Stereoselective synthesis of a MCHr1 antagonist

Article abstract of DOI:10.1021/jo701894v

(Chemical Equation Presented) Melanin-concentrating hormone (MCH) is implicated in the feeding behavior in mammals affording a potential target to control overeating in people. Compound 1 (AMG 076) has been identified as a potent MCHr1 antagonist for the

Full text of DOI:10.1021/jo701894v

Stereoselective Synthesis of a MCHr1 Antagonist
Denise Andersen,*,† Thomas Storz,†,| Pingli Liu,# Xin Wang,† Leping Li,‡ Pingchen Fan,‡
Xiaoqi Chen,‡ Alan Allgeier,† Alain Burgos,§ Jason Tedrow,† Jean Baum,† Ying Chen,†
Rich Crockett,† Liang Huang,† Rashid Syed,† Robert D. Larsen,† and Mike Martinelli†
Chemistry Process Research and DeVelopment, Amgen Inc., Thousand Oaks, California 19320, Medicinal
Chemistry, Amgen SF, South San Francisco, California 94080, Process Chemistry, Incycte Corporation,
Wilmington, Delaware 19880, and Pharmaceutical Products PPG-Sipsy, Z.I. La Croix Cadeau B.P. 79,
F-49242 AVrille´ Cedex, France
danders555@yahoo.com
ReceiVed September 7, 2007
Melanin-concentrating hormone (MCH) is implicated in the feeding behavior in mammals affording a
potential target to control overeating in people. Compound 1 (AMG 076) has been identified as a potent
MCHr1 antagonist for the treatment of obesity. A synthesis suitable for the large-scale preparation of
this lead candidate was developed to support preclinical studies. A Robinson annulation of benzylpiperidone
and resolution of the desired enone from a mixture of the diastereomers afforded key intermediate 6
after a stereoselective hydrogenation. Subsequent Fischer indole synthesis with hydrazine 5 then provided
the advanced intermediate, indole 2. Two complementary reductive amination strategies employing either
aldehyde 3 or lactol 4 led to the synthesis of title compound 1.
Introduction
combined effects of reduced food intake and increased metabo-
lism. As a result, a large scientific effort has been placed on
the development of MCHr1 antagonists.4 Discovery efforts
Melanin-concentrating hormone (MCH), a cyclic nonade-
capeptide, plays a key role in eating behavior and energy
homeostasis.1 In rodents, intracerebraventricular administration
of MCH leads to increased food intake2 while MCH-deficient
or MCH-knockout mice are lean, hypophagic, and hypermeta-
bolic.3 The action of MCH is believed to be the result of binding
to MCH receptor 1 (MCHr1), which is a member of the
G-protein-coupled receptors and is highly expressed in the lateral
hypothalamus and zona incerta of the CNS.1 Antagonism of
MCHr1 has the potential to lead to weight loss through the
(2) (a) Gomori, A.; Ishihara, A.; Ito, M.; Mashiko, S.; Matsushita, H.;
Yumoto, M.; Tananka, T.; Tokita, S.; Moriya, M.; Iwassa, H.; Kanatani,
A. Am. J. Physiol. Endocrinol. Metab. 2003, 284, E583. (b) Ito, M.; Gomori,
A.; Ishihara, A.; Oda, Z.; Mashiko, S.; Natsushita, H.; Yumoto, M.; Sano,
H.; Moriya, M.; Kanatani, A. Am. J. Physiol. Endocrinol. Metab. 2003,
284, E940. (c) Kennedy, A. R.; Todd, J. F.; Stanley, S. A.; Abbott, C. R.;
Small, C. J.; Ghateri, M. A.; Bloom, S. R. Endocrinology 2001, 142, 3265.
(d) Sherman, L. P.; Camacho, R. E.; Sloan, S. D.; Zhou, Bednarek, M. A.;
Hereniuk, D. L.; Feighner, S. D.; Tan, C. P.; Howard, A. D.; Van der Ploeg,
L. H.; MacIntyre, D. E.; Hickey, G. J.; Strack, A. M. Eur. J. Pharmacol.
2003, 15, 475.
(3) (a) Shimada, M.; Tritos, D.; Aitken, A.; Donella-Deana, A.; Hem-
mings, B. A.; Parker, P. J. Nature 1998, 396, 670. (b) Shimada, M.; Tritos,
D.; Aitken, A.; Donella-Deana, A.; Hemmings, B. A.; Parker, P. J. Eur. J.
Biochem. 1982, 124, 21. (c) Chen, Y.; Hu, C.; Hsu, C. K.; Zhang, Q.; Bi,
C.; Asnicar, M.; Hsiung, H. M.; Fox, N.; Slieker, L. J.; Yang, D. D.; Heiman,
M. L.; Shi, Y. Endocrinology 2002, 143, 2469. (d) Marsh, D. J.; Weingarth,
D. T.; Novi, D. E.; Chen, H. Y.; Trumbauer, M. E.; Chen, A. S.; Guan, X.
M.; Jiang, M. M.; Feng, Y.; Camacho, R. E.; Shen, Z.; Frazier, E. G.; Yu,
H.; Metzger, J. M.; Kuca, S. J.; Shearman, L. P.; Gopal-Truter, S.; Macneil,
D. J.; Strack, A. M.; MacIntntyre, D. E.; Van der Ploeg, L. H.; Qian, S.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 3240.
† Chemistry Process Research and Development, Amgen Inc.
‡ Medicinal Chemistry, Amgen SF.
# Process Chemistry, PIncycte Corporation.
§ Pharmaceutical Products PPG-Sipsy.
| Current address: Wyeth Research, Chemical Development, 401 N. Middle-
town Road, Pearl River, NY 10965.
(1) (a) Cambers, J.; Ames, R. S.; Bergsma, D.; Muir, A.; Fitzgerald, L.
R.; Hervieu, G.; Dytko, G. M.; Foley, J. J.; Martin, J.; Liu, W-S.; Park, J.;
Ellis, C.; Ganguly, S.; Konchar, S.; Cluderays, J.; Leslie, R.; Wilson, S.;
Sarau, H. M. Nature 1999, 400, 261. (b) Saito, Y.; Nothacker, H-P.; Wang,
Z.; Lin, S.; Leslie, F.; Civelli, O. Nature 1999, 400, 265.
10.1021/jo701894v CCC: $37.00 © 2007 American Chemical Society
Published on Web 11/13/2007
9648
J. Org. Chem. 2007, 72, 9648-9655

StereoselectiVe Synthesis of a MCHr1 Antagonist
SCHEME 2. Preparation of Octahydroisoquinolinone 6
FIGURE 1. MCHr1 antagonist AMG 076.
SCHEME 1. Retrosynthetic Analysis of AMG 076
amination with cyclohexylacetaldehyde derivative 3 or 4
completed the synthesis of AMG 076. This sequence of steps
was sufficient for preparing initial quantities of material for
testing but the process was not capable of supporting further
preclinical and clinical development. Specifically, the indoliza-
tion and the means for converting 2 to 1 were major bottlenecks.
Furthermore, preparation the octahydroisoquinolinone presented
a unique challenge in establishing the relative and absolute
stereocenters for the drug candidate. Herein, we describe the
development of an effective and productive synthesis of the
MCHr1 antagonist AMG 076 (1).
resulted in the identification of compound 1 (AMG 076) as a
potent, bioavailable MCHr1 antagonist (Figure 1).5
The synthetic route to compound 1 has proceeded through
three key building blocks (Scheme 1). A Fischer-indolization
between hydrazine 5 and octahydroisoquinolinone 66 provided
the indole ring of penultimate intermediate 2. Reductive
Results and Discussion
Octahydroisoquinolinone Preparation. The traditional meth-
odology for the synthesis of such a bicyclic enone would be
the Robinson annulation.7 However, the relative stereochemistry
to be expected with pentenone was uncertain.8 The protocol was
applied to the starting material, N-benzyl-4-piperidone (7), which
was converted to enamine 8 before alkylation by using pyrro-
lidine in refluxing toluene over a Dean-Stark trap to remove
water (Scheme 2).9 Rapid heating of the toluene was not an
option since pyrrolidine could be lost in the distillate. However,
by gradually increasing the temperature of the reaction mixture
to 84 °C and holding the temperature for 1 h, the pyrrolidine
was contained in the mixture. The resultant adduct was formed
completely with only 1.5 equiv of the reagent. The heating was
(4) (a) Tavares, F. X.; Al-Barazanji, K. A.; Bishop, M. J.; Britt, C. S.;
Carlton, D. L.; Cooper, J. P.; Feldman, P. L.; Garrido, D. M.; Goertz, A.
