A practical synthesis of a potent and selective diacylglycerol acyltransferase-1 (DGAT-1) inhibitor

Article abstract of DOI:10.1055/s-0034-1378653

A practical synthesis of a potent, selective, and orally efficacious diacylglycerol acyltransferase-1 (DGAT-1) inhibitor, is described. This synthesis is suitable for multi-kilogram scale with high regioselectivity and stereoselectivity. The synthesis inv

Full text of DOI:10.1055/s-0034-1378653

PAPER
▌3047
paper
A Practical Synthesis of a Potent and Selective Diacylglycerol Acyltransferase-1
(DGAT-1) Inhibitor
Synthesis of a Selective DGAT-1 Inhibitor
Youngsook Shin, Philippe Bergeron, Brian M. Fox, Frank Kayser, Lawrence R. McGee, Sharon McKendry,
Dustin L. McMinn, Marc Labelle, Timothy D. Cushing*
Department of Therapeutic Discovery, Amgen Inc., 1120 Veterans Boulevard, South San Francisco, CA 94080, USA
Fax +1(650)8379369; E-mail: tcushing@amgen.com
Received: 12.06.2014; Accepted: 21.07.2014
CO2H
Abstract: A practical synthesis of a potent, selective, and orally ef-
ficacious diacylglycerol acyltransferase-1 (DGAT-1) inhibitor, is
described. This synthesis is suitable for multi-kilogram scale with
high regioselectivity and stereoselectivity. The synthesis involves a
Knoevenagel condensation with Meldrum’s acid followed by the
stereoselective addition of phenyl cuprate, regioselective Friedel–
Crafts acylation, cyclization, and a regioselective reduction through
an enol triflate with catalytic platinum oxide to provide the desired
compound in 5.2% yield over 12 steps.
NH2
DGAT-1 potency
N
biochemical IC50 = 0.004 0.002 μM
cellular IC50 = 0.003 0.001 μM
PK
N
F3C
N
O
mouse CL = 0.45 L/h/kg, F = 77%
1
rat
CL = 0.21 L/h/kg, F = 98%
Key words: diacylglycerol acyltransferase-1, hydrogenation,
Knoevenagel condensation, Friedel–Crafts acylation, Friedel–
Crafts alkylation, stereoselectivity, regioselectivitiy
cyno CL = 0.02 L/h/kg, F = 99%
Figure 1 Profile of DGAT-1 inhibitor 1
A second synthetic route solved the regiochemical issue
using 5-bromoindanone (11) as a handle, but the stereose-
lectivity was not fully addressed (Scheme 2).5 The key
stereoselective hydrogenation yielded only 18% of the de-
sired trans-isomer 17 from 15, and contained 6% of the
undesired cis-isomer after crystallization. Although, this
route rapidly delivered 1 in 4% yield over 11 steps on
gram scale, the yields diminished to 1.6% on a larger
scale. Additionally, several steps were problematic, such
as the use of carbon monoxide and the costly 5-bromoin-
danone. We describe here a final alternative route, suit-
able for multi-kilogram scale with high regio- and
stereoselectivity.
The enzyme diacylglycerol acyltransferase (DGAT) is a
microsomal enzyme that plays a central role in the metab-
olism of cellular glycolipids.1 DGAT-1 is one of two
known isoforms, which catalyzes the formation of tri-
acylglycerides from diacylglycerol and a fatty acid acyl
CoA.2 DGAT-1 is implicated in playing a role in the de-
velopment of obesity and insulin resistance.3 For exam-
ple, under a high-fat diet, DGAT-1 knockout mice are
resistant to weight gain with concomitant increased leptin
sensitivity. Fat tissue explanted from DGAT-1 knockout
mice convey resistance to weight gain in wild-type mice.
Additionally, high-fat fed adipose transgenic mice show
increased adiposity. Therefore, inhibition of DGAT-1
could have a therapeutic potential for people suffering
from obesity, dyslipidemia, and diabetes.3,4 In an earlier
report, we disclosed the discovery of a potent, selective
and orally efficacious DGAT-1 inhibitor 1, which demon-
strated promising in vivo efficacy in a mouse diet-induced
obesity (DIO) model and an acceptable safety profile in
rat and cynomolgus monkey toxicity studies (Figure 1).5
Key features of the new route are the use of inexpensive
and readily available phenol 24, stereoselective conver-
sion of 23 into 22, and the regioselective acylation of 21
(Scheme 3).7
There is significant literature precedent for high-pressure
hydrogenation, even at medium or large scale. However,
the conversion of 24 into 25 on this scale under these con-
ditions was beyond our capacity to achieve. It was hypo-
thesized that the hydrogenation could be run at lower
pressure (1 bar) by simply increasing the reaction time
and temperature (Table 1).8 Thus, under a blanket of hy-
drogen at 80 °C with 10 mol% of palladium catalyst, ke-
tone 25 was formed, but with a significant amount of
alcohol by-product 26. However, the catalytic hydrogena-
tion of 24 to 25 proceeded cleanly using only 1.5 mol% of
palladium catalyst at 90 °C with hydrogen sparging.
Sparging with hydrogen through the reaction mixture was
critical to avoid the formation of the alcohol by-product
26. No alcohol by-product 26 was observed on either a
500 gram or 3.0 kilogram scale with sparging. Unfortu-
Previously, compound 1 had been prepared by either of
two distinct routes. The first route evolved from the needs
of the medicinal chemistry campaign. It involved scarce
precursors, toxic intermediates, and required several steps
of chromatography, including a preparative HPLC to ob-
tain 1 as the pure trans-isomer (Scheme 1).5a,6 Additional-
ly, it was not suitable for scale-up because it was neither
stereoselective nor regioselective.
SYNTHESIS 2014, 46, 3047–3058
Advanced online publication: 25.08.2014
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0034-1378653; Art ID: ss-2014-m0255-op
© Georg Thieme Verlag Stuttgart · New York

