Development of a scalable chiral synthesis of MK-3281, an inhibitor of the hepatitis C virus NS5B polymerase

Article abstract of DOI:10.1055/s-0030-1260790

The development of a scalable chiral synthesis for the HCV NS5B inhibitor MK-3281 is being reported. Several alternative routes were explored and are being described. Georg Thieme Verlag Stuttgart ? New York.

Full text of DOI:10.1055/s-0030-1260790

LETTER
1527
Development of a Scalable Chiral Synthesis of MK-3281, an Inhibitor of the
Hepatitis C Virus NS5B Polymerase
Scalable
C
hira
t
l
Synth
e
e
sis of MK
f
-3281,
a
a
n
Inhibitor
n
of the HC
V
i
S5B
P
a
olymerase Colarusso,*1 Immacolata Conte, Marcello Di Filippo, Caterina Ercolani, Angela C. Mackay,
Maria Cecilia Palumbi, Maria del Rosario Rico Ferreira, Ian Stansfield, Simone Zaramella, Frank Narjes,
Jörg Habermann*1
Department of Medicinal Chemistry, IRBM P. Angeletti S.p.A., Merck Research Laboratories Rome,
Via Pontina km 30.600, 00040 Pomezia (Rome), Italy
Fax +39(06)91093225; E-mail: s.colarusso@irbm.it; E-mail: jhabermann@epo.org
Received 24 February 2011
Our initial route started from phenol 25 and used readily
Abstract: The development of a scalable chiral synthesis for the
available glycidyl nosylates or tosylates to establish the
HCV NS5B inhibitor MK-3281 is being reported. Several alterna-
stereocenter on the benzoxazocine ring. Compound 1 was
obtained in nine steps and 7% overall yield (Scheme 1).
Three steps in the synthetic sequence proved especially
problematic for scale-up. The formation of the benzox-
azocine 4 from epoxide 3 was accomplished only in 55%
isolated yield even after some optimization (unpublished
results). To obtain amine 6, the alcohol was converted into
a tosylate and then reacted with trimethylsilyl azide in the
presence of tetrabutylammonium triphenyldifluorosilicate
(TBAT) to give azide 5.6 Although slow, the azide forma-
tion proceeded with good yield and high enantiomeric ex-
cess, but a severe loss in yield (<50%) was observed when
the reaction was run on a gramscale. Alternative reactions
conditions had failed to give good yields and high enanti-
oselectivity.5 The introduction of the N,N-dimethylamino-
ethyl side chain necessitated preformation of the imine,
and 7 was isolated in only 46% yield. Furthermore, puri-
fication by flash chromatography was required after al-
most every reaction step, and the final product had to be
purified by RP-HPLC and was obtained with an enantio-
meric excess in the range of 92–96%.
tive routes were explored and are being described.
Key words: chiral pool, MK-3281, HCV NS5B polymerase inhib-
itor, medicinal chemistry, heterocycles
About 3% of the world population is chronically infected
with hepatitis C virus (HCV). The virus infects the liver
and as such is a major cause of acute hepatitis and chronic
liver disease, including cirrhosis and liver cancer.2 The vi-
rally encoded nonstructural proteins of the hepatitis C vi-
rus are attractive targets for direct antiviral drug therapy.3
As such, the hepatitis C virus NS5B polymerase is a key
enzyme necessary for the replication of the viral genome
and has recently been the focus of intense research.4 Dur-
ing our research in the field we have identified the indole
derivative MK-3281 (1, Figure 1) as a candidate for pre-
clinical development.5
N
N
(R)
O
A clear improvement in the synthetic efficacy was re-
quired to rapidly produce multigram amounts of material
in high optical purity. In this letter we describe the evolu-
tion of different strategies which led to the successful
multigram synthesis of MK-3281.
HOOC
N
Initially, we explored an ‘achiral’ route from the ketone 8
(Scheme 2), with final resolution of the enantiomers of 1,
to produce a few grams of material in a relatively short
time frame. Alcohol 4 was oxidized to 8 with Dess–Mar-
tin periodinane, and the crude material was used directly.
