.; Clothier, M. F.; Pickering, D. A. J. Org. Chem. 1997, 62, 7836-
`7840.
`(18) Williams, R. M.; Glinka, T.; Kwast, E.; Coffman, H.; Stille, J. K. J. Am.
`Chem. Soc. 1990, 112, 808-821.
`(19) Cushing, T. D.; Sanz-Cervera, J. F.; Williams, R. M. J. Am. Chem. Soc.
`1996, 118, 557-579.
`(20) Williams, R. M.; Cao, J.; Tsujishima. H. Angew. Chem., Int. Ed. 2000, 39,
`2540-2544.
`
`Synthesis of an r-Alkylated-(cid:226)-Hydroxyproline
`
`Despite the apparent similarity in the structures of paraher-
`quamides A and B, synthesis of the former turned out to be a
`significantly more challenging endeavor owing to the presence
`of the unusual (cid:226)-hydroxy-(cid:226)-methyl proline residue. In the
`semisynthesis of paraherquamide A from marcfortine A (3), the
`final step was addition of methylmagnesium bromide to 14-
`oxoparaherquamide B (14).17 We planned to use this same
`methodology to complete our total synthesis and to construct
`14 using a similar strategy to that used for paraherquamide B,
`that is, coupling of suitably functionalized indole (19) and
`diketopiperazine (18) units and then an intramolecular SN2¢
`cyclization followed by palladium-mediated closure of the
`seventh ring, and finally oxidation and rearrangement of the
`2,3-disubstituted indole to the spiro-oxindole of 14-oxopara-
`herquamide B19 (Scheme 1).
`New methodology was now required to prepare a suitably
`functionalized R-alkylated-(cid:226)-hydroxyproline residue. A variety
`of methods were investigated for the asymmetric construction
`of this class of compound, leading to the development of a
`potentially general synthetic method which uses dianion alky-
`lation of the readily available N-t-BOC-(cid:226)-hydroxyproline ethyl
`ester derivative 12 with net retention of stereochemistry.21 This
`methodology has now successfully been applied to a concise
`asymmetric and stereocontrolled total synthesis of paraherqua-
`mide A.
`Epoxide 20, which is commercially available or made by
`epoxidation of
`isoprene with mCPBA, was treated with
`n-Bu4NI and TBSCl to provide iodide 21 as a mixture of geo-
`metrical isomers (E:Z (cid:25) 6:1) in 58% overall yield. Diester 22
`was prepared in two steps from ethyl glycinate and ethyl acry-
`late, and then a Dieckmann cyclization was conducted, using a
`slight modification of the procedure described by Rapoport,22
`
`(21) Williams, R. M.; Cao, J. Tetrahedron Lett. 1996, 37, 5441-5444.
`(22) (a) Blake, J.; Willson, C. D.; Rapoport, H. J. Am. Chem. Soc. 1964, 86,
`5293-5299. For more regioselective methods, see: (b) Yamada, Y.; Ishii,
`T.; Kimura, M.; Hosaka, K. Tetrahedron Lett. 1981, 22, 1353-1354. (c)
`Sibi, M. P.; Christensen, J. W.; Kim, S.-G.; Eggen, M.; Stessman, C.; Oien,
`L. Tetrahedron Lett. 1995, 36, 6209-6212.
`
`J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003 12173
`
` P. 2
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`A R T I C L E S
`
`Williams et al.
`
`Scheme3. Assignment of Absolute Chemistry of 25
`
`Figure2. Assignment of relative stereochemistry of 25.
`Scheme2. Synthesis of R-Alkylated-(cid:226)-Hydroxyproline 25
`
`Scheme4. Preparation of the Diketopiperazine 34
`
`to yield racemic (cid:226)-ketoester 23 (Scheme 2). Baker’s yeast
`reduction afforded the optically active (cid:226)-hydroxyester 24 with
`an enantiomeric ratio of ca. 95:5 as described by Knight et al.23
`Alkylation of the dianion of 24 with substituted allyl iodide 21
`proceeded with retention of stereochemistry and excellent
`diastereoselectivity under the conditions previously developed.21
`The desired R-alkylated product 25 was obtained in 58-70%
`isolated yield with little or no O-monoalkylation or O-,C-
`dialkylation taking place. It was interesting to note during large
`scale synthesis of 25 that the amount of HMPA required in the
`alkylation reaction ranged from 1.4 to 13.6 equiv depending
`on the batch of 24 that was used, despite the batches being
`apparently identical by 1H NMR, IR, TLC, and optical rotation.
