Stereocontrolled preparation of a nonpeptidal (-)-spirobicyclic NK-1 receptor antagonist

Article abstract of DOI:10.1021/jo0157472

The synthesis of a spirobicyclic NK-1 receptor (Substance-P) antagonist 1 antipode is described. Retrosynthetic analysis reveals an allylic halide A bearing the cyclopropoxy-substituted aryl group and a 2-phenyl-3-piperidone B. The stereochemistry in the spirobicyclic system bearing three chiral centers is initially set via a highly diastereoselective zinc-mediated coupling of the allylic bromide 23 to the optically active ketopiperidine 3. The remaining benzylic asymmetric center is set by a diastereoselective hydroboration followed by cyclization to the spirobicyclic system.

Full text of DOI:10.1021/jo0157472

J . Org. Chem. 2002, 67, 1093-1101
1093
Ster eocon tr olled P r ep a r a tion of a Non p ep tid a l (-)-Sp ir obicyclic
NK-1 Recep tor An ta gon ist
Peter E. Maligres,* Marjorie, M. Waters, J aemoon Lee, Robert A. Reamer, and David Askin
Department of Process Research, Merck Research Laboratories, Merck & Co., Inc., P.O. Box 2000,
Rahway, NJ 07065
Michael S. Ashwood and Mark Cameron
Department of Process Research, Merck Sharp & Dohme Research Laboratories, Merck & Co., Inc.,
Hertford Road, Hoddesdon, Hertfordshire, EN11 9BU, UK
peter_maligres@merck.com
Received May 10, 2001
The synthesis of a spirobicyclic NK-1 receptor (Substance-P) antagonist 1 antipode is described.
Retrosynthetic analysis reveals an allylic halide A bearing the cyclopropoxy-substituted aryl group
and a 2-phenyl-3-piperidone B. The stereochemistry in the spirobicyclic system bearing three chiral
centers is initially set via a highly diastereoselective zinc-mediated coupling of the allylic bromide
23 to the optically active ketopiperidine 3. The remaining benzylic asymmetric center is set by a
diastereoselective hydroboration followed by cyclization to the spirobicyclic system.
Sch em e 1
In tr od u ction
The neurokinin NK-1 receptor antagonists are an area
of continued interest in the pharmaceutical field.1 The
observation of antidepressant activity exhibited by a
neuropeptide antagonist developed by Merck has moti-
vated further interest in these Substance-P antagonists.2
The spirobicyclic NK-1 receptor antagonist (+)-1 has been
identified as a candidate required for our clinical pro-
gram.3
the key intermediate piperidinone B with an aryl three-
carbon fragment A already bearing the cyclopropoxy
function (Scheme 1).4 The resulting homoallylic alcohol
C would essentially be an isomer of the spirobicyclic
system. Attainment of this goal was initially hampered
by the sensitivity of the cyclopropoxy group to palladium
catalysts and other aggressive Lewis acids.5
Resu lts a n d Discu ssion
This novel 6-phenyl-1-oxa-7-azaspiro[4,5]decane struc-
ture poses several synthetic challenges including the
installation of three chiral centers embedded within the
spirobicyclic system and the unusual cyclopropoxy and
trifluoromethoxy substituents. We were primarily inter-
ested in developing a convergent route that would unite
To rapidly explore the feasibility of conversion of a
homoallylic alcohol with structure C to the spirobicyclic
(3) (a) Elliott, J . M.; Cascieri, M. A.; Chicchi, G. G.; Curtis, N. R.;
Hargreaves, R.; Harrison, T.; Hollingworth, G. J .; Kulagowski, J .;
Kurtz, M. M.; Owen, S.; Rupniak, N. M. J .; Rycroft, W.; Sadowski, S.;
Seward, E.; Swain, C. J .; Tattersall, F. D.; Williams, B.; Williams, A.
R. A New Class of Spirocyclic NK1 Antagonists. Book of Abstracts,
Tachykinins 2000; Physiology, Drug Discovery and Clinical Applica-
tions, La Grande Motte, France, 2000; poster P84. (b) Baker, R.; Curtis,
N. R.; Elliott, J . M.; Harrison, T.; Hollingworth, G. J .; J ackson, P. S.;
Kulagowski, J . J .; Seward, E. M.; Swain, C. J .; Williams, B. J . WO
97/49710, 1997; Chem Abstr. 1998, 128, 88921.
(4) (a) Wallace, D. J .; Goodman, J . M.; Kennedy, D. J .; Davies, A.
J .; Cameron, J .; Cowden, C. J .; Ashwood, M. S.; Cottrell, I. F.; Dolling,
U.-H.; Reider, P. J . J . Org. Lett. 2001, 3, 671-674. (b) Kulagowski, J .
J .; Curtis, N. R.; Swain, C. J .; Williams, B. J . Org. Lett. 2001, 3, 667-
670. (c) Wallace, D. J .; Cowden, C. J .; Kennedy, D. J .; Ashwood, M. S.;
Cottrell, I. F.; Dolling, U.-H. Tetrahedron Lett. 2000, 41, 2027-2029.
(d) Wallace, D. J .; Bulger, P. G.; Kennedy, D. J .; Ashwood, M. S.;
Cottrell, I. F.; Dolling, U.-H. Synlett, in press.
(1) (a) Swain, C. J . In Progress in Medicinal Chemistry; Ellis, G. P.,
Luscombe, D. K., Oxford, A. W., Eds.; Elsevier Science BV: Amster-
dam, 1998; Vol. 35, pp 57-81. (b) Harrison, T.; Williams, B. J .; Swain,
C. J .; Ball, R. G. Bioorg. Med. Chem. Lett. 1994, 4, 2545-2550. (c)
Snider, R. M.; Constantine, J . W.; Lowe, J . A., III; Longo, K. P.; Lebel,
W. S.; Woody, H. A.; Drozda, S. E.; Desai, M. C.; Vinick, F. J .; Spencer,
R. W.; Hess, H.-J . Science 1991, 251, 435-437.
(2) Kramer, M. S.; Cutler, N.; Feighner, J .; Shrivastava, R.; Carman,
J .; Sramek, J . J .; Reines, S. A.; Liu, G.; Snavely, D.; Wyatt-Knowles,
E.; Hale, J . J .; Mills, S. G.; MacCoss, M.; Swain, C. J .; Harrison, T.;
Hill, R. G.; Hefti, F.; Scolnick, E. M.; Cascieri, M. A.; Chicchi, G. G.;
Sadowski, S.; Williams, A. R.; Hewson, L.; Smith, D.; Carlson, E. J .;
Hargreaves, R. J .; Rupniak, N. M. J . Science 1998, 281, 1640-1645.
10.1021/jo0157472 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/22/2002

1094 J . Org. Chem., Vol. 67, No. 4, 2002
Maligres et al.
Sch em e 2
chromatography to provide 5a (42%) and 5b (20%).
Oxidation (PCC, CH2Cl2) of each diastereomer 5a and 5b
provided lactones 6a and 6b, respectively.11 Treatment
of either pure lactone 6a or 6b with DBU gave a mixture
of 6a and 6b in a 60:40 ratio, respectively, in both cases
as judged by HPLC analysis.12 Both diols 5a and 5b were
converted to their corresponding methanesulfonates 7a
and 7b which were cyclized to the spirobicyclic ethers
8a and 8b, respectively.13 The relative stereochemistry
of 8a and 8b was ascertained by observation of NOE
interactions within their spirobicyclic frameworks as was
done in the case of 6a and 6b.14
A substantial improvement of hydroboration stereo-
selectivity could open the path to an efficient and
convergent route to the desired spirobicyclic product.
Such a sequence involving the coupling of 2 with (R)-3
followed by diastereoselective hydroboration of 4 would
allow the asymmetry to be transferred from (R)-3 ulti-
(8) Competitive allylic homocoupling was responsible for low yields
observed when the 2 was first converted to its organozinc derivative.
Addition of 2 to zinc in the presence of 3 alleviated this problem, but
reaction proceeded only to ca. 80% completion. Separation of 4 from
the crude product containing unreacted 3 was not possible by silica
gel chromatography but treatment of the crude product with NaBH4
in MeOH converted 3 to the more polar corresponding alcohol which
was easily removed from the product by chromatography. 4: 1H NMR
(250 MHz, CDCl3) δ 7.44-7.23 (m, 10 H), 5.44 (d, J ) 1.6, 1 H), 5.26
(d, J ) 1.6, 1 H), 5.14 (s, 1 H), 4.06 (m, 1 H), 3.10 (m, 1 H), 3.06 (d, J
) 14.0, 1 H), 2.97 (d, J ) 14.0, 1 H), 1.93-1.64 (m, 4 H), 1.39 (s, 9H);
13C NMR (62.9 MHz, CDCl3) δ 155.4, 145.0, 142.5, 139.2, 129.4, 129.0,
128.5, 128.0, 127.7, 127.1, 126.6, 125.4, 118.6, 79.9, 72.8, 63.0, 44.1,
38.8, 31.3, 28.4, 21.8.
system, a simpler model system was prepared. Conden-
sation of 2-phenylallyl bromide (2)6 as its organozinc
derivative with (2R)-ketone 37,1b in Scheme 2 gave
homoallylic alcohol 4 in low yields. Alternatively, addition
of 2 to a mixture of 3 and zinc dust in THF furnished 4
in 74% yield.8 As expected, attack of the ketone 3
proceeded with virtually complete anti selectivity with
respect to the 2-phenyl group as shown by NOE inter-
actions.9
(9) The relative stereochemistry of a similar addition product derived
from 3 (ref 4a-c) was proven by the NOE interactions. Similar NOE
interactions were observed for 4 as shown.