Two consecutive reductive aminations were necessary to
obtain bisamine 10, since 7 did not react efficiently with
secondary amines. Reduction to the racemic alcohol 4 was
always the major product. Both amines 9 and 10 were
formed cleanly, and this allowed their purification by sim-
ple trituration of their respective hydrochloride salts with
diethyl ether. Final hydrolysis of 10 afforded a material
which was pure enough to be directly subjected to enanti-
omeric separation by SFC on a chiral phase. Compound 1
was obtained in 4% overall yield, 99.5% purity and >99%
ee starting from phenol 2. This route allowed for the prep-
aration of 1–2 grams of final compound in a single cycle.
1
Figure 1 HCV NS5B polymerase inhibitor MK-3281
To allow for efficacy studies in different animal models of
HCV infection, and in particular in HCV-infected chim-
panzees, we needed a robust and scalable synthesis of
MK-3281 (1) which could afford multigram amounts of
material. In this letter we describe the evolution of differ-
ent strategies which led to the successful multigram syn-
thesis of MK-3281.
SYNLETT 2011, No. 11, pp 1527–1532
Advanced online publication: 15.06.2011
DOI: 10.1055/s-0030-1260790; Art ID: B05011ST
© Georg Thieme Verlag Stuttgart · New York
x
x
.x
x
.2
0
1
1

1528
S. Colarusso et al.
LETTER
O
O
HO
H
(S)-glycidyl nosylate,
CsF, DMF
Cs2CO3, dioxane,
100 °C
H
MeOOC
MeOOC
N
N
82%
55%
2
3
HO
N3
1) TosCl, pyridine
92%
O
O
MeOOC
MeOOC
N
N
H2, Pd/C, MeOH
2) TMSN3, TBAT,
THF, 70 °C
84%
4
5
Boc
H2N
N
NH
N
H
N-Boc-aminoacetaldehyde,
HC(OMe)3; NaCNBH3,
AcOH, MeOH
O
O
MeOOC
MeOOC
N
1
46%
6
7
Scheme 1 The oxirane route to MK-3281
This route, although overall not higher yielding, was fast- final separation of the enantiomers and the loss of half of
er than our initial synthesis, but still not applicable for the the material.
preparation of 1 in a scale >10 g. This was partially due to
For our second-generation approach we thought of replac-
low throughput of the SFC apparatus available. Further-
ing the epoxide with two chiral nitrogen-containing build-
more, the introduction of the amine side chain looked still
ing blocks, which would already contain the amine in the
desired configuration and lead to key intermediate 6 in a
more convergent fashion.
cumbersome and proceeded with loss of material. We also
had not resolved the low yield in the formation of the
medium-sized ring in 4. Clearly, we also wanted to avoid
O
1) N,N-dimethylethylenediamine,
NaBH(OAc)3, AcOH, DCE
2) HCl, Et2O
O
MeOOC
N
DMP, CH2Cl2
96%
4
72% over two steps
8
N
N
* 2 HCl
* 2 HCl
NH
N
1) KOH, dioxane–H2O
1) HCHO (37% in H2O)
NaBH3CN, NaOAc, CH2Cl2
2) HCl, Et2O
2) SFC separation (Chiralpak AD-H,
35% i-PrOH + 0.4% DEA)
3) ISOLUTE C18
O
O
MeOOC
MeOOC
N
N
1
51% over two steps
23% over three steps
9
10
Scheme 2 Achiral synthesis of MK-3281
Synlett 2011, No. 11, 1527–1532 © Thieme Stuttgart · New York

LETTER
Scalable Chiral Synthesis of MK-3281, an Inhibitor of the HCV NS5B Polymerase
1529
1) NaBH4, CaCl2,
THF–EtOH (1:1), 0 °C
2) NosCl, Et3N,
Cbz
NHCbz
OH
2,2-dimethoxypropane
p-TsOH, toluene
N
Cbz
O2N
DMAP, CH2Cl2
N
O
O
O
O
71%
O
80%
S
O
O
O
O
11
12
13
Scheme 3 Ex-chiral-pool intermediate 13 from D-serine
Oxazolidine nosylate 13 was obtained in three steps start- Since the iodide showed poor shelf stability, we preferred
ing from commercially available Cbz-protected D-serine the bromide. Treatment of 19 with sodium hydride in
methyl ester 11.7 The intermediate oxazolidine 12 was re- DMF resulted cleanly in formation of the eight-membered
duced to the alcohol with in situ formed calcium boro- ring. After cleavage of the Cbz group amine 6 was ob-
hydride, which was converted into nosylate 13 tained in 62% isolated yield. When the aziridine interme-
(Scheme 3).8,9 The reaction sequence proved to be fairly diate 17 was employed (Scheme 5, route B), the
robust and high yielding (57% over three steps).