`The reasons for this phenomenon are presently unclear.24
`The assignment of the relative stereochemistry of 25 was
`obtained by comparison of the 1H NMR and optical rotation
`data of 25 to those of 26, which was obtained by alkylation of
`24 with 1,4-dibromobutane. The relative stereochemistry of 26
`was assigned unambiguously through single-crystal X-ray
`analysis (Figure 2).21 The absolute stereochemistry of 25 was
`confirmed by Barton deoxygenation and conversion to dike-
`topiperazine (+)-29 as illustrated in Scheme 3. This same
`diketopiperazine could be obtained, as the enantiomer, from 30.
`This compound has previously been converted to (+)-paraher-
`quamide B, a substance whose absolute stereochemistry has been
`confirmed.19
`
`Synthesis of a Functionalized Diketopiperazine
`
`It was necessary to convert the substituted proline (25) into
`a suitably functionalized diketopiperazine for a similar Somei-
`Kametani coupling reaction to that used in our total synthesis
`of paraherquamide B. Initial studies on this system were carried
`out with the secondary alcohol protected as a benzyl ether.
`
`(23) (a) Cooper, J.; Gallagher, P. T.; Knight, D. W. J. Chem. Soc., Perkin Trans.
`1 1993, 1313-1317. For alternative baker’s yeast reductions, see: (b) Sibi,
`M. P.; Christensen, J. W. Tetrahedron Lett. 1990, 31, 5689-5692. (c) Bhide,
`R.; Mortezaei, R.; Scilimati, A.; Sih, C. J. Tetrahedron Lett. 1990, 31,
`4827-4830.
`(24) Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624-1654.
`
`12174 J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003
`
`However, because of problems with selectivity and purification
`later in the synthesis, the less bulky and more polar methoxy-
`methyl (MOM) protecting group was used in the final synthetic
`route. After MOM protection of the alcohol, the N-t-BOC group
`was smoothly removed with ZnBr2 in dichloromethane25 and
`the exposed secondary amine (31) was acetylated with bro-
`moacetyl bromide under Schotten-Baumann conditions (Scheme
`4). Treatment of the bromoacetamide with methanolic ammonia
`afforded the corresponding glycinamide (32) which was directly
`subjected to cyclization in the presence of sodium hydride in
`toluene/HMPA to afford the bicyclic compound 33 in 75%
`overall yield from 25. An interesting observation about the ease
`of closure of hydroxylated diketopiperazines was made during
`this study. When there is no hydroxyl substituent (e.g., in 28)
`or the protected hydroxyl group is trans to the ester, the
`diketopiperazine typically forms spontaneously from the ami-
`noester in methanol at room temperature. On the other hand, a
`cis-isomer such as 31 can be isolated as the aminoester from
`the amination reaction, and formation of the diketopiperazine
`requires much more forcing conditions. On amination of a
`
`(25) Nigam, S. C.; Mann, A.; Taddei, M.; Wermuth, C.-G. Synth. Commun.
`1989, 19, 3139-3142.
`
` P. 3
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`Total Synthesis of Paraherquamide A
`
`Scheme5. Preparation of the Gramine Derivative 51
`
`A R T I C L E S
`
`mixture of diastereomeric bromoacetamides 35, the aminoester
`36 and the diketopiperazine 37 are produced. This is pre-
`sumably because the cis-diketopiperazine is significantly more
`sterically hindered. After diketopiperazine formation, a one-
`pot double carbomethoxylation reaction was performed by the
`sequential addition of n-BuLi in THF followed by addition of
`methylchloroformate, which carbomethoxylates the amide ni-
`trogen. Subsequent addition of more methylchloroformate
`followed by LHMDS afforded 34 in 93% yield as an (cid:24)6:1
`mixture of E and Z isomers, with the newly created stereogenic
`center as a single stereoisomer (relative configuration was not
`assigned).
