Two asymmetric syntheses of AMG 221, an inhibitor of 11β- hydroxysteroid dehydrogenase type 1

Article abstract of DOI:10.1021/jo900287b

(Chemical Equation Presented) Two asymmetric syntheses of AMG 221 (2), an inhibitor of 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) discovered in our laboratories, are reported. One of the syntheses utilizes chiral trimethylsilyl cyanohydrin 12 as s

Full text of DOI:10.1021/jo900287b

Two Asymmetric Syntheses of AMG 221, an Inhibitor of
11ꢀ-Hydroxysteroid Dehydrogenase Type 1
Seb Caille,* Sheng Cui, Tsang-Lin Hwang, Xiang Wang, and Margaret M. Faul
Chemical Process R&D, Amgen Inc., One Amgen Center DriVe, Thousand Oaks, California 91320
scaille@amgen.com
ReceiVed February 9, 2009
Two asymmetric syntheses of AMG 221 (2), an inhibitor of 11ꢀ-hydroxysteroid dehydrogenase type 1
(11ꢀ-HSD1) discovered in our laboratories, are reported. One of the syntheses utilizes chiral trimethylsilyl
cyanohydrin 12 as starting material and the other utilizes its enantiomer ent-12. The displacement approach
involves the conversion of 12 to 2 via a six-step sequence, occurs with net inversion of configuration,
and employs amine 6 as starting material. This route features a novel approach toward chiral
dialkylsubstituted R-mercaptoacids. The cyclization approach entails the synthesis of 2 from ent-12 in 2
steps, takes place with net retention of configuration, and uses thiourea 8 as starting material. The final
step of this route exemplifies a novel synthesis of chiral C-5 dialkylsubstituted 2-aminothiazolones from
chiral R-hydroxyacids and thioureas. Insights into the mechanism of this transformation and study of the
effect of the medium on the stereochemical outcome of the reaction are presented.
Introduction
The 2-aminothiazolone core structure represents a widely
exploited pharmacophore in the pharmaceutical industry.1 While
the majority of biologically active 2-aminothiazolones display
a C-5 alkylidene,2 some C-5 dialkylsubstituted 2-aminothiazo-
lones are also biologically relevant. An example of the former
group of substances is 5-(2,4-dihydroxybenzylidene)-2-(phe-
nylimino)-1,3-thiazolidin (DBPT),3 a potential therapeutic agent
for colorectal cancer (Figure 1). Herbicidal agent 14 is a
representative of the latter ensemble of compounds. During the
course of a recent development program, we became interested
in elaborating strategies for the synthesis of AMG 221 (2), an
FIGURE 1. Examples of 2-aminothiazolones.
inhibitor of 11ꢀ-hydroxysteroid dehydrogenase type 1 (11ꢀ-
HSD1) discovered in our laboratories.5 Inhibitors of 11ꢀ-HSD1
are potential therapeutic agents for the treatment of type 2
diabetes.6
While several methods for the assembly of achiral C-5
dialkylsubstituted 2-aminothiazolones have been described, the
synthesis of their chiral counterparts has received little attention.
The alkylation of anions of C-5 monoalkylsubstituted 2-ami-
nothiazolones constitutes a commonly used approach to generate
C-5 dialkylsubstituted 2-aminothiazolones.4,7 The addition of
(1) Pulici, M.; Quartieri, F. Tetrahedron Lett. 2005, 46, 2387–2391.
(2) Singh, S. P.; Parmar, S. S.; Raman, K.; Stenberg, V. I. Chem. ReV. 1981,
81, 175–203, and references cited therein.
(3) Teraishi, F.; Wu, S.; Zhang, L.; Guo, W.; Davis, J. J.; Dong, F.; Fang,
B. Cancer Res. 2005, 65, 6380–6387.
(4) Suzuki, M.; Morita, K.; Yukioka, H.; Miki, N.; Mizutani, A. J. Pesticide
Sci. 2003, 28, 37–43.
(5) Henriksson, M.; Homan, E.; Johansson, L.; Vallgarda, J.; Williams, M.;
Bercot, E.; Fotsch, C. H.; Li, A.; Cai, G.; Hungate, R. W.; Yuan, C.; Tegley,
C.; St. Jean, D.; Han, N.; Huang, Q.; Liu, Q.; Bartberger, M. D.; Moniz, G. A.;
Frizzle, M. J. Patent WO 2005116002.
(6) Kershaw, E. E.; Morton, N. M.; Dhillon, H.; Ramage, L.; Seckl, J. R.;
Flier, J. S. Diabetes 2005, 54, 1023–1031.
10.1021/jo900287b CCC: $40.75
Published on Web 04/24/2009
2009 American Chemical Society
J. Org. Chem. 2009, 74, 3833–3842 3833

Caille et al.
SCHEME 1. Synthetic Strategies Toward 2-Aminothiazolone 2
amines to 2-thiothiazolones has also been exploited to obtain
these molecules.8 Another method described to synthesize these
compounds is the condensation of thioureas and R-haloacids
or esters, though the transformation has a limited substrate scope
in the case of tertiary halides.9 Tertiary R-bromoacid bromides
have been used to circumvent this problem by rendering the
halogen displacement an intramolecular process.4,10
reaction of thiourea 8 with the acid chloride formed from 7
followed by mesylation of the tertiary alcohol group. The
synthesis of 2 from 9 was predicted to have a reasonable chance
of success due to the intramolecular nature of the proposed SN2
displacement.10
Results
A flexible strategy for the synthesis of 2 in which simple
chiral cyanohydrin 3 and either amine 611 or thiourea 811 could
be utilized as starting material was contemplated (Scheme 1).
In the displacement approach, the simple chiral building block
4 was expected to be obtained from chiral cyanohydrin 3 via a
three-step sequence involving activation of the hydroxy group
of 3 as a mesylate, SN2 displacement using a nucleophilic sulfur
reagent,12 and hydrolysis of the nitrile function to yield the
corresponding carboxylic acid 4. 2-Thiothiazolone 5 would be
generated from 4 by condensation with MeSCN.13 Finally, AMG
221 (2) was proposed to be synthesized from 5 by treatment
with S-(exo)-2-aminonorbornane (6).8,14 In the cyclization
approach, acid 7 would be produced by hydrolysis of the nitrile
function of 3. Acylthiourea 9 was projected to be generated by
Displacement Approach. The enantioselective preparation
of trimethylsilyl cyanohydrin 12 catalyzed by the aluminum
chiral Salen complex 11 and N,N-dimethylaniline N-oxide has
been reported by Feng and co-workers.15 We elected to utilize
this efficient and reliable transformation for the generation of
12 (85-92% ee, Scheme 2).16 Unfortunately, hydrolysis of the
trimethylsilylether group of 12 utilizing HCl (1 M in MeOH)
or NaOH (1 M in MeOH) resulted in complete racemization
and recovery of (()-3. Alternatively, the treatment of 12 with
0.05 equiv of CSA in 2-methyltetrahydrofuran (2-MeTHF)
permitted trimethylsilyl ether hydrolysis while maintaining
configurational stability. Mesylation of the crude material (3)
proceeded smoothly and 13 was isolated in 92% yield over the
last two transformations.
Several nucleophilic sulfur reagents were surveyed to effect
the SN2 displacement of the mesylate group of 13. The use of
KSAc, t-BuSNa, and EtOCS2K was unsuccessful. In these cases,
elimination to form the corresponding R,ꢀ-unsaturated nitrile
was the dominant reaction pathway. Thioacetic acid has been
reported to bring about such a transformation for a related
example in the presence of 2,4,6-collidine.17 Although these
experimental conditions (AcSH, 2,4,6-collidine, toluene, 50 °C,
(7) For additional examples see: (a) St. Jean, D. J.; Yuan, C.; Bercot, E. A.;
Cupples, R.; Chen, M.; Fretland, J.; Hale, C.; Hungate, R. W.; Komorowski,
R.; Veniant, M.; Wang, M.; Zhang, X.; Fotsch, C. J. Med. Chem. 2007, 50,
429–432. (b) Chande, M. S.; Suryanarayan, V. Tetrahedron Lett. 2002, 43, 5173–
5175.