S.; Grizzle, M. K.; Hertzog, D. L.; Ignar, D. M.; Lang, D. G.; McIntryre,
M. S.; Ott, R. J.; Peat, A. J.; Zhou, H.-Q. J. Med. Chem. 2006, 49, 7108.
(b) Tavares, F. X.; Al-Barazanji, K. A.; Bigham, E. C.; Bishop, M. J.; Britt,
C. S.; Carlton, D. L.; Feldman, P. L.; Goertz, A. S.; Grizzle, M. K.; Guo,
Y. C.; Handlon, A. L.; Hertzog, D. L.; Ignar, D. M.; Lang, D. G.; Ott, R.
J.; Peat, A. J.; Zhou, H.-Q. J. Med. Chem. 2006, 49, 7095. (c) Takekawa,
S.; Asami, A.; Ishihara, Y.; Terauchi, J.; Kato, K.; Shimomura, Y.; Mori,
M.; Murakoshi, H.; Kato, K.; Suzuki, N.; Nishimura, O.; Fujino, M. Eur.
J. Pharmacol. 2002, 483, 129. (d) Kowalski, T. J.; Farley, C.; Cohen-
Williams, M. E.; Varty, G.; Spar, B. D. Eur. J. Pharmacol. 2004, 497, 41.
(e) Borowsky, B.; Durkin, M. M.; Ogozalek, K.; Marzabadi, M. R.; DeLeon,
J.; Lagu, B. Nat. Med. 2002, 8, 825. (f) Arienzo, R.; Clack, D. E.; Cramp,
S.; Daly, S.; Dyke, H. J.; Lockey, P.; Norman, D.; Roach, A. G.; Stuttle,
K.; Tomlinson, M.; Wong, M.; Wren, S. P. Bioorg. Med. Chem. Lett. 2004,
14, 4099. (g) Clark, D. E.; Higgs, C.; Wren, S. P.; Dyke, H. J.; Wong, M.;
Norman, D.; Lockey, P.; Roach, A. G. J. Med. Chem. 2004, 47, 3962. (h)
Souers, A. J.; Wodka, D.; Gao, J.; Lewis, J. C.; Vasudevan, A.; Gentles,
R.; Brodjian, S.; Dayton, B.; Ogiela, C. A.; Collins, C. A.; Kym, P. R.
Bioorg. Med. Chem. Lett. 2004, 14, 4879. (i) Souers, A. J.; Wodka, D.;
Gao, J.; Lewis, J. C.; Vasudevan, A.; Brodjian, S.; Dayton, B.; Ogiels, C.
A.; Fry, D.; Hernandez, L. E.; Marsh, K. C.; Collins, C. A.; Kym, P. R.
Bioorg. Med. Chem. Lett. 2004, 14, 4883. (j) Ulven, T.; Frimurer, T. M.;
Receveur, J.-M.; Little, P. B.; Rist, O.; Norregaard, P. K.; Hogberg, T. J.
Med. Chem. 2005, 48, 5684. (k) Soures, A. J.; Gao, J.; Brune, M.; Bush,
E.; Wodka, D.; Vasudevan, A.; Dayton, B.; Shapiro, R.; Hernandez, L. E.;
Collins, C. A.; Kym, P. R. J. Med. Chem. 2005, 48, 1318.
(6) For an alternative strategy for the formation of a similar, tetracyclic
isoquinolinone system, see: Le Goffic, F.; Gouyetts, A.; Ahoud, A.
Tetrahedron 1973, 29, 3357.
(7) Gawley, R. E. Synthesis 1976, 777.
(8) Annulation of N-methyl-4-piperidone with pentenone was reported
to give the N-Me analogue of 9. However, the stereochemistry was not
defined, see: (a) Bergman, J.; Goonewardena, H. Acta. Chem. Scand. Ser.
B 1980, 34, 763. For the reaction of 2-methylcyclohexanones with
pentenone, see: (b) Kikughi, M.; Yoshikoshi, A. Bull. Chem. Soc. Jpn.
1981, 54, 3420. (c) Coates, R. M.; Shaw, J. E. J. Am. Chem. Soc. 1970, 92,
5657. Robinson annulation of cyclohexanone with â-aryl methyl vinyl
ketones has been reported but the stereochemistry was not defined, see:
(d) Eiden, F.; Gmeiner, P. Arch. Pharm. 1988, 321, 397. (e) Eiden, F.;
Gmeiner, P. Liebigs Ann. Chem. 1988, 125.
(9) (a) Djerassi, C.; Pettit, G. R. J. Org. Chem. 1957, 22, 393. (b)
Danishefsky, S.; Cavanaugh, R. J. Org. Chem. 1968, 33, 2959. (c) The
reaction was monitored by gas chromatography for the consumption of
N-benzyl-4-piperidone. It was critical to the success of this process to leave
<2.5% of the starting material at this stage.
(5) (a) Leping, L; Fan, P.; Chen, X; Mihalic, J.; Fu, Y.; Dai, K.; Liang,
L.; Reed, M.; Wright, M.; Timmermans, P.; Chen, J-L.; Laen, J. Abstracts
of Papers, 231st National Meeting of the American Chemical Society, March
26-30, 2006; American Chemical Society: Washington, DC, 2006. (b)
Chen, X.; Chen, X.; Fan, P.; Jaen, J.; Li, L.; Mihalic, J. T. Patent WO
04/043958 A1, 2004.
J. Org. Chem, Vol. 72, No. 25, 2007 9649

Andersen et al.
SCHEME 3. Fischer-Indolization to 2
then increased to reflux to complete the dehydration. The
Michael addition and cyclization were accomplished by heating
the crude enamine with 3-penten-2-one10 in dioxane at 90 °C.
The reaction mixture contained alkylated piperidone intermedi-
ates and pyrrolidine adducts of the cyclized product.11 Addition
of aqueous acetic acid/sodium acetate converted these interme-
diates to the enone 9, which was a racemic mixture of the syn-
and anti-isomers of the methyl and bridgehead positions. The
enone was also a mixture of the R,â- and â,γ-isomers. Analysis
of the crude product showed a 17:76:6:1 mixture by gas
chromatography of the four possible isomers (compounds 9a-
d). The relative stereochemistry for the enone isomers was not
determined, rather, the mixture was used directly in the
subsequent chemical resolution step.
Treatment of the mixture of 9a-d with di-p-toluoyl-L-tartaric
acid (L-DTTA) in 80% ethanol/water resulted in the crystal-
lization of the desired diastereomeric salt as an off-white
solid.12 Analysis of the optical purity revealed that the desired
(8S,8aR)-enantiomer of 9 was obtained with 96.7% ee in 25-
28% overall yield from N-benzyl-4-piperidone, which was
sufficient for further processing.13 Fortuitously the Robinson
annulation procedure was suitable for establishing the relative
stereochemistry of the two key chiral centers and we could easily
access the pure enantiomer through resolution.
zine 5 was performed in dioxane at 70 °C with sulfuric acid,16
which gratifyingly provided an ∼8:1 mixture of desired linear
indole 2 and undesired angular indole 10 in 92% yield (Scheme
3).17 An extensive screen of a variety of acids and solvents did
not improve the ratio. Depending on the reaction temperature
variable amounts of the hydrolyzed byproducts 11 and 12, along
with HF, were generated.18 Control of the reaction temperature
between 67 and 72 °C resulted in full conversion while
minimizing the hydrolysis of the CF3 group to <5%. To avoid
the hydrolysis side reaction, and perhaps improve the isomer
ratio, an alternative to acid-catalyzed indolization of a ketone
with a hydrazine would be the palladium-catalyzed indolization
of 2-haloanilines, which is run under basic conditions.19
Unfortunately, coupling of isoquinolinone 6 and 2-iodo-4-
trifluoromethylaniline only gave a 1:1 ratio of the indole
products.