3048
Y. Shin et al.
PAPER
CO2Me
O
O
O
b
d
a
c
2
3
4
5
O
CO2Me
CO2Me
CO2Me
Br
Br
Br
f
e
+
Br
O
O
6
7a
7b
CO2R
CO2H
F3C
N
NH2
CO2R
N
h
N
NH2
N
NH2
N
O
+
N
O
N
N
N
F3C
F3C
O
9a R = Me
10a R = H
9b R = Me
10b R = H
1
g
g
Scheme 1 First route to 1, nonregioselective and nonstereoselective. Reagents and conditions: (a) 2,2-bis(2-bromoethyl)-1,3-dioxolane,
KHMDS, THF, 0 °C to r.t., 85%; (b) 3 M aq H2SO4, MeOH, reflux, 90%; (c) (MeO)2P(O)CH2CO2Me, NaH, THF, 0 °C to r.t., 90%; (d) H2,
10% Pd/C, EtOAc, 95%; (e) AlCl3, CH2Cl2, 0 °C to r.t., quant; (f) 5,6-diamino-2-(trifluoromethyl)pyrimidin-4-ol (8),5 MeOH, H2O, 2 M HCl,
reflux; (g) LiI, DMF, 135 °C, (h) prep HPLC, 20% over three steps.
nately, the reaction did not go to completion on the 3.0 ki- The first step in the synthesis of key intermediate 21 is the
logram scale providing 69% of 25.
Knoevenagel condensation of 25 with Meldrum’s acid to
produce 23.9 The standard method using ammonium ace-
tate and acetic acid with molecular sieves in toluene gave
23 in 65% yield.7 Alternatively, the ketone 25 could be
treated with piperidine to form the enamine 27, which was
further reacted with Meldrum’s acid to produce 23 in 51%
yield depending on scale. Ketone 25 could also be treated
with piperidine and a catalytic amount of p-toluenesulfon-
ic acid in refluxing toluene with a Dean–Stark trap to form
enamine 27, which was then allowed to condense with
Meldrum’s acid in dichloromethane at room temperature
followed by elimination with aqueous hydrochloric acid
to give 23 in 75% yield on a 500 gram scale and 51% yield
on a 1.4 kilogram scale from 25 (Scheme 4).
Table 1 Optimization of Hydrogenation of 24 to 25
CO2Me
CO2Me
+
CO2Me
H2
Pd/C (10%)
64–80 h
OH
O
OH
26
24
25
Entrya Amount H2
Pd
Temp
Yield
26
(g) 24
(mol%) (°C)
(%) of 25b
1
2
3
4
250
<500
500
balloon
balloon
sparging
sparging
10
1.5
80
90
90
90
N/Ac
N/Ac
80
yes
yes
no
In order to achieve the stereoselective addition on 23 to
22, several conditions were tried (Table 2). The conjugate
addition with the cuprate reagent generated from phenyl-
lithium and cuprous iodide gave 22 in 22% yield as a 1:1
mixture of cis/trans-isomers.10 Grignard conjugate addi-
tion11 gave excellent stereoselectivity, but the yield was
not reproducible. The best stereoselectivity was achieved
in 87–94% yield by using the higher order cuprate as the
nucleophile.7,12 The isopropylidene group of 22 was then
hydrolyzed in refluxing 80% aqueous acetic acid, fol-
lowed by thermal decarboxylation in refluxing xylene to
1.5
1.5
3000
69d
no
a Solvent: AcOH.
b Isolated yield after distillation.
c Not accurate (N/A) due to by-product 26.
d The reaction stopped at approximately 80% conversion even after 80
h.
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York

PAPER
Synthesis of a Selective DGAT-1 Inhibitor
3049
O
b
O
O
O
O
a
c
d
Br
Br
Br
Br
11
12
13
14
CO2Me
CO2Me
CO2Me
e
f
g
BnO
HO2C
Br
O
15
16
17
CO2H
CO2Me
CO2Me
h
i, j
NH2
N
Br
N
O
N
O
O
F3C
18
19
1
Scheme 2 Second route to 1, regioselective, but nonstereoselective. Reagents and conditions: (a) MeOCHPPh3+ Cl–, t-BuOK, THF, DMSO,
0 °C; (b) p-TsOH, 1,4-dioxane, H2O, 39% over two steps; (c) H2, 10% Pd/C, AcOH, 99%; (d) (EtO)2P(O)CH2CO2Me, t-BuOK, THF, 97%; (e)
CO, BnOH, Et3N, toluene, Pd(PPh3)4, 90 °C; (f) H2, 10% Pd/C, EtOH, r.t., 18% for trans-isomer over two steps; (g) i. (COCl)2, CH2Cl2, ii. i-
PrMgCl, CuCN, THF, 99%; (h) CuBr2, EtOAc, CHCl3, 57%; (i) 5,6-diamino-2-(trifluoromethyl)pyrimidin-4-ol (8),5 MeOH, H2O, aq 2 M HCl,
reflux, 24 h, 56%; (j) LiI, DMF, 135 °C, 75%.
NH2
NH2
N
CO2H
F3C
N
8
OH
CO2Me
+
NH2
N
CO2H
N
O
N
HO
O
F3C
21
regioselective
acylation
1
O
20
CO2Me
CO2Me
CO2Me
O
O
O
O
O
O
O
OH
O
24
22
stereoselective
addition
23
Scheme 3 Retrosynthesis of 1
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 3047–3058

3050
Y. Shin et al.
PAPER
CO2Me
CO2Me
CO2Me
a
b
O
O
O
N
O
O
27
25
23
CO2Me
CO2Me
CO2Me
d
c
O
O
O
O
O
O
O
O
HO
O
23
22
21
Scheme 4 Synthesis of 21 from 25. Reagents and conditions: (a) piperidine, p-TsOH, toluene, reflux; (b) i. Meldrum’s acid, CH2Cl2, r.t., ii.
aq HCl, 75%; (c) see Table 2; (d) i. 80% AcOH, H2O, reflux, ii. xylene, reflux, 64% over two steps.
provide acid 21 in 64% yield. The acid 21 was very crys- of the less-hindered ketone to a methylene group should
talline and the desired trans-isomer 21 was easily purified provide 19.
from the minor cis-isomer impurity.
Following route A, the acid 21 was reduced to the corre-
sponding alcohol 32 using one equivalent of borane-tetra-
hydrofuran complex in tetrahydrofuran at –18 °C in over
92% yield. However, if more than two equivalents of bo-
Table 2 Stereoselective Addition to 22
CO2Me
rane are used, both acid and methyl ester are reduced to al-
CO2Me
cohols. The alcohol 32 was converted into chloride 28 in
78% yield. This was followed by the Friedel–Crafts acy-
O
O
lation of chloride 28 with 2-bromoisobutyryl bromide and
aluminum chloride in dichloromethane to give 29, regio-
selectively in 90% yield, with no Friedel–Crafts alkyla-
tion of the chloride. At this point, in order to cyclize to the
indane 19, two conditions were tried. The acylated chloro
compound 29 was treated with aluminum chloride in re-
fluxing 1,2-dichloroethane or aluminum chloride with so-
dium chloride at 150 °C. Unfortunately, the unwanted
indanone 33 was the only product, formed through the
conversion of 19 into the α,β-unsaturated ketone interme-
diate with simultaneous intramolecular cyclization
(Scheme 6).13
O
O
O
O
O
O
22
23
Entry Conditions
Yield (%)a trans-22/cis-22
1
2
3
PhLi, CuI, Et2O
22
1:1
11:1
10:1
PhMgBr, THF
24–87b
87–94
PhMgBr, CuCN, LiCl, THF
a Isolated yield after precipitation.
b Yield was not reproducible.
In order to avoid this unwanted ring closure, 28 was acy-
lated using isobutyryl chloride to give 29. However, the
desired indane 18 was not obtained even under a variety
of conditions (Scheme 7).
Two alternative regiospecific routes A and B to indane 19
from acid 21 were envisioned (Scheme 5). Route A in-
volved the chemoselective reduction of the acid 21 to a
primary alcohol followed by conversion into the chloride
28. Subsequent Friedel–Crafts para-acylation of 28
should produce 29, followed by intramolecular Friedel–
Crafts alkylation to form indane intermediate 19. Route B,
on the other hand, involved initial Friedel–Crafts para-ac-
ylation of 21 to 30 followed by intramolecular Friedel–
Crafts acylation to form 31. The regioselective reduction
To effect this intramolecular Friedel–Crafts alkylation, al-
cohol 32 was converted into mesylate 34 using methane-
sulfonyl chloride and triethylamine in 99% yield. We
envisioned that the mesylate would react by way of
Friedel–Crafts alkylation, but would still be less reactive
than the Friedel–Crafts acylation with isobutyryl chloride.
Unfortunately, this Friedel–Crafts acylation/alkylation of
34 with isobutyryl bromide and aluminum chloride in di-
chloromethane at 0 °C gave the two para-directed regio-
isomers 35a and 35b, reminiscent of 7a and 7b along with
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York