tetracyclic intermediate 20 was formed smoothly in a one-
pot process in high yield from phenol 2, using CsF in
DMF for the alkylation, followed by addition of potassi-
um tert-butoxide to the reaction mixture to induce cycliza-
tion. Acidic cleavage of the Boc group gave amine 6 in
85% isolated yield. The optical purity of amine 6 was sat-
isfactory for both routes, as evidentiated by the high ee
(>99%) obtained for compound 1.
The synthesis of aziridine nosylate 17 required five steps
starting from Boc-protected L-serine (Scheme 4), which
after treatment with TBDMS chloride and ester reduction
led to alcohol 15. The alcohol was then treated under Mit-
sonobu conditions with diisopropyl azodicarboxylate and
triphenylphosphine, which allowed the ring closure to the
protected aziridine 16.10 Cleavage of the silicon protective
group and activation of the alcohol as nosylate gave inter- The use of aziridine 17 was clearly more efficient both in
mediate 17. The deprotection–activation sequence needed terms of yield (85% vs. 36%), number of steps, and the
to be carried out with strict temperature control to avoid number of chromatographic purifications required (1 vs.
opening of the aziridine. Both products 13 and 17 exhibit- 3) and was therefore judged more amenable to produce
ed reasonable storage stability, when kept in a freezer.
large quantities of key intermediate 6.
Relying on this methodology, the remaining steps for the
synthesis of 1 were accomplished as described in
Scheme 6. Amine 6 was formylated with 2,2,2-trifluoro-
ethyl formate11 and selectively reduced in situ with borane
dimethylsulfide complex giving the monomethylated
amine 21 in 80% yield as its hydrochloride salt. The use
of this methodology avoided impurities such as dimethy-
lation of the amine, which had been observed using aque-
ous formaldehyde. Reductive amination with N-Boc-
aminoacetaldehyde was carried out under hydrogenation
conditions at elevated temperature, which was superior
when compared with classical reductive amination condi-
tions using complex hydrides. Acidic cleavage of the Boc
group from 22 delivered 23, which was purified by crys-
tallization of the hydrochloride salt from diethyl ether.
Reductive amination with formaldehyde gave 24, which
was then subjected to basic hydrolysis of the ester. After
workup, compound 112 was purified by trituration of the
hydrochloride salt with boiling acetonitrile.
1) TBDMS-Cl, imidazole, CH2Cl2
NHBoc
OH
2) NaBH4, CaCl2,
THF–EtOH (1:1)
NHBoc
OTBDMS
O
HO
72%
15
O
14
1) TBAF, THF–Et2O (1:1)
2) nosyl chloride, Et3N,
DMAP, CH2Cl2, 0 °C
Boc
N
DIAD, Ph3P,
THF–MeCN (6:1)
68%
57%
OTBDMS
16
Boc
O
N
O
S
NO2
O
17
Scheme 4 Ex-chiral-pool intermediate 17 from L-serine
Both chiral building blocks were then employed in the
second-generation chiral routes to key intermediate 6.
Phenol 2 was thus reacted with oxazolidine 13 (Scheme 5,
route A) in the presence of cesium fluoride as base, and
the oxazolidine subsequently opened under acidic condi-
tions to afford alcohol 18. Activation of 18 for the ring
closure using nosyl chloride gave unsatisfactory results
due to the extremely sluggish nosylate formation. The
transformation into the corresponding iodide (with iodine,
imidazole, and triphenylphosphine in toluene–acetoni-
trile, 74%) or the bromide 19 (under Appel conditions
with carbon tetrabromide and triphenylphosphine in
dichloromethane) resulted in clean product formation.
The synthesis was concluded in about 40% overall yield
starting from phenol 2. Only four chromatographic steps
were required to afford the final product which was ob-
tained in excellent purity (>99.5%) and enantiopurity (ee
>99%). A single synthetic run produced 10 g of interme-
diate 20 whilst final hydrolysis was feasible also on a
20 g scale with normal lab equipment.