After salt-break of the enone 9 with aqueous sodium
carbonate, hydrogenation of the enone over 10% Pd(OH)2 on
carbon in the presence of Boc2O effected not only reduction of
the enone to set the final chiral center but in situ debenzylation
followed by N-acylation to afford the desired product 6. After
removal of the catalyst, crystallization of the product by addition
of heptane provided the octahydroisoquinolinone 6 in 69% yield
and >99% ee.14
Indole Synthesis. The Fischer-indole reaction15 between
octahydroisoquinolinone 6 and 4-trifluoromethylphenylhydra-
The indole isomers cocrystallized from the product mixture
after pH adjustment (>12) with no appreciable enrichment of
2. Crystallization of 2 as either the (R)-2-chloromandelic acid
or phosphoric acid salt effectively rejected the indole byproduct
10. Ultimately, the latter salt was selected as optimal. Thus,
formation of the phosphate in methanol followed by salt-break
(10) The geometric purity of the pentenone is >95% by 1H NMR. It
was essential to check the quality of the 3-penten-2-one used carefully. A
thorough analytical analysis revealed that it can contain variable amounts
of acetic acid and water. These impurities caused hydrolysis of enamine 8,
regenerating N-benzyl-4-piperidone, which lowered the yield of the tartrate
salt dramatically. Experimentation revealed that <0.10 wt % of acetic acid
and <1.0 wt % of water were tolerated.
(11) The structures were surmised by the GC/MS and literature precedent,
see: Stork, G.; Brizzolara, A.; Landesman, H.; Szmuskovicz, J.; Terrell,
R. J. Am. Chem. Soc. 1963, 55, 207. All the intermediates were converted
to the enone after addition of acetic acid.
(14) Minor amounts of the other diastereomers 6a and 6b were rejected
in the filtrate. Compounds 6a and 6b were each isolated in ∼0.25% yield
1
by chromatography and identified by H NMR. The remaining isomer 6c
was not isolated. The structure of 6c was assumed based on the mass
obtained from mass spectrometry and retention time comparison to the three
characterized isomers 6, 6a, and 6b. The stereochemistry of the methyl
group was key to the selectivity in the hydrogenation to produce the anti-
relationship between the two bridgehead positions. For comparison, the des-
methyl-N-benozyl analogue of 9 was converted to a mixture of the syn,anti-
isomers: Rastogi, S. N.; Bindra, J. S.; Rai, S. N. Indian J. Chem. 1972, 10,
673.
(15) Hughes, D. L. Org. Prep. Proced. Int. 1993, 25, 667.
(16) Liazarzabura, M. E.; Shuttleworth, S. J. Tetrahedron Lett. 2004,
45, 4781.
(12) The amount of piperidone was detrimental to the isolated yield of
the tartrate salt of 9. With >10% of N-benzyl-4-piperidone, the resolution
yield of 9 dropped from 28% overall yield to <12%, and in some cases,
the tartrate salt failed to crystallize. In the event that the N-benzyl-4-
piperidone was outside the set specification, unreacted material could be
removed by washing the crude enone mixture with 1 M aqueous citric acid
at pH 5.4.
(13) A second slurry in 80% ethanol/water could upgrade the material
to 98.5% ee, if required. However, material of >96% ee will produce
isoquinolinone 6 with >99% ee in the subsequent transformations.
(17) The Fischer-indolization of analogous decalinone systems has been
reported to be selective for the formation of the linear isomer over the
angular isomer: (a) Miller, F. M.; Lohr, R. A., Jr. J. Am. Chem. Soc. 1978,
43, 3388. (b) Rastogi, S. N.; Bindra, J. S.; Rai, S. N.; Anand, N. Indian J.
Chem. 1972, 10, 673. (c) Christoffers, J. Synthesis 2006, 318.
(18) At temperature of >72 °C there was significant glass erosion from
HF evolution (>1.5 mg/m3).
(19) Chen, C.; Lieberman, D. R.; Larsen, R. D.; Verhoeven, T. R.; Reider,
P. J. J. Org. Chem. 1997, 62, 2576.
9650 J. Org. Chem., Vol. 72, No. 25, 2007

StereoselectiVe Synthesis of a MCHr1 Antagonist
SCHEME 4. Coupling of Side Chain
SCHEME 5. tert-Butyl Ester Aldehyde Synthesis
SCHEME 6. Coupling of Indole 2 with tert-Butylester
Aldehyde 3
yielded 72% of pure indole 2 (Scheme 3). A crystal structure
of the (R)-2-chloromandelate salt established the absolute
stereochemistry of the molecule as 4aR,11R,11aS.
Reductive Amination of Indole 2 with tert-Butyl Ester
Aldehyde 3. The final coupling of the indoloisoquinoline 2 to
the cyclohexyl-(1-carboxylic acid)-ethyl moiety relied on a
reductive amination strategy. In preliminary studies the ester
aldehyde 13 was prepared by allylation of the methyl ester of
cyclohexylcarboxylic acid followed by ozonolysis (Scheme 4).
However, the product suffered from poor stability and could
not be isolated in pure form. Simply standing at room temper-
ature resulted in the formation of degradants including the lactol
4. Alternatively, the nitrile analogue 1-(2-oxoethyl)cyclohex-
anecarbonitrile 14, which was easily prepared by alkylation of
cyclohexylcarbonitrile followed by hydrolysis, was stable.
Coupling proceeded quite well with NaBH(OAc)3. However,
the nitrile group proved to be too hindered for hydrolysis to
the desired carboxylic acid moiety under either acidic or basic
conditions.
On the other hand, the tert-butyl ester group proved to be
more stable to degradation and was easily removed after
coupling. Deprotonation of 1620 with LDA and alkylation with
allyl bromide provided the allyl derivative 17 as an oil (Scheme
5). Alternatively, alkylation of 16 with 1,4 dibromo-3-butene
provided bis-adduct 18 as a crystalline solid in >90% yield.21
Ozonolysis of either compound 17 or 18 in isopropanol provided
tert-butyl 1-(2-oxoethyl)cyclohexane-carboxylate 3 as a light-
yellow oil. A number of reducing agents, including Me2S,
Na2S2O5, NH2CSNH2, and P(OMe)3, were tested for quenching
the intermediate ozonide. The cleanest product was obtained
with P(OMe)3.22
with NaBH(OAc)3 at room temperature to afford the desired
product 21 and corresponding regioisomer 22. The pure product
(21) The formaldehyde from the ozonolysis of compound 17 was not
easily separated. Despite wiped-film distillation of aldehyde 3 [see: (a)
Cvengrosˇ, J.; Valko, M.; Polla´k, Sˇ.; Lutisˇan, J. Chem. Papers 1999, 53,
417] small amounts of formaldehyde always contaminated the product,
which over time produced other byproducts. Aldehyde 3 was also prone to
air oxidation. To improve stability, 1000 ppm of triethanolamine [see: (b)
Ishihara, T.; Yamamoto, A.; Ohshima, M.; Aiba, N. Patent EP 0148648.