Attempts at direct isomerization of 4 to the spirobicy-
clic system gave complex mixtures and were hampered
by the acid lability of 4.10 Hydroboration/oxidation of 4
with BH3‚THF or BH3‚Me2S gave a 2:1 diastereomeric
mixture of diols which was separated by silica gel
(10) Treatment of 4 with HCl, TsOH, ZnCl2, AlCl3, TiCl4, ZrCl4, Cp2-
ZrCl2, Cp2ZrHCl, BF3‚OEt2, PdCl2, Pd(OAc)2, NiBr2, CuCl2, CuCl, Cu-
(OTf)2, KOt-Bu, KOt-Bu/CuI, or BuLi lead to complex mixtures or to
no reaction.
(11) Oxidation of 5a and 5b was performed with 2 equiv of PCC in
CH2Cl2 in the presence of 3 Å molecular sieves to provide lactones 6a
(71%) and 6b (52%), respectively. Swern oxidation (DMSO, (COCl)2,
CH2Cl2, then Et3N) gave similar results, and none of the corresponding
lactols were observed in either oxidation.
(12) Both lactones were equilibrated to the diastereomeric mixtures
by treatment of each pure lactone 6a and 6b (2 mg) with DBU (2 mg)
in MeCN (0.05 mL) for 30 min at 20 °C. The mixtures were directly
analyzed by HPLC.
(13) Diols 5a and 5b were separately converted to 7a and 7b with
1.2 equiv of MsCl and 1.5 equiv of DIEA in CH2Cl2 at -20 °C followed
by aq NaHSO4 workup. Crude 7a and 7b were treated with 1.5 equiv
of NaHMDS in THF at 0 °C which afforded 8a (70%) and 8b (60%)
after aq NaHSO4 workup and silica gel chromatography. 8a : 1H NMR
(250 MHz, CDCl3) δ 7.64 (m, 2 H), 7.34-7.23 (m, 8 H), 5.35 (s, 1 H),
4.32 (t, J ) 8.1, 1 H), 3.98 (m, 1 H), 3.91 (t, J ) 9.0, 1 H), 3.58 (m, 1
H), 2.76 (m, 1 H), 2.69 (dd, J ) 12.7, 7.6, 1 H), 2.26 (td, J ) 12.7, 5.3,
1 H), 1.90 (dd, J ) 12.7, 10.7, 1 H), 1.84-1.63 (m, 4 H), 1.48 (s, 9 H);
13C NMR (62.9 MHz, CDCl3) δ 155.4, 141.3, 138.9, 129.0, 128.7, 128.1,
127.3, 126.8, 126.7, 84.9, 80.0, 74.1, 60.3, 45.8, 445, 38.9, 32.8, 28.5,
23.8.
(5) The cyclopropoxy moiety is isomerized to the allyloxy group in
the presence of palladium or strong acid. The resulting aryl allyl ether
undergoes cyclization to the dihydropyran. Hollingworth, G. J .; Dinnell,
K.; Dickinson, L. C.; Elliott, J . M.; Kulagowski, J . J .; Swain, C. J .;
Thomson, G. Tetrahedron Lett. 1999, 40, 2633-2636.
(6) Bromide 2 was generated in 95% yield by treatment of 2-
phenylallyl alcohol with PPh3Br2 in CH2Cl2 followed by dilution with
pentane, filtration through a plug of silica gel, and isolation of 2 by
vacuum distillation. Allylic bromide 2 has been prepared mixed with
isomeric 1-bromo-2-phenylpropene by bromination of commercially
available R-methylstyrene with NBS: (a) Vaccher, C.; Berthelot, P.;
Flouquet, N.; Vaccher, M.-P.; Debaert, M. Synth. Commun. 1993, 23,
671-679. (b) Mulzer, J .; Bruntrup, G.; Kuhl, U.; Hartz, G. Chem. Ber.
1982, 115, 3453. 2-Phenylallyl alcohol can be obtained by treatment
of PhCOCl with ClCH2I and MeLi‚LiBr: (c) Barluenga, J .; Concellon,
J . M.; Fernandezsimon, J . L.; Yus, M. J . Chem. Soc., Chem. Commun.
1988, 8, 536-537; or from oxidation of R-methylstyrene with t-BuOOH
in the presence of SeO2 (0.2 mol %).
(7) (2R)-Ketone 3 was employed for the synthesis of (3S,5S,6R)-
(-)-1. (2R)-Ketone 3 was prepared from a sample of enantiomerically
enriched intermediate (2R,3R)-cis-3-hydroxy-2-phenylpiperidine kindly
provided by Dr. R. D. Wilson by recrystallization as the dibenzoyl
D-tartrate salt followed by protection as the N-Boc derivative and
Swern oxidation to (2R)-3. (2S)-Ketone 3 is the material required for
the NK-1 receptor antagonist (3R,5R,6S)-(+)-1 and (2R)-3 was the
byproduct available from the resolution of racemic 3 to prepare (2S)-
3. For an asymmetric synthesis of 3, see: Lee, J .; Hoang, T.; Lewis,
S.; Weissman, S. A.; Askin, D.; Volante, R. P.; Reider, P. J . Tetrahedron
Lett. 2001, 42, 6223-6225.
(14) NOE interaction was observed between the 6-piperidine ben-
zylic proton and the 3-tetrahydrofuran benzylic proton.

Nonpeptidic (-)-Spirobicyclic NK-1 Receptor Antagonist
J . Org. Chem., Vol. 67, No. 4, 2002 1095
Ta ble 1. Hyd r obor a tion of 4 a n d 9
Sch em e 4
substrate
conditionsa
ratio 5a /5bb
4
4
4
4
4
9
9
BH3‚THF, 0 °C
62/38
61/39
55/45
66/33
60/40
83/17
70/30
BH3‚Me2S/THF, 0 °C
catecholborane/THF, 60 °C
9-BBN/THF, 60 °C
9-BBN/toluene, 60 °C
BH3‚THF, 0 °C
9-BBN/THF, 60 °C
a
The borane (3 equiv) was added to the substrate (0.1 mmol)
in the specified solvent (0.3 mL) at 0 °C, and the mixture was
b
raised to the specified temperature for 1 h. Ratios were deter-
mined by HPLC analysis after quenching with 0.3 mL 2:1 5 M aq
NaOH/30% H2O2. In the case of 9 the quenched reaction mixtures
were treated with 0.3 mL of 1 M TBAF in THF.
Sch em e 3
substituted allylic halide A. A synthetic route to aryl
iodide 14 starting from commercially available 4-tri-
fluoromethoxyphenol in Scheme 4 has been developed
during initial explorations toward installation of the
cyclopropoxy functionality early in the synthetic se-
quence. The development of milder conditions for the
formation of the spirobicyclic system now could allow
such a convergent synthesis. Optimization of the route
to 14 allowed the rapid preparation of large quantities
of 14 in 67% overall yield from 4-trifluoromethoxy-
phenol.17
The aryl iodide 14 could be smoothly converted to its
Grignard reagent 15 by treatment with Mg in methyl
tert-butyl ether (MTBE) or diethyl ether (complete con-
version by HPLC after quenching an aliquot in water).
Interestingly, the Grignard reagent formation did not
occur in THF or DME; it was found that the Grignard
reagent prepared in MTBE was completely precipitated
from solution by both THF and DME.18 Low yields of the
allylic alcohol 16 could be isolated by transition metal
catalyzed additions of 15 to the alkoxides of propargyl
alcohol and 2-chloropropenol (Scheme 5) while addition
to their THP ethers only gave complex mixtures.19
The addition of 15 to chloroacetone and dichloroacetone
in Scheme 6 proceeded in nearly quantitative yield to
furnish the chlorohydrins 17 and 18, respectively. Acid-
catalyzed elimination of tertiary alcohol 17 led to a 2:1
mixture of 19b and 19a in favor of the undesired vinyl
halide 19b.20 Conversion of 18 to the methanesulfonate
20a , acetate 20b, and trichloride 20c followed by reduc-
mately to the C-5-spirocenter of the spirobicyclic system
and then in turn to the tetrahydrofuran C-3 center. The
exploration of several borane reagents revealed BH3‚THF
gave the highest stereoselectivity (2:1) for the desired diol
5a (Table 1).