In summary, we have accomplished the development of a
robust, scalable, and high-yielding route to the hepatitis C
virus NS5B-polymerase inhibitor MK-3281. We have
been able to improve synthetic efficiency using readily
Synlett 2011, No. 11, 1527–1532 © Thieme Stuttgart · New York

1530
S. Colarusso et al.
LETTER
Cbz
Cbz
HO
Br
HN
O
HN
O
1) 13, CsF,
DMF,45 °C
2) HCl, MeOH
H
N
CBr4, Ph3P,
K2CO3, CH2Cl2
H
N
MeOOC
MeOOC
68–82%
77%
route A
18
19
HO
H
1) NaH, DMF
2) H2, Pd/C,
EtOAc–MeOH
MeOOC
N
62%
2
BocHN
N
H2N
route B
O
O
1) 17, CsF,
MeOOC
MeOOC
N
DMF, 30 °C
2) KOt-Bu, DMF
CH2Cl2–TFA
quant.
85%
20
6
Scheme 5
HN
N
N-Boc-aminoacetaldehyde,
NaOAc, AcOH,
H2, Pd/C, MeOH, 55 °C
O
1) 2,2,2-trifluoroethyl formate,
THF
MeOOC
6
82%
2) BH3⋅SMe2, THF
79%
21
H
* 2 HCl
O
N
N
N
H2N
Boc
MeOOC
O
HCHO (37% in H2O),
NaOAc, H2, Pd/C, MeOH
1) TFA, CH2Cl2
2) HCl, Et2O
MeOOC
N
N
quant.
80%
22
23
N
N
1) NaOH, MeOH–
THF–H2O (1:1:0.8)
2) HCl, MeOH
O
MeOOC
N
1
91%
24
Scheme 6
Synlett 2011, No. 11, 1527–1532 © Thieme Stuttgart · New York

LETTER
Scalable Chiral Synthesis of MK-3281, an Inhibitor of the HCV NS5B Polymerase
1531
sat. aq NaHCO3 solution and with brine, then dried over
Na2SO4. After filtration, the volatiles were evaporated in
vacuo to leave a colorless oil. The material was supported on
silica gel and loaded on top of 1 kg of silica gel on a sintered
frit. of CH2Cl2 (5 L) was allowed to flow through, then the
product was eluted with of EtOAc (5 L). The solvent was
evaporated in vacuo leaving 15 as clear oil (66 g, 72%). 1H
NMR (400 MHz, CDCl3): d = 5.15 (br s, 1 H), 3.89–3.78 (m,
3 H), 3.75–3.63 (m, 2 H), 2.71 (br s, 1 H), 1.47 (s, 9 H), 0.92
(s, 9 H), 0.1 (s, 6 H).
available chiral starting materials. In the course of our
studies we have been able to greatly improve the overall
yield and to minimize notably the purification effort. The
final product was obtained with chemical and optical pu-
rity. Our studies were a useful starting point for the devel-
opment of a process route to the compound, which has
been object of a recent publication.13
Acknowledgment
Preparation of 16
A solution of Ph3P (30.9 g, 118 mmol) in THF–MeCN (9:1,
1160 mL) was treated dropwise over 15 min with DIAD
(22.91 mL, 118 mmol) at 0 °C. After 15 min, a solution of 15
(24 g, 79 mmol) in THF (120 mL) was added dropwise over
30 min. A white solid formed upon addition. The ice–water
bath was removed and stirring continued overnight at r.t. 1H
NMR control showed formation of product and consumption
of the starting material after 24 h and represented the most
unambiguous way to assess completeness of the reaction. All
volatiles were evaporated in vacuo, and the residual material
was treated with PE–CH2Cl2 (95:5) to precipitate most of the
triphenylphosphine oxide. The precipitate was filtered off
and washed with small portions of the same solvent mixture
until no more desired product was detectable by TLC in the
filtrate. The filtrate was evaporated in vacuo leaving ca. 40 g
of crude product. Column chromatography on 600 g silica
gel (PE–EtOAc = 95:5, 6 L) afforded after evaporation of the
eluent 16.05 g (68%) of 16. 1H NMR (400 MHz, CDCl3):
d = 3.81 (dd, 1 H, J = 4.73, 11.39 Hz), 3.63 (dd, 1 H,
J = 4.87, 11.39 Hz), 2.61–2.54 (m, 1 H), 2.26 (d, 1 H,
J = 6.08 Hz), 2.07 (d, 1 H, J = 3.59 Hz), 1.46 (s, 9 H), 0.92
(s, 9 H), 0.09 (s, 3 H), 0.08 (s, 3 H).