(c) Weber, J.; Falk, V.; Kniep, C. Patent DE 2917789] was added to the
distilled product, which was isolated in 78% yield and >95% purity as a
light yellow oil. In contrast, no formaldehyde was generated in the
ozonolysis of intermediate 18 to aldehyde 3. Material prepared from this
route was free of side products and was isolated in >95% yield and >99%
GC purity without the need for wiped-film distillation. As mentioned, one
problem with using the aldehyde 3 derived from 17 was the presence of
small amounts of formaldehyde. This led to the formation of the N-Me
derivative 19. This impurity could be avoided by using 3 prepared from
18.
In the subsequent reductive amination step (Scheme 6), the
crude 8:1 mixture of indole 2 and 10 was used directly as the
undesired isomer 22 was easily rejected in the crystallization
of 21. The indoles and purified aldehyde 3 in THF were treated
(20) Chandrasekaran, S.; Tuner, J. V. Synth. Commun. 1982, 12, 727.
J. Org. Chem, Vol. 72, No. 25, 2007 9651

Andersen et al.
SCHEME 8. Coupling of Indole 2 with Lactol 4
SCHEME 7. Lactol Synthesis
21 was isolated by crystallization from ethyl acetate as the
benzenesulfonic acid salt in 80% yield. Deprotection of the tert-
butyl ester group was carried out by the addition of 0.25 equiv
of benzenesulfonic acid to the salt 21 in acetonitrile and heating
the mixture at 70 °C. AMG 076 was isolated as the benzene-
sulfonate salt in 81% yield. An excess of benzenesulfonic acid
was required to improve the rate of the deprotection, as well as
to prevent partial disproportionation of the desired salt to the
free base.
very poor recoveries. Extraction of the product also was not
practical as the product is highly insoluble in the typical solvents.
Catalytic hydrogenation was evaluated as this could offer
improved separation of the product from the spent reagent
(Scheme 8). From a screen of a number of homogeneous and
heterogeneous catalysts, hydrogenation over Pt/C in isopropanol
gave the most effective conversion.25 The crystallization of the
zwitterion 1 as it formed in the coupling was overcome by the
addition of acetic acid. Thus, in a 70:30 mixture of IPA-AcOH,
the zwitterion remained in solution throughout the hydrogena-
tion. The spent catalyst was easily separated by filtration from
the reaction mixture. The product was crystallized by a solvent
switch to 1:15 AcOH/water. This produced a milky solution of
the zwitterion as observed previously. However, with this system
the fine particles could be reproducibly agglomerated into easily
filterable particles by the slow addition of 3-7% of toluene
with vigorous stirring to afford 1 in 86% yield. Salt formation
with BSA in 75% aqueous isopropanol gave the desired final
product 1 in 84% yield.
Reductive Amination of Indole 2 with Lactol 4. Although
lactol 4 was a degradant of methyl ester 13, the compound had
benefits over 3. The intermediate would avoid the stability issues
with the ester aldehydes, as well as a deprotection step (Scheme
7). Under forcing conditions with a strong acid, the methyl ester
was not converted cleanly to 4. Rather, the desired lactol was
synthesized directly from cyclohexane carboxylic acid (23).23
To avoid the exothermic deprotonation, a THF solution of 23
was slowly added to LDA at -10 °C. To ensure complete
dianion formation, it was crucial to stir the reaction mixture at
30-40 °C for several hours. The subsequent exothermic
alkylation of the dianion with allyl bromide was performed at
-10 to +10 °C. A catalytic amount of sodium iodide was
needed for full consumption of the anion, otherwise, the reaction
stalled at 80-90% conversion. The crude acid24 was obtained
as a clear yellow-orange oil in 87% yield. Allyl ester formation
or C,O-dialkylation was not observed. Ozonolysis of the allyl
group was carried out in MeOH at -45 °C. Trimethylphosphite
was used to reduce the ozonide at -30 to -40 °C to afford
lactol 4, which was isolated as a white solid in 62% yield.
Five-membered-ring lactols such as 4 have been reported to
be readily reduced to the corresponding lactone with a variety
of reducing agents.24 We were gratified that analogous to the
coupling of indole 2 with aldehyde 3 (Scheme 6), reductive
amination with lactol 4 by using NaBH(OAc)3 in THF ef-
fectively produced compound 1. Little over-reduction to the
butyrolactone was observed. Product isolation, however, was
very difficult. Compound 1 is a zwitterion, which formed a
cloudy mixture of ultrafine particles upon aqueous quench. The
solid could not be effectively isolated by filtration, leading to
Conclusions
Key to developing a productive synthesis of AMG 076 (1)
were the preparation of chiral isoquinolinone 6, the regioselec-
tivity in the Fischer-indolization, the preparation of a suitable
cyclohexylacetaldehyde equivalent, and a final reductive ami-
nation that allowed effective isolation of the drug substance.
Two alternative building blocks 3 and 4 were employed for
coupling the cyclohexylethyl side chain with the core hetero-
cycle 2. Although either endgame was suitable, the route through
the lactol was optimal based on stability, robustness, and
throughput. Kilogram quantities of 1 of were prepared utilizing
this chemistry.
(22) Performing the ozonolysis step in methanol led to dimethyl acetal
20 byproduct formation since the pH of the product mixture became acidic
due to the decomposition of P(OMe)3. By running the reaction in isopropanol
and adding a mild base such as Na2CO3 prior to the reducing agent, acetal
formation was prevented.
Experimental Section
(8S,8aR)-2-Benzyl-8-methyl-1,3,4,7,8,8a-hexahydroisoquino-
lin-6(2H)-one di-p-toluoyl-L-tartaric Acid Salt (9). A mixture of
N-benzyl-4-piperidone (41.95 kg, 221 mol), toluene (145 kg, 126
L), and pyrrolidine (23.75 kg, 334 mol) was slowly heated to 84
°C over 2 h and held for 1 h, then the mixture was heated to reflux.
Water was continuously removed by azeotropic distillation over a
(23) Direct alkylation of carboxylic acids has been reported previously:
(a) Creger, P. L. J. Org. Chem. 1972, 37, 1907. (b) Miller, R. D.; Goelitz,
P. J. Org. Chem. 1981, 46, 1616.
(24) For instance, see: (a) Sinhababu, A. K.; Borchardt, R. T. J. Org.
Chem. 1983, 48, 2356. (b) El-Shishtawy, R. M.; Fukunishi, K. Synthesis
1994, 1411. (c) Tanino, H.; Fukuishi, K.; Ushiyama, M.; Okada, K.
Tetrahedron 2004, 60, 3273.
(25) Only alcohol solvents performed well in this reaction. The N-ethyl
byproduct of 2 was observed in ethanol due to the generation of acetaldehyde
from oxidation of the solvent. Fortunately, isopropanol resulted in <1% of
the corresponding N-alkylated byproduct. In t-BuOH and sec-BuOH the
reactions were prohibitively sluggish.
9652 J. Org. Chem., Vol. 72, No. 25, 2007

StereoselectiVe Synthesis of a MCHr1 Antagonist
Dean-Stark trap. The consumption of starting material was
monitored by GC. When the amount of N-benzyl-4-piperidone was
<2.5%, the toluene was distilled to <25% from the reaction mixture
at 30 Torr and 50-60 °C. 1,4-Dioxane (174 kg, 174 L) and
3-penten-2-one (29.30 kg, 267 mol) were then charged to the thick,
dark oil and the solution was stirred at 90 °C for 14 h. Ap-
proximately 50% of the dioxane was then removed by vacuum
distillation and a mixture of sodium acetate (17.35 kg, 211 mol),
acetic acid (33.80 kg, 563 mol), and water (35.05 kg, 1945 mol)
was added slowly, while maintaining the temperature at <50 °C.