`
`Improved Synthesis of the Gramine Derivative
`
`With this functionalized diketopiperazine in hand, we turned
`our attention to improvement of the synthesis of the dioxepin-
`containing indole fragment that we originally described in
`1990.26 The original route provides compound 51 in 14 steps
`with no chromatography required until the ninth step. However,
`further optimization was necessary to achieve a more rapid and
`efficient large-scale synthesis. The route we have developed is
`illustrated in Scheme 5. Vanillin (38) was acetylated with acetic
`anhydride and then treated with fuming nitric acid to afford
`39, the desired regioisomer, and 40, the undesired isomer, in
`an (cid:24)10:1 ratio. Initially, these regioisomers were separated by
`hydrolysis of the acetate group and isolation of the desired phe-
`nol isomer by crystallization.27 Analysis of the product mixture
`by TLC revealed that 39 had a lower Rf and 40 had exactly the
`same Rf as that of the starting material, and it was possible to
`isolate 39 by flash chromatography. However, neither purifica-
`tion method proved optimal for a large-scale protocol. The new
`
`(26) Williams, R. M.; Cushing, T. D. Tetrahedron Lett. 1990, 31, 6325-6328.
`(27) Raiford, L. C.; Stoesser, W. C. J. Am. Chem. Soc. 1928, 50, 2556-2563.
`
`approach circumvents these problems. Instead, we directly used
`the mixture of nitrobenzaldehydes 39 and 40. After a three-
`step transformation,28 39 provided the desired acid 41, and 40
`provided the undesired acid 42. Reduction of the nitro group
`was originally carried out in 95% yield by hydrogenation over
`palladium on carbon at 40 psi and 80 (cid:176) C. However, this protocol
`could prove awkward on a large scale, so an alternative approach
`was developed using iron and NH4Cl29 which, while the yield
`(74%) is more moderate, proved easier to scale-up. On reduction
`to the corresponding amines, the amine intermediate from 41
`cyclized to oxindole 43, but 42 was simply reduced to amino
`acid 44, which cannot undergo an intramolecular cyclization
`reaction due to geometric restriction. On extraction of the reac-
`tion mixture, the amino acid (44) was removed with aqueous
`acid leaving the oxindole (43) in the organic phase. Demethy-
`lation then proceeded smoothly as already described to give
`45.30
`Prenylation of 45 is partially selective for the 7-hydroxy
`position due to the greater acidity of this hydroxyl group.
`However, under the prenylation conditions originally developed
`for paraherquamide B, small amounts of the 6-prenyloxy and
`6,7-diprenyloxy isomers were also formed, and the three
`compounds are difficult to separate by flash chromatography.
`In this modification of our original route, replacement of the
`base with Cs2CO3 improves the selectivity and yield of 46.
`
`(28) (a) MacDonald, S. F. J. Chem. Soc. 1948, 376-378. (b) Beer, R. J. S.;
`Clarke, K.; Davenport, H. F.; Robertson, A. J. Chem. Soc. 1951, 2029-
`2032.
`(29) Wissner, A.; Berger, D. M.; Boschelli, D. H.; Floyd, M. B., Jr.; Greenberger,
`L. M.; Gruber, B. C.; Johnson, B. D.; Mamuya, N.; Nilakantan, R.; Reich,
`M. F.; Shen, R.; Tsou, H.-R.; Upeslacis, E.; Wang, Y. F.; Wu, B.; Ye, F.;
`Zhang, N. J. Med. Chem. 2000, 43, 3244-3256.
`(30) Since this route was developed, McWhorter and Savall have published a
`route to 45 which is shorter and higher yielding, but the applicability of
`their synthesis on a large scale has not yet been demonstrated. Savall, B.
`M.; McWhorter, W. W. J. Org. Chem. 1996, 61, 8696-8697.
`
`J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003 12175
`
` P. 4
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`A R T I C L E S
`
`Scheme6. Coupling of the Indole and Diketopiperazine
`
`Williams et al.