We gratefully thank Anna Alfieri and Silvia Pesci for analytical and
NMR support.
References and Notes
(1) New address: Stefania Colarusso, IRBM Science Park Srl,
Via Pontina km. 30.600, 00040 Pomezia (Rome), Italy; Jörg
Habermann, European Patent Office, Bayerstrasse 34,
80335 München, Germany.
(2) (a) Cohen, J. Science 1999, 285, 26. (b) Lauer, G. M.;
Walker, B. D. N. Engl. J. Med. 2001, 345, 41. (c) WHO
Factsheet No. 164: Hepatitis C, accessed 09 May 2007, http:/
/www.who.int/mediacentre/factsheets/fs164/en/print.html.
(d) Brown, R. S. Nature (London) 2005, 436, 97.
(3) (a) Gordon, C. P.; Keller, P. A. J. Med. Chem. 2005, 48, 1.
(b) De Francesco, R.; Migliaccio, G. Nature (London) 2005,
463, 953. (c) Kwong, A. D.; McNair, L.; Jacobsen, I.;
George, S. Curr. Opin. Pharmacol. 2008, 8, 522.
(4) (a) Beaulieu, P. L. Expert Opin. Ther. Pat. 2009, 19, 145.
(b) Koch, U.; Narjes, F. Curr. Top. Med. Chem. 2007, 7,
1302.
(5) Narjes, F.; Crescenzi, B.; Ferrara, M.; Habermann, J.;
Colarusso, S.; Rico Ferreira, Md. R.; StansfieldI, ; Mackay,
A.; Conte, I.; Ercolani, C.; Zaramella, S.; Palumbi, M.-C.;
Meuleman, P.; Leroux-Roels, G.; Giuliano, C.; Fiore, F.; Di
Marco, S.; Baiocco, P.; Koch, U.; Migliaccio, G.; Altamura,
S.; Laufer, R.; De Francesco, R.; Rowley, M. J. Med. Chem.
2011, 54, 289.
(6) Soli, E. D.; Manoso, A. S.; Patterson, M. C.; DeShong, P.;
Favor, D. A.; Hirschmann, R.; Smith, A. B. III. J. Org.
Chem. 1999, 64, 3171.
(7) Garner, P.; Park, J. M. J. Org. Chem. 1987, 52, 2361.
(8) Narasimhan, S.; Prasad, K. G.; Madhavan, S. Synth.
Commun. 1995, 25, 1689.
(9) Mahler, G.; Serra, G.; Manta, E. Synth. Commun. 2005, 35,
1481.
(10) (a) Ho, M.; Chung, J. K. K.; Tang, N. Tetrahedron Lett.
1993, 34, 6513. (b) Travins, J. M.; Etzkorn, F. A.
Tetrahedron Lett. 1998, 39, 9389.
Preparation of 17
A solution of 16 (25 g, 87 mmol) in THF (250 mL) and Et2O
(250 mL) was cooled to 0 °C and treated dropwise over 20
min with 1 M TBAF in THF (91 mL, 91 mmol). The
resulting solution was stirred at 0 °C for 30 min. The
reaction mixture was quenched by the addition of sat. aq
NaHCO3 solution (500 ml) and extracted into Et2O–PE (4:1,
300 mL + 200 mL). The organic layers were collected,
washed with brine, dried over Na2SO4, and after filtration
evaporated in vacuo, keeping the water bath at r.t. The
residual oil (ca. 42 g) was dissolved in dry CH2Cl2 (500 mL)
and Et3N (15.8 mL, 113 mmol) and cooled to 0 °C. DMAP
(1.06 g, 8.7 mmol) was added, followed by 4-nitro-
benzenesulfonyl chloride (21.2 g, 96 mmol). The resulting
orange heterogeneous mixture was stirred at r.t. overnight.