(8) Caille, S.; Bercot, E. A.; Cui, S.; Faul, M. M. J. Org. Chem. 2008, 73,
2003–2006.
(9) The majority of examples reported involve the reaction of R-bromoma-
lonates with thioureas. For examples see: (a) Kochikyan, T. Synth. Commun.
2004, 34, 4219–4225. (b) Hurst, D. T.; Stacey, A. D.; Nethercleft, M.; Rahim,
A.; Harnden, M. R. Aust. J. Chem. 1988, 41, 1221–1229.
(10) For an additional example see: Skinner, G. S.; Elmslie, J. S.; Gabbert,
J. D. J. Am. Chem. Soc. 1959, 81, 3756–3759.
(14) Couplings of 2-thiothiazolones and amines to generate C-5 dialkylsub-
stituted 2-aminothiazolones are reported in ref 6. For general examples (not
dialkylsubstituted substrates) see: (a) Unangst, P. C.; Connor, D. T.; Cetenko,
W. A.; Sorenson, R. J.; Kostlan, C. R.; Sircar, J. C.; Wright, C. D.; Schrier,
D. J.; Dyer, R. D. J. Med. Chem. 1994, 37, 322–328. (b) Khodair, A. I.
J. Heterocycl. Chem. 2002, 39, 1153–1160.
(15) Chen, F.-X.; Zhou, H.; Liu, X.; Qin, B.; Feng, X.; Zhang, G.; Jiang, Y.
Chem. Eur. J. 2004, 10, 4790–4797.
(16) Feng and co-workers utilized the chiral Salen ligand and Et3Al as separate
reagents in the reaction to generate the complex 11 in situ. We have found that
complex 11 can be prepared and isolated for convenient use as a preformed
catalyst.
(11) For synthesis of 6, see: Huang, J.; Bunel, E.; Allgeier, A.; Tedrow, J.;
Storz, T.; Preston, J.; Correll, T.; Manley, D.; Soukup, T.; Jensen, R.; Syed, R.;
Moniz, G.; Larsen, R.; Martinelli, M.; Reider, P. J. Tetrahedron Lett. 2005, 46,
7831–7834. For synthesis of 8, see ref 5. A manuscript describing the syntheses
of 6 and 8 in >99% ee and >100 Kg scale is in preparation.
(12) This displacement represents a considerable synthetic challenge due to
the sterically hindered nature of the stereogenic center of 3 (and of the
corresponding mesylate).
A
suitable nucleophilic sulfur reagent for this
transformation would hopefully be identified.
(13) Analogous cyclizations utilizing R-mercaptoacid derivatives and cy-
anamides to generate 2-aminothiazolones have been reported. For examples see:
(a) Kretov, A. E.; Bespalyi, A. S. Zh. Obshch. Khim. 1963, 33, 3323–3325. (b)
D’Angeli, F.; Santinello, I. Farmaco, Ed. Sci. 1957, 12, 960–968.
(17) Effenberger, F.; Gaupp, S. Tetrahedron: Asymmetry 1999, 10, 1765–
1775.
3834 J. Org. Chem. Vol. 74, No. 10, 2009

Two Asymmetric Syntheses of AMG 221
SCHEME 2. Synthesis of Mesylate 13
TABLE 1. Stereoselectivity of NaSH Displacement in Buffered
Reaction Mediaa
SCHEME 3. Introduction of Thiol Moiety
24 h) gave the desired thioacetate product in 35% yield (>50%
of 13 unreacted), the reaction rate was too low to be practical.
The desired SN2 displacement was conducted with NaSH ·H2O18
(6 equiv) in H2O at 45 °C and the transformation was complete
after 12 h (Scheme 3). However, the resultant thiocyanohydrin
had poor stability19 after acidification of the aqueous reaction
medium (pH ∼2). Moreover, this intermediate was unstable to
the acidic conditions necessary for hydrolysis of the nitrile
function (concentrated HCl, 80-95 °C). These obstacles were
circumvented by a stepwise hydrolysis of the nitrile group. Thus,
addition of KOH (70 equiv) to the aqueous reaction mixture
after completion of the displacement step (NaSH ·H2O, H2O,
45 °C, 12 h) and aging at 95 °C for 8 h afforded amide 14
(40-42% yield, 75% ee).20 This intermediate was isolated and
treated with aqueous concentrated HCl at 85 °C for 24 h to
provide carboxylic acid 4 (92-95% from 14). It should be noted
that direct hydrolysis of 14 to generate 4 under basic conditions
(KOH, 95 °C) was feasible. However, this process required 72 h
and produced a lower yield (20%), due to the formation of
elimination product 15 (Table 1).
The erosion of chirality observed in the displacement step
(85% ee to 75% ee) was studied in further detail. Toward this
end the displacement step was run for 48 h. It was determined
that the NaSH displacement was incomplete after 4 h at 45 °C
(32% of 13 not consumed by GC) but reached completion after
10 h. The mixture was sampled at different time intervals during
the experiment and the aliquots were treated with KOH at 95
°C for 4 h21 (entries 1-6, Table 1). The percent ee values of
14 (after KOH treatment) increased during the course of the
time for
entry step 1 (h)
additive for
step 1
14
14
4
15
(% Y) (% ee) (% Y) (% Y)
1
2
3
4
5
6
7
8
1
2
4
10
24
48
18
18
none
none
none
none
none
none
HCI
AcOH/NaOAc
11
25
44
54
52
48
53
51
51
70
75
75
75
75
81
81
2
4
4
2
4
5
3
3
69
49
29
11
12
14
11
13
a Reaction conditions: 20 mL of H2O per g of 13 utilized. Assay
yields determined by GC versus an authentic standard are reported.
NaSH step from 51% ee (entry 1, 1 h) to 75% ee (entry 3, 4 h
and entry 4, 10 h). Moreover, the thiocyanohydrin intermediate
was determined to be configurationally stable under the reaction
conditions as demonstrated by the unchanged percent ee value
of 14 (75%) for additional cycling times of 12 h (entry 5) or
36 h (entry 6). The pH of the NaSH aqueous solution was
measured throughout the course of the reaction and determined
to change from ∼11 to ∼8.5 in the first 2 h of the transformation
and be constant for the remaining 10 h of the NaSH treatment.
This finding led to the hypothesis that the displacement reaction
was less stereoselective at higher pH (11) than at lower pH (8.5).
Thus, the NaSH step was performed with either HCl (entry 7)
or a AcOH/NaOAc buffer (entry 8) as additives to maintain
aqueous solutions of pH ∼8.5 throughout the course of the
reactions. As a result, the percent ee values of 14 were
significantly improved (entry 7, 81% ee and entry 8, 81% ee).
AMG 221 (2) was synthesized by reaction of R-mercaptoacid
4 with MeSCN13 (3.0 equiv) in the presence of 3 Å molecular
sieves at 110 °C followed by treatment of the resultant
2-thiothiazolone 5 (Scheme 1) with S-(exo)-2-aminonorbornane
(6, 99% ee) in MeOH (overall yield 45-47%, Scheme 4).14,22
The absolute configuration of 2 was established using a series
of single-crystal X-ray analyses23 and consequently the absolute
configurations of 4, 12, and 13 were determined to be as drawn.
(18) For synthesis of tertiary R-mercaptoesters from tertiary R-haloesters with
NaSH · H2O see: Itsuda, H.; Kawamura, M.; Kato, K.; Kimura, S. Patent JP
63010755.
(19) Thioacetonitrile has been reported to be an unstable species, see:
Gaumont, A. C.; Wazneh, L.; Denis, J. M. Tetrahedron 1991, 47, 4927–4940.
(20) A baseline separated chiral GC assay was developed for 14. No such
assay was available for 4. It was assumed that no important change in % ee
occurred going from 14 to 4. This assumption was proven correct after synthesis
of 2 from 4. The reported yield of 14 (40-42%) includes 4-6% of 4 (if 4 is
excluded the isolated yield of 14 is 34-38%) generated in the KOH step and
separated by chromatography. One plausible reason for the low isolated yield
of 14 is the difficulty in extracting this species from water (high water solubility,
see the Experimental Section, ref 38).