PAPER
Synthesis of a Selective DGAT-1 Inhibitor
3051
CO2Me
CO2Me
Friedel–Crafts
acylation
1) chemoselective
reduction
cyclization
(Friedel–Crafts
alkylation)
CO2Me
Br
2) chlorination
Cl
CO2Me
O
Cl
route A
28
29a
Br
CO2Me
CO2Me
O
route B
19
HO
O
regioselective
reduction
cyclization
Friedel–Crafts
acylation
21
(Friedel–Crafts
acylation)
Br
Br
HO
30
O
O
O
O
31
Scheme 5 Proposed regiospecific routes A and B to indane 19
CO2Me
CO2Me
CO2Me
b
a
HO
O
OH
32
Cl
28
21
O
CO2Me
CO2Me
CO2Me
Br
Br
d or e
c
Br
Br
Cl
O
O
O
29a
33
19
Scheme 6 Route A: Chemoselective reduction and Friedel–Crafts acylation and alkylation to 19. Reagents and conditions: (a) THF·BH3, THF,
–18 °C to r.t., 92%; (b) SOCl2, pyridine, benzene, reflux, 78%; (c) AlCl3, CH2Cl2, r.t., 90%; (d) AlCl3, DCE, reflux; (e) AlCl3, NaCl, 150 °C.
O
CO2Me
CO2Me
CO2Me
Cl
a
b, c, or d
Cl
Cl
28
O
O
29b
18
Scheme 7 Route A: Friedel–Crafts acylation and alkylation with chloride. Reagents and conditions: (a) AlCl3, CH2Cl2, r.t.; (b) AlCl3, DCE,
reflux; (c) AlCl3, NaCl, 150 °C; (d) H2SO4, 100 °C.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 3047–3058

3052
Y. Shin et al.
PAPER
O
b
CO2Me
CO2Me
CO2Me
Br
CO2Me
a
O
OH
32
OMs
34
O
35a
35b
Scheme 8 Route A: Friedel–Crafts acylation and alkylation with mesylate. Reagents and conditions: (a) MsCl, Et3N, CH2Cl2, 0 °C, 99%; (b)
AlCl3, CH2Cl2, 0 °C, 49%.
CO2Me
O
CO2Me
CO2Me
Cl
b
a
c
O
O
OH
O
OH
O
32
36
37
CO2Me
CO2Me
CO2Me
d
+
OMs
Cl
O
O
O
38
18
29
Scheme 9 Route A: Alternative approach to 18. Reagents and conditions: (a) AlCl3, CH2Cl2, 0 °C to r.t., 58%; (b) NaOMe, MeOH, r.t.; (c)
MsCl, Et3N, CH2Cl2, 0 °C, 48% over two steps; (d) AlCl3, CH2Cl2, reflux.
an unknown by-product. This result indicated that the in- amide in dichloromethane at room temperature to give
tramolecular Friedel–Crafts alkylation preceded the acid chloride 39. This acid chloride 39 was immediately
Friedel–Crafts acylation (Scheme 8).
treated with aluminum chloride in refluxing 1,2-dichlo-
Direct Friedel–Crafts acylation of alcohol 32 using two
equivalents of isobutyryl chloride and aluminum chloride
in dichloromethane was attempted producing the doubly
acylated product 36 in 58% yield. The alcohol 37 was ob-
tained by deblocking with sodium methoxide to provide
37, which was subsequently converted into 38 in 48%
yield over two steps. Mesylate 38 was transformed to 18
with aluminum chloride in dichloromethane, but the
chloro by-product 29 and an unknown by-product were
obtained as major products (Scheme 9). After these efforts
on route A, it became clear that aluminum chloride-medi-
ated intramolecular Friedel–Crafts alkylation would not
be efficient in this relatively electron-poor aryl system.
O
CO2Me
CO2Me
Br
Br
b
a
Br
HO
30
O
HO
O
O
21
CO2Me
c
CO2Me
The successful conversion of 21 into spiroindanone 31
was done in three steps. First, the acid 21 was converted
into 30 through Friedel–Crafts para-acylation. The order
of addition of reagents was important to avoid the intra-
molecular cyclization by the carboxylic acid moiety in 21
before the desired Friedel–Crafts acylation. The crude
product 30 was treated with two equivalents of oxalyl
chloride and a catalytic amount of N,N-dimethylform-
Br
Br
Cl
O
O
O
O
39
31
Scheme 10 Route B: Successful conversion of 21 into 31. Reagents
and conditions: (a) AlCl3, CH2Cl2, –15 °C; (b) (COCl)2, DMF,
CH2Cl2, r.t.; (c) AlCl3, DCE, reflux, 63% over three steps.
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York

PAPER
Synthesis of a Selective DGAT-1 Inhibitor
3053
roethane to give spiroindanone 31, which was easily iso- flate 42 was carried on to the next step without purifica-
lated as a pure solid from the crude reaction mixture by tion. Intermediate 42 was reduced with catalytic platinum
simple precipitation. These three steps were performed oxide in the presence of excess lithium carbonate in meth-
without any purification to give 31 in 63% yield on a 350 anol under one atmosphere of hydrogen to give the debro-
gram scale (Scheme 10).
minated by-product 18 in 82% yield from spiroindanone
31 after crystallization on a 460 gram scale (Scheme 11).
Unfortunately, the bromine could not be preserved under
the reduction conditions. However, α-bromination is
known to be a trivial reaction, so 18 was used as an inter-
mediate for the next reactions.5 The treatment of 18 with
20% sodium hydroxide in refluxing 1,4-dioxane and wa-
ter followed by acidification with aqueous 2 M hydrochlo-
ric acid gave crude acid 20 in 99% yield on a 336 gram
scale. The acid 20 was successfully brominated with cop-
per bromide in refluxing acetonitrile to form 43 in 95%
yield after precipitation on a 232 gram scale (Scheme 12).
In order to reduce the spiroindanone 31 regioselectively,
several different approaches were attempted. The treat-
ment of sprioindanone 31 with excess triethylsilane in tri-
fluoroacetic acid at room temperature reduced 2-bromo-2-
methylpropanoyl ketone to a secondary alcohol by-prod-
uct 40, instead of 19. When the reaction was continued in
refluxing trifluoroacetic acid, both ketones were reduced
to give the putative methylene by-product 41. The reduc-
tion of 31 with tert-butylamine borane complex in di-
chloromethane at 0 °C also gave the presumed secondary
alcohol by-product 40 (Table 3).
Compound 43 was then treated with 8 in refluxing aceto-
nitrile and water for five days to give compound 1 in 72%
yield after precipitation on a 335.9 gram scale (Scheme
13).
The successful regioselective reduction of 31 was realized
by taking advantage of the enolizable ketone in 31. The
enol triflate 42 was prepared by treatment of spiroinda-
none 31 with trifluoromethanesulfonic anhydride in the
presence of 2,6-di-tert-butyl-4-methylpyridine and 2,4,6- Relatively inexpensive 24 was converted into 1 in 5.2–
collidine in dichloromethane at room temperature. We 14.7% yield (depending on scale) over 12 steps. Key fea-
had tried to use isobutyryl bromide instead of 2-bromo 2- tures include: catalytic hydrogenation, Knoevenagel con-
methylpropanoyl bromide to effect these transformations densation with Meldrum’s acid, phenyl cuprate addition,
but were unable to prevent the bis-enol triflate formation. deprotection and decarboxylation, Friedel–Crafts acyla-
The bromo group functioned as a blocking group for this tion and cyclization, enol triflate formation with subse-
step. The solvents were removed and the crude enol tri- quent reduction, ester hydrolysis, bromination, and final
Table 3 Route B: Studies on Regioselective Reduction of 31
CO2Me
CO2Me
CO2Me
CO2Me
Br
Br
Br
O
O
O
O
OH
31
19
40
41
Entry
Conditions
19a
40a
41a
1
2
3
Et3SiH, TFA, r.t.
none
none
trace
√
Et3SiH, TFA, reflux
t-BuNH2·BH3, CH2Cl2, 0 °C
√
√
a Not isolated, putative structures based on MS.
CO2Me
CO2Me
CO2Me
b
a
Br
Br
O
OTf
O
O
O
42
31
18
Scheme 11 Route B: Successful regioselective reduction of 31. Reagents and conditions: (a) Tf2O, 2,6-di-tert-butyl-4-methylpyridine, 2,4,6-
collidine, CH2Cl2, r.t.; (b) H2, PtO2, Li2CO3, MeOH, r.t., 81% over two steps.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 3047–3058