After dilution with CH2Cl2 the mixture was washed with sat.
aq NaHCO3 solution, H2O, and brine. After drying over
Na2SO4, filtration, and evaporation in vacuo a residue was
obtained which was purified by chromatography (PE–
EtOAc, 8:2, 750 g silica gel) to afford 19.89 g (57%) of 17
as an off-white solid. 1H NMR (400 MHz, CDCl3): d = 8.48–
8.44 (m, 2 H), 8.22–8.18 (m, 2 H), 4.37 (dd, 1 H, J = 4.20,
11.30 Hz), 3.99 (dd, 1 H, J = 7.36, 11.30 Hz), 2.78–2.71 (m,
1 H), 2.30 (d, 1 H, J = 6.42 Hz), 2.05 (d, 1 H, J = 3.58 Hz),
1.35 (s, 9 H).
(11) Hill, D. R.; Hsiao, C.-N.; Kurukulasuriya, R.; Wittenberger,
S. J. Org. Lett. 2002, 4, 111.
(12) Experimental Procedure for the Synthesis of 1
Preparation of 15
A suspension of CaCl2 (49.9 g, 450 mmol) in dry THF (400
mL) was cooled to 0 °C, and NaBH4 (34.0 g, 900 mmol) was
added under stirring. To the resulting suspension, a solution
of Boc-Ser(TBDMS)OMe (100 g, 300 mmol) in dry EtOH
(400 mL) was added dropwise over the course of 2 h.
Stirring was then continued over night. The mixture was
poured onto crushed ice (600 g) and a sat. aq NH4Cl solution
(600 mL) was added. This mixture was left stirring until
evolution of gas had ceased, then EtOAc was added. The
phases were separated, the aqueous phase was treated with
1 N HCl to dissolve all material, then extracted twice with
EtOAc. The combined organic phases were extracted with
Preparation of 20
A solution of 25 (19 g, 54.4 mmol) in DMF (360 mL) was
treated with CsF (33.0 g, 218 mmol) in one portion. The
resulting fluorescent yellow mixture was stirred 20 min at r.t.
then a solution of 17 (24.9 g, 69.6 mmol in 130 mL of DMF)
was added dropwise over 30 min. The resulting orange clear
solution was stirred at 30 °C overnight. The reaction mixture
was cooled to 0 °C, and powdered KOt-Bu (8.54 g, 76
mmol) was added slowly to the reaction mixture. After 1.5 h
Synlett 2011, No. 11, 1527–1532 © Thieme Stuttgart · New York

1532
S. Colarusso et al.
LETTER
the reaction was quenched by the addition of sat. aq NH4Cl
solution and the product extracted into EtOAc. The
combined organic layers were washed with H2O and brine,
dried over Na2SO4, filtered, and evaporated in vacuo. The
residual material was purified by chromatography (PE–
EtOAc, 80:20, 600–700 g silica gel) affording 23.2 g (85%)
of 20. C30H36N2O5; MS (ES+): m/z = 527 [M + Na]+. 1H
NMR (400 MHz, DMSO-d6): d = 8.36 (s, 1 H), 7.93–7.86
(m, 1 H), 7.73–7.62 (m, 1 H), 7.58–7.46 (m, 1 H), 7.35–7.11
(m, 3 H), 4.40–4.25 (m, 2 H), 3.90–3.65 (m, 6 H), 2.75–2.64
(m, 1 H), 2.03–1.28 (m, 20 H).
1.32 (m, 3 H), 1.37 (s, 9 H).
Preparation of 23
TFA (120 mL) was added at r.t. to a solution of 22 (16.4 g,
15.3 mmol) in dry CH2Cl2 (370 mL), and the mixture was
stirred for 1.5 h. Evaporation to dryness gave a residue that
was dissolved in CH2Cl2 (450 mL), treated with 2 M HCl–
Et2O (65 mL) and evaporated again. This treatment was
repeated twice. The residue obtained was dried under high
vacuum for 2 h. The light brown powder (23, 13.3 g, 87%)
was used as such. C28H35N3O2·2HCl; MS (ES+): m/z = 462
[M + H]+.
Preparation of 6
Preparation of 24
Compound 20 (14 g, 27.7 mmol) was treated with CH2Cl2–
TFA (220 mL, 9:1) at r.t. for 1 h. The reaction was diluted
with CH2Cl2, and sat. aq NaHCO3 was slowly added. The
mixture was extracted exhaustively with CH2Cl2. The
combined organic phases were washed with brine, dried over
Na2SO4, and filtered. After evaporation in vacuo 11.21 g
(quant.) of off-white 6 were obtained which were used
without further purification. C25H28N2O3; MS (ES+):
m/z = 405 [M + H]+.