The temperature of the reaction mixture was then adjusted to 90
°C and the solution was stirred for 2 h. The remaining dioxane
was removed by vacuum filtration and the pH was adjusted to pH
8 with sodium hydroxide (50.25 kg, 353 mol) in water (45.25 kg,
2511 mol). Isopropyl acetate (110 kg, 126 L) was charged into the
mixture, the phases were separated, and the aqueous phase was
extracted once more with isopropyl acetate (110 kg, 126 L). The
pooled organic phases were concentrated by vacuum distillation
and the crude product mixture was analyzed by GC. The crude
enone mixture contained 2.4% of N-benzyl-4-piperidone. This
material was left in the reactor and was used as is in the resolution.
The crude enone was charged with ethanol (70 kg, 100 L) and water
(23 kg, 23 L).
A mixture of di-p-toluoyl-L-tartaric acid (73.0 kg, 188.9 mol),
water (30 kg, 30 L), and ethanol (95 kg, 120 L) was heated to 40
°C to ensure complete dissolution. The prepared solution was added
to the enone solution at 70 °C. During the addition a precipitate
formed. The suspension was cooled to 20 °C over 4 h and aged for
8 h. The product was isolated by filtration and washed with 80:20
ethanol-water (2 × 71 kg). The wet filter cake was sampled and
analyzed by chiral HPLC (96.7% ee). The salt was added to 80:20
ethanol-water (203 kg) and the mixture was heated at reflux for 1
h, cooled to 20 °C over 4 h, and aged for 8 h. The product was
isolated by filtration, washed with 80:20 ethanol-water (70 kg),
and vacuum-dried to give 79 kg of technical grade salt (28% overall
yield from 7), which contained ∼5% water: 98.5% ee; chemical
purity 99.5%; 1H NMR (400 MHz, DMSO-d6) δ 13.09 (br s, 1H),
7.29 (d, J ) 7.6 Hz, 4H), 6.79-6.70 (m, 9H), 5.18 (m, 3H), 3.21
(d, J ) 13.1 Hz, 1H), 3.07 (d, J ) 13.1 Hz, 1H), 2.76 (dd, J )
11.3, 5.5 Hz, 1H), 2.41 (m, 1H), 1.80 (s, 6H) superimposed on
1.91-1.72 (m, 3H), 1.62-1.46 (m, 3H), 1.34 (t, J ) 11.3 Hz, 1H),
1.19-1.10 (m, 1H), 0.34 (d, J ) 6.6 Hz, 3H); 13C NMR (100 MHz,
DMSO-d6) δ 198.0, 167.6, 164.7, 161.7, 144.3, 135.9, 129.5, 129.4,
128.2, 127.3, 127.6, 126.1, 124.2, 71.6, 60.6, 56.5, 51.5, 44.5, 42.2,
32.6, 32.1, 21.2, 18.8; IR (neat) 1719, 1670, 1611 cm-1. Anal. Calcd
for C37H39NO9: C, 69.25; H, 6.13; N, 2.18. Found: C, 68.97; H,
6.13; N, 2.16.
(8S,8aR)-2-Benzyl-8-methyl-1,3,4,7,8,8a-hexahydroisoquino-
lin-6(2H)-one (9). An aliquot of the isolated di-p-toluoyl-L-tartaric
acid salt 9 was treated with 10% aqueous sodium carbonate in water
to provide a salt-free, analytical standard of compound 9 as a light
yellow oil: 1H NMR (400 MHz, CDCl3) δ 7.39-7.32 (m, 5H),
7.30 (m, 1H), 5.84 (br s, 1H), 3.65 (d, J ) 13.3 Hz, 1H), 3.52 (d,
J ) 13.3 Hz, 1H), 3.30 (ddd, J ) 11.1, 5.6, 1.9 Hz, 1H), 2.99
(dddd, J ) 11.0, 5.6, 2.1, 2.1 Hz, 1H), 2.65-2.50 (m, 1H), 2.40-
2.35 (m, 2H), 2.32-2.27 (m, 1H), 2.14 (dd, J ) 15.9, 13.5 Hz,
1H), 2.04 (ddd, J ) 11.0, 9.5, 2.9 Hz, 1H), 1.90-1.80 (m, 1H),
1.76 (t, J ) 11.0 Hz, 1H), 1.04 (d, J ) 6.5 Hz, 3H); 13C NMR
(100 MHz, CDCl3) δ 199.4, 163.2, 138.0, 128.9, 128.3, 127.2,
124.6, 62.6, 58.3, 52.8, 45.2, 44.0, 34.6, 32.8, 19.3; IR (neat) 1715,
1677 cm-1. HRMS calcd for C17H22NO 256.1701 [M + H]+, found
256.1686 [M + H]+.
(4aR,8S,8aR)-8-methyl-6-oxo-octahydroisoquinoline-2-car-
boxylic Acid tert-Butyl Ester (6). A reactor was charged sequen-
tially with a 4.7 wt/wt solution of sodium carbonate in water (210
kg, 141.5 mol), resolved di-p-toluoyl-L-tartaric acid salt 9 (78.2
kg, 66.3 wt %, 84.2 mol), and EtOAc (156 kg, 169 L). The biphasic
mixture was stirred at 25 °C for 1 h, the layers were separated,
and the aqueous layer was extracted with a second portion of EtOAc
(182 kg, 197 L). The organic layers were combined, washed with
a 4.7 wt/wt solution of sodium carbonate in water (2 × 51 kg,
22.6 mol), and concentrated to 300 L by distillation. This sequence
was repeated for an additional 89.4 mol of di-p-toluoyl-L-tartaric
acid salt 9 and the free base solutions were combined and used
directly in the next step.
Di-tert-butyl carbonate (41.6 kg, 190.3 mol) and 10% Pd(OH)2
(3.4 kg, 2.4 mol) were charged into the crude reaction mixture.
The reaction vessel was pressurized with hydrogen (3 bar) and the
mixture was stirred at 40 °C. Conversion to product was monitored
by HPLC (12-17 h). The Pd catalyst was removed by adding Celite
(1.5 kg) to the reaction mixture and filtering through a bed of Celite
(0.7 kg). The excess Boc anhydride was destroyed by refluxing
the filtrate with water (38.9 kg, 2159 mol) for 8 h and discarding
the aqueous phase. The organic layer was concentrated to 130 L
and heptane was added (184 L). The reaction mixture was
concentrated (25-35 °C, 30 Torr) to a final volume of 85 L and
seeded to induce crystallization. The slurry was cooled to 0 °C,
aged for 30 min, filtered, and washed with cold heptane (16 kg).
The wet cake was dried to give 31.8 kg (69% yield) of ketone 6 as
a white solid: mp (DSC) 86.7 °C; 1H NMR (400 MHz, CDCl3) δ
4.44 (br s, 1H), 4.16 (br s, 1H), 2.69 (br t, J ) 11.5 Hz, 1H), 2.38
(dddd, J ) 14.2, 14.2, 4.6, 2.0 Hz, 2H), 2.33-2.23 (m, 1H), 2.11
(ddd, J ) 13.0, 13.0, 2.5 Hz, 2H), 1.65-1.50 (m, 2H), 1.47 (s,
9H), 1.32 (ddd, J ) 13.0, 13.0, 4.6 Hz, 1H), 1.27-1.21 (m, 1H),
1.05 (d, J ) 6.6 Hz, 3 H); 13C NMR (100 MHz, CDCl3) δ 209.2,
154.6, 79.5, 49.6, 47.7, 47.1, 46.0, 43.9, 41.2, 35.9, 32.9, 28.3, 19.0;
IR (neat) 1664, 1124 cm-1. HRMS calcd for C11H18NO3 212.1287
[M - tert-butyl + H]+, found 212.1272 [M - tert-butyl + H]+.
Anal. Calcd for C11H18NO3: C, 67.38; H, 9.42; N, 5.24. Found:
C, 67.43; H, 9.52; N, 5.18.