`
`Extraction into base during the workup procedure also removes
`the diprenylated byproduct which allowed for easier purification.
`A major problem in our first generation synthesis of the
`gramine derivative was during reduction of the oxindole to the
`indole, when over-reduction to the indoline occurred in variable
`quantities giving a ratio of 4:1 to 2:1 of indole/indoline. Attempts
`were made, without success, to effect a more selective reduction
`of the oxindole. However, the problem was solved in an indirect
`fashion as it proved possible to oxidize the indoline byproduct
`to the indole with DDQ31 in greater than 90% yield.
`Formation of TBS ethers on hindered alcohols is known to
`be very sensitive to the concentration of the reaction mixture.
`The silylation reaction was optimized by concentrating the
`reaction mixture to give an improved yield of 95% from 82%.
`Finally indole 50 was converted to the gramine derivative 51
`under standard conditions. The advantages of this new approach
`are significant in terms of increased yield, lower cost, and faster
`synthesis on a large scale.
`
`Coupling of the Indole and Diketopiperazine
`Somei-Kametani coupling32 of diketopiperazine 34 with the
`gramine derivative 51 in the presence of tri-n-butylphosphine
`gave a separable mixture of two diastereomers 52 and 53 in a
`
`(31) He, F.; Foxman, B. M.; Snider, B. B. J. Am. Chem. Soc. 1998, 120, 6417-
`6418.
`(32) (a) Somei, M.; Karasawa, Y.; Kaneko, C. Heterocycles 1981, 16, 941-
`949. (b) Kametani, T.; Kanaya, N.; Ihara, M. J. Chem. Soc., Perkin Trans.
`1 1981, 959-963.
`
`12176 J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003
`
`3:1 ratio, each as a mixture of four diastereomers (Scheme 6).33
`Decarbomethoxylation was effected by treatment of 52 and 53
`individually with LiCl in hot, aqueous HMPA to provide, in
`both cases, a mixture of 54 (anti-isomer) and 55 (syn-isomer),
`which could now be separated into the E and Z isomers, each
`of which as a mixture of two diastereomers (epimeric at the
`dioxepin secondary alcohol). However, as separation of the
`geometric isomers proved to be difficult, the compounds were
`usually carried through the synthetic sequence as a mixture and
`separated only for analytical purposes. Protection of the second-
`ary amide as the corresponding methyl
`lactim ether was
`accomplished by treating 54 and 55 with trimethyloxonium
`tetrafluoroborate and Cs2CO3 in dichloromethane. Model studies
`had shown that Cs2CO3 was a more efficient base than
`Na2CO3 for this reaction, as it leads to a lower incidence of
`TBS cleavage and N-methylation. Next, the indole nitrogen was
`protected as the corresponding N-t-BOC derivative by treatment
`with di-tert-butyl dicarbonate in the presence of DMAP, and
`then the silyl ethers were removed with tetrabutylammonium
`fluoride (TBAF) to provide 58 (anti) and 59 (syn). From this
`point onward, the E and Z isomers were utilized separately.
`Unfortunately, the Corey procedure,34 which had been successful
`
`(33) The stereochemistry at the newly formed stereogenic centers in 52 and 53,
`and in all subsequent compounds, was assigned on the basis of 1H NMR
`data. In compounds where the indole substituent is on the same face of the
`diketopiperazine as the MOM ether, the signal for the methoxy group is at
`significantly higher field than in the situation where these two substituents
`are on opposite faces. This is due to the proximity of the methoxy group
`to the shielding effects of the aromatic system.
`(34) Corey, E. J.; Kim, C. U.; Takeda, M. Tetrahedron Lett. 1972, 13, 4339-
`4342.
`
` P. 5
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`Total Synthesis of Paraherquamide A
`
`A R T I C L E S
`
`in the synthesis of paraherquamide B for conversion of an allylic
`alcohol to the corresponding chloride, proved unreliable when
`applied to the paraherquamide A system. Under the conditions
`used previously, cleavage of the lactim ether and chlorination
`at
`the 2-position of the indole were observed. Extensive
`investigation into suitable conditions was carried out, and it was
`eventually found that selective conversion of the primary
`alcohols 58 and 59 to the corresponding mesylates was possible
`in the presence of the hindered base collidine. Displacement of
`mesylate by a chloride ion under these reaction conditions was
`very slow so Bu3BnNCl (as an external chloride source) and a
`polar solvent were added to accelerate the reaction, allowing
`formation of the allylic chlorides (60 and 61) in up to 90% yield.