(22) Mixtures of 2 and its diastereomer (having the opposite absolute
configuration at C-5 on the 2-aminothiazolone ring, compound 16) of >85/15
DR could be upgraded to >99.5/0.5 DR through a single recrystallization of a
mixture of the corresponding salts (made with a simple achiral acid). The
complete details associated with this upgrade will be included in a separate paper,
which is currently in preparation. Alternatively, the same two compounds could
be separated by chiral chromatography.
(21) The aliquots were treated with KOH to generate 14 and allow for
quantitative achiral and chiral analyses (GC). The assay yields of 14 were
typically 10-20% higher than the corresponding isolated yields.
J. Org. Chem. Vol. 74, No. 10, 2009 3835

Caille et al.
SCHEME 4. Synthesis of 2 from Carboxylic Acid 4
SCHEME 5. Single Step Formation of 2-Aminothiazolones
16/2 from Acid 7
Cyclization Approach. Our study of the second proposed
asymmetric approach to 2 commenced with the hydrolysis of
trimethylsilyl cyanohydrin 12 utilizing aqueous concentrated
HCl at 85 °C to provide 7 in a moderate 45-50% yield (Scheme
5). The absolute configuration of 7 was ascertained by single-
crystal X-ray analysis of the salt produced from 7 and (R)-(+)-
R-methylbenzylamine.24 The synthesis of an acylthiourea
product via the coupling of acid 7 and thiourea 8 (99% ee) was
planned.25,26 To achieve this purpose, 7 was treated with SOCl2
in DMF for 2 h, 8 and i-Pr2EtN (2 equiv) were added, and the
resultant reaction mixture was aged for 12 h. Unexpectedly,
this experiment directly provided 2-aminothiazolones 16 and 2
in a 61% isolated yield for both isomers and a 74/26 DR (16/
2), along with 18% unreacted 8. Some loss of chiral information
was observed in this cyclization as the percent ee of starting
material 7 was 89% (95/5 ER). Moreover, the stereochemical
outcome for the major diastereomer resulted from a net retention
of configuration.
The effect of the solvent utilized on the newly discovered
ring formation process was examined using racemic 7. The
transformation was successful in DMF and DMAC and afforded
2/16 mixtures of 50/50 DR (entries 1 and 2, Table 2). These
results excluded the possibility that the stereochemical outcome
of the ring formation process is controlled by the chirality of
thiourea 8 (99% ee). The reaction was unsuccessful with THF,
CH3CN, or DMSO as solvents (entries 3, 4, and 5). An
experiment conducted in DMSO afforded a complex mixture
of products (no 2/16 or 8).
2/16 along with a complex mixture of unidentified products.
Other tertiary amine bases were successfully utilized in the
process to provide 2/16 in assay yields of 62% (N-methylmor-
pholine, entry 4) and 66% (Et3N, entry 5). In these experiments,
the major diastereomer produced was 2 as could be predicted
from the result obtained with 7 (Scheme 5). Furthermore, the
2/16 DR range was relatively narrow (from 73/27 to 81/19) and
again represented a significant loss of stereochemical informa-
tion relative to ent-7 (87% ee, 94/6 ER). Reaction in the absence
of base led to the recovery of starting material 8 (entry 6).
In these early studies, DMF and DMAC had been the only
effective solvents for the cyclization process (Table 2, vide
supra). In an effort to probe the transformation in less polar
media, mixtures of solvents were utilized. DMF was employed
as a minor component along with THF or 2-MeTHF (Table 4).
Gratifyingly, these combinations of solvents afforded 2/16
mixtures with DR values ranging from 88/12 to 91/9,22 which
represented losses of 2-5% relative to ent-7 (87% ee, 94/6 ER)
and constituted a marked improvement with regard to earlier
results (Table 3, vide supra). The assay yields were lower than
those of the same experiments performed in DMF (Table 3,
entries 1 and 3). The preeminent result of this series was
obtained using PO(OMe)Cl2 in 25% DMF/2-MeTHF (v/v) (entry
3, 49%).
The transformation was explored utilizing ent-7 (87% ee) as
starting material while changing the R-hydroxyacid activator
and tertiary amine base employed. Thionyl chloride (entry 1,
62%, Table 3) was successfully replaced with PO(OMe)Cl2
(entry 3, 59%) to effect the desired transformation while the
use of SOBr2 (entry 2, 16%) afforded a lower assay yield of
(23) The absolute configuration of amine 6 was ascertained using single-
crystal X-ray analysis of the phthalamic acid salt presented below, prepared from
6 and a phthalamic acid derivative of known absolute configuration (R). 2 was
synthesized from 6 and the absolute configuration of 2 at C-5 was thus determined
by single-crystal X-ray analysis of 2. For this last analysis, although the angles
refine to 90°, the data show that the crystal is truly monoclinic, with the R(int)
for the orthorhombic system of 0.47 and that of the other two monoclinic systems
0.509 and 0.519, respectively, with the correct monoclinic setting value of 0.020.
The study of the effect of the reaction medium on this
transformation was further pursued. The data presented in Table
5 reflect the results of a single experiment separated into three
different aliquots after the first part of the reaction (treatment
of ent-7 with PO(OMe)Cl2 in DMF for 2 h). The aliquots were
charged with the appropriate solvent, 8, and i-Pr2EtN, and stirred
for an additional 12 h. The measured erosion of chirality
correlated well with the polarity of the solvent utilized in the
second part of the experiment; the more polar solvent (DMSO,
entry 3) afforded a 68/32 mixture of stereoisomers and the least
polar solvent (2-MeTHF, entry 1) provided a 84/16 mixture of
stereoisomers. From these results it can be concluded that the
erosion of chirality in the first part of the experiment is minimal
(no greater than 88/12 ER (77% ee) ent-7 to 84/16 DR 2/16)
and that the polarity of the medium in the second part of the
experiment influences the stereochemical outcome of the
reaction in a significant manner. The reaction conditions
(24) This salt has been synthesized before (Mori, K.; Ebata, T.; Takechi, S.
Tetrahedron 1984, 40, 1761–1766.) but no single-crystal X-ray analysis was
performed. The DR of the salt was upgraded to 98.3/1.7 by recrystallization
(twice hexanes/IPA and once THF) prior to single-crystal X-ray analysis.
(25) For an example of such a coupling, see: Sun, C.; Zhang, X.; Huang,
H.; Zhou, P. Bioorg. Med. Chem. 2006, 14, 8574–8581.
(26) 8 was generated from 6, thus establishing its absolute configuration.
3836 J. Org. Chem. Vol. 74, No. 10, 2009

Two Asymmetric Syntheses of AMG 221
TABLE 2. Solvent Screen in Formation of 2/16 from Racemic 7a
TABLE 4. Solvent Mixture Screen in Formation of 2/16 from
ent-7a
entry
solvent
8 (%)
2/16 (%)
DR
1
2
3
4
5
DMF
DMAC
THF
CH3CN
DMSO
15
21
88
91
0
62b
55c
0
0
0
50/50
50/50
NA
NA
NA
entry
reagent
SOCl2
PO(OMe)Cl2
PO(OMe)Cl2
solvents (v/v)
2/16 (%)
DR
1
2
3
25% DMF/2-MeTHF
25% DMF/THF
25% DMF/2-MeTHF
31
47
49
88/12
89/11
91/9
a Reaction conditions: 10 mL of solvent per g of ent-7 utilized. Both
parts of the experiments ran at 23 °C. Assay yields determined by
HPLC versus an authentic standard are reported.
a Reaction conditions: 10 mL of solvent per g of 7 utilized. Both parts
of the experiments ran at 23 °C. b Isolated yield. c Assay yield
determined by HPLC versus an authentic standard.