(4aR,11R,11aS)-11-Methyl-9-(trifluoromethyl)-2,3,4,4a,5,6,-
11,11a-octahydro-1H-pyrido[4,3-b]carbazole (2) and (4aR,5S,-
11cS)-5-Methyl-10-(trifluoromethyl)-2,3,4,4a,5,6,7,11c-octahydro-
1H-pyrido[3,4-c]carbazole (10). The ketone 6 (195.4 g, 0.731 mol)
and 4-(trifluoromethyl)phenylhydrazine 5 (180.6 g, 0.738 mol) were
mixed in dioxane (1.4 L). A solution of premixed concentrated
sulfuric acid (125 mL, 2.34 mol) and water (6.5 mL, 0.36 mol)
was added slowly to the yellow heterogeneous reaction mixture
while keeping the temperature of the vessel at <45 °C. When the
addition was complete, the reaction mixture was heated over 1 h
to 70 °C. Off-gassing of carbon dioxide and isobutylene from the
loss of the Boc group was observed at the beginning of the reaction.
The consumption of the hydrazone from the initial condensation
of ketone 6 and hydrazine 5 was monitored by HPLC and after
∼19 h there was <1%. The reaction was cooled to rt over 3 h and
NaOH (2.48 L of a 1.9 N solution in water, 4.75 mol) was added.
A beige precipitate formed and the mixture was aged at rt for 2 h.
The slurry was filtered, washed with water (3 × 0.5 L), and dried
at 60 °C with a nitrogen bleed for 24 h to give 207.4 g (92% yield)
of an 8.1:1 mixture of indoles 2 and 10.
For the coupling of the indole with the tert-butyl ester 3 the
crude indole mixture can be used as is. For coupling of 2 with the
lactol 4 the indole isomer 10 must be removed.
The crude mixture of indoles 2 and 10 (207.8 g, 0.67 mol) was
heated in methanol (2.0 L) at 60 °C to give a homogeneous solution.
Concentrated phosphoric acid (75 mL, 1.1 mol) was added over
20 min to the solution while maintaining the temperature at <65
°C. The reaction mixture was cooled over 2 h to 25 °C. The
phosphate salt of 2 was filtered, washed with methanol (2 × 0.4
L), and dried under vacuum with a nitrogen stream for 1 h. A
mixture of the crude phosphoric acid salt of compound 2, water
(3.2 L), and THF (0.8 L) was warmed to 35 °C and NaOH (142
mL of a 10 N solution in water, 1.42 mol) was added. The beige
slurry was stirred vigorously for 2 h then filtered and the filter
cake was washed with water (3 × 0.2 L). The wet filter cake was
dried to give 149 g (72% yield, 99.7 LCAP) of purified indole 2:
1
mp (DSC) 232.4 °C; H NMR (300 MHz, MeOH-d4) δ 7.77 (s,
1H), 7.37 (d, 8.5 Hz, 1H), 7.26 (dd, J ) 8.5, 1.3 Hz, 1H), 3.36
J. Org. Chem, Vol. 72, No. 25, 2007 9653

Andersen et al.
(dd, J ) 12.4, 3.7 Hz, 1H), 2.98 (br d, J ) 12.3 Hz, 1H), 2.63 (dd,
J ) 16.1, 4.3 Hz, 1H), 2.56-2.47 (m, 2H), 2.34 (ddd, J ) 16.1,
10.7, 2.1 Hz, 1H), 2.23 (dd, J ) 12.3, 11.1 Hz, 1H), 1.75-1.70
(m, 1H), 1.50-1.40 (m, 1H), 1.37 (d, J ) 6.5 Hz, 3H), 1.29 (dd,
J ) 12.0, 4.1 Hz, 1H), 1.24-1.11 (m, 1H); 13C NMR (75 MHz,
MeOH-d4) δ 139.3, 136.9, 127.8, 127.3 (q, JC-F ) 270 Hz), 121.4
(q, JC-F ) 31 Hz), 117.7 (q, JC-F ) 4 Hz), 117.2 (q, JC-F ) 5 Hz),
114.6, 112.0, 51.5, 48.2, 47.0, 38.5, 34.5, 33.1, 31.3, 20.0; IR (neat)
3100, 1105 cm-1. HRMS calcd for C17H20F3N2: 309.1573 [M +
H]+,found 309.1567 [M + H]+. Anal. Calcd for C17H19F3N2: C,
66.22; H, 6.21; N, 9.06. Found: C, 66.19; H, 6.24; N, 9.06.
An analytical standard of the regioisomer 10 was prepared by
concentrating the original mother liquor from the phosphoric acid
salt formation and rinsing the precipitate well with methanol: mp
(DSC) 249.3 °C; 1H NMR (400 MHz, CDCl3) δ 7.89 (s, 1H), 7.33
(s, 2H), 3.36 (dd, J ) 11.6, 3.0 Hz, 1H), 3.25 (m, 1H), 2.89 (td, J
) 12.0, 2.0 Hz, 1H), 2.79 (dd, J ) 16.3, 5.4 Hz, 1H), 2.69 (m,
2H), 2.52 (m, 2H), 1.87-1.75 (m, 2H), 1.35 (qd, J ) 10.7, 3.0 Hz,
1H), 1.08 (d, J ) 6.5 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ
137.5, 135.7, 126.8, 125.7, 120.4 (q, JC-F ) 32 Hz), 116.6, 116.2,
112.0, 110.5, 47.2, 42.8, 39.6, 32.2, 31.7, 31.6, 18.2, 3.70; IR (neat)
3150, 1105 cm-1. HRMS calcd for C17H20F3N2: 309.1573 [M +
H]+, found 309.1576 [M + H]+.
sodium triacetoxyborohydride (55.0 g, 259 mol) was added. The
mixture was stirred at rt for 4 h, diluted with EtOAc (200 mL),
and washed with 10% aqueous sodium carbonate (2 × 140 mL),
water (2 × 150 mL), and brine (2 × 150 mL). The product solution
was treated directly with a solution of PhSO3H (31 g, 146 mmol)
in EtOAc (130 mL) at rt over 1 h at which time a white precipitate
had formed. The product was isolated by filtration and dried to
constant mass to give 70 g (80% yield) of compound 21 as a fluffy
1
white solid: mp (DSC) 226.1 °C; H NMR (400 MHz, CDCl3) δ
11.30 (s, 1H), 9.12 (br s, 1H), 7.81 (s, 1H), 7.61 (d, J ) 2.0 Hz,
2H), 7.46 (d, J ) 8.5 Hz, 1H), 7.31 (m, 4H), 3.83 (d, J ) 11.0 Hz,
1H), 3.57 (d, J ) 11.5 Hz, 1H), 3.13 (m, 2H), 3.05-2.77 (m, 5H),
2.70 (t, J ) 7.5 Hz, 1H), 2.45 (m, 1H), 2.08 (m, 1H), 1.98-1.84
(m, 6H), 1.45 (s, 9H), 1.42 (d, J ) 6.5 Hz, 3H), 1.36-1.19 (m,
7H); 13C NMR (100 MHz,CDCl3) 173.8, 148.1, 137.4, 135.4, 128.1,
127.4, 125.5 (q, JC-F ) 271.4 Hz), 125.4, 125.2, 118.8 (q, JC-F
)
30.4 Hz), 116.4, 115.5 (JC-F ) 3.5 Hz), 111.8, 111.2, 80.0, 55.0,
52.2, 50.9, 45.0, 43.2, 33.9, 33.3, 32.7, 32.3, 30.8, 29.1, 28.3, 27.5,
25.0, 22.3, 22.2, 19.0; IR (neat) 3200, 1714, 1115 cm-1. HRMS
calcd for C26H34F3N2O2 463.2572 [M - tert-butyl + H]+, found
463.2588 [M - tert-butyl + H]+.