`This is a simple, practical, and reproducible method for
`preparing allylic chlorides in molecules bearing labile functional
`groups. Finally, careful reprotection of the secondary alcohols
`with tert-butyldimethylsilyl triflate in the presence of 2,6-lutidine
`afforded the key allylic chlorides 62 and 63.
`SN2¢ Cyclization and Closure of the Seventh Ring
`The stage was now set for the critical intramolecular SN2¢
`cyclization, that sets the relative stereochemistry at C-20 during
`formation of the bicyclo[2.2.2]diazaoctane ring nucleus. Based
`on precedent from the paraherquamide B synthesis,19 63E was
`treated with NaH in refluxing benzene. However, the reaction
`was very slow and gave the desired cyclization product 64 in
`only 25% yield, accompanied by products from competing
`pathways. The acidic proton in 63E is more sterically hindered
`than in the corresponding substrate for the paraherquamide B
`synthesis, due to the presence of the MOM ether. Since NaH
`likely exists as heterogeneous clusters in benzene,
`it was
`expected that use of a more coordinating solvent may break up
`the clusters and render deprotonation more facile. Conveniently,
`use of NaH in refluxing THF afforded the desired SN2¢
`cyclization product 64 in 87% yield from 63E exclusively as
`the desired syn-isomer.35 This remarkably diastereoselective
`intramolecular SN2¢ cyclization reaction proceeds, in a nonpolar
`solvent like THF, via a tight, intramolecular ion-pair driven
`cyclization (“closed” transition state)36 as shown in Scheme 7.
`Compound 62E also underwent the same transformation to give
`64 in 82% yield. In both cases, the product was sometimes
`accompanied by a small amount of Boc-deprotected cyclized
`product which could be reprotected under standard conditions.
`In addition, it was interesting to note that the Z-isomer, 63Z,
`provides the same cyclization product, again with exclusive syn
`selectivity, in 50% yield.
`Closure of the seventh ring was attempted using PdCl2 and
`AgBF4 in acetonitrile followed by NaBH4 to reduce the incipient
`heptacyclic (cid:243)-palladium adduct,37 reaction conditions which had
`
`(35) The syn/anti relationship in this case refers to the relative stereochemistry
`between the C-20 stereogenic center (see paraherquamide numbering) and
`the proline residue.
`
`(36) (a) Denmark, S. E.; Henke, B. R. J. Am. Chem. Soc. 1989, 111, 8032-
`8034. (b) Denmark, S. E.; Henke, B. R. J. Am. Chem. Soc. 1991, 113,
`2177-2194.
`
`Scheme7. Formation of the Heptacycle
`
`been successful in the paraherquamide B synthesis.19 However,
`the only products isolated under the same conditions with 64
`were those appearing to arise from removal of the N-t-BOC,
`MOM and lactim ether protecting groups, presumably by HBF4
`generated in situ. To buffer the reaction mixture, propylene oxide
`was added as an acid scavenger and the reaction now proceeded
`to give the desired 2,3-disubstituted indole (65) in 85% yield.
`
`Completion of the Synthesis
`
`Conditions could not be found which would allow direct and
`high-yielding conversion of the lactim ether (65) to the amide.