Attempts to remove the N-Boc group from the piperi-
dine ring of 4 in order to alter the conformation of the
system prior to hydroboration lead only to complex
mixtures due to the apparent acid lability of 4.10 Conver-
sion of the tertiary alcohol in 4 to various alkyl, acyl, and
silyl derivatives was investigated with the intent of
changing the steric environment in the vicinity of the
olefin and the ability of the tertiary alcohol to complex
with the borane. The best results were obtained with the
trimethylsilyl ether derivative; thus, 4 was converted to
the trimethylsilyl ether 9 (Scheme 3).15 Exposure of 9 to
excess BH3‚THF provided a 5:1 ratio of 5a and 5b after
silyl ether cleavage with TBAF (Table 1). The use of less
than 3 equiv of borane gave variable results (2:1 ratio of
5a and 5b and incomplete conversion). It was also found
that the silyl ether cleavage was faster with 5a than with
5b.16
(17) An improved protocol in which Et2Zn is slowly added to a
mixture containing ClH2I and 13 generates the more reactive ClCH2-
ZnCH2Cl rather than ClCH2ZnEt and generates the unstable carbenoid
slowly to allow it to react with 13, providing cyclopropane 14 in
virtually quantitative yield. Denmark, S. E.; Edwards, J . P. J . Org.
Chem. 1991, 56, 6974-6981.
(18) The Grignard reagent precipitated by THF was found to react
very sluggishly with ketones.
(19) (a) Organ, M. G.; Murray, A. P. J . Org. Chem. 1997, 62, 1523-
1526. (b) Negishi, E.-I.; Zhang, Y.; Cederbaum, F. E.; Webb, M. B. J .
Org. Chem. 1986, 51, 4080-4082. Dppp ) 1,3-Bis(diphenylphosphino)-
propane. The copper catalyzed addition of HC∞CCH2OMgBr to 15 gave
up to 8% of the product resulting from a second addition of HCtCCH2-
OMgBr.
Encouraged by these results, we turned our attention
to the preparation of the fully functionalized 2-aryl-
(15) The steric congestion in the vicinity of the tertiary alcohol in 4
made derivatization difficult. Acetylation of 4 (Ac2O, DMAP) provided
the corresponding acetate only as a minor product which gave complex
mixtures upon hydroboration with BH3‚THF. Alkylation of 4 with MeI,
Me2SO4, or BnBr in the presence of NaH or NaOH was sluggish. Milder
silylation protocols with TMSCl and various amine bases did not
convert 4 to 9. Treatment of 4 with TMSOTf and 2,4,6-collidine in THF
at 0 °C provided 9 in 75% yield.
(16) It was important to effect complete cleavage of the silyl ether
prior to determining the hydroboration selectivity, but this desilylation
rate difference could be taken advantage of to enrich the product
mixtures in 5a .
(20) Kuo, S.-C.; Hou, D.; Zhan, Z.-Y. US Pat. 5349099, 1994.

1096 J . Org. Chem., Vol. 67, No. 4, 2002
Maligres et al.
Sch em e 5
Sch em e 7
ties by conversion of 16 to its methanesulfonate 22
followed by nucleophilic displacement with bromide.
With the key retrosynthetic fragment 23 in hand, the
coupling was probed. Since (2R)-ketone 3 was available
in large quantities from a resolution7 of the racemate
while the (2S)-ketone was in short supply at the time of
this study, we opted to use the 2R antipode in our
investigations to ultimately prepare the antipodal NK-1
receptor antagonist (-)-1. Furthermore, only (+)-1 had
been synthesized thus far, and our work could provide
the yet to be prepared (-)-1 antipode while at the same
time potentially resulting in a new route to optically
active 1. Addition of the preformed organozinc derivative
of 23 to (2R)-37 gave the desired homoallylic alcohol 24a
albeit in moderate yields (allylic homocoupling) after
flash column chromatography.24 Addition of 23 to a
mixture of Rieke zinc and (R)-3 in THF gave somewhat
better results, but in all these cases HPLC analysis of
24a (Chiralpak AD, supercritical CO2/MeOH) indicated
extensive racemization (10-50%). Superior results were
obtained by gradual addition of 23 to a mixture of
ordinary zinc dust and (R)-3 (98%ee) in DMF at 10-15
°C (Scheme 8). This procedure gave a 99% yield of pure
24a (no detectable allylic homocoupling by HPLC analy-
sis) without the need of chromatographic purification and
<2% racemization. The mild conditions of the zinc-
mediated coupling had no effect on the cyclopropoxy
group. An X-ray diffraction study of 24a confirmed the
structure, and the absolute configuration of 24a was
confirmed by conversion to its p-bromobenzoate 24b
followed by an X-ray diffraction study.25 Furthermore, the
X-ray structure of 24a shown in Scheme 8 clearly
illustrates the axial orientation of the phenyl and allylic
substituents on the piperidine ring due to the steric bulk
of the N-Boc group.
Sch em e 6
tive elimination with zinc or magnesium resulted in dark
complex reaction mixtures.21
A much more efficient conversion of 14 to 16 was
realized in the sequence of steps shown in Scheme 7.22
The dichlorohydrin 18 was converted to the epichloro-
hydrin 21 with KOt-Bu and 21 was converted to 16 (80%
overall from 14) by treatment with LiI in an enolizable
ketone such as methyl isobutyl ketone (MIBK), diethyl
ketone, or cyclohexanone at 100 °C.23 Allylic bromide 23
was obtained from 16 in 95% yield either by treatment
with PPh3/Br2 or more conveniently in multigram quanti-
Hydroboration of 24a with BH3‚THF or BH3‚Me2S gave
a 2:1 diastereomeric mixture of diols 25a and 25b as in
(24) Reactions in which 23 was first converted to the organozinc
derivative did not proceed to completion (ca 50%). No reaction between
23 and magnesium in THF to form the Grignard reagent was observed.
(25) p-Bromobenzoate ester 24b was prepared by reaction of 24a
(0.20 mmol) with p-bromobenzoyl chloride (0.22 mmol) and Et3N (0.25
mmol) in CH2Cl2 (0.5 mL) at 20 °C for 15 min followed by aqueous
workup and purification by flash column chromatography (silica, 5%
MTBE in hexanes) to provide 24b (94%). Crystals suitable for X-ray
crystallography were obtained by crystallization of 24b from 2:1
MeOH-CHCl3. The X-ray structure of 24b shows the trifluoromethyl
group to be disordered. Thermal ellipsoids are drawn at the 30%
probability level in Scheme 8, and only selected hydrogen atoms are
shown for clarity. X-ray data was obtained on a Bruker AXS instru-
ment with data collection using a Bruker SMART program and data
reduction using a Bruker SHELXTL program. Structure solution and
refinement were performed with SHELXS-97 and SHELXL-97, re-
spectively.
(21) Hosokawa, A.; Tanaka, K. J pn. Pat. 09188637, 1997.
(22) (a) Blundell, P.; Ganguly, A. K.; Girijavallabhan, V. M. Synlett
1994, 4, 263-265. (b) Attempted reductive elimination of 21 with Te,
Na2Te, Na2Se, or Na2S resulted in sluggish reactions. Polson, G.;
Dittmer, D. C. Tetrahedron Lett. 1986, 27, 5579-5582.
(23) Reaction is sluggish below 90 °C, i.e., when methyl ethyl ketone
at reflux is used. The use of other iodine scavengers such as phenol,
zinc dust, or magnesium powder instead of a ketone gave no detectable
product 16.

Nonpeptidic (-)-Spirobicyclic NK-1 Receptor Antagonist
J . Org. Chem., Vol. 67, No. 4, 2002 1097
Sch em e 8
Sch em e 9
Sch em e 10
Ta ble 2. Hyd r obor a tion of 24a , 24b, 26a , a n d 26b
substrate
conditionsa
ratio 5a /5bb
24a
24a
24a
24a
24b
26a
26a
26a
26a
26a
26a
26a
26b
BH3‚THF, 25 °C
67/33
76/24
71/29
78/22
62/38c
80/20
90/10
87/13
64/36
72/28
60/40
70/30
67/33
BH3‚THF, -10 °C
9-BBN/THF, 25 °C
9-BBN/THF, 0 °C
BH3‚THF, 25 °C
BH3‚THF, 25 °C
BH3‚THF, -10 °C
BH3‚Me2S/THF, 0 °C
1.5 equiv of BH3‚Me2S/THF, 0 °C
9-BBN/THF, 25 °C
(+)-IPC2BH/THF, 20 °C
(-)-IPC2BH/THF, 20 °C
BH3‚THF, 20 °C
hydroboration of 26a with BH3‚THF at -10 °C which
gave a 9:1 diastereomeric mixture of diols 25a and 25b
after desilylation. Interestingly, the use of excess borane
was critical to maintaining a high diastereoselectivity in
the hydroboration of 26a as was observed with the
hydroboration of model 9. The desilylation of the major
diastereomer 27a was faster than 27b as observed in the
model case. Consequently, cautious silyl ether cleavage
with HF in aqueous MeCN provided diol 25a in 80% yield
containing <1% of the minor diastereomer 25b after
simple chromatographic purification.
a
The borane (5 equiv) was added to the substrate (0.1 mmol)
in the specified solvent (0.3 mL) at -10 °C, and the mixture was
raised to the specified temperature for 18 h unless otherwise
stated. bRatios were determined by HPLC analysis after quenching
with 0.3 mL 2:1 5 M aq NaOH/30% H2O2. In the case of 26a and
26b the quenched reaction mixtures were treated with 0.3 mL of
c
1 M TBAF in THF. The PBB ester was partly cleaved (ca. 25%,
HPLC) during the hydroboration (reduction of ester to 4-bromo-
benzyl alcohol) and completely cleaved by the aq NaOH/H2O2
quench.
the model system. The use of other boranes and solvents
did not appreciably affect the diastereoselectivity of the
hydroboration (Table 2). Silylation of 24a proved to be
more difficult than with the unsubstituted model 4.