TABLE 5. Effect of Solvent Polarity in Formation of 2/16 from
TABLE 3. Activator and Base Screen in Formation of 2/16 from
ent-7a
ent-7a
entry
activator
SOCl2
SOBr2
PO(OMe)Cl2
SOCl2
base
2/16 (%)
DR
entry
solvent
dielectric constant (εr)
2/16 (%)
DR
1
2
3
4
5
6
iPr2EtN
iPr2EtN
iPr2EtN
62b
16c
59b
62c
66c
0
80/20
73/27
79/21
79/21
81/19
NA
1
2
3
2-MeTHF
acetone
DMSO
7.0
21.0
47.2
73b
63c
78c
84/16
77/23
68/32
N-methylmorpholine
SOCl2
SOCl2
Et3N
none
a Reaction conditions: 7.5 mL of DMF and 22.5 mL of other solvent
used per g of ent-7. Both parts of the experiments ran at 23 °C.
b Isolated yield. c Assay yield determined by HPLC versus an authentic
standard.
a Reaction conditions: 10 mL of DMF per g of ent-7 utilized. Both
parts of the experiments ran at 23 °C. b Isolated yield. c Assay yield
determined by HPLC versus an authentic standard.
%). Racemic salt 17 was subjected to the reaction conditions
employed in the second part of the cyclization transformation
[iPr2EtN (2 equiv), HCl (1 M solution in Et2O, 1 equiv), 8 (0.5
equiv, 99% ee) in DMF] and 2-aminothiazolones 2 and 16 were
obtained as major products albeit in moderate yield (46% for
both isomers).
The observed stereochemical outcome of the cyclization (see
Tables 4 and 5, vide supra) was of net retention of configuration.
We intended to probe this result further by performing the
cyclization with a different leaving group than that employed
thus far. To this end, mesylate 18 was synthesized via the route
detailed in Scheme 7. The carboxylic acid function of ent-7
described in entry 1 represent the optimized procedure to
synthesize 2 from ent-7.
Efforts were directed at clarifying the mechanism of the
2-aminothiazolone formation process. Toward this end, racemic
727 was treated with SOCl2 [or PO(OMe)Cl2]28 in DMF at 23
°C. Addition of excess 2-MeTHF to the reaction mixture resulted
in the precipitation of intermediate 17,29 as depicted in Scheme
6. The white solid (17) was also produced by addition of racemic
7 to a pre-stirred solution of SOCl2 in DMF (23 °C, 2 h), aging
of the resultant mixture for 2.5 h, and addition of antisolvent
(2-MeTHF). The structure of 17 was determined utilizing a
series of NMR experiments (13C, COSY, HSQC, HMBC,
1H-15N HMBC).30 The chloride content of the salt was
measured to be 13 wt % by AgNO3 titration (theoretical 15 wt
1
(30) The H-13C HMBC displays a 3-bond correlation between Ha and Cb
but not between Ha and Ca. The chemical shift of Ha is 9.08 ppm and that of Cc
is 167.5 ppm. These NMR experiments were performed with a 600 MHz
instrument equipped with a cryoprobe. Additionally, an HRMS was obtained
for salt 17 (organic portion) and the salt was acidic on litmus test (MeOH
solution).
(27) Racemic 7 was employed for this mechanistic study due to the
availability of the starting material.
(28) Intermediate 17 was generated and observed by 1H NMR in DMF-d7
by treatment of
7 with both SOCl2 and PO(OMe)Cl2. However, when
PO(OMe)Cl2 was utilized the isolation of 17 as a solid from the reaction mixture
was complicated by the presence of an additional phosphorus species [PO(OMe)O-
HCl] and the solid obtained could not be filtered easily.
(29) Alkoxyiminium ions have been synthesized from DMF. For examples
see: Barluenga, J.; Campos, P. J.; Gonzalez-Nunez, E.; Asensio, G. Synthesis
1985, 426–428. Sforza, S.; Dossena, A.; Corradini, R.; Virgili, E.; Marchelli, R.
Tetrahedron Lett. 1998, 39, 711–714. Dimethyl substituted alkoxyiminium ions
have been utilized as leaving groups. For examples see: Hanessian, S.; Plessas,
N. R. Chem. Commun. 1967, 22, 1152–1155. Barrett, A. G. M.; Braddock, D. C.;
James, R. A.; Koike, N.; Procopiou, P. A. J. Org. Chem. 1998, 63, 6273–6280.
J. Org. Chem. Vol. 74, No. 10, 2009 3837

Caille et al.
SCHEME 6. Formation of 2/16 from Racemic 7 via Chloride Salt 17
SCHEME 7. Synthesis of 2/16 via Mesylate 18
SCHEME 8. Mechanistic Proposal for Conversion of Salt 17 to 2/16
was benzylated (BnBr, Cs2CO3) and the tertiary alcohol was
derivatized as a mesylate utilizing Ms2O and pyridine. Following
hydrogenolysis of the benzyl ester linkage, acyl thiourea 18 was
obtained by sequential treatment of the resultant carboxylic acid
with SOCl2, KSCN, and 6 (66% yield from ent-7). This isolated
intermediate (18) was reacted with Cs2CO3 to afford 2 (62%).
The loss of stereochemical information observed in this series
of reactions was minimal [77% ee (89/11 ER) ent-7 to 87/13
DR 2/16] and a net retention of configuration was observed.
proceed and 8 remained unreacted. This result may be explained
by the failure of 19 to form without base. The carbonyl group
of 19 is susceptible to nucleophilic attack by 8 to give rise to
intermediate 20. A species having the molecular weight (340)
of intermediate 20 was observed by mass spectroscopy (chemi-
cal ionization) upon reaction of 17 with 8 in the presence of
iPr2EtN (2 equiv) but in the absence of HCl, thereby lending
further evidence for this mechanistic explanation.32 One possible
alternative mechanism to explain the conversion of 17 to 2/16
(31) Aza analogues of proposed intermediate 19 have been generated
employing DMF and isolated, see: (a) Cavicchioni, G.; D’Angeli, F.; Casolari,
A.; Orlandini, P. Synthesis 1988, 94, 7–950. (b) Scrimin, P.; D’angeli, F.;
Veronese, A. C.; Baioni, V. Tetrahedron Lett. 1983, 24, 4473–4476.
Discussion
A mechanistic proposal for the transformation of racemic salt
17 into 2/16 is presented in Scheme 8 (dioxolanone pathway).
In this mechanism, dioxolanone 1931 is produced by deproto-
nation of the carboxylic acid function of 17 with iPr2EtN
followed by 5-membered ring formation. As described in Table
3, in the absence of a tertiary amine base the reaction did not
3838 J. Org. Chem. Vol. 74, No. 10, 2009

Two Asymmetric Syntheses of AMG 221
12 (ent-12) as well as thiourea 8 were employed as chiral starting
materials in the cyclization approach. Two linear steps were
necessary to synthesize 2 from ent-12 and a net retention of
configuration was observed. The final reaction of this sequence
represents a novel one-step synthesis of a chiral C-5 dialkyl-
substituted 2-aminothiazolone from a chiral dialkylsubstituted
R-hydroxyacid and a thiourea occurring with minimal loss of
stereochemical information when an adequate reaction medium
is utilized. Insights into the mechanism of this transformation,
including the structure determination of intermediate 17, were
also provided.
FIGURE 2. Proposed epoxy-imine intermediate 23.
is also depicted in Scheme 8 (R-lactone pathway) and involves
the intermediacy of R-lactone33 21. Although it is conceivable
that carboxylic acid 22 would be generated via this mechanism,
the condensation reaction necessary to produce 2/16 typically
occurs at 80-90 °C and not 23 °C.34 Moreover, this mechanism
does not account for the observation of a 340 amu intermediate
by mass spectroscopy. Therefore, this second mechanism is not
considered to be a likely pathway to generate 2/16 from 17.
Considering the data presented in Table 4 (vide supra) and
Scheme 7, a common stereochemical outcome for these
transformations was observed. This result was independent of
the nature of the leaving group utilized for these two examples.
A single intermediate is put forward in Figure 2 to account for
these results. The observed stereochemical outcome of net
retention can be rationalized by a double inversion mechanism
proceeding via epoxy-imine intermediate 23,35 which is disposed
toward internal attack of the pendant thiourea moiety. The
covalent character of the epoxide C-O bond in 23 may depend
on the polarity of the medium employed (Tables 4 and 5). More
polar solvents are expected to better solvate the C+ and O-
charges and cause 23 to possess a less covalent C-O bond.
This weaker C-O bond would facilitate rotation around the
epoxide C-C bond before the cyclization of 23 to form 2 and
result in greater loss of stereochemical information.