3054
Y. Shin et al.
PAPER
CO2Me
CO2H
CO2H
b
a
Br
O
O
O
18
20
43
Scheme 12 Route B: Ester hydrolysis and bromination. Reagents and conditions: (a) i. 20% NaOH, 1,4-dioxane, H2O, reflux, ii. aq 2 M HCl,
99%; (b) CuBr2, MeCN, reflux, 95%.
NH2
CO2H
NH2
OH
CO2H
N
F3C
N
8
NH2
N
MeCN–H2O
90 °C, 5 d
72%
Br
N
O
N
O
F3C
43
1
Scheme 13 Route B: Cyclization to 1
of N2 gas. AcOH (6.0 L) was then slowly added via a large addition
funnel. The resulting mixture was stirred and sparged with more N2
(large bubbler) for some time. The N2 gas was then replaced with H2
and the sparging continued as the reaction mixture was heated to
90 °C with continued H2 sparging for 80 h. At this time, the reaction
mixture was cooled down to r.t. and the reaction vessel vacuum
purged (6 ×) with N2 gas. The heterogeneous solution was removed
via a tygon tube and carefully filtered through Celite and washed
with CH2Cl2 (4 L) at all times keeping the filter cake moist. The re-
sulting solution was concentrated in vacuo to provide a viscous liq-
uid. The viscous liquid was purified by distillation (100 °C/0.5 mm
Hg) to provide 25 as a colorless oil; yield: 2.11 kg (69%).
condensation with 8. This synthesis has great advantages,
providing a good yield, no wasteful, time-consuming col-
umn chromatography purification steps and with excel-
lent regioselectivity and stereoselectivity. The synthesis is
suitable for multi-kilogram scale (Scheme 14).
All solvents and chemicals used were reagent grade. Anhydrous
solvents were purchased from Aldrich and used as such. Analytical
TLC and flash chromatography were performed on Merck silica gel
60 (230–400 mesh). Removal of solvents was conducted by using a
rotary evaporator and residual solvent was removed from nonvola-
tile compounds using a vacuum manifold maintained at approxi-
mately 1 Torr. All yields reported are isolated yields. Preparative
reversed-phase high pressure liquid chromatography (RP-HPLC)
was performed using an Agilent 1100 series HPLC and Phenomen-
ex Gemini C18 column (5 μm, 100 mm × 30 mm i.d.), eluting with
a binary solvent system A and B using a gradient elution (A: H2O
with 0.1% TFA, B: MeCN with 0.1% TFA) with UV detection at
220 nm. All final compounds were purified to ≥95% purity as deter-
mined by an Agilent 1100 series HPLC with UV detection at 220
nm using the following method: Zorbax SB-C8 column (3.5 μm,
150 mm × 4.6 mm i.d.), eluting with a binary solvent system A and
B using a 5–95% B (0–15 min) gradient elution (A: H2O with 0.1%
TFA, B: MeCN with 0.1% TFA); flow rate 1.5 mL/min. Mass spec-
tral data was recorded on an Agilent 1100 series LCMS with UV de-
tection at 254 nm. NMR spectra were recorded on a Varian Gemini
400 MHz or Bruker Avance 500 MHz NMR spectrometer. Chemi-
cal shifts (δ) are reported in parts per million (ppm) relative to re-
sidual undeuterated solvent as internal reference and coupling
constants (J) are reported in Hz. Standard abbreviations are used to
denote the splitting patterns.
1H NMR (500 MHz, CDCl3): δ = 3.69 (s, 3 H), 2.38 (m, 4 H), 2.33
(m, 2 H), 2.27 (m, 1 H), 2.09 (m, 2 H), 1.47 (m, 2 H).
MS (ESI+): m/z = 171.1 (M + 1).
Anal. Calcd for C9H14O3: C, 63.51; H, 8.29. Found: C, 63.16; H,
8.51.
Methyl 2-[4-(2,2-Dimethyl-4,6-dioxo-1,3-dioxan-5-ylidene)cy-
clohexyl]acetate (23)
A 12 L 3-necked flask equipped with two Dean–Stark traps was
charged with toluene (2.7 L), 25 (1.4 kg, 8.2 mol), piperidine (1.1
L, 11.1 mol) and p-TsOH (47 g, 0.25 mol). The mixture was heated
to reflux. After 7 h, GC analysis indicated that 60% of 25 remained.
After a further 1.5 h at reflux, the reaction mixture was allowed to
cool down overnight and the reflux was resumed the following
morning. The reaction was monitored by GC and after a further 6 h,
the reaction had reached a point where no further conversion of
starting material into enamine appeared to be occurring; 17% of
compound 25 remained by GC. The reaction mixture was allowed
to cool to r.t. overnight and the solvent was evaporated in vacuo (45
Torr/58 °C). Toluene (600 mL) was added and evaporated in vacuo
to further remove additional H2O. This process was repeated twice
to give the enamine 27. CH2Cl2 (12 L) was added to the crude enam-
ine 27. Following dissolution, a solution of Meldrum’s acid (1.18
kg, 8.2 mol) in CH2Cl2 (2 L) was added over a period of 10 min (the
internal temperature of the reaction did not rise more than 3–4 °C
Methyl 2-(4-Oxocyclohexyl)acetate (25)
A 20 L reaction vessel was vacuum purged of air and flushed (6 ×)
with N2. To this vessel blanketed in N2, 10% Pd/C (300.0 g) was
carefully added followed by methyl 4-hydroxyphenylacetate (24;
3.0 kg, 18.05 mol) at all times maintaining a slight positive pressure
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York