Treatment of 24a with silyl triflates in the presence of
bases such as 2,4,6-collidine, DIEA, and LiHMDS gave
mixtures of products, but treatment with TMSCl in the
presence of LiHMDS provided a high yield of the steri-
cally congested silyl ether 26a (Scheme 9). The triethyl-
silyl ether 26b could also be prepared by this method but
in somewhat lower yield (70%) due to a lower conversion
rate. The most favorable results were obtained from the
Conversion to the spirobicyclic system 29a was ac-
complished in 88% yield (Scheme 10) via cyclization of
the methanesulfonate derivative 28a with NaHMDS in
THF at -20 °C. A moderate diastereoselectivity was
observed for the cyclization, favoring the cyclization of
28a over 28b, thus affording a further opportunity for
rejection of the undesired diol 25b. The major diastereo-
mer 29a was found to be identical to an authentic sample
of 29a (NMR and HPLC analysis) except it had the
opposite optical configuration (chiral HPLC and optical
rotation).3b Cleavage of the N-Boc protecting group
provided 1 in 95% yield which was also found to be

1098 J . Org. Chem., Vol. 67, No. 4, 2002
Maligres et al.
was extracted with MTBE (500 mL). The organic phase (upper
layer) was washed with water (500 mL), 3% aq Na2SO3 (500
mL), and saturated aq NaCl (200 mL). A single portion of
MTBE (100 mL) was used to sequentially extract the aqueous
phases from the acidified reaction, water wash, 3% Na2SO3
wash, and saturated NaCl wash from the aforementioned
extractions, respectively. The combined organic extracts were
dried over MgSO4, treated with Darco G-60 (1 g) for 30 min at
20 °C, and filtered through a pad composed of silica (5 g)
layered over neutral alumina (5 g), using MTBE (60 mL) to
wash the pad. The filtrate was concentrated via distillation
through a Snyder fractioning column (100 °C bath, 760 f 10
Torr). The residue was fractionally vacuum distilled at 1 Torr
and three fractions were collected: 6.6 g (bp 70-80 °C,
4-trifluoromethoxyphenol 14.9 area%, 11 84.2 area%, D <0.1
area%, 4% yield) and 146.2 g (bp 80-85 °C, 4-trifluoromethoxy-
phenol 0.6 area%, 11 99.1 area%, D <0.1 area%, 80% yield).
The main fraction #2 was obtained as a yellow oil which
solidified at 15-20 °C. Caution: persistent stench. HPLC
retention times: 4-trifluoromethoxyphenol ) 2.7 min, 11 )
3.7 min, 2,6-diiodo-4-trifluoromethoxyphenol D ) 5.0 min.
HPLC conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5 µm
column, gradient elution with 60 to 80% MeCN in 0.025% aq
H3PO4 over 7 min then isocratic elution with 80% MeCN in
0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with detection
identical to an authentic sample but with the expected
opposite optical configuration.3b The optical purity of
(-)-1 generated by this synthetic route was found to be
97% ee by HPLC analysis after derivatization with
Marfey’s reagent.26
Con clu sion
In conclusion, a new convergent synthesis of 1 has been
identified allowing its rapid preparation from readily
available raw materials. This route provides 1 in 40%
overall yield from commercially available 4-trifluoro-
methoxyphenol and requires only three chromatographic
purifications. The asymmetry of ketopiperidine 3 is
transmitted to the adjacent asymmetric center via a
highly diastereoselective zinc mediated coupling of an
allylic bromide 23 to 3. This asymmetry is in turn
transmitted to the C-3 center via a diastereoselective
hydroboration followed by cyclization to the spirobicyclic
system. The diastereoselectivity of the hydroboration and
the selective reactivity for the desired diastereomers
observed in the subsequent conversions allow the enrich-
ment of the final product to >99% diastereomeric purity
with minimal effort in purification. Finally, this envi-
ronmentally friendly synthesis requires no heavy metal
compounds or cryogenic temperatures, produces little
waste, and requires few isolations.
1
at 220 nm. 11: Rf ) 0.25 (4:1 hexane-MTBE); H NMR (400
MHz, CDCl3) δ 7.54 (dd, J ) 2.8, 0.8, 1 H), 7.16 (ddd, J ) 8.9,
2.8, 0.8, 1 H), 7.00 (d, J ) 8.9, 1 H), 5.34 (s, 1 H); 13C NMR
(100 MHz, CDCl3) δ 154.0, 142.6, 130.9, 123.3, 120.5 (q, J )
258, OCF3), 115.2, 84.6; FTIR (thin film) νmax 3492, 1482, 1259,
1218, 1194, 1173. Anal. Calcd for C7H4F3IO2: C, 27.66; H, 1.33;
F, 18.75; I, 41.74. Found: C, 27.39; H, 1.18; F, 18.78; I, 41.87.
1-(2-Ch lor oe t h oxy)-2-iod o-4-t r iflu or om e t h oxyb e n -
zen e (12). A mixture of phenol 11 (136.8 g, 0.450 mol),
2-chloroethyl p-toluenesulfonate (107.7 g, 0.459 mol), DMF
(450 mL), and K2CO3 (124.4 g, 0.900 mol) was aged for 16 h
at 50 °C. Water (900 mL), MTBE (300 mL), and hexanes (150
mL) were added, and the mixture was stirred for 5 min and
was allowed to settle for 5 min. The organic phase was washed
with water (900 mL). The aqueous phase was back-extracted
with a mixture of MTBE (100 mL) and hexanes (50 mL). The
organic back-extract was washed with the 900 mL water that
was used to wash the first organic extract. The two organic
phases were combined, washed with saturated aq NaCl, dried
over MgSO4, and filtered through a pad of silica (20 g), washing
the pad with a mixture of MTBE (40 mL) and hexanes (20
mL). The combined filtrates were evaporated on the rotary
evaporator to give crude chloroethyl ether 12 as a pale yellow
oil (165 g) which solidified after several days to a pale yellow
crystalline solid. HPLC profile of crude 12: 3.3 area% 2-
chloroethyl p-toluenesulfonate, 82.3 area% 12, 12.6 area%
dimeric byproduct E. Retention times: 11 ) 3.7 min, 12 )
6.5 min, 2-chloroethyl p-toluenesulfonate ) 3.1 min, dimeric
byproduct E ) 12.6 min. HPLC conditions: 150 × 4.6 mm HP
Eclipse XDB-C8 5 µm column, gradient elution with 60 f
80% MeCN in 0.025% aq H3PO4 over 7 min then isocratic
elution with 80% MeCN in 0.025% aq H3PO4 for 5 min at 1
mL/min, 25 °C with detection at 220 nm. 12: mp ) 41-42 °C;
Exp er im en ta l Section
Gen er a l. Reagents were used as received unless otherwise
stated; 3 Å molecular sieves were used to dry solvents for
anhydrous reactions. Unless otherwise noted, all manipula-
tions were carried out under an inert atmosphere of nitrogen
gas. In general, glassware was not specially dried prior to use.
High-pressure liquid chromatography (HPLC) analysis was
performed with a Hewlett-Packard series 1100 HPLC system.
Analytical TLC was performed using Merck Kieselgel G60 F254
precoated plates (0.25 mm) followed by visualization with UV
light (254 nm), staining with iodine vapor or staining with a
solution of 14% ammonium molybdate and 0.5% ceric sulfate
in 10% aqueous sulfuric acid and then heat. Flash column
chromatography was performed using silica gel (Merck, 70-
230 mesh ASTM). Optical rotations were obtained with a
Perkin-Elmer 241 polarimeter. Melting points were obtained
with a Barnstead/Thermolyne MEL-TEMP apparatus and are
1
uncorrected. H and 13C NMR chemical shifts are reported in
ppm; coupling constants are reported in hertz. 1H and 13C
spectra were recorded on Bruker AM and AMX systems. IR
spectra were recorded on a Nicolet Magna-IR 550 spectropho-
tometer. Elemental analyses were obtained from Quantitative
Technologies Inc, Whitehouse, NJ .