Experimental Section
Aluminum Complex 11.36 AlEt3 (2.0 M in hexanes, 2.5 mL,
0.005 mol) was slowly added via syringe to a solution of 2-((E)-
((1S,2S)-2-((E)-5-bromo-2-hydroxybenzylideneamino)-1,2-diphe-
nylethylimino)methyl)-4-bromophenol (2.9 g, 0.005 mol) in THF
(10 mL) under an atmosphere of nitrogen. Gas evolution was
observed, a yellow precipitate formed, and the reaction was
exothermic. The mixture was cooled to 23 °C and heptane was ad-
ded (20 mL). The suspension was stirred for 15 min, filtered, and
the solid was washed with hexanes (15 mL) to afford 2.75 g (87%)
of 11 as a yellow powder. Decomposition temperature ∼280 °C.
1H NMR [400 MHz, (CD3)2NCDO)] δ 8.43 (s, 1 H), 8.08 (s, 1 H),
7.30-7.65 (m, 14 H), 6.80-6.88 (m, 2 H), 5.49 (d, 1 H, J ) 12
Hz), 5.40 (d, 1H, J ) 12 Hz), 0.87 (s, 3 H, J ) 8 Hz), -0.05-0.20
(m, 2H); 13C NMR [100 MHz, (CD3)2NCDO] δ 172.0, 165.9, 165.7,
165.1, 162.5, 138.9, 138.9, 138.1, 136.2, 136.1, 134.4, 130.5, 129.6,
129.5, 129.4, 128.8, 123.9, 123.7, 121.4, 121.1, 107.0, 106.4, 72.0,
70.8, 10.1, 3.8 (br signal); IR (neat) 2850, 1625, 1535, 1462, 1378,
1311, 1183, 829, 780, 697 cm-1; exact mass [C30H25AlBr2N2O2 +
H]+ calcd 631.0176, measured 631.0157.
(R)-2,3-Dimethyl-2-(trimethylsilyloxy)butanenitrile (12). TM-
SCN (28.8 g, 0.29 mol) and N,N-dimethylaniline oxide (0.2 g,
0.0015 mol) were dissolved in THF (75 mL) and the resultant
solution was stirred for 1 h at 23 °C under an atmosphere of
nitrogen. 3-Methylbutan-2-one (50.0 g, 0.58 mol) was added via
syringe and the mixture was cooled to -30 °C. Complex 11 (1.82
g, 0.0029 mol) was added and the reaction mixture was stirred for
24 h. The mixture was warmed to 23 °C and filtered. The filtrate
was concentrated (30 mmHg) and the residue was distilled under
reduced pressure (30 mmHg, 80 °C) to yield 47.2 g (88%) of 12
(85% ee) as an oil. Chiral GC analysis: Cyclosil-B, 30 m × 250
µm × 0.25 µm, 1 µL injection, 2.2 mL/min, 21.5 psi, 50 to 51 °C
over 25 min, FID detector, 12 at 23.6 min, ent-12 at 23.9 min.
[R]23D +12.2 (c 2.10, CDCl3); 1H NMR (400 MHz, CDCl3) δ 1.86
(septet, 1 H, J ) 4 Hz), 1.53 (s, 3 H), 1.04 (d, 3 H, J ) 4 Hz), 1.02
(d, 3 H, J ) 4 Hz), 0.25 (s, 9 H); 13C NMR (100 MHz, CDCl3) δ
121.5, 73.4, 39.1, 26.0, 17.1, 16.9, 1.15; IR (neat) 2969, 1375, 1254,
1160, 991, 841, 755 cm-1; exact mass [C9H19NOSi + Na]+ calcd
208.1128, measured 208.1130.
Both routes to AMG 221 (2) described here exemplify novel
approaches to chiral C-5 dialkylsubstituted 2-aminothiazolones.
Clearly, the linear sequence from ketone 10 to 2-aminothiaz-
olone 2 is shorter in the case of the cyclization approach (3
steps) than in the case of the displacement approach (7 steps),
which represents a substantial advantage for application on large
scale.
Summary
Asymmetric strategies to synthesize AMG 221 (2) from either
amine S-6 or thiourea S-8 have been described. In the displace-
ment approach, trimethylsilyl cyanohydrin 12 and amine 6 were
utilized as chiral starting materials and 2 was generated in six
linear steps with a net inversion of configuration relative to
building block 12. This route features a novel approach toward
chiral dialkylsubstituted R-mercaptoacids. The enantiomer of
(R)-2-Cyano-3-methylbutan-2-ylmethanesulfonate (13). 12
(11.0 g, 0.059 mol, 85% ee) was dissolved in 2-MeTHF (110 mL)
under an atmosphere of nitrogen at 23 °C. Water (4.4 mL) and
CSA (0.68 g, 0.00295 mol) were added and the solution was stirred
for 3 h. The reaction mixture was treated with saturated aqueous
NaHCO3 (100 mL), the phases were separated, and the aqueous
phase was extracted with 2-MeTHF (2 × 50 mL). The combined
organic phases were washed with brine, dried (Na2SO4), and
(32) This species was converted to 2 and 16 by chromatography on silica
gel or treatment with 1 M HCl in THF.
(33) For evidence of an R-lactone intermediate, see: Pirrung, M. C.; Brown,
W. L. J. Am. Chem. Soc. 1990, 112, 6388–6389. For addition of a nucleophile
to an R-lactone intermediate, see: Adam, W.; Alzerreca, A.; Liu, J.-C.; Yany, F.
J. Am. Chem. Soc. 1977, 99, 5768–5773.
(34) For examples see: (a) St. Jean, D. J.; Yuan, C.; Bercot, E. A.; Cupples,
R.; Chen, M.; Fretland, J.; Hale, C.; Hungate, R. W.; Komorowski, R.; Veniant,
M.; Wang, M.; Zhang, X.; Fotsch, C. J. Med. Chem. 2007, 50, 429–432. (b)
Leite, A. C. L.; Lima, R. S.; Moreira, D. R.; Cardoso, M. V.; Brito, A. C. G.;
Santos, L. M. F.; Hernandes, M. Z.; Kiperstok, A. C.; Lima, R. S.; Soares,
M. B. P. Bioorg. Med. Chem. 2006, 14, 3749–3757.
(35) For a proposed epoxy-imine intermediate, see: Cohen, A. D.; Showalter,
B. M.; Toscano, J. P. Org. Lett. 2004, 6, 401–403. For an isolated epoxy-imine
intermediate, see: Ziegler, E.; Kollenz, G.; Ott, W. Justus Liebigs Ann. Chem.
1976, 11, 2071–2082.
(36) Complex ent-11 was synthesized utilizing an identical procedure and
the opposite enantiomer of the chiral Salen ligand.
(37) We have encountered difficulties with regard to the completion of the
subsequent mesylation reaction when the residue did not remain under high
vacuum (∼1 mmHg) for the described amount of time (12 h). If any problem is
encountered with completion of the mesylation, the cyanohydrin intermediate
can be distilled (20 mmHg, 110 °C). We have utilized this alternative procedure
and it resolved any issues observed with completion of the mesylation reaction.