NaOAc (6.15 g, 75 mmol) and 37% aq HCHO (4.3 mL, 57.5
mmol) were added to a stirred solution of 29 (13.36 g, 25
mmol) in dry MeOH (200 mL), and the mixture was stirred
at r.t. for 10 min. Pd/C (10 mol%, 2.66 g, 2.5 mmol) was
added portionwise. The mixture was thoroughly degassed
and H2 atmosphere (balloon) was applied. The reaction was
stirred for 20 h at r.t. The catalyst was filtered off over Celite,
and the filtrate was concentrated in vacuo. The residual
material was dissolved in EtOAc, washed with sat. aq
NaHCO3 and brine, dried over Na2SO4, and concentrated in
vacuo to give 29 g of crude product. Chromatography (silica
gel, EtOAc–MeOH–Et3N = 100:6:2) afforded a main
fraction that was triturated with Et2O–pentane. After
filtration a pale yellow solid was obtained which was dried
in vacuo to yield pure methyl ester 24 (9.8 g, 80%).
C30H39N3O3; MS (ES+): m/z = 490 [M + H]+. 1H NMR (400
MHz, DMSO-d6): d = 8.12 (s, 1 H), 7.90 (d, 1 H, J = 8.48
Hz), 7.68 (d, 1 H, J = 8.48 Hz), 7.58–7.51 (m, 1 H), 7.36–
7.26 (m, 3 H), 4.56 (d, 1 H, J = 14.96 Hz), 4.29 (dd, 1 H,
J = 4.91, 12.06 Hz), 4.05–3.96 (m, 1 H), 3.87 (s, 3 H), 3.76–
3.67 (m, 1 H), 3.07–2.98 (m, 1 H), 2.81–2.63 (m, 3 H), 2.36
(s, 3 H), 2.38–2.32 (m, 2 H), 2.17 (s, 6 H), 2.04–1.64 (m, 6
H), 1.60–1.12 (m, 4 H).
Preparation of 21
A solution of 6 (18.4 g, 45 mmol) in THF (131 mL) was
treated dropwise with 2,2,2-trifluoroethyl formate (7 mL, 55
mmol) and stirred overnight at r.t. All volatiles were
evaporated in vacuo, and the residual material was dissolved
in THF (400 mL). Borane dimethylsulfide complex in THF
(114 mL, 228 mmol) was added dropwise. The resulting
yellow solution was stirred at r.t. for 20 h. The reaction was
quenched by the careful addition of HCl–MeOH (1.25 M,
100 mL), and the resulting solution was heated in an open
flask for 2 h to destroy all the boron adducts and remove
B(OMe)3. All remaining volatiles were then evaporated in
vacuo. The residual material was partitioned between sat. aq
NaHCO3 and EtOAc. The combined organic layers were
washed with brine, dried over Na2SO4, and evaporated to
afford a residue which was purified by chromatography
(EtOAc–PE, 8:2 + 1% Et3N). After evaporation of the
eluents in vacuo 21 was obtained as a colorless solid (14.9 g,
79%). C26H30N2O3; MS (ES+): m/z = 419 [M + H]+. 1H NMR
(400 MHz, DMSO-d6): d = 8.15 (s, 1 H), 7.89 (d, 1 H,
J = 8.49 Hz), 7.68 (dd, 1 H, J = 1.35, 8.49 Hz), 7.55–7.49
(m, 1 H), 7.34–7.24 (m, 3 H), 4.51 (d, 1 H, J = 15.23 Hz),
4.27 (dd, 1 H, J = 4.33, 12.13 Hz), 3.88 (s, 3 H), 3.73 (dd, 1
H, J = 8.42, 12.13 Hz), 3.50 (dd, 1 H, J = 10.28, 15.23 Hz),
2.87–2.77 (m, 1 H), 2.74–2.63 (m, 1 H), 2.48 (s, 3 H), 2.14–
1.50 (m, 8 H), 1.44–1.10 (m, 3 H).