2-Iod o-4-tr iflu or om eth oxyp h en ol (11). A homogeneous
solution of iodine (175.1 g, 0.690 mol) and NaI (269.8 g, 1.80
mol) in water (600 mL) was added to a solution of 4-trifluoro-
methoxyphenol (106.9 g, 0.600 mol) and 40% aq Me2NH (412
mL, 3.30 mol) over 45 min maintaining the temperature below
0 °C. The mixture was aged for 64 h at -5 to 0 °C.
1
Rf ) 0.36 (4:1 hexane-MTBE); H NMR (400 MHz, CDCl3) δ
7.68 (dd, J ) 2.7, 0.8, 1 H), 7.20 (ddd, J ) 9.0, 2.7, 0.8, 1 H),
6.81 (d, J ) 9.0, 1 H), 4.27 (t, J ) 6.0, 2 H), 3.87 (t, J ) 6.0, 2
H); 13C NMR (100 MHz, CDCl3) δ 156.0, 143.4, 132.2, 122.2,
120.5 (q, J ) 258, OCF3), 112.6, 86.4, 70.0, 41.4; FTIR (thin
film) νmax 1485, 1251, 1218, 1192, 1170 cm-1. E (isolated below
1
in the preparation of 13): mp 95-97 °C; H NMR (250 MHz,
CDCl3) δ 7.66 (d, J ) 2.6, 2 H), 7.21 (dd, J ) 9.0, 2.6, 2 H),
(26) Derivatization of 1 with Marfey’s reagent (N-R-(2,4-dinitro-5-
fluorophenyl)-L-alaninamide) was performed by heating a mixture of
1 mg of 1, 2 mg of Marfey’s reagent, 0.1 mL of 5% aq NaHCO3, and 1
mL of MeCN at 55 °C for 5 min. The resulting mixture was then
analyzed directly by HPLC. (a) Marfey, P. Carlsberg Res. Commun.
1984, 49, 591. (b) Bru¨ckner, H.; Keller-Hoehl, C. Chromatographia
1990, 30, 621. (c) Pete´r, A.; Olajos, E.; Casimir, R.; Tourwe´, D.;
Broxterman, G. B.; Kaptein, B.; Armstrong, D. W. J . Chromatogr. A
2000, 871, 105-113.
The mixture was filtered at 5 °C, and the filter cake
containing D‚Me2NH salt was washed with 100 mL water at
5 °C. The filtrate was acidified to pH ) 5-6 by the addition of
concd HCl (45 mL) over 10 min with cooling to maintain the
temperature below 30 °C. The resulting two-phase mixture

Nonpeptidic (-)-Spirobicyclic NK-1 Receptor Antagonist
J . Org. Chem., Vol. 67, No. 4, 2002 1099
6.95 (d, J ) 9.0, 2 H), 4.43 (s, 4 H); 13C NMR (62.9 MHz, CDCl3)
δ 156.3, 143.2, 132.5, 122.3, 120.4 (q, J ) 257, OCF3), 112.9
column, gradient elution with 60 f 80% MeCN in 0.025% aq
H3PO4 over 7 min then isocratic elution with 80% MeCN in
0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with detection
86.4, 68.5; FTIR (thin film) νmax 1492, 1256, 1213, 1178 cm-1
.
1
Anal. Calcd for C16H10F6I2O4: C, 30.31; H, 1.59; F, 17.98; I,
40.03. Found: C, 30.45; H, 1.44; F, 17.80; I, 40.14.
at 220 nm. 14: Rf ) 0.43 (9:1 hexane-MTBE); H NMR (250
MHz, CDCl3) δ 7.62 (d, J ) 2.3, 1 H), 7.18 (m, 2 H), 3.78 (m,
1 H), 0.84 (m, 4 H); 13C NMR (62.9 MHz, CDCl3) δ 156.7, 142.9,
132.2, 122.1, 120.2 (q, J ) 270, OCF3), 112.8, 85.1, 52.5, 6.4;
FTIR (thin film) νmax 1481, 1252, 1190, 1036, 810 cm-1. Anal.
Calcd for C10H8F3IO2: C, 34.91; H, 2.34; F, 16.56; I, 36.88.
Found: C, 34.86; H, 2.23; F, 16.51; I, 37.10.
Dich lor oh yd r in 18. A mixture of 14 (10.32 g, 30.0 mmol),
magnesium turnings (0.766 g, 31.5 mmol) and anhydrous
MTBE (60 mL) were stirred for 3 h while moderating the
exotherm to a gentle reflux with a water bath. The mixture
was cooled to -20 °C, and a solution of 1,3-dichloroacetone
(4.00 g, 31.5 mmol) in anhydrous MTBE (15 mL) was added
over 5 min. The mixture was allowed to warm to 5 °C over 20
min and was quenched with saturated aq NH4Cl (30 mL) and
water (3 mL). The biphasic mixture was filtered from unre-
acted magnesium through a cotton plug using pentane (50 mL)
to complete the transfer. The organic phase was separated
from the filtrate, dried (MgSO4), treated with Darco G60 (50
mg, 15 min), and filtered through a plug composed of silica (5
g) layered over neutral alumina (5 g). The plug was washed
with 1:1 hexanes-MTBE (80 mL), and the combined filtrates
were evaporated to provide the crude dichlorohydrin 18 (10.48
g). HPLC retention times: 18 ) 5.5 min, 4-cyclopropoxy-1-
trifluoromethoxybenzene F ) 6.0 min. HPLC conditions: 150
× 4.6 mm HP Eclipse XDB-C8 5 µm column, gradient elution
with 60 f 80% MeCN in 0.025% aq H3PO4 over 7 min then
isocratic elution with 80% MeCN in 0.025% aq H3PO4 for 5
min at 1 mL/min, 25 °C with detection at 220 nm. 18: Rf )
1-(Vin yloxy)-2-iod o-4-tr iflu or om eth oxyben zen e (13).
Solid KOt-Bu (62.7 g, 0.559 mol) was added over 5 min to a
solution of 12 (164 g, 0.447 mol crude from above alkylation)
in THF (675 mL), maintaining the temperature below 0 °C,
and the mixture was aged at 20 °C for 20 h. Water (1.2 L) was
added to the mixture which was transferred to a 3 L separatory
funnel and was extracted with a mixture of hexanes (600 mL)
and MTBE (300 mL). The organic phase was washed with
water (1 L). The aqueous phase was back-extracted with 2:1
hexanes-MTBE (180 mL). The organic back-extract was
washed with the 1 L water that was used to wash the first
organic extract. The two organic phases were combined and
washed with saturated aq NaCl (200 mL). The organic phase
was dried over K2CO3 and filtered through a pad composed of
silica (20 g) layered over neutral alumina (10 g), using 2:1
hexanes-MTBE (100 mL) to wash the pad. The filtrate was
concentrated via distillation through a Snyder fractioning
column (100 °C bath, 760 f 10 Torr). The residue was diluted
with 2-methylbutane (100 mL) and aged at -20 °C for 18 h.
The precipitated crystalline dimeric byproduct E (14 g) was
filtered off, and the filtrate was concentrated through a Snyder
column as above. The residue was flash vacuum distilled
through a vacuum jacketed Vigreux column (0.20-0.25 Torr,
120-130 °C bath, 80-90 °C still head) into an ice-cooled
receiver. Vinyl ether 13 was collected as a colorless liquid (117
g, 80% from 2, >98 area% purity by HPLC analysis). TLC
retention factors: 11 ) 0.38, 12 ) 0.53, 13 ) 0.90, B ) 0.47,
2-chloroethyl toluenesulfonate ) 0.18 (2:1 hexanes-CH2Cl2,
silica). HPLC retention times: 12 ) 6.5 min, 13 ) 6.8 min,
dimeric byproduct E ) 12.0 min (vide supra). HPLC condi-
tions: 150 × 4.6 mm HP Eclipse XDB-C8 5 µm column,
gradient elution with 60 f 80% MeCN in 0.025% aq H3PO4
over 7 min then isocratic elution with 80% MeCN in 0.025%
aq H3PO4 for 5 min at 1 mL/min, 25 °C with detection at 220
1
0.39 (4:1 hexane-MTBE); H NMR (250 MHz, CDCl3) δ 7.52
(d, J ) 2.8, 1 H), 7.29 (d, J ) 8.9, 1 H), 7.20 (m, 1 H), 4.06 (dd,
J ) 23.2, 11.3, 4 H), 3.79 (m, 1 H), 3.25 (s, 1 H), 0.83 (m, 4H);
13C NMR (62.9 MHz, CDCl3) δ 153.5, 143.3, 129.2, 122.2, 122.1,
120.5 (q, J ) 257, OCF3), 113.9, 75.6, 51.7, 49.3, 6.6; FTIR
(thin film) νmax 1485, 1260, 1243, 1221, 1185, 1160 cm-1
.