J. Org. Chem. Vol. 74, No. 10, 2009 3839

Caille et al.
concentrated under reduced pressure (∼1 mmHg for 12 h).37 The
residue was dissolved in 2-MeTHF (100 mL) under an atmosphere
of nitrogen at 23 °C. Et3N (12.34 mL, 0.088 mol) and MsCl (6.78
mL, 0.088 mol) were added via syringes and the reaction mixture
was stirred for 3 h. The reaction was exothermic and precipitates
were formed. The mixture was treated with saturated aqueous
NaHCO3 (100 mL), the phases were separated, and the aqueous
phase was extracted with 2-MeTHF (3 × 50 mL). The combined
organic phases were washed with brine, dried (Na2SO4), and
concentrated under reduced pressure. Chromatographic purification
(70 g silica gel, 10-30% EtOAc/hexanes) of the residual material
yielded 10.36 g (92%) of 13 (85% ee) as an oil. Chiral GC analysis:
Cyclosil-B, 30 m × 250 µm × 0.25 µm, 1 µL injection, 2.2 mL/
min, 29.3 psi, 150 to 180 °C over 20 min, FID detector, 13 at 11.5
min, enantiomer of 13 at 11.8 min. [R]23D +15.0 (c 1.25, CDCl3);
1H NMR (400 MHz, CDCl3) δ 3.17 (s, 3 H), 2.24 (septet, 1 H, J
) 8 Hz), 1.89 (s, 3 H), 1.14 (t, 6 H, J ) 8 Hz); 13C NMR (100
MHz, CDCl3) δ 116.7, 82.6, 39.7, 37.8, 23.1, 16.7, 16.6; IR (neat)
2979, 1466, 1358, 1180, 1048, 901, 805 cm-1; exact mass
[C7H13NO3S + Na]+ calcd 214.0508, measured 214.0510.
dried (Na2SO4), and concentrated under reduced pressure. Chro-
matographic purification (10 g silica gel, 10-20% EtOAc/hexanes)
of the residual material yielded 5. This material was dissolved in
MeOH (11 mL) and 6 (1.22 g, 0.011 mol, 99% ee) was added under
an atmosphere of nitrogen. The solution was stirred for 4 h at 23
°C and concentrated. Chromatographic purification (10 g silica gel,
10-40% EtOAc/hexanes) of the residue yielded 0.87 g (45%) of
a 2/16 mixture (90.1/9.9) as a solid. Chiral HPLC analysis: OD-H
column, 250 mm × 4.6 mm, 5 µm; 1.5 mL/min; 5 µL; 25 °C;
0.025% diethylamine/6% ethanol/94% hexane, isocratic; 16 at 4.34
min, 2 at 6.85 min. 1H NMR (400 MHz, CDCl3, 90.15/9.85 mixture
of diastereomers, signals for the major diastereomer) δ 3.33-3.40
(m, 1 H), 2.36-2.45 (m, 2 H), 2.21 (septet, 1 H, J ) 8 Hz),
1.84-1.91 (m, 1 H), 1.60-1.83 (m, 1 H), 1.42-1.68 (m, 3 H),
1.62 (s, 3 H), 1.13-1.30 (m, 4 H), 1.05 (d, 3 H, J ) 8 Hz), 0.90
(d, 3 H, J ) 8 Hz); 13C NMR (100 MHz, CDCl3, 90.15/9.85 mixture
of diastereomers, signals for the major diastereomer) δ 191.1, 180.9,
70.9, 59.5, 43.0, 38.5, 35.9, 35.7, 35.6, 28.2, 26.6, 25.6, 19.0, 18.4;
IR (neat) 3168, 2959, 2869, 1696, 1585, 1440, 1327, 1256, 1090,
1017, 829 cm-1; exact mass [C14H22N2OS + H]+ calcd 267.1526,
measured 267.1525.
(S)-2-Mercapto-2,3-dimethylbutanoic Acid (4). NaSH·H2O
(23.2 g, 0.314 mol) was dissolved in water (200 mL) and the
solution was warmed to 45 °C under an atmosphere of nitrogen.
The pH of the aqueous solution was adjusted to 8.7 by addition of
1.2 mL of aqueous concentrated HCl. 13 (10.0 g, 0.052 mol, 85%
ee) was added via pipet and the reaction mixture was stirred for
24 h. To the resultant solution was added KOH (200 g, 3.57 mol)
as a solid and the mixture was warmed to 95 °C. The solution was
stirred for 10 h and cooled to 23 °C. The mixture was chilled (0
°C) and aqueous concentrated HCl (∼400 mL) was added via pipet
until the pH of the aqueous mixture was ∼2. The internal
temperature of the aqueous mixture during this process was kept
under 40 °C. The heterogeneous mixture was diluted with 200 mL
of water (all salts were solubilized) and extracted using IPAC (3
× 500 mL).38 The combined organic phases were washed with
brine, dried (Na2SO4), and concentrated under reduced pressure.
Chromatographic purification (40 g silica gel, 5-60% EtOAc/
hexanes) of the residue yielded 2.54 g (33%) of 14 (81% ee) and
0.38 g (5%) of 4.39 Chiral GC analysis of 14: Cyclosil-B, 30 m ×
250 µm × 0.25 µm, 1 µL injection, 2.2 mL/min, 29.3 psi, 150 to
180 °C over 20 min, FID detector, 14 at 12.3 min, enantiomer of
14 at 12.5 min. Amide 14 (2.5 g, 0.017 mol) was dissolved in
aqueous concentrated HCl (36 mL) and the resultant mixture was
warmed to 85 °C under an atmosphere of nitrogen. The mixture
was stirred for 24 h, cooled to 23 °C, and extracted using IPAC (3
× 100 mL). The combined organic phases were washed with brine,
dried (Na2SO4), and concentrated under reduced pressure. Chro-
matographic purification (15 g silica gel, 10-20% EtOAc/hexanes)
(S)-2,3-Dimethyl-2-(trimethylsilyloxy)butanenitrile (ent-12).
The procedure is identical with that used for the synthesis of 12
except that catalyst ent-11 was utilized. Chiral GC analysis for ent-
12 (87% ee): Cyclosil-B, 30 m × 250 µm × 0.25 µm, 1 µL
injection, 2.2 mL/min, 21.5 psi, 50 to 51 °C over 25 min, FID
detector, 12 at 23.4 min, ent-12 at 24.0 min. [R]23D -12.13 (c 1.70,
1
CDCl3); H NMR (400 MHz, CDCl3) δ 1.86 (septet, 1 H, J ) 4
Hz), 1.53 (s, 3 H), 1.04 (d, 3 H, J ) 4 Hz), 1.02 (d, 3 H, J ) 4
Hz), 0.25 (s, 9 H); 13C NMR (100 MHz, CDCl3) δ 121.5, 73.4,
39.1, 26.0, 17.1, 16.9, 1.15; IR (neat) 2969, 1375, 1254, 1160, 992,
841, 755 cm-1; exact mass [C9H19NOSi + Na]+ calcd 208.1128,
measured 208.1129.
(S)-2-Hydroxy-2,3-dimethylbutanoic Acid (ent-7). Aqueous
concentrated HCl (60 mL) was warmed to 85 °C under an
atmosphere of nitrogen. ent-12 (5.0 g, 0.027 mol, 87% ee) was
added and the mixture was stirred for 16 h. The solution was cooled
to 23 °C and extracted using IPAC (3 × 75 mL). The combined
organic phases were washed with brine, dried (Na2SO4), and
concentrated under reduced pressure. Chromatographic purification
(30 g silica gel, 10-50% EtOAc/hexanes) of the residual material
yielded 1.61 g (45%) of ent-7 (87% ee) as a solid. The chiral GC
analysis was performed using the corresponding ethyl ester:40
Cyclosil-B, 30 m × 250 µm × 0.25 µm, 1 µL injection, 2.8 mL/
min, 28.2 psi, 80 to 105 °C over 17 min, FID detector, ethyl ester
of 7 at 6.52 min, ethyl ester of ent-7 at 6.60 min. Mp 47-49 °C;
[R]23D +5.70 (c 1.00, CHCl3); 1H NMR (400 MHz, CDCl3) δ 2.02
(septet, 1 H, J ) 8 Hz), 1.44 (s, 3 H), 1.00 (d, 3 H, J ) 8 Hz), 0.93
(d, 3 H, J ) 8 Hz); 13C NMR (100 MHz, CDCl3) δ 182.1, 77.1,
35.5, 23.3, 17.2, 15.8; IR (neat) 3433, 2973, 2882, 1725, 1460,
1377, 1247, 1164, 1120, 1045, 948, 855, 737 cm-1; exact mass
[C6H12O3 + Na]+ calcd 155.0678, measured 155.0679.
of the residual material yielded 2.32 g (92% from 14) of 4 as a
1
solid. Mp 78-80 °C; [R]23 +3.2 (c 2.60, CDCl3), H NMR (400
D
MHz, CDCl3) δ 2.25 (septet, 1 H, J ) 4 Hz), 2.22 (s, 1 H), 1.43
(s, 3 H), 1.09 (d, 3 H, J ) 4 Hz), 0.98 (d, 3 H, J ) 4 Hz); 13C
NMR (100 MHz, CDCl3) δ 181.5, 53.9, 36.4, 20.2, 18.2, 17.3; IR
(neat) 2968, 2877, 1693, 1404, 1276, 1110, 925 cm-1; exact mass
[C6H12O2S + Na]+ calcd 171.0450, measured 171.0449.