Preparation of 1
Freshly prepared 1 M aq NaOH (85 mL, 85 mmol) was
added dropwise under N2 atmosphere to a solution of 24
(20.38 g, 41.6 mmol) in THF (100 mL) and MeOH (100
mL), and the mixture was stirred at 60 °C for 4 h. The
mixture was concentrated in vacuo to ca. 10% of its volume,
then H2O (200 mL) was added, followed by dropwise
addition of 1 N aq HCl (85 mL). The aqueous layer was
extracted with n-BuOH (1 × 500 mL, then 2 × 250 mL), and
the organic layer was washed twice with small amounts of
ice-cold water, dried over Na2SO4, and concentrated to give
a residue which was dissolved in CH2Cl2 and filtered.
Evaporation to dryness afforded a residue that was taken into
MeCN, and 1 N aq HCl (100 mL) was added. The resulting
solution was evaporated to dryness, and this operation was
repeated twice, using 40 mL of 1 N HCl each time. The
material was dried under high vacuum overnight to remove
H2O, then was triturated with hot MeCN, and filtered to give
20.8 g (91%) of 1 (bishydrochloride salt, 20.8 g, >99% ee,
99.2% purity). C29H37N3O3·2HCl. MS (ES+): m/z = 476 [M
+ H]+; [a]D20 +55.5 (c 1.0, MeCN:H2O, 1:1). 1H NMR (free
base, 400 MHz, DMSO-d6): d = 8.11 (s, 1 H), 7.86 (d, 1 H,
J = 8.45 Hz), 7.66 (d, 1 H, J = 8.45 Hz), 7.56–7.49 (m, 1 H),
7.34–7.25 (m, 3 H), 4.54 (d, 1 H, J = 14.69 Hz), 4.28 (dd, 1
H, J = 4.62, 12.27 Hz), 4.00 (dd, 1 H, J = 9.65, 12.27 Hz),
3.71 (dd, 1 H, J = 10.06, 15.09 Hz), 3.08–2.97 (m, 1 H),
2.81–2.63 (m, 3 H), 2.43–2.37 (m, 2 H), 2.37 (s, 3 H), 2.20
(s, 6 H), 2.04–1.64 (m, 6 H), 1.58–1.09 (m, 5 H). Anal. Calcd
for C29H39Cl2N3O3: C, 63.50; H, 7.17; N, 7.66. Found: C,
63.04; H, 6.89; N, 7.24.
Preparation of 22
To a solution of Boc-amino acetaldehyde (5.7 g, 35.5 mmol)
in dry MeOH (93 mL) was added a mixture of 21 (14.9 g,
35.5 mmol), AcOH (4.1 mL, 71.0 mmol), and NaOAc (2.9
g, 35.5 mmol) in dry MeOH (260 mL), and the mixture was
stirred at r.t. for 15 min. Then Pd/C (10 mol%, 5.67 g, 5.34
mmol) was added, the mixture was degassed thoroughly, and
H2 atmosphere was applied. The resulting mixture was
stirred under H2 atmosphere at 63 °C for 5 h, then at 55 °C
overnight. The mixture was cooled to r.t., degassed, and
flushed with argon. Filtration through Celite with MeOH
and EtOAc as solvent afforded after evaporation in vacuo a
residue (26.9 g) that was purified by chromatography (PE–
EtOAc = 2.5:1 to 1.5:1) to give 22 (16.4 g, 82%).
C33H43N3O5; MS (ES+): m/z = 562 [M + H]+. 1H NMR (400
MHz, CDCl3): d = 8.11 (s, 1 H), 7.90 (d, 1 H, J = 8.36 Hz),
7.68 (d, 1 H, J = 8.36 Hz), 7.57–7.51 (m, 1 H), 7.34–7.26 (m,
3 H), 6.70–6.63 (m, 1 H), 4.56 (d, 1 H, J = 14.78 Hz), 4.26
(dd, 1 H, J = 4.86, 12.26 Hz), 3.97 (dd, 1 H, J = 9.92, 12.26
Hz), 3.88 (s, 3 H), 3.74–3.62 (m, 1 H), 3.06–2.94 (m, 3 H),
2.75–2.56 (m, 3 H), 2.41 (s, 3 H), 2.01–1.63 (m, 7 H), 1.44–
(13) Scott, J. P.; Alam, M.; Bremeyer, N.; Goodyear, A.; Lam, T.;
Wilson, R. T.; Zhou, G. Org. Process Res. Dev. 2011, 15, in
press; DOI: 10.1021/op200002u.
Synlett 2011, No. 11, 1527–1532 © Thieme Stuttgart · New York