PAPER
Synthesis of a Selective DGAT-1 Inhibitor
3055
during addition). The reaction was allowed to stir for 10 min and
was found to be complete by GC analysis. Aq 3 M HCl (2.7 L) was
added and the reaction mixture stirred vigorously for 15 min and
monitored by GC. The aqueous phase was removed and a 5% aq
solution of NaHCO3 (2 L) was added and the mixture stirred vigor-
ously. The aqueous phase was removed and this process was repeat-
ed twice using a 2.5% aq solution of NaHCO3 (2 × 2 L). The organic
solvent was then evaporated in vacuo (170 Torr/30 °C). Toluene (1
L) was added and the solvent again removed in vacuo (82 Torr/40
°C). This process was repeated with an additional amount of toluene
(800 mL). The resultant orange oil was left standing for 90 min. Af-
ter this time, crystals had formed in the flask, which were scratched
to speed up crystallization. This resulted in solid crashing out of the
oil to provide a yellow oil/solid mixture. EtOAc (600 mL) was add-
ed and the mixture stirred with a mechanical stirrer to provide a
thick suspension. Heptane (2 × 600 mL) was added, and after stir-
ring for 15 min a dark yellow, sticky solid was obtained. EtOAc
(400 mL) was added and the solid became white while the solvent
turned from pale yellow to orange. The mixture was stirred for 30
min and the precipitate was filtered, washed with 10% EtOAc in
heptane (600 mL) and dried in vacuo to provide 23 as an off-white
solid; yield: 1.23 kg (51%).
GC Analysis: Settings: 80 °C for 2 min, then 20 °C/min increase to
300 °C; starting ketone retention time: 2.8 min.; enamine retention
time: 6.2 min.; Meldrum’s acid adduct retention time: 2 peaks at 8.4
min.; 23: peak appears at 5.08 min, but product was found to be
somewhat unstable to GC conditions.
1H NMR (500 MHz, CDCl3): δ = 3.70–3.75 (m, 5 H), 2.22–2.34 (m,
5 H), 2.14–2.16 (m, 2 H), 1.75 (s, 3 H), 1.74 (s, 3 H), 1.35–1.41 (m,
2 H).
MS (ESI+): m/z = 319.0 (M + 23).
Methyl 2-[(trans)-4-(2,2-Dimethyl-4,6-dioxo-1,3-dioxan-5-yl)-4-
phenylcyclohexyl)acetate (22)
A 20 L 5-necked reactor equipped with a cooling jacket and pneu-
matic overhead stirring was charged under N2 atmosphere with
CuCN (430 g, 4.80 mol), LiCl (407 g, 9.60 mol), and THF (4 L).
CO2Me
CO2Me
CO2Me
CO2Me
b
d
a
c
O
O
O
O
O
O
OH
O
O
O
24
25
23
CO2Me
22
O
e
CO2Me
CO2Me
Br
Br
f
g
Br
Br
HO
O
Cl
O
HO
O
O
O
21
30
39
CO2Me
CO2Me
CO2Me
h
i
j
Br
Br
O
OTf
O
O
O
31
42
18
CO2H
CO2H
CO2H
k
l
NH2
N
Br
N
O
N
O
O
F3C
20
43
1
Scheme 14 Improved synthetic route to 1. Reagents and conditions: (a) H2, Pd/C, AcOH, 90 °C, 69%; (b) i. piperidine, p-TsOH, toluene, re-
flux, ii. Meldrum’s acid, CH2Cl2, r.t., iii. aq HCl, 51%; (c) CuCN, LiCl, THF, PhMgBr, 0 °C to r.t., 81% for trans-isomer; (d) i. 80% AcOH,
H2O, reflux, ii. xylene, reflux, 64% over two steps; (e) AlCl3, CH2Cl2, –15 °C; (f) (COCl)2, DMF, CH2Cl2, r.t., (g) AlCl3, DCE, reflux, 63%
over three steps; (h) Tf2O, 2,6-di-tert-butyl-4-methylpyridine, 2,4,6-collidine, CH2Cl2, r.t.; (i) H2, PtO2, Li2CO3, MeOH, r.t., 81% over two
steps; (j) i. aq 20% NaOH, 1,4-dioxane, H2O, reflux, 2 h, ii. aq 2 M HCl, 99%; (k) CuBr2, MeCN, reflux, 95%; (l) 5,6-diamino-2-(trifluoro-
methyl)pyrimidin-4-ol (8), MeCN, H2O, 90 °C, 72%.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 3047–3058