Ep ich lor oh yd r in 21. A solution of dichlorohydrin 18 (10.13
g, 29.0 mmol) in THF (29 mL) was cooled to -20 °C, and
KOtBu (3.58 g, 31.9 mmol) was added as a solid over 2 min.
The mixture was allowed to warm to 20 °C over 20 min. MTBE
(60 mL), hexanes (60 mL), water (30 mL), and brine (30 mL)
were added, and the mixture was extracted. The organic
extract was washed with brine (60 mL), dried (MgSO4), treated
with Darco G60 (200 mg, 15 min), and filtered through a plug
composed of silica (5 g) layered over neutral alumina (5 g).
The plug was washed with 1:1 MTBE-hexanes (50 mL) and
the combined filtrates were evaporated to provide the crude
epichlorohydrin 21 (8.95 g). HPLC retention time: 21 ) 6.3
min. HPLC conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5
µm column, gradient elution with 60 f 80% MeCN in 0.025%
aq H3PO4 over 7 min then isocratic elution with 80% MeCN
in 0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with
1
nm. 13: Rf ) 0.53 (9:1 hexane-MTBE); H NMR (250 MHz,
CDCl3) δ 7.68 (m, 1 H), 7.20 (m, 1 H), 6.96 (d, J ) 8.8, 1 H),
6.54 (dd, J ) 13.6, 6.0, 1 H), 4.80 (dd, J ) 13.6, 2.0, 1 H), 4.57
(dd, J ) 6.0, 2.0, 1 H); 13C NMR (62.9 MHz, CDCl3) δ 154.6,
147.7, 144.7, 132.3, 122.3, 120.4 (q, J ) 258, OCF3), 117.1, 96.8,
87.0; FTIR (thin film) νmax 1644, 1475, 1252, 1218, 1191, 1167
cm-1. Anal. Calcd for C9H6F3IO2: C, 32.75; H, 1.83; F, 17.27;
I, 38.45. Found: C, 32.78; H, 1.70; F, 16.89; I, 38.58.
1
detection at 220 nm. 21: Rf ) 0.43 (4:1 hexane-MTBE); H
NMR (250 MHz, CDCl3) δ 7.28-7.14 (m, 3 H), 4.14 (d, J )
11.8, 1 H), 3.79 (m, 1 H), 3.61 (d, J ) 11.8, 1 H), 3.15 (d, J )
5.1, 1 H), 2.81 (d, J ) 5.1, 1 H), 0.81 (m, 4H); 13C NMR (62.9
MHz, CDCl3) δ 154.9, 142.7, 126.5, 122.2, 122.1, 120.5 (q, J )
257, OCF3), 113.1, 57.8, 53.8, 51.5, 47.7, 6.5; FTIR (thin film)
1-(Cyclop r op oxy)-2-iod o-4-t r iflu or om et h oxyb en zen e
(14). Via addition funnel, Et2Zn (59.3 g, 0.480 mol) was added
to a mixture of 13 (99.0 g, 0.300 mol), ClCH2I (169.3 g, 0.960
mol), and ClCH2CH2Cl (450 mL) over 2 h, maintaining the
temperature at -5 to 0 °C. The mixture was warmed to 20 °C
and aged at 20-25 °C for 15 min. The mixture was cooled back
to 0 °C, and saturated aq NH4Cl (600 mL), concd aq NH4OH
(100 mL), and hexanes (600 mL) were added. The organic
phase was washed with saturated aq NH4Cl (100 mL). The
aqueous phase was back-extracted with hexanes (100 mL). The
organic back-extract was washed with the 100 mL of saturated
aq NH4Cl that was used to wash the first organic extract. The
two organic phases were combined, dried over K2CO3, and
filtered through a pad composed of silica (20 g) layered over
neutral alumina (10 g), using 99:1 hexanes-MTBE (200 mL)
to wash the pad. The filtrate was evaporated to a constant
weight of 103.2 g (>99% yield, >99 area% purity by HPLC
analysis). HPLC retention times: 13 ) 6.8 min, 14 ) 7.5 min.
HPLC conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5 µm
νmax 1494, 1356, 1247, 1218, 1185, 1162, 972 cm-1
.
Allylic Alcoh ol 16. A mixture of epichlorohydrin 21 (7.72
g, 25.0 mmol), LiI (5.02 g, 37.5 mmol), and 3-pentanone (10
mL) was heated to 100 °C for 30 min. The mixture was cooled
to 25 °C diluted with 1:1 MTBE-hexanes (120 mL) and
filtered. The filtrate was washed with saturated aq NH4Cl (60
mL), water (60 mL), 39% aq NaHSO3 (2 × 60 mL), water (60
mL), and brine (60 mL). The organic phase was dried (MgSO4),
treated with Darco G60 (200 mg, 15 min), and filtered through
a plug composed of silica (5 g) layered over neutral alumina
(5 g). The plug was washed with 1:1 MTBE-hexanes (50 mL),
and the combined filtrates were evaporated. The residue was
purified by flash column chromatography (silica, 30% MTBE
in hexanes) to provide allylic alcohol 16 (5.48 g, 80% from 14).
HPLC retention time: 16 ) 4.4 min. HPLC conditions: 150

1100 J . Org. Chem., Vol. 67, No. 4, 2002
Maligres et al.
× 4.6 mm HP Eclipse XDB-C8 5 µm column, gradient elution
with 60 f 80% MeCN in 0.025% aq H3PO4 over 7 min then
isocratic elution with 80% MeCN in 0.025% aq H3PO4 for 5
min at 1 mL/min, 25 °C with detection at 220 nm. 16: Rf )
20 min, maintaining the temperature <16 °C and using DMF
(2 mL) to complete the transfer. The mixture was aged at 15
°C for 1 h, diluted with water, saturated aq NH4Cl, MTBE,
and hexanes (15 mL each), and filtered from any remaining
zinc through a cotton plug. The organic phase was separated
from the filtrate and was washed with a mixture of water (15
mL) and saturated aq NH4Cl (15 mL) and then with water
(40 mL). The aqueous phases were extracted with 1:1 MTBE-
hexanes (20 mL). The combined organic phases were dried
(MgSO4) and evaporated. The residue (97 area% purity by
HPLC and NMR analysis) was purified by flash column
chromatography (silica, 25% MTBE in hexanes) to provide
homoallylic alcohol 24a as a white crystalline solid (7.84 g,
99%, >99 area% purity by HPLC and NMR analysis). Crystals
suitable for X-ray diffraction analysis was prepared by recrys-
tallization of a sample of 24a from hexane. HPLC retention
times: 24a ) 9.7 min, (R)-Boc-2-phenyl-3-piperidone (3) ) 4.7
min. HPLC conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5
µm column, gradient elution with 60 f 80% MeCN in 0.025%
aq H3PO4 over 7 min then isocratic elution with 80% MeCN
in 0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with
detection at 220 nm. Chiral supercritical phase chromatogra-
phy (SFC) indicated 24a (after chromatography but not
crystallized) to be >96%ee. SFC conditions: Chiralpak AD 250
× 4.6 mm column, isocratic elution, 2% MeOH in supercritical
CO2 for 2 min then gradient elution with 2 f 10% MeOH in
supercritical CO2 over 8 min at 250 bar, 1.5 mL/min flow at
50 °C with detection at 220 nm. SFC Retention times: (2R,3S)-
24a ) 8.3 min, (2S,3R)-24a ) 9.4 min. 24a : mp 78-80 °C; Rf
) 0.23 (4:1 hexane-MTBE); 1H NMR (250 MHz, CDCl3) δ
7.41-7.07 (m, 8 H), 5.32 (d, J ) 1.7, 1 H), 5.24 (d, J ) 1.7, 1
H), 4.62 (s, 1 H), 4.03 (dt, J ) 12.9, 3.3, 1 H), 3.72 (m, 1 H),
3.08 (m, 1 H), 2.94 (dd, J ) 17.5, 14.1, 2 H), 1.81 (m, 1 H),
1.59 (m, 3 H), 1.37 (s, 9 H), 0.74 (m, 4H); 13C NMR (62.9 MHz,
CDCl3) δ 155.4, 154.1, 142.9, 142.7, 139.3, 133.8, 129.3, 128.1,
127.2, 122.9, 121.6, 120.9, 120.6 (q, J ) 254, OCF3), 113.5, 79.9,
72.7, 63.3, 51.5, 45.1, 39.0, 31.2, 28.4, 21.6, 6.4; FTIR (thin
film) νmax 3455, 1688, 1492, 1416, 1245, 1214, 1162 cm-1. Anal.
Calcd for C29H34F3NO5: C, 65.28; H, 6.42; N, 2.63. Found: C,
65.11; H, 6.36; N, 2.51.