`However, use of 0.1 M aqueous HCl
`in THF gave the
`corresponding ring-opened amine methyl ester (66) which was
`recyclized to the bicyclo[2.2.2]diazaoctane (67) by treatment
`of 66 with catalytic 2-hydroxypyridine in hot toluene. Chemose-
`lective reduction of the tertiary amide in the presence of the
`secondary amide to give 68 could be effected by treatment of
`the diamide 67 with the AlH3-Me2NEt complex followed by
`quenching with sodium cyanoborohydride, methanol, and acetic
`acid, as used in the synthesis of paraherquamide B. However,
`use of excess diisobutylaluminum hydride (DIBAL-H) in
`dichloromethane was a simpler experimental procedure and gave
`improved yields of 68.38 N-Methylation of the secondary amide
`proceeded smoothly and was followed by cleavage of the MOM
`ether with bromocatecholborane.39 Oxidation of the secondary
`alcohol with Dess-Martin periodinane40 and concomitant
`removal of the N-t-BOC group and TBS ether with TFA gave
`ketone 70 (Scheme 8).
`The final critical oxidative spirocyclization of the 2,3-
`disubstituted indole was effected by a two-step procedure.
`
`(37) Trost, B. M.; Fortunak, J. M. D. Organometallics 1982, 1, 7-13.
`(38) Fukuyama, T.; Liu, G. Pure Appl. Chem. 1997, 69, 501-505.
`(39) Boeckman, R. K., Jr.; Potenza, J. C. Tetrahedron Lett. 1985, 26, 1411-
`1414.
`(40) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155-4156. (b)
`Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-7287.
`
`J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003 12177
`
` P. 6
`
`UT Ex. 2026
`SteadyMed v. United Therapeutics
`IPR2016-00006
`
`

`
`A R T I C L E S
`
`Scheme8. Manipulation of the Heptacycle
`
`Williams et al.
`
`Scheme9. Spirocyclization and Completion of the Synthesis
`
`Treatment of 70 with tert-butyl hypochlorite in pyridine
`provided a labile 3-chloroindolenine, from which it was found
`necessary to rigorously remove, by azeotroping with benzene,
`all of the pyridine prior to the next step. Pinacol-type rear-
`rangement with TsOH in aqueous THF then generated the
`desired spiro-oxindole (73). From our investigations during the
`paraherquamide B synthesis, it was found that use of a sterically
`demanding amine such as pyridine gives the best stereoselec-
`tivity during the chlorination reaction. It is assumed that addition
`of chlorine to 70 proceeds from the least hindered face of the
`indole giving the R-chloroindolenine 71. Hydration of the imine
`functionality, interestingly, must also occur from the same
`R-face, that is, syn-to the relatively large chlorine atom, to
`furnish the syn-chlorohydrin 72 which subsequently rearranges
`stereospecifically to the desired spiro-oxindole 73 (Scheme 9).
`Dehydration of the seven-membered ring in 73 with methyl
`triphenoxyphosphonium iodide (MTPI) in DMPU afforded 14-
`oxoparaherquamide B (14) in moderate yield.17 This intermedi-
`ate has been previously obtained semisynthetically from marc-
`fortine A by a group from Pharmacia-Upjohn, and comparison
`of the authentic and synthetic materials confirmed the identity
`of this substance. Addition of methylmagnesium bromide to the
`ketone group of 14 has been previously described to give
`paraherquamide A along with the corresponding C-14 epimer
`in around 50% yield.17a Employment of this protocol using
`
`12178 J. AM. CHEM. SOC. 9 VOL. 125, NO. 40, 2003
`
`MeMgBr with our synthetic ketone gave (-)-paraherquamide
`A (1) as the exclusive product (the C-14 epimer was not
`detected) in 42% yield. This product was identical to a natural
`sample of paraherquamide A by 1H NMR, 13C NMR, IR, exact
`mass, and mobility on TLC (Rf). A synthetic sample was
`recrystallized from ether and had mp 250 (cid:176)C (dec), [ R]D
`25)
`-22 (c ) 0.2, MeOH). Natural paraherquamide A recrystallized
`from ether under the same conditions yielded a sample with
`mp 250 (cid:176)C (dec) and [ R]D
`25 ) -21 (c ) 0.2, MeOH). All
`intermediates up to the final product have an enantiomeric ratio
`of approximately 97.5:2.5; the final synthetic paraherquamide
`A upon recrystallization from ether is approximately optically
`pure.
`We have reported here the first total synthesis of paraher-
`quamide A, the most biologically potent member of this family
`of compounds. This asymmetric synthesis proceeds in 46 steps
`from commercially a