3056
Y. Shin et al.
PAPER
The mixture was stirred for 20 min to afford scant LiCl granules
against a green solution. The solution was brought to –4 °C where-
upon cannulation of a 3 M solution of PhMgBr in THF (3.20 L, 9.60
mol) was initiated. The rate of addition and cooling was controlled
to maintain a temperature between –4 and 5 °C and required a total
of 2.5 h. The dark viscous solution was stirred an additional 35 min
between –6 and 5 °C, after which the temperature was allowed to
rise to 17 °C over 1 h. The cuprate solution was then cooled to
–15 °C. To the cuprate solution was added a 1.68 M solution of 23
in THF (2.25 L, 3.78 mol) over a period of 45 min at a rate sufficient
to maintain the temperature between –6 and 5 °C. The reaction was
complete within 5 min as evidenced by TLC [sat. aq NH4Cl ex-
traction against EtOAc, eluted with 30% EtOAc in hexane, TLC:
PMA stain, Rf (product) = 0.5]. The reaction contents were poured
into sat. aq NH4Cl (12 L) and stirred vigorously while sparging with
air for 45 min. EtOAc (6 L) was added to the resulting purple sus-
pension followed by paper filtration. The solids were resuspended
in additional EtOAc (6 L) and filtered. The organics from the sec-
ond filtration were used to re-extract the sequestered aqueous layer
from the first filtration. The combined organics were evaporated in
vacuo. The solids were resuspended in EtOAc (6 L) and extracted
against 0.5 M HCl (1.5 L) then brine (1.5 L). The organics were
dried with MgSO4 (250 g) and evaporated in vacuo. Suspension of
the resulting solids in hexanes with vigorous stirring followed by
filtration and drying under vacuum afforded 22 as a white solid;
yield: 1.15 kg (81%). 1H NMR spectra matched that of the previous-
ly reported spectra indicating a 7% presence of the cis-isomer.
slightly impure material. At this time, crop 2 and crop 3 were com-
bined and recrystallized a final time [hexane (4 L)–Et2O (700 mL)]
to provide 21 (crop 4, 86.5 g); final yield of 21: 292.5 g (64%);
white solid.
1H NMR (500 MHz, CDCl3): δ = 7.28–7.35 (m, 4 H), 7.17–7.23 (m,
1 H), 3.68 (s, 3 H), 2.70 (s, 2 H), 2.31 (d, J = 7.2 Hz, 2 H), 2.23 (d,
J = 12.8 Hz, 2 H), 1.84 (dtt, J = 14.4, 7.3, 3.3 Hz, 1 H), 1.66–1.79
(m, 4 H), 1.30–1.40 (m, 2 H).
13C NMR (151 MHz CDCl3): δ = 177.2, 173.3, 147.3, 128.1, 126.1,
125.6, 51.5, 41.5, 41.0, 39.2, 35.0,33.8, 28.2.
MS (ESI–): m/z = 290 and 289 (M – 1).
2-{(trans)-1-[4-(2-Bromo-2-methylpropanoyl)phenyl]-4-
(2-methoxy-2-oxoethyl)cyclohexyl}acetic Acid (30)
A 12 L 3-necked flask equipped with a mechanical stirrer and an ad-
dition funnel was placed under a N2 atmosphere, charged with a sus-
pension of AlCl3 (539.36 g, 4.05 mol) and CH2Cl2 (2 L), and cooled
to –15 °C in an EtOH–H2O (1:4) bath with dry ice. The addition
funnel was charged with 2-bromo-2-methylpropionyl bromide (324
mL, 2.62 mol) and added dropwise over 20 min. The reaction mix-
ture was stirred at –15 °C for 35 min. A clean addition funnel was
placed, charged with a solution of 21 (345.43 g, 1.19 mol) in CH2Cl2
(2 L), and the solution was added dropwise over 40 min. The reac-
tion temperature was then allowed to warm to 15 °C over 5 h with
stirring. The mixture was poured onto ice water (4 L). The aqueous
layer was extracted with CH2Cl2 (2 × 2 L), and the combined organ-
ic layers were washed with brine (1 × 4 L), dried over MgSO4 (100
g) filtered, and concentrated in vacuo to give 30 as a syrup (522.67
g), which was used directly in the following step.
1H NMR (500 MHz, CDCl3): δ = 0.71 (s, 3 H, major isomer only,
93%), 0.79 (s, 3 H, minor isomer only), 1.5 (s, 3 H, minor isomer
only), 1.51 (s, 3 H, major isomer only), 1.54–1.60 (m, 2 H, major
and minor isomers), 1.68–1.84 (m, 2 H, major and minor isomers),
1.86–2.00 (m, 2 H, major and minor isomers), 2.04–2.06 (m, 2 H,
minor isomer only), 2.36 (d, J = 7 Hz, 2 H, major isomer only), 2.90
(br d, J = 13 Hz, 2 H major isomer only), 2.96 (br s, 2 H, minor iso-
mer only), 3.63 (s, 3 H, minor isomer only), 3.71 (s, 3 H, major iso-
mer only), 3.86 (br s, 1 H), 7.24–7.44 (m, 5 H, major and minor
isomers).
1H NMR (500 MHz, CDCl3): δ = 8.12 (d, J = 8.6 Hz, 2 H), 7.4 (d,
J = 8.6 Hz, 2 H), 3.69 (s, 3 H), 2.77 (s, 2 H), 2.31 (d, J = 6.95 Hz, 2
H), 2.24 (d, J = 12.98 Hz, 2 H), 2.04 (s, 6 H), 1.85–1.70 (m, 5 H),
1.36 (m, 2 H).
MS (ESI–): m/z = 437 and 439 (M – 1).
Methyl 2-{(trans)-4-[4-(2-Bromo-2-methylpropanoyl)phenyl]-
4-(2-chloro-2-oxoethyl)cyclohexyl}acetate (39)
MS (ESI–): m/z = 373.1 (M – 1).
A 3 L-flask was charged with a solution of the crude 30 (522.67 g)
in CH2Cl2 (1 L) and cooled to 0 °C in an ice water bath. Oxalyl chlo-
ride (208 mL, 2.38 mol) and DMF (3 drops) were added dropwise.
The ice bath was removed, and the reaction mixture was stirred at
r.t. After 2 h at r.t., LC-MS showed the reaction to be complete. The
solvent was removed in vacuo to provide the acid chloride 39 as a
foamy syrup (544.62 g), which was used directly in the following
step.
2-[(trans)-4-(2-Methoxy-2-oxoethyl)-1-phenylcyclohexyl]acetic
Acid (21)
A 12 L 3-necked flask equipped with a mechanical stirrer and two
reflux condensers was charged with AcOH (3.53 L) and H2O (880
mL) and heated until the solution was gently refluxing. At this time,
22 (588 g, 1.57 mol) was added quickly by removing one condenser
and replacing it with a large funnel. The mixture was refluxed for
25 min, and the heat source was removed to allow the solution to
cool. After 15 min of cooling, the volatiles were removed in vacuo.
The tan residue was triturated with hexane (1.75 L) and Et2O (250
mL). This mixture was stirred for 1 h and filtered. The filter cake
was dried and found to weigh 450 g. This material was used without
further purification. The intermediate dicarboxylic acid (450 g, 1.35
mol) was charged in a 12 L 3-necked flask equipped with a mechan-
ical stirrer and two reflux condensers with p-xylene (5.2 L). The
mixture was heated to reflux in 0.5 h, and continued at full reflux
for 3.5 h. The heating was stopped and the contents were allowed to
cool. After 2 h, most of the xylene was removed in vacuo, the
clumpy residue was removed from the receiving flask using Et2O
(1.2 L) and transferred to a 6 L Erlenmeyer flask. This was used to
effect crystallization with the addition of more Et2O (2 L total vol-
ume). The solution was gently heated until the volume was reduced
to 1.3 L (as measured from the flask markings). This was cooled and
stirred slowly overnight to affect recrystallization. The crystals
were collected by filtration to provide 21 (crop 1, 206 g). During the
suction filtration, additional crystals were formed in the filtrate and
the solution was allowed to stand overnight at which point these
crystals were collected to give 21 (crop 2, 99 g). The remaining
mother liquor was evaporated and the crude dried residue chroma-
tographed (4:1 hexane–EtOAc) to provide 21 (crop 3, 55.75 g) of
Methyl 2-{(trans)-5′-(2-Bromo-2-methylpropanoyl)-3′-oxo-
2′,3′-dihydrospiro[cyclohexane-1,1′-inden]-4-yl}acetate (31)
A 5 L 3-necked flask equipped with a condenser and an addition
funnel was charged with a suspension of AlCl3 (380.72 g, 2.86 mol)
in CH2Cl2 (1 L) and cooled to 0 °C in an ice water bath. The addition
funnel was charged with a solution of the acid chloride 39 in DCE
(1.5 L) and added dropwise over 15 min. The reaction mixture was
then heated at reflux. After 1.5 h at reflux, LC-MS showed the reac-
tion to be complete. The mixture was poured onto ice-water (3 L).
The aqueous layer was extracted with CH2Cl2 (2 × 2 L), and the
combined organic layers were washed with brine (1 × 3 L), dried
over MgSO4 (200 g), filtered through a silica gel pad (650 g), and
the silica gel pad was washed with EtOAc–hexane (1:2, 7 L). The
filtrate was then concentrated in vacuo to give a brown syrup. The
desired product 31 was isolated as a pure solid from the crude mix-
ture after precipitation from Et2O (450 mL) and hexane (1.8 L) with
stirring. The precipitates were filtered, and washed with hexane (2
L) to give 31 as a yellow solid; yield: 314.78 g (63% over three
steps).
1H NMR (600 MHz, CDCl3): δ = 8.46–8.56 (m, 1 H), 8.34 (dd, J =
8.2, 1.8 Hz, 1 H), 7.57 (d, J = 7.7 Hz, 1 H), 3.70 (s, 3 H), 2.62 (s, 2
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York