1
0.53 (1:1 hexane-MTBE); H NMR (250 MHz, CDCl3) δ 7.24
(d, J ) 9.9, 1 H), 7.11 (m, 2 H), 5.43 (dd, J ) 2.7, 1.3, 1 H),
5.23 (d, J ) 1.3, 1 H), 4.36 (s, 2 H), 3.74 (m, 1 H), 2.18 (br s,
1 H), 0.80 (m, 4H); 13C NMR (62.9 MHz, CDCl3) δ 154.4, 146.8,
142.9, 130.5, 123.2, 121.2, 120.5 (q, J ) 256, OCF3), 116.2,
113.4, 65.2, 51.5, 6.3; FTIR (thin film) νmax 3375, 1492, 1246,
1220, 1185, 1160 cm-1
.
Allylic Meth a n esu lfon a te 22. Methanesulfonyl chloride
(2.41 g, 21.0 mmol) was added over 15 min to a mixture of
allylic alcohol 16 (5.48 g, 20.0 mmol) and DIEA (2.84 g, 22.0
mmol) in CH2Cl2 (20 mL), maintaining the temperature below
-40 °C. The mixture was warmed to 0 °C over 15 min, and
water (35 mL), pentane (30 mL), and ether (30 mL) were
added. The organic phase was washed with 5% aq HBr (2 ×
25 mL), water (25 mL), and saturated aq Na2SO4 (25 mL). The
organic phase was dried (MgSO4), treated with Darco G60 (200
mg, 15 min), and filtered through a plug composed of silica (5
g) layered over neutral alumina (5 g). The plug was washed
with 1:1 pentane-ether (50 mL), and the combined filtrates
were evaporated to provide the crude allylic methanesulfonate
22 (7.05 g). HPLC retention time: 22 ) 5.9 min. HPLC
conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5 µm column,
gradient elution with 60 f 80% MeCN in 0.025% aq H3PO4
over 7 min and then isocratic elution with 80% MeCN in
0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with detection
1
at 220 nm. 22: Rf ) 0.50 (1:1 hexane-MTBE); H NMR (400
MHz, CDCl3) δ 7.27-7.17 (m, 2 H), 7.09 (m, 1 H), 5.55 (d, J )
0.6, 1 H), 5.40 (s, 1 H), 5.06 (s, 2 H), 3.76 (m, 1 H), 2.90 (s, 3
H), 0.82 (m, 4H); 13C NMR (100 MHz, CDCl3) δ 154.7, 142.9,
140.6, 128.3, 123.5, 122.1, 120.6 (q, J ) 256, OCF3), 120.0,
113.6, 71.1, 51.7, 37.9, 6.4; FTIR (thin film) νmax 1492, 1358,
1248, 1218, 1176, 971 cm-1
.
Allylic Br om id e 23. A mixture of crude allylic methane-
sulfonate 22 (7.05 g, 20.0 mmol) and LiBr (3.47 g, 40.0 mmol)
in DMF (20 mL) was stirred for 16 h at 10 °C and was diluted
with water (50 mL), pentane (49 mL), and ether (1 mL). The
organic phase was separated and washed with 0.5% aq Na2-
SO4 (2 × 50 mL) and saturated aq Na2SO4 (25 mL). The
organic phase was dried (MgSO4), treated with Darco G60 (200
mg, 15 min), and filtered through a plug composed of silica (5
g) layered over neutral alumina (5 g). The plug was washed
with 29:1 pentane-ether (50 mL), and the combined filtrates
were evaporated to provide the crude allylic bromide 23 (6.47
g, 96% from 16). HPLC retention times: 23 ) 7.7 min. HPLC
conditions: 150 × 4.6 mm HP Eclipse XDB-C8 5 µm column,
gradient elution with 60 f 80% MeCN in 0.025% aq H3PO4
over 7 min and then isocratic elution with 80% MeCN in
0.025% aq H3PO4 for 5 min at 1 mL/min, 25 °C with detection
at 220 nm. 23: Rf ) 0.40 (19:1 hexane-MTBE); 1H NMR (250
MHz, CDCl3) δ 7.24 (d, J ) 9.0, 1 H), 7.16 (m, 2 H), 5.50 (d, J
) 1.0, 1 H), 5.24 (d, J ) 1.0, 1 H), 4.37 (s, 2 H), 3.75 (m, 1 H),
0.80 (m, 4H); 13C NMR (100 MHz, CDCl3) δ 154.6, 143.7, 142.8,
129.4, 123.8, 121.7, 120.6 (q, J ) 256, OCF3), 120.1, 113.4, 51.6,
35.2, 6.4; FTIR (thin film) νmax 1492, 1245, 1221, 1190, 1160
TMS Eth er 26a . Chlorotrimethylsilane (TMSCl, 1.01 mL,
0.869 g, 8.00 mmol) and a 1 M solution of LiHMDS (8.0 mL,
8.0 mmol) in THF were added simultaneously over 30 min to
a solution of 24a (2.13 g, 4.00 mmol) in THF (8 mL) maintained
at 20 °C, and the mixture was aged for 1 h. The mixture was
evaporated to half its volume, diluted with water (40 mL), and
extracted with hexanes (40 mL). The organic extract was
washed with aq NaH2PO4 (2 × 40 mL) and filtered through a
plug of 2:1 by wt silica gel-MgSO4 (5 g). The plug was washed
with 2:1 hexanes-MTBE (20 mL), and the combined filtrates
were evaporated to provide TMS ether 26a (2.30 g, 95%).
HPLC retention time: 26a ) 11.3 min, HPLC conditions: 150
× 4.6 mm ACE 3 C18 column, isocratic elution with 90%
MeCN in 0.025% aq H3PO4 at 1 mL/min, 25 °C with detection
at 220 nm. 26a : Rf ) 0.31 (4:1 hexane-MTBE); 1H NMR (400
MHz, CDCl3) δ 7.27-7.11 (m, 8 H), 5.29 (d, J ) 2.0, 1 H), 5.25
(d, J ) 2.0, 1 H), 5.10 (s, 1 H), 4.13 (dd, J ) 13.4, 3.6, 1 H),
3.73 (m, 1 H), 3.39 (m, 1 H), 3.04 (d, J ) 13.7, 1 H), 2.80 (d, J
) 13.7, 1 H), 1.89-1.74 (m, 3 H), 1.63 (m, 1 H), 1.27 (s, 9 H),
0.80 (m, 4 H), -0.44 (s, 9 H); 13C NMR (100 MHz, CDCl3) δ
155.8, 154.2, 143.9, 142.8, 141.2, 134.8, 129.4, 127.1, 126.3,
123.1, 121.5, 120.6 (q, J ) 256, OCF3), 120.2, 113.1, 79.5, 76.4,
61.4, 51.3, 44.9, 40.0, 32.1, 28.2, 21.4, 6.4, 6.2, 1.7; FTIR (thin
cm-1
.
Meth od 2. Bromine (352 mg, 2.20 mmol) was added to a
mixture of PPh3 (629 mg, 2.40 mmol) and CH2Cl2 (3 mL) over
10 min, maintaining the temperature below -10 °C. Allylic
alcohol 16 (548 mg, 2.00 mmol) in CH2Cl2 (1 mL) was added,
and the mixture was allowed to warm to 10 °C over 20 min.
The mixture was diluted with pentane (25 mL) and filtered
through a plug composed of silica (5 g) layered over neutral
alumina (5 g). The plug was washed with 29:1 pentane-ether
(30 mL), and the combined filtrates were evaporated to provide
the crude allylic bromide 23 (641 mg, 95% from 16).
Hom oa llylic Alcoh ol 24a . TMSCl (0.050 mL) was added
to a mixture of (R)-piperidine 3 (4.08 g, 14.83 mmol), zinc dust
(1.16 g, 17.8 mmol), and DMF (13 mL), the mixture was aged
for 10 min at 40 °C, dibromoethane (0.050 mL) was added,
and the mixture was again aged for 10 min at 40 °C. Allylic
bromide 23 (5.10 g, 15.1 mmol) was added to the mixture over
film) νmax 1693, 1362, 1245, 1164, 978, 837 cm-1
.
Diol 25a . A solution of 26a (1.82 g, 3.00 mmol) in THF (6
mL) was added over 15 min to a 1 M solution of borane in
THF (15.0 mL, 15.0 mmol) maintained at -10 °C. The mixture
was allowed to gradually warm to 20 °C over 16 h and
evaporated to dryness. The residue was dissolved in THF (6
mL), and a mixture of 2 M aq NaOH (10 mL) and 30% H2O2
(5 mL) was added over 10 min, maintaining the temperature
at 10 °C. The mixture was aged 1 h at 20 °C and 15 min at 45
°C. The mixture was cooled to 20 °C, diluted with water (15
mL), and extracted with MTBE (2 × 20 mL). The combined
organic extracts were washed with water (10 mL), 10% aq Na2-

Nonpeptidic (-)-Spirobicyclic NK-1 Receptor Antagonist
J . Org. Chem., Vol. 67, No. 4, 2002 1101
SO3 (10 mL), and brine (10 mL). The organic extract was
filtered through a plug of 2:1 by wt silica gel-MgSO4 (5 g).