1-[2-((4aR,11R,11aS)-11-Methyl-9-trifluoromethyl-1,3,4,4a,5,6,-
11,11a-octahydropyrido[4,3-b]carbazol-2-yl)ethyl]cyclohexan-
ecarboxylic Acid‚Benzenesulfonic Acid (1). Salt 21 (50 g, 74
mmol) and PhSO3H (13 g, 81 mmol) were added to acetonitrile
(500 mL). The vessel was made inert with three vacuum/nitrogen
cycles and heated to 70 °C. After 5.5 h, toluene (500 mL) was
added over 1 h and the reaction mixture was then cooled to ambient
temperature over 1 h and 0 °C over 1 h. The slurry was stirred at
0 °C for a further 1 h and filtered. The filter cake was washed with
a 1:1 mixture of acetonitrile and toluene (2 × 100 mL) and dried
to give 37.4 g (81%) of the title compound 1 as a white solid:
1-Allylcyclohexanecarboxylic Acid tert-Butyl Ester (17). n-
Butyllithium (195 mL of a 1.6 M solution in hexanes, 0.31 mol)
was added to a solution of diisopropylamine (35.6 g, 0.35 mol) in
THF (195 mL) at -20 to 0 °C over 40 min and held at 0 °C for 30
min. The solution was cooled to -20 °C and tert-butyl cyclohex-
anecarboxylate (16) (50.0 g, 0.27 mol) was added. The reaction
mixture was stirred at 0 °C for 30 min and cooled to -20 °C. Allyl
bromide (26.0 g, 0.30 mol) was added and the reaction mixture
was then warmed to rt over 90 min. The reaction was monitored
by GC for the consumption of tert-butyl cyclohexanecarboxylate.
When the level was <1% the crude reaction mixture was cooled
to 0 °C and the reaction was quenched by adding it to a solution of
citric acid (67 g, 0.35 mol) in water (150 mL). The layers were
separated and the organic phase was washed with 10% aqueous
sodium carbonate solution, water (150 mL), and brine (150 mL)
and concentrated to give 58 g (95% yield) of 17 as a yellowish oil:
1
water (Karl Fischer) 1.4%, mp (DSC) 226.1 °C; H NMR (400
MHz, DMSO-d6) δ 12.55 (br s, 1H), 11.32 (s, 1H), 9.15 (br s,
1H), 7.81 (s, 1H), 7.64-7.61 (m, 2H), 7.47 (d, J ) 8.6 Hz, 1H),
7.31 (m, 4H), 3.84 (d, J ) 10.8 Hz, 1H), 3.58 (d, J ) 11.4 Hz,
1H), 3.16 (m, 1H), 3.05-2.79 (m, 4H), 2.67 (t, J ) 7.3 Hz, 1H),
2.43 (m, 1H), 2.07 (d, J ) 13.7 Hz, 1H), 1.93 (m, 3H), 1.73 (m,
1H), 1.60-1.50 (m, 5H), 1.40 (d, J ) 6.5 Hz, 3H), 1.36-1.26 (m,
4H); 13C NMR (100 MHz,DMSO-d6) 176.7, 148.1, 137.6, 135.6,
128.5, 127.6, 125.7 (q, JC-F ) 271.4 Hz), 125.6, 125.5, 119.0 (q,
JC-F ) 30.3 Hz), 116.5, 115.7 (q, JC-F ) 3.5 Hz), 112.0, 111.5,
55.3, 52.5, 51.1, 44.5, 43.4, 34.2, 33.2, 33.0, 32.1, 31.0, 29.3, 28.5,
25.2, 22.5, 19.2; IR (neat) 3463, 1735, 1694 cm-1. Anal. Calcd for
C32H19F3N2O5S + 0.5H2O: C, 61.03; H, 6.40; N, 4.45. Found: C,
61.34; H, 6.47; N, 4.26.
1-Allylcyclohexanecarboxylic Acid (23). A reactor was charged
with THF (3 L) and made inert with three vacuum/nitrogen cycles,
then diisopropylamine (2.05 kg, 20.26 mol) was added. The reaction
was cooled -30 to -40 °C and n-butyllithium (7.18 L of a 2.5 M
solution in hexanes, 17.95 mol) was added at a rate to maintain a
temperature of <0 °C. The solution was then stirred at -10 °C for
30 min and a solution of cyclohexanecarboxylic acid (1.0 kg, 7.8
mol) in THF (1.3 L) was added over 1 h. The reaction mixture
was heated to reflux (∼42 °C) and held for 8 h. The reaction
completion was determined by quenching an aliquot with allyl
bromide and analysis for cyclohexylcarboxylic acid by GC. When
the level was <5% the slurry was cooled to -20 °C and NaI (117.0
g, 0.78 mol) was added in one portion. Subsequently, allyl bromide
(1.01 L, 11.7 mol) was added at a rate such that the temperature
did not exceed 10 °C. When the addition was complete, the reaction
mixture was warmed to rt and stirred until the conversion was
>90% by GC. The mixture was cooled to -10 °C and water (3 L)
and MTBE (5 L) were added. The layers were separated and the
organic layer was extracted with 10% aqueous sodium carbonate
(2.0 L). The aqueous layers were combined and cooled to 10 °C
and the pH of the mixture was adjusted to 1-2 with 10% HCl
(∼5.0 L). The product separated out as an oil and was extracted
with MTBE (2 × 2.0 L). The combined organic phases were washed
1
bp 47-49 °C (1 Torr); H NMR (300 MHz, CDCl3) δ 5.72 (m,
1H), 5.04 (m, 1H), 5.00 (m, 1H), 2.20 (d, J ) 7.4 Hz, 2H), 2.00
(br d, 13.4 Hz, 2H), 1.55 (m, 3H), 1.44 (s, 9H), 1.35 (m, 2H), 1.20
(m, 3H); 13C NMR (75 MHz, CDCl3) δ 175.4, 134.0, 117.3, 79.8,
47.1, 44.9, 34.0, 28.2, 26.0, 23.2; IR (neat) 1719, 1128 cm-1. Anal.
Calcd for C14H24O2: C, 74.95; H, 10.78. Found: C, 74.83; H, 10.93.
1-(2-Oxo-ethyl)cyclohexanecarboxylic Acid tert-Butyl Ester
(3). Ozone gas was passed through a solution of allylcyclohexan-
ecarboxylate 17 (100 g, 0.44 mol) in isopropanol (500 mL) at -25
°C for 4 h. Nitrogen was then passed through the solution for 30
min at -25 °C to remove any excess ozone. Solid sodium carbonate
(47.3 g, 0.45 mol) was added first to the reaction mixture followed
by trimethylphosphite (56.0 g, 0.45 mol) over 60 min at -15 °C.
The slurry was then filtered into a mixture of MTBE (1 L) and 5%
aqueous sodium carbonate (1 L). The layers were separated and
the organic phase was washed with water (2 × 1 L) and brine (100
mL). The crude solution of the product was concentrated and then
purified by wiped-film distillation to afford 85 g (85% yield) of
aldehyde 3 as a slightly yellow oil: 1H NMR (300 MHz, CDCl3)
δ 9.75 (t, J ) 2.4 Hz, 1H), 2.54 (d, J ) 2.5 Hz, 2H), 2.00 (m, 2H),
1.52 (m, 6H), 1.45 (s, 9H), 1.38 (m, 2H); 13C NMR (75 MHz,
CDCl3) δ 201.1, 174.9. 80.9, 51.5, 45.1, 34.0, 28.0, 25.6, 22.4.
The thermal instability of aldehyde 3 precluded further characteriza-
tion.
1-[2-((4aR,11R,11aS)-11-Methyl-9-trifluoromethyl-1,3,4,4a,5,6,-
11,11a-octahydropyrido[4,3-b]carbazol-2-yl)ethyl]cyclohexan-
ecarboxylic Acid tert-Butyl Ester Benzenesulfonic Acid Salt (21).