PAPER
Synthesis of a Selective DGAT-1 Inhibitor
3057
H), 2.30 (d, J = 7.2 Hz, 2 H), 2.04 (s, 6 H), 1.92–2.02 (m, 1 H), 1.84–
1.92 (m, 4 H), 1.61 (d, J = 12.8 Hz, 2 H), 1.15–1.24 (m, 2 H).
13C NMR (151 MHz CDCl3): δ = 204.5, 195.9, 173.1, 166.9, 136.4,
135.4, 134.5, 125.4, 124.0, 60.1, 51.5, 48.3, 43.0, 41.4, 37.8, 34.0,
31.4, 30.0.
liquor from the crop 1 recrystallization was evaporated to a viscous
oil and also run through a plug of silica gel (500 g) eluted with 10:1
hexane–EtOAc to provide 18 (crop 3, 36.1 g). At this point all three
crops were combined and a final recrystallization was performed
with hexane (300 mL) to provide 18; yield: 294.4 g (81% over two
steps); white solid. The mother liquor from this last crystallization
was chromatographed to give an additional 17.9 g of very pure ad-
ditional product 18.
1H NMR (500 MHz, CDCl3): δ = 7.81 (d, J = 6.5 Hz, 1 H), 7.80 (s,
1 H), 7.20 (d, J = 7.6 Hz, 1 H), 3.68–3.70 (m, 3 H), 3.52–3.56 (m, 1
H), 2.92 (t, J = 7.3 Hz, 2 H), 2.28 (d, J = 6.9 Hz, 2 H), 2.02 (t, J =
7.3 Hz, 2 H), 1.89 (td, J = 7.3, 3.9 Hz, 1 H), 1.77 (d, J = 12.7 Hz, 2
H), 1.58–1.70 (m, 4 H), 1.18–1.26 (m, 8 H).
MS (ESI+): m/z = 421 and 423 (M + 1).
Methyl 2-{(trans)-5′-(2-Bromo-2-methylpropanoyl)-3′-{[(tri-
fluoromethyl)sulfonyl]oxy}spiro[cyclohexane-1,1′-inden]-
4-yl}acetate (42)
A 5 L 3-necked flask equipped with a mechanical stirrer, a reflux
condenser, and an addition funnel was charged with 31 (463 g, 1.10
mol) and CH2Cl2 (1.85 L). The mixture was stirred, and over the
course of 7 min, triflic anhydride (200 mL, 1.154 mol) was added
with an addition funnel. This was followed by the addition in the
same way of 2,4-di-tert-butyl-4-methylpyridine (90.2 g, 0.44 mol)
over 5 min, keeping the temperature between 25–29 °C with a large
cooling bath. Finally, 2,4,6-trimethylpyridine (102 mL, 0.77 mol)
was added in the same manner in 5 min while maintaining the tem-
perature between 29–40 °C with the cooling bath. After 5 min, the
temperature had dropped to 30 °C and the cooling bath was re-
moved, the solution was stirred for an additional 1 h, after which the
TLC (eluent: EtOAc–hexane, 1:4) indicated the reaction was com-
plete. The solution was diluted with CH2Cl2 (1 L) and transferred to
a large plastic bucket where it was washed with HCl (0.05 M, 1 L).
The CH2Cl2 solution was further diluted to 4.5 L total volume and
washed with brine (2 L). The resulting solution was dried (Na2SO4)
further, filtered, and the CH2Cl2 was removed in vacuo to provide
42 as a brown syrup (608.73 g, 1.10 mol), which when dried in vac-
uo was used without further purification. (A significant amount of
2,6-di-tert-butyl-4-methylpyridine remained).
MS (ESI+): m/z = 329 (M + 1).
2-{(trans)-5′-Isobutyryl-2′,3′-dihydrospiro[cyclohexane-1,1′-in-
den]-4-yl}acetic Acid (20)
A solution of 18 (90 g, 0.274 mol) in 1,4-dioxane (275 mL) and 20%
aq NaOH (90 mL) was heated at reflux for 1.5 h and then cooled to
r.t. Aq 2 M HCl (225 mL) was added and the mixture stirred until it
had cooled to r.t. EtOAc (90 mL) was added and the layers were
separated. The aqueous layer was extracted with EtOAc (2 × 300
mL) and the organics were pooled, washed with brine (500 mL),
dried (MgSO4), filtered, and concentrated in vacuo to provide 20 as
an off-white solid; yield: 85.65 g (99%); mp 129–131 °C.
1H NMR (500 MHz, CDCl3): δ = 7.81 (d, J = 7.0 Hz, 1 H), 7.81 (s,
1 H), 7.20 (d, J = 7.3 Hz, 1 H), 3.55 (quin, J = 6.9 Hz, 1 H), 2.93 (t,
J = 7.3 Hz, 2 H), 2.33 (d, J = 6.9 Hz, 2 H), 2.03 (t, J = 7.3 Hz, 2 H),
1.92 (td, J = 7.3, 3.9 Hz, 1 H), 1.82 (d, J = 13.7 Hz, 2 H), 1.59–1.71
(m, 4 H), 1.16–1.30 (m, 8 H).
MS (ESI+): m/z = 315.1 (M + 1).
1H NMR (500 MHz, CDCl3): δ = 8.16 (d, J = 9.4 Hz, 1 H), 8.15 (s,
1 H), 7.42 (d, J = 7.8 Hz, 1 H), 6.71 (s, 1 H), 3.71 (s, 3 H), 2.36 (d,
J = 6.9 Hz, 2 H), 1.95–2.08 (m, 11 H), 1.45 (d, J = 13.7 Hz, 2 H),
1.28–1.37 (m, 2 H).
2-{(trans)-5′-(2-Bromo-2-methylpropanoyl)-2′,3′-dihydro-
spiro[cyclohexane-1,1′-inden]-4-l}acetic Acid (43)
A mixture of 20 (232 g, 0.738 mol) and CuBr2 (371 g, 1.66 mol) in
MeCN (1.5 L) was heated at 90 °C under N2 atmosphere for 3 h.
LCMS indicated that a small amount of starting material remained.
Additional CuBr2 (32.97 g, 0.2 equiv) was added and the reaction
mixture was heated at 90 °C for 2 h. The reaction mixture was
cooled to r.t. and aq 2 M HCl (3 L) was added in 1 L portions. The
resulting precipitate was collected by vacuum filtration and washed
with concd HCl (2 × 300 mL) to remove any remaining CuBr, and
then with H2O (500 mL). The solid was dried in vacuo to provide 43
as an off-white solid; yield: 275.8 g (95%); mp 176–178 °C.
MS (ESI+): m/z = 553 and 555 (M + 1).
Methyl 2-{(trans)-5′-Isobutyryl-2′,3′-dihydrospiro[cyclohex-
ane-1,1′-inden]-4-yl}acetate (18)
The intermediate triflate 42 (608.73 g, 1.10 mol) was placed in a
large (20 L) conical reaction vessel equipped with three-way ports.
The triflate was dissolved in MeOH (10 L) and was stirred vigor-
ously with magnetic stirring. After the dissolution, Li2CO3 (406 g,
5.5 mol) and PtO2 (13.5 g, 0.055 mol) were added. The tan reaction
mixture was then carefully vacuum-purged with N2 three times.
This was then followed by three vacuum-purge cycles with H2. The
mixture was then stirred over a blanket of H2 gas with three large
balloons of H2, the balloons had to be recharged several times. After
2.5 h, the H2 was replaced with N2 by vacuum purge. The heteroge-
neous mixture was carefully filtered through a large plug of Celite
(>1 kg) into a large filtering flask, being ever mindful at all times
that the filter cake remain moist. The Celite was washed with
MeOH (6 L), again never allowing the filter cake to become dry. Af-
ter the final rinse, the still saturated fritted funnel was removed and
placed in another large filtering flask and quenched with H2O (3–4
L). The Pt/Celite was never allowed to become dry in this process.
The large volume of MeOH was removed by evaporation and the re-
sulting solid stirred in hexane (6 L) overnight. This mixture was
then filtered. The solid was crushed, resuspended in hexane (6 L)
and filtered. The two filtrates were combined and the hexane evap-
orated, the resulting crude green-yellow solid was dissolved in hot
hexane (200 mL) and recrystallized overnight. The solid was isolat-
ed to give 18 (crop 1, 265.4 g). The undissolved solids from the sec-
ond filtration were taken up in hot hexane (1.5 L) and this
suspension was filtered. After evaporation, a dark brown solid was
obtained, which was redissolved in hexane (200 mL) and run
through a plug of silica gel eluting with 10:1 hexane–EtOAc to pro-
vide 18 after evaporation a white solid (crop 2, 21.6 g). The mother
1H NMR (500 MHz, CDCl3): δ = 8.04 (d, J = 8.2 Hz, 1 H), 7.99 (s,
1 H), 7.19 (d, J = 8.0 Hz, 1 H), 2.95 (t, J = 7.3 Hz, 2 H), 2.35 (d, J =
7.0 Hz, 2 H), 2.06 (s, 6 H), 2.04 (t, J = 7.5 Hz, 2 H), 1.94 (m, 1 H),
1.84 (m, 2 H), 1.67 (m, 4 H), 1.29 (dq, J = 3.7, 12.7 Hz, 2 H).
MS (ESI+): m/z = 393.0 and 395.0 (M + 1).
Anal. Calcd for C20H25BrO3: C, 61.07; H, 6.41; Br, 20.32. Found:
C, 61.05; H, 6.37; Br, 20.35.
2-{(trans)-5′-[4-Amino-7,7-dimethyl-2-(trifluoromethyl)-7H-
pyrimido[4,5-b][1,4]oxazin-6-yl]-2′,3′-dihydrospiro[cyclohex-
ane-1,1′-inden]-4-yl}acetic Acid (1)
A mixture of 43 (335.9 g, 0.854 mol) and 5,6-diamino-2-(trifluoro-
methyl)pyrimidin-4-ol (8; 198.9 g, 1.02 mol) in MeCN (850 mL)
and H2O (250 mL) was heated at 90 °C for 5 days. H2O (1.02 L) was
added and the reaction mixture was slowly cooled to r.t. over 4 h.
The mixture was then cooled to 0 °C while stirring magnetically for
2 h. The resulting precipitate was collected by vacuum filtration and
washed with 1:1 MeCN–H2O (1 L) and dried in vacuo to provide 1
as an off-white solid; yield: 300 g (72%); mp 254–255 °C.
1H NMR (500 MHz, CDCl3): δ = 7.48 (s, 1 H), 7.46 (d, J = 8.1 Hz,
1 H), 7.21 (d, J = 7.9 Hz, 1 H), 6.05 (s, 2 H), 2.96 (t, J = 7.3 Hz, 2
H), 2.36 (d, J = 7.0 Hz, 2 H), 2.06 (t, J = 7.4 Hz, 2 H), 1.94 (m, 1
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 3047–3058