The plug was washed with 2:1 hexanes-MTBE (20 mL), and
the combined filtrates were evaporated. HPLC analysis of the
residue indicated the residue (1.68 g) to be a 6:1:3 mixture of
27a , 27b, and 25a (25b was not detected). HPLC retention
times: 27a ) 4.0 min, 27b ) 4.2 min, 25a ) 13.2 min, 25b )
14.0 min, HPLC conditions: 150 × 4.6 mm ACE 3 C18 column,
isocratic elution with 80% MeCN in 0.025% aq H3PO4 at 1 mL/
min, 25 °C with detection at 220 nm. The crude residue was
dissolved in MeCN (10 mL), and 48% aqueous HF (0.20 mL)
was added at 20 °C. The mixture was aged at 20 °C for 3 h.
MTBE (10 mL) and 3M aq K2CO3 (20 mL) were added, the
organic phase was separated, and the organic phase was
evaporated. HPLC analysis indicated the residue to be a 90:
3:7 mixture of 25a , 25b, and 27b (27a was not detected). The
residue was purified by flash column chromatography (silica,
30% MTBE in hexanes) to provide diol 25a (1.32 g, 80%, >99
area% by HPLC and NMR analysis). HPLC retention times:
25a ) 6.9 min, 25b ) 7.4 min, HPLC conditions: 150 × 4.6
mm ACE 3 C18 column, isocratic elution with 70% MeCN in
0.025% aq H3PO4 at 1 mL/min, 25 °C with detection at 220
nm. Typically crude mixtures resulting from the hydroboration
of 26a could be completely desilylated with excess 48% aq HF
or TBAF to provide a 9:1 diastereomeric mixture of 25a and
25b.
residue was purified by flash column chromatography (silica,
20% MTBE in hexanes) to provide 29a (470 mg, 88%) and 29b
(15 mg, 3%). HPLC retention times: 29a ) 13.3 min, 29b )
12.9 min, HPLC conditions: 150 × 4.6 mm ACE 3 C18 column,
isocratic elution with 80% MeCN in 0.025% aq H3PO4 at 1 mL/
min, 25 °C with detection at 220 nm. Samples of 29a and 29b
exhibited spectral data in agreement to those reported in the
literature.3b
(3S,5S,6R)-29a m a jor d ia ster eom er : Rf ) 0.40 (4:1
hexane-MTBE); 13C NMR (100 MHz, CDCl3) δ 155.5, 155.4,
143.0, 139.0, 131.3, 129.1, 128.1, 126.8, 120.6 (q, J ) 255,
OCF3), 120.3, 119.9, 113.1, 84.5, 79.9, 72.4, 60.4, 51.3, 43.4,
39.0, 38.3, 32.8, 28.5, 23.8, 6.4; FTIR (thin film) νmax 1689,
1496, 1415, 1366, 1247, 1157 cm-1. Anal. Calcd for C29H34F3-
NO5: C, 65.28; H, 6.42; N, 2.63. Found: C, 65.06; H, 6.37; N,
2.50.
(3R,5S,6R)-29b m in or d ia ster eom er : Rf ) 0.30 (4:1
hexane-MTBE); 13C NMR (100 MHz, CDCl3) δ 155.4, 155.3,
143.0, 139.0, 131.0, 129.0, 128.0, 126.9, 120.6 (q, J ) 257,
OCF3), 120.2, 119.9, 113.0, 84.2, 79.9, 72.4, 62.7, 51.3, 42.9,
38.4, 37.6, 31.5, 28.4, 23.3, 6.4; FTIR (thin film) νmax 1690,
1496, 1414, 1362, 1246, 1162 cm-1
.
(3S,5S,6R)-(-)-6-P h en yl-1-oxa-7-azaspir o[4,5]decan e (-)-
1. A mixture of 29a (400 mg, 0.750 mmol), EtOAc (3 mL) and
48% aq HBr (0.40 mL) was aged for 1 h at 25 °C and was
evaporated to dryness. The residue was dissolved in CHCl3 (3
mL) and washed with 2 M aq Na2CO3 (3 mL). The aqueous
phase was extracted with CHCl3 (3 mL), and the combined
organic extracts were evaporated to dryness to give (-)-1 (309
mg, 95%). HPLC analysis of the Marfey’s reagent (N-R-(2,4-
dinitro-5-fluorophenyl)-L-alaninamide) derivative26 indicated
(-)-1 to be optically pure (97% ee). HPLC retention times:
(+)-1 Marfey’s derivative ) 9.5 min, (-)-1 Marfey’s derivative
) 10.2 min, HPLC conditions: 150 × 4.6 mm ACE 3 C18
column, isocratic elution with 70% MeCN in 0.025% aq H3-
PO4 at 1 mL/min, 25 °C with detection at 360 nm. The spectral
data exhibited by (-)-1 was identical to that reported in the
literature3b except for the expected opposite optical rotation.
25a m a jor d ia ster eom er : Rf ) 0.35 (1:1 hexane-MTBE);
1H NMR (400 MHz, CDCl3) δ 7.47 (d, J ) 8.6, 2 H), 7.34-7.20
(m, 4 H), 7.06 (m, 2 H), 5.15 (s, 1 H), 4.04 (m, 1 H), 3.78-3.66
(m, 3 H), 3.54 (quint, J ) 6.2, 1 H), 3.15 (m, 1 H), 2.29 (br s,
2 H), 2.25 (dd, J ) 14.7, 5.7, 1 H), 2.07 (dd, J ) 14.7, 6.6, 1 H),
1.97 (m, 1 H), 1.76 (m, 3 H), 1.28 (s, 9 H), 0.80 (m, 4H); 13C
NMR (100 MHz, CDCl3) δ 155.4, 154.9, 143.1, 139.1, 133.5,
129.5, 128.2, 127.4, 121.9, 120.6 (q, J ) 255, OCF3), 119.8,
113.5, 79.9, 72.3, 66.9, 62.1, 51.4, 40.8, 39.3, 37.4, 33.1, 28.2,
21.9, 6.5; FTIR (thin film) νmax 3381, 1688, 1661, 1246, 1163
cm-1
.
25b m in or d ia ster eom er : Rf ) 0.28 (1:1 hexane-MTBE);
1H NMR (400 MHz, CDCl3) δ 7.48 (d, J ) 7.1, 2 H), 7.36-7.27
(m, 3 H), 7.23 (d, J ) 8.9, 1 H), 7.15 (d, J ) 2.7, 1 H), 7.07 (m,
1 H), 5.01 (s, 1 H), 4.05 (m, 1 H), 3.83-3.72 (m, 3 H), 3.18 (m,
1 H), 2.56 (br s, 2 H), 2.24 (dd, J ) 15.0, 4.1, 1 H), 2.06 (dd, J
) 15.0, 8.3, 1 H), 1.99 (dd, J ) 13.4, 3.7, 1 H), 1.88 (m, 1 H),
1.71 (m, 2 H), 1.33 (s, 9 H), 0.85-0.72 (m, 4H); 13C NMR (100
MHz, CDCl3) δ 155.6, 154.3, 143.2, 138.8, 134.2, 129.5, 128.4,
127.6, 121.2, 120.6 (q, J ) 255, OCF3), 119.8, 113.2, 80.1, 72.7,
66.3, 64.8, 51.4, 41.9, 39.8, 35.7, 30.3, 28.3, 21.6, 6.4; FTIR
(-)-1: [R]23 ) -12.8° (c ) 3.05, MeOH); authentic sample of
D
(+)-1: [R]23 ) +12.7° (c ) 3.00, MeOH).
D
Ack n ow led gm en t. The authors are grateful to Dr
David Hands and Dr. Rob. D. Wilson of the Department
of Process Research, Merck Sharp & Dohme Research
Laboratories, Merck & Co., Inc. for generously supplying
a 1 kg sample of (2R,3R)-cis-3-hydroxy-2-phenylpiperi-
dine which enabled the rapid development of the new
synthetic route to (-)-1. We thank Dr. Kevin Campos
for his crystallographic support.
(thin film) νmax 3395, 1688, 1662, 1244, 1163 cm-1
.
N-Boc 6-P h en yl-1-oxa -7-a za sp ir o[4,5]-d eca n es 29a a n d
29b. Methanesulfonyl chloride (143 mg, 1.25 mmol) was added
over 5 min to a mixture of diol 25a (552 mg, 1.00 mmol) and
DIEA (194 mg, 1.50 mmol) in THF (1.5 mL) maintained at
-20 °C. The mixture was warmed to 0 °C over 1 h. A solution
of 1 M NaHMDS in THF (2.5 mL, 2.5 mmol) was added in
three portions over 1 h to the mixture maintained at 20 °C.
The mixture was diluted with 2 M aq NaHSO4 (10 mL) and
extracted with MTBE (2 × 10 mL). The combined organic
extracts were washed with brine (5 mL) and evaporated. The
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallo-
1
graphic data for compounds 24a and 24b. Copies of H NMR
and 13C NMR spectra for compounds 16, 25a , and 25b. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O0157472