The crude 8:1 mixture of indoles 2 and 10 (40.0 g, 130 mmol) and
aldehyde 3 (38.2 g, 169 mmol) were mixed in THF (160 mL). The
resultant mixture was stirred at rt for 1 h, then cooled to 10 °C and
9654 J. Org. Chem., Vol. 72, No. 25, 2007

StereoselectiVe Synthesis of a MCHr1 Antagonist
with 5 N NaCl solution (2 × 2.0 L) and evaporated to dryness to
yield 1.28 kg of 24 (97% yield, 91.4% GC purity) as a clear,
yellowish-orange oil: 1H NMR (600 MHz, CDCl3) δ 11.80 (br s,
1H), 5.73 (m, 1H), 5.07 (m, 1H), 5.03 (m, 1H), 2.29 (m, 2H), 2.05
(m, 2H), 1.65-1.52 (m, 3H), 1.47-1.37 (m, 3H), 1.30-1.23 (m,
4H); 13C NMR (100 MHz, CDCl3) δ 183.6, 133.4, 117.9, 47.2,
44.4, 33.5, 25.8, 23.1; IR (neat) 3078, 1697, 1641 cm-1. HRMS
calcd for C10H17O2 169.1223 [M + H]+, found 169.1222 [M +
H]+.
IPA (22.5 mL), and acetic acid was then added to the catalyst
suspension. The mixture was vigorously stirred under 20 psi of
hydrogen gas at 40 °C for 24 h at which time the conversion of
indole 2 to product was >99%. The spent catalyst was removed
by filtration and the filter cake was washed with AcOH (2 × 10
mL). The homogeneous reaction mixture was then concentrated to
ca. half of the original volume and water (375 mL) was added.
The resultant milky white suspension was vigorously stirred for 1
h at 40 °C and cooled to rt. Toluene (15.7 mL) was slowly added
to the rapidly stirred emulsion over 45 min. This resulted in the
agglomeration of the fine suspension of particles and filtration
furnished 12.8 g (86%) of zwitterion 1 as white granules: mp (DSC)
(() 3-Hydoxy-2-oxa-spiro[4.5]decan-1-one (4). A 2 L, 3-necked
Morton flask equipped with a thermocouple probe, a gas transfer
tube, and a nitrogen inlet was charged sequentially with 1-allyl-
cyclohexane carboxylic acid 24 (100 g, 0.594 mol) and methanol
(750 mL) and cooled to -45 °C. Ozone was bubbled into the
solution through the gas transfer tube until the blue color of ozone
persisted (∼3 h). The reaction progress can also be monitored by
quenching an aliquot into an excess of P(OMe)3 in methanol at
-78 °C and monitoring the conversion by GC. Nitrogen was then
bubbled into the solution until the reaction mixture was nearly
colorless (∼45 min) and then trimethylphosphite (84.1 mL, 0.713
mol) in methanol (84 mL) was added at -45 to -20 °C. After the
addition was complete, the solution was warmed to rt and stirred
for 10 h. The solvent was switched to MTBE and the mixture was
washed with water (3 × 300 mL). The solvent was switched again
to cyclohexane and the volume was adjusted to 900 mL. The
mixture was heated to reflux (∼80 °C), cooled to 35 °C, and seeded
to induce crystallization. The resultant slurry was aged at 0 °C for
2 h. The product was isolated by filtration, washed with cold
cyclohexane (200 mL), and dried in a vacuum oven (30 °C, 30
Torr) for 16 h to give 63.1 g (62.3% yield) of lactol 4 as a white
1
276.3 °C; H NMR (600 MHz, TFA-d1 + DMSO-d6) δ 11.29 (s,
1H), 9.35 (br s, 1H), 7.79 (s, 1H), 7.44 (d, J ) 8.5 Hz, 1H), 7.29
(d, J ) 8.5 Hz, 1H), 3.85 (br d, J ) 12.0 Hz, 1H), 3.58 (br d, J )
11.4 Hz, 1H), 3.17 (m, 1H), 3.04 (m, 1H), 2.95 (t, J ) 11.4 Hz,
1H), 2.87 (t, J ) 12.0 Hz, 1H), 2.82 (dd, J ) 16.1, 4.4 Hz, 1H),
2.67 (qui, J ) 6.5 Hz, 1H), 2.44 (dd, J ) 16.1, 12.6 Hz, 1H), 2.06
(d, J ) 12.0 Hz, 1H), 1.95 (m, 4H), 1.71 (m, 1H), 1.55 (m, 5H),
1.42 (d, J ) 6.5 Hz, 3H), 1.31 (m, 5H); 13C NMR (150 MHz, TFA-
d1 + DMSO-d6) 176.8, 137.7, 135.6, 125.9 (q, JC-F ) 270.7 Hz),
125.8, 119.4 (q, JC-F ) 30.8 Hz), 116.8, 115.9, 112.2, 111.6, 55.4,
52.7, 51.2, 44.7, 43.7, 34.4, 33.4, 33.4, 33.3, 31.2, 29.5, 28.7, 25.4,
22.6, 22.6, 19.3; IR (KBr) 3253, 3059, 1668, 1626 cm-1. HRMS
calcd for C26H34F3N2O2 463.2567 [M + H]+, found 463.2575 [M
+ H]+.
1-[2-((4aR,11R,11aS)-11-Methyl-9-trifluoromethyl-1,3,4,4a,5,6,-
11,11a-octahydropyrido[4,3-b]carbazol-2-yl)ethyl]cyclohexan-
ecarboxylic Acid‚Benzenesulfonic Acid (1). Benzenesulfonic acid
(4.9 g, 31.1 mmol) in water (15.6 mL) at 70 °C was slowly added
to a slurry of zwitterion 1 (12.0 g, 25.9 mmol) in IPA (65 mL).
Additional water (49 mL) was added to the mixture and stirring
was continued at 70 °C. The homogeneous solution was cooled
over 2 h to 55 °C and seeded to induce crystallization. The solution
was aged for 10 h at 55 °C and then was cooled to 5 °C and aged
for a further 4 h. The product was isolated by filtration and washed
with 50% IPA-water (2 × 75 mL) and dried to give 13.5 g (84%)
of the salt 1, which was identical with the sample synthesized from
1
solid: mp (DSC) 74.7 °C; H NMR (400 MHz, ACN-d3) δ 5.16
(br s, 1H), 5.76 (dd, J ) 6.0, 3.6 Hz, 1H), 2.33 (dd, J ) 13.6, 6.0
Hz, 1H), 1.95 (dd, J ) 13.6, 3.6 Hz, 1H), 1.68 (m, 3H), 1.64 (m,
1H), 1.62 (m, 1H), 1.58 (m, 1H), 1.50 (br d, J ) 13.0 Hz, 1H),
1.34 (m, 2H); 13C NMR (100 MHz, ACN-d3) δ 181.9, 97.5, 45.4,
41.2, 34.8, 34.6, 22.9, 22.8, 26.1; IR (KBr) 3465, 3314, 1763, 1731
cm-1. HRMS calcd for C9H15O3 171.1022 [M + H]+, found
171.1026 [M + H]+.
1-[2-((4aR,11R,11aS)-11-Methyl-9-trifluoromethyl-1,3,4,4a,5,6,-
11,11a-octahydropyrido[4,3-b]carbazol-2-yl)ethyl]cyclohexan-
ecarboxylic Acid (1). A suspension of 5% Pt/C (2.5 g, 50% wet
weight, 0.33 mmol) in acetic acid (22.5 mL) was purged sequen-
tially with nitrogen and hydrogen. The mixture was stirred at 40
°C and pressurized to 20 psi with hydrogen gas for 1 h to pre-
hydrogenate the catalyst. A homogeneous mixture of indole 2 (10.00
g, 32.4 mmol) in IPA (105 mL), lactol 4 (6.07 g, 35.6 mmol) in
1
the tert-butyl ester aldehyde 3 (mp, H NMR).
Supporting Information Available: Experimental methods,
X-ray crystallographic data, and NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO701894V
J. Org. Chem, Vol. 72, No. 25, 2007 9655