3058
Y. Shin et al.
PAPER
H), 1.86 (m, 2 H), 1.76 (s, 6 H), 1.69 (m, 4 H), 1.30 (dq, J = 3.8, 12.6
Hz, 2 H).
McMinn, D. L.; Ogawa, N.; Rubenstein, S. M.; Sagawa, S.;
Sugimoto, K.; Suzuki, M.; Tanaka, M.; Ye, G.; Yoshida, A.;
Zhang, J. Patent WO2004047755 A2, 20031121, 2004.
(b) Fox, B. M.; Sugimoto, K.; Iio, K.; Yoshida, A.; Zhang,
J.; Li, K.; Hao, X.; Labelle, M.; Smith, M.-L.; Rubenstein, S.
M.; Ye, G.; McMinn, D.; Jackson, S.; Choi, R.; Shan, B.;
Ma, J.; Miao, S.; Matsui, T.; Ogawa, N.; Suzuki, M.;
Kobayashi, A.; Ozeki, H.; Okuma, C.; Ishii, Y.; Tomimoto,
D.; Furakawa, N.; Tanaka, M.; Matsushita, M.; Takahashi,
M.; Inaba, T.; Sagawa, S.; Kayser, F. J. Med. Chem. 2014,
57, 3464.
MS (ESI+): m/z = 489.1 (M + 1).
Anal. Calcd for C25H27F3N4O3: C, 61.47; H, 5.57; N, 11.47. Found:
C, 61.39; H, 5.59; N, 11.38.
Acknowledgment
We would like to thank the research project teams at Japan Tobacco
and Amgen who were instrumental in the discovery and preclinical
development of compound 1.
(6) Li, K.; Fox, B. M.; Zhang, J.; Iio, K.; Suzuki, M.; Soichiro,
I.; Ozeki, H.; Kobayashi, A.; Ma, J.; Hao, A.; Labelle, M.;
McMinn, D.; Smith, M.; Sagawa, S.; Furukawa, N.; Inaba,
T.; Matsushita, M.; Matsui, T.; Ogawa, N.; Okuma, C.;
Sugimoto, K.; Tanaka, M.; Yoshida, A.; Ishii, Y.;
Tomimoto, D.; Jackson, S.; Choi, R.; Shan, B.; Srivastava,
A.; Kayser, F. Abstracts of Papers, 245th ACS National
Meeting & Exposition, New Orleans, April 7–11, 2013;
American Chemical Society: Washington DC, 2013, MEDI–
410.
References
(1) Cases, S.; Smith, S. J.; Zheng, Y.-W.; Myers, H. M.; Lear, S.
R.; Sande, E.; Novak, S.; Collins, C.; Welch, C. B.; Lusis, A.
J.; Erickson, S. K.; Farese, R. V. Jr. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 13018.
(2) (a) Bell, R. M.; Coleman, R. A. Ann. Rev. Biochem. 1980,
49, 459. (b) Lehner, R.; Kuksis, A. Prog. Lipid Res. 1996,
35, 169. (c) Dow, R. L. RSC Drug Discovery Ser. 2012, 27,
215. (d) Cases, S.; Stone, S. J.; Zhou, P.; Yen, E.; Tow, B.;
Lardizabal, K. D.; Voelker, T.; Farese, R. V. Jr. J. Biol.
Chem. 2001, 276, 38870.
(3) (a) Hubbard, B. K.; Enyedy, I.; Gilmore, T. A.; Serrano-Wu,
M. H. Expert Opin. Ther. Pat. 2007, 17, 1331. (b) Zammit,
V. A.; Buckett, L. K.; Turnbull, A. V.; Wure, H.; Proven, A.
Pharmacol. Ther. 2008, 118, 295.
(4) (a) Birch, A. M.; Buckett, L. K.; Turnbull, A. V. Curr. Opin.
Drug Discovery Dev. 2010, 13, 489. (b) Chen Hubert, C.;
Farese Robert, V. Jr. Arterioscl. Thromb. Vasc. Biol. 2005,
25, 482. (c) King, A. J.; Judd, A. S.; Souers, A. J. Expert
Opin. Ther. Pat. 2010, 20, 19.
(7) Davis, A. P.; Egan, T. J.; Orchard, M. G.; Cunningham, D.;
McArdle, P. Tetrahedron 1992, 48, 8725.
(8) (a) Yadav, G. C.; Dorwal, H. N.; Valavala, S.; Sharma, V. K.
Patent WO 2007138435 A2, 20070524, 2007. (b) Kirsch, P.;
Tarumi, K. Angew. Chem. Int. Ed. 1998, 37, 484.
(c) Marvell, E. N.; Sturmer, D.; Rowell, C. Tetrahedron
1966, 22, 861.
(9) Jones, G. Org. React. 1967, 15, 204.
(10) Castro, J.; Moyano, A.; Pericas, M. A.; Riera, A.; Greene, A.
E.; Alvarez-Larena, A.; Piniella, J. F. J. Org. Chem. 1996,
61, 9016.
(11) Haslego, M. L.; Smith, F. X. Synth. Commun. 1980, 10, 421.
(12) Lipshutz, B. H.; Sengupta, S. Org. React. 1992, 41, 135.
(13) Layer, R. W.; MacGregor, I. R. J. Org. Chem. 1956, 21,
1120.
(5) (a) Fox, B. M.; Furukawa, N.; Hao, X.; Iio, K.; Inaba, T.;
Jackson, S. M.; Kayser, F.; Labelle, M.; Li, K.; Matsui, T.;
Synthesis 2014, 46, 3047–3058
© Georg Thieme Verlag Stuttgart · New York