2-Aminothiazolones 2/16. ent-7 (1.0 g, 0.0076 mol, 77% ee)41
was dissolved in DMF (7.5 mL) under an atmosphere of nitrogen.
PO(OMe)Cl2 (0.96 mL, 0.0081 mol, 85% pure) was added via
syringe and the solution was stirred at 23 °C for 2 h. The reaction
mixture was divided into three equal portions (∼2.5 mL) and one
of the portions was diluted with 2-MeTHF (7.5 mL) under an
atmosphere of nitrogen. 8 (0.29 g, 0.0017 mol, 99% ee) and iPr2EtN
(0.87 mL, 0.005 mol) were immediately added (8 as a solid and
iPr2EtN via syringe) and the resultant mixture was stirred for 12 h
at 23 °C. The mixture was treated with saturated aqueous NaHCO3
(15 mL), the phases were separated, and the aqueous phase was
extracted with EtOAc (3 × 20 mL). The combined organic phases
were washed with brine, dried (Na2SO4), and concentrated under
2-Aminothiazolones 2/16. 4 (1.1 g, 0.0073 mol, 81% ee) was
dissolved in toluene (11 mL) under an atmosphere of nitrogen.
Activated 3A sieves (5.5 g) and MeSCN (1.5 mL, 0.022 mol) were
added and the resultant mixture was warmed to 110 °C. The mixture
was stirred for 5 h and cooled to 23 °C. The mixture was treated
with saturated aqueous NaHCO3 (20 mL), the phases were
separated, and the aqueous phase was extracted with EtOAc (3 ×
15 mL). The combined organic phases were washed with brine,
(38) A fourth extraction with IPAC still yielded ∼4% of 14. This finding
supports the hypothesis that the low isolated yield in this step is in part due to
the water solubility of 14.
(40) This ester was synthesized from 7 utilizing the following reaction
conditions: EtI, K2CO3, DMF.
(41) This lot of ent-7 was of 77% ee. The difference in % ee relative with
the previous lot (87% ee) arose in the formation of ent-12.
(39) The yield of 4 was calculated using 1H NMR integration ratios obtained
from a mixture of 4 and 15. The chromatographic separation of 4 and acid 15
is very challenging.
3840 J. Org. Chem. Vol. 74, No. 10, 2009

Two Asymmetric Syntheses of AMG 221
reduced pressure. Chromatographic purification (5 g silica gel,
10-30% EtOAc/hexanes) of the residual material yielded 0.33 g
(73%) of a 2/16 mixture (84/16) as a solid. Chiral HPLC analysis:
OD-H column, 250 mm × 4.6 mm, 5 µm; 1.5 mL/min; 5 µL; 25
°C; 0.025% diethylamine/6% ethanol/94% hexane, isocratic; 16 at
4.45 min, 2 at 5.99 min.42 1H NMR (400 MHz, CDCl3, 84/16
mixture of diastereomers, signals for the major diastereomer) δ
3.33-3.40 (m, 1 H), 2.36-2.45 (m, 2 H), 2.21 (septet, 1 H, J ) 8
Hz), 1.84-1.91 (m, 1 H), 1.60-1.83 (m, 1 H), 1.42-1.68 (m, 3
H), 1.62 (s, 3 H), 1.13-1.30 (m, 4 H), 1.05 (d, 3 H, J ) 8 Hz),
0.90 (d, 3 H, J ) 8 Hz); 13C NMR (100 MHz, CDCl3, 84/16 mixture
of diastereomers, signals for the major diastereomer) δ 191.1, 180.9,
70.9, 59.5, 43.0, 38.5, 35.9, 35.7, 35.6, 28.2, 26.6, 25.6, 19.0, 18.4;
IR (neat) 3168, 2957, 1696, 1587, 1440, 1327, 1256, 1090, 1017,
834 cm-1; exact mass [C14H22N2OS + H]+ calcd 267.1526,
measured 267.1524.
in DMF (22.5 mL) and BnBr (1.9 mL, 0.0158 mol) was added via
syringe. The reaction mixture was stirred for 14 h at 23 °C. EtOAc
(80 mL) and water (50 mL) were added and the phases were
separated. The organic phase was washed with brine (3 × 50 mL),
dried (Na2SO4), and concentrated under reduced pressure. The
residual material was dissolved in pyridine (15 mL) and the resultant
solution was cooled to 0 °C under an atmosphere of nitrogen.
DMAP (185 mg, 0.0015 mol) was added as a solid. A CH2Cl2 (20
mL) solution of Ms2O (4.0 g, 0.023 mol) was added via syringe
over a period of 20 min. The mixture was warmed to 23 °C and
stirred for 1 h. CH2Cl2 (100 mL) and water (50 mL) were added
and the phases were separated. The organic phase was washed with
aqueous 1 M HCl (50 mL), saturated aqueous NaHCO3 (50 mL),
and brine (50 mL). The organic phase was dried (Na2SO4) and
concentrated under reduced pressure. The residue was dissolved
in 200 mL of 30% EtOAc/hexanes, the resultant solution was
filtered through 20 g of silica gel, and the solvents were evaporated
under reduced pressure. The residual material was dissolved in
EtOAc (50 mL) and Pd/C (237 mg, 10 wt %, 0.00023 mol) was
added as a solid. The reaction mixture was stirred under 1 atm of
H2 at 23 °C for 48 h. The mixture was filtered through Celite and
the Celite pad was rinsed with EtOAc (50 mL). The filtrate was
concentrated under reduced pressure to afford 2.6 g of the crude
carboxylic acid-mesylate intermediate as a white solid. An
analytically pure sample of this material was obtained by recrys-
tallization from hexanes. Mp 45-47 °C; 1H NMR (400 MHz,
CDCl3) δ 7.91 (br s, 1 H), 3.14 (s, 3 H), 2.17 (m, 1 H), 1.80 (s, 3
H), 1.04 (d, 3 H, J ) 4 Hz), 1.00 (d, 3 H, J ) 4 Hz); 13C NMR
(100 MHz, CDCl3) δ 175.92, 91.19, 40.36, 36.75, 19.45, 16.83,
15.59; IR (neat) 2974, 2884, 2556, 1707, 1330, 1173, 905, 552
cm-1; exact mass [C7H14O5S - CH4O3S]+ calcd 114.0681, mea-
sured 114.0682. SOCl2 (3.6 mL, 0.049 mol) and DMF (0.1 mL)
were added via syringes to the crude carboxylic acid-mesylate
under an atmosphere of nitrogen. The reaction mixture was warmed
to 70 °C, stirred for 1 h at that temperature, cooled to 23 °C, and
concentrated under reduced pressure. The residual material was
dissolved in acetone (25 mL) and the resultant solution was cooled
to 0 °C under an atmosphere of nitrogen. KSCN (1.31 g, 0.0135
mol) was added as a solid. The mixture was warmed to 50 °C and
stirred for 5 min at that temperature. The reaction mixture was
cooled to 23 °C and a solution of 6 (1.36 g, 0.0123 mol, 99% ee)
in acetone (9 mL) was added via syringe over a period of 20 min.
The mixture was warmed to 50 °C and stirred at that temperature
for 30 min. The suspension was cooled to 23 °C, filtered through
Celite, and the Celite pad was rinsed with EtOAc (20 mL). The
filtrate was concentrated under reduced pressure. Flash chroma-
tography (20 g silica gel, 0-40% EtOAc/hexanes) of the residue
afforded 3.64 g (66% from ent-7) of 18 (mixture of diastereomers)
as an oil. 1H NMR (400 MHz, CDCl3, major diastereomer) δ 10.17
(br s, 1 H), 8.78 (br s, 1 H), 4.08 (m, 1 H), 3.19 (s, 3 H), 2.45 (m,
1 H), 2.35 (m, 1 H), 2.20 (m, 1 H), 1.92 (m, 1 H), 1.83 (s, 3 H),
1.57 (m, 1 H), 1.53 (m, 1 H), 1.40 (m, 2 H), 1.29 (m, 2 H), 1.20
(m, 1 H), 1.05 (d, 3 H, J ) 4 Hz), 0.98 (d, 3 H, J ) 4 Hz); 13C
NMR (100 MHz, CDCl3, major diastereomer) δ 177.32, 171.19,
93.31, 59.15, 41.78, 40.86, 39.95, 36.71, 36.02, 35.93, 28.16, 26.35,
17.79, 16.99, 16.64; IR (neat) 3318, 2960, 1685, 1525, 1352, 1209,
897, 521 cm-1; exact mass [C15H26N2O4S2 + H]+ calcd 363.1412,
measured 363.1407.
Racemic Alkoxyiminium Salt 17. Racemic 743 (0.5 g, 0.0037
mol) was dissolved in DMF (5 mL) under an atmosphere of
nitrogen. SOCl2 (0.28 mL, 0.0039 mol) was added via syringe and
the solution was stirred at 23 °C for 2.5 h. 2-MeTHF (50 mL) was
added and the suspension was filtered. The solid was rinsed with
2-MeTHF (10 mL) and dried on the filter cake for 2 h to afford
0.66-0.70 g (80-85%) of racemic 17 as a white solid. Mp
1
112-114 °C; H NMR (400 MHz, CD3OD) δ 9.08 (s, 1 H), 3.42
(br s, 6 H), 2.34 (septet, 1 H, J ) 8 Hz), 1.83 (s, 3 H), 1.13 (d, 3
H, J ) 8 Hz), 1.06 (d, 3 H, J ) 8 Hz); 13C NMR (150 MHz,
CD3OD) δ 173.6, 167.5, 95.0, 39.3 (broad signal), 36.7, 21.4, 17.8,
1
16.2. As detailed in footnote 25, the H-13C HMBC displays a
3-bond correlation between Ha and Cb but not between Ha and Ca.
IR (neat) 2974, 2353, 1690, 1245, 1123 cm-1; exact mass (organic
portion) [C9H18NO3]+ calcd 188.1281, measured 188.1282. 17 was
acidic when tested on litmus as a MeOH solution. Chloride content
of 17 (theory 15 wt %) was measured by AgNO3 titration in MeOH/
H2O (Titroline instrument): 13 wt %.
2-Aminothiazolones 2/16. Racemic 17 (0.6 g, 0.0027 mol) was
dissolved in DMF (7.0 mL) under an atmosphere of nitrogen. 8
(0.23 g, 0.00135 mol, 99% ee), iPr2EtN (0.94 mL, 0.0054 mol),
and HCl (2.7 mL, 0.0027 mol, 1 M solution in Et2O) were added
(8 as a solid, iPr2EtN and HCl via syringes) and the solution was
stirred at 23 °C for 12 h. The mixture was treated with saturated
aqueous NaHCO3 (20 mL), the phases were separated, and the
aqueous phase was extracted with EtOAc (3 × 20 mL). The
combined organic phases were washed with brine, dried (Na2SO4),
and concentrated under reduced pressure. Chromatographic puri-
fication (10 g silica gel, 10-40% EtOAc/hexanes) of the residual
material yielded 0.17 g (46%) of a 2/16 mixture (50/50) as a solid.
Chiral HPLC analysis: OD-H column, 250 mm × 4.6 mm, 5 µm;
1.5 mL/min; 5 µL; 25 °C; 0.025% diethylamine/6% ethanol/94%
1
hexane, isocratic; 16 at 4.34 min, 2 at 6.85 min. H NMR (400
MHz, CDCl3, 50/50 mixture of diastereomers, signals for both
diastereomers) δ 9.60-10.03 (br s, 1 H), 3.33-3.39 (m, 1 H),
2.32-2.48 (m, 2 H), 2.18-2.24 (m, 1 H), 1.86-1.97 (m, 1 H),
1.73-1.82 (m, 1 H), 1.45-1.68 (m, 6 H), 1.10-1.29 (m, 3 H),
1.00-1.07 (m, 3 H), 0.85-0.94 (m, 3 H); 13C NMR (100 MHz,
CDCl3, 50/50 mixture of diastereomers, signals for both diastere-
omers) δ 191.0, 190.9, 181.0, 70.7, 70.6, 59.6, 59.5, 42.9, 42.7,
38.4, 38.3, 35.9, 35.9, 35.7, 35.6, 35.6, 35.5, 28.2, 26.6, 26.5, 25.6,
18.9, 18.3.
Mesylate 18. ent-7 (1.98 g, 0.015 mol, 77% ee)41 was dissolved
in MeOH (28 mL) and water (5.6 mL) under an atmosphere of
nitrogen. Cs2CO3 (2.44 g, 0.0075 mol) was added as a solid, the
reaction mixture was stirred for 0.5 h at 23 °C, and the solvents
were evaporated under reduced pressure. The residue was dissolved
2-Aminothiazolones 2/16. 18 (73 mg, 0.0002 mol) was dissolved
in MeCN (1 mL) under an atmosphere of nitrogen and Cs2CO3 (98
mg, 0.0003 mol) was added as a solid. The mixture was stirred at
23 °C for 24 h, filtered through Celite, and the Celite pad was rinsed
with EtOAc (5 mL). The filtrate was concentrated under reduced
pressure. Flash chromatography (4 g silica gel, 5-30% acetone/
hexanes) yielded 33 mg (62% from 18) of a 2/16 mixture (87/13)
as a solid. Chiral HPLC analysis: OD-H column, 250 mm × 4.6
mm, 5 µm; 1.5 mL/min; 5 µL; 25 °C; 0.025% diethylamine/6%
(42) The retention times of 2 and 16 were slightly shifted in this trace relative
to other data. The identity of the peaks was confirmed employing a 50/50 mixture
of 2/16.
(43) Racemic 7 was prepared according to the procedure described in the
following report: Mori, K.; Ebata, T.; Takechi, S. Tetrahedron 1984, 40, 1761–
1766.
1
ethanol/94% hexane, isocratic; 16 at 4.35 min, 2 at 6.84 min. H
NMR (400 MHz, CDCl3, 90.15/9.85 mixture of diastereomers,
J. Org. Chem. Vol. 74, No. 10, 2009 3841

Caille et al.
signals for the major diastereomer) δ 3.33-3.40 (m, 1 H),
2.36-2.45 (m, 2 H), 2.21 (septet, 1 H, J ) 8 Hz), 1.84-1.91 (m,
1 H), 1.60-1.83 (m, 1 H), 1.42-1.68 (m, 3 H), 1.62 (s, 3 H),
1.13-1.30 (m, 4 H), 1.05 (d, 3 H, J ) 8 Hz), 0.90 (d, 3 H, J ) 8
Hz); 13C NMR (100 MHz, CDCl3, 90.15/9.85 mixture of diaster-
eomers, signals for the major diastereomer) δ 191.1, 180.9, 70.9,
59.5, 43.0, 38.5, 35.9, 35.7, 35.6, 28.2, 26.6, 25.6, 19.0, 18.4; IR
(neat) 3168, 2959, 2869, 1696, 1585, 1440, 1327, 1256, 1090, 1017,
829 cm-1; exact mass [C14H22N2OS + H]+ calcd 267.1526,
measured 267.1525.
Neil Langille for help with the preparation of this manuscript.
We thank Yoo Chul, Paul Schnier, Tiffany Correll, Jason
Simiens, Randy Jensen, David Yeung, and Lan Hong for
analytical support. We thank Richard Staples for performing
all of the single-crystal X-ray analyses included.
Supporting Information Available: Copies of 1H, 13C,
1
COSY, HMQC, HMBC, and H-15N HMBC NMR spectra
along with crystallographic information files (CIFs). This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. We thank Karl Hansen, Shawn Walker,
Matt Bio, Robert Larsen, and Emilio Bunel for helpful discus-
sions. We thank Eric Bercot, Chris Borths, Brenda Burke, and
JO900287B
3842 J. Org. Chem. Vol. 74, No. 10, 2009