Practical synthesis of a peptidomimetic thrombin inhibitor

Article abstract of DOI:10.1016/j.tet.2009.01.078

Compound 1 is a low molecular weight thrombin inhibitor developed for treatment of deep vein thrombosis and cardiovascular diseases. We herein report our efforts to develop a robust, efficient and reproducible process suitable for large-scale synthesis of

Full text of DOI:10.1016/j.tet.2009.01.078

Tetrahedron 65 (2009) 2689–2694
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Practical synthesis of a peptidomimetic thrombin inhibitor
*
´
Serge Wilmouth , Christel Bufferne-Perret, Christine Flouzat, Gregoire Humblot,
Christiane Ray, Nadine Simbille
Schering-Plough France, Preclinical Development Center, 20 rue Henri Goudier, B.P. 140, 63203 Riom Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Compound 1 is a low molecular weight thrombin inhibitor developed for treatment of deep vein
thrombosis and cardiovascular diseases. We herein report our efforts to develop a robust, efficient and
reproducible process suitable for large-scale synthesis of compound 1.
Received 12 December 2008
Received in revised form 15 January 2009
Accepted 16 January 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Available online 23 January 2009
Keywords:
Low molecular weight thrombin inhibitor
Peptidomimetic
a
-Hydroxy amide oxidation
Impurity identification
Process development and scale-up
1. Introduction
development. The medicinal chemistry route (Scheme 1) goes
through two building blocks 3 and 8, which are readily available
Compound 1 is a low molecular weight thrombin inhibitor with
a high selectivity ratio in the inhibition of factor Xa.1 The compound
is a peptidomimetic containing a transition state analogue, which
forms a complex with Ser195 of thrombin, inducing prolonged
coagulation time and inhibition of platelet aggregation.2 Developed
for treatment of deep vein thrombosis and arterial indications, the
hydrochloride salt of 1 was selected development candidate as
a racemate, since the active (S)-enantiomer rapidly racemizes in
solution. A hydrogen/deuterium exchange experiment performed
from commercially available raw materials. Intermediate 3 is syn-
thesized in three steps from Z-Lys(Boc)OH 4 by using known in-
house procedure,3 while intermediate 5 is synthesized in three
steps from 2-hydroxy-6-methyl-pyridine-3-carboxylic acid
2
according to the literature prescription.4
2. Results and discussions
The process starting from building blocks 3 and 5 as described in
Scheme 1, was not deemed suitable for larger-scale production for
several reasons: (a) the Dess–Martin reagent must be avoided on
at pH 7.4 for 1 day shows that 72% of
aC–H is converted into aC–D.
large scale because of safety concerns; (b) aminolysis of
a-enol
NH2
O
ester 6 upon treatment with ammonia is not suitable for scaling-up.
The reaction is hampered by a major solubility issue, which may
result in stirring problem; (c) the reverse phase chromatography
prescription to purify crude material 1 lacks robustness for an ap-
plication to large scale; (d) the overall yield of this process over the
last seven steps (12%) is rather low; (e) moreover, preliminary de-
velopment work raises an additional issue: Synthesis of in-
termediate 6 by using described reaction conditions resulted in the
formation of two main impurities, 9 and 10 (Fig. 1).
Impurity 9, identified as a bis-hydroxy derivative is formed by
reaction of 6 with excess free amine 5. Compound 10 a bis-sul-
fonamide impurity, originates from the esterification of 3 with
alcohol 6. These impurities amount up to 30% in the reaction
mixture and any efforts failed to wash them from hydroxy ester 6.
O
O
S
O
N
NH2
N
H
N
H
O
O
(+/-)-1.HCl
We herein report our process development efforts to scale-up an
alternative route of synthesis of compound 1 suitable for further
* Corresponding author. Tel.: þ33 (0)473 334 994; fax: þ33 (0)473 333 934.
E-mail address: serge.wilmouth@spcorp.com (S. Wilmouth).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.01.078

2690
S. Wilmouth et al. / Tetrahedron 65 (2009) 2689–2694
O
N
NH
HO
O
S
OH
N
H
O
NHBoc
2
O
O
3
O
O
O
a
NHBoc
b
NHBoc
O
O
S
N
O
N
H
N
H
O
O
OH
6
O
CbzHN
O
H2N
OH
4
5
OH
NHBoc
NH2
NH2
NHBoc
NH2
O
O
O
O
O
O
d, e, f
c
O
O
O
O
S
N
S
N
O
O
S
N
N
H
N
H
NH2
N
N
H
N
N
H
H
H
O
OH
O
O
O
O
1
7
8
Scheme 1. Medicinal chemistry route. Reagents and conditions: (a) N-(3-dimethylaminopropyl)-N0-ethylcarbodiimine (EDC), hydroxybenzotriazole (HOBt), N,N-diisopropyl-
ethylamine, DMF; (b) methanol, NH3; (c) Dess–Martin periodinane, dichloromethane; (d) TFA, dichloromethane; (e) phosphate buffer pH 8; (f) (1) purification by reverse phase
chromatography, (2) ion exchange chromatography.
A first alternative route was sought to circumvent the issues
described above. This approach (Scheme 2) makes the synthesis
route more convergent, as three steps instead of five remains after
peptidic coupling reaction.
When 11 was submitted to the Medicinal Chemistry procedure
for amide formation (15N ammonia solution in MeOH or 15N
methanolic ammonia solution), the solubility issue was not en-
9 (MIM 867)
NHBOC NHBOC
countered, but a major impurity identified by NMR as the oxazo-
lidinone structure 15 (Scheme 3) was detected up to 20%.
Formation of 15 can be rationalized by intramolecular cyclization of
the hydroxyl with the carbamate protecting group.
O
O
O
O
S
O
N
OMe
N
H
N
H
N
H
O
OH
OH
Removal of this impurity by crystallization from diisopropyl-
ether gave 12 in 81% yield and 99% HPLC purity. In addition, the
reaction proved to be sluggish and doesn’t go to completion even
after three days. Prolonged reaction time and processing time se-
riously limit the throughput of this step. A modified Weinreb pro-
cedure (AlMe3, NH3, CH2Cl2/THF)5 was tested to speed up the
reaction, which has resulted in a complete conversion within 12 h
at 40 ꢀC, without side product formation. Presumably, ammonia
trapped as an aluminate complex act as a deactivated nucleophile,
10 (MIM 927)
NHBOC
O
O
O
S
O
N
OMe
N
N
H
H
NHBoc
O
O
NHBoc
O
NH3
O
N
O
MeOH
HN
O
NH2
CbzHN
O
OMe
NH
S
O
O
O
OH
15
5
Figure 1. Identified impurities.
Scheme 3. Formation of impurity 15.
NHBoc
O
NHBoc
NHBoc
NHBoc
O
O
O
CbzHN
CbzHN
CbzHN
OMe
NH2
NH2
H2N
NH2
OH
OH
O
O
11
12
13
14
NHBoc
O
O
O
O
O
S
N
+
N
H
OH
O
O
S
N
N
H
N
H
NH2
O
O
O
3
8
Scheme 2. First alternative route.

S. Wilmouth et al. / Tetrahedron 65 (2009) 2689–2694
2691
which is unable to promote the cyclization. Compound 12 was
isolated in very high purity (>99%) by precipitation in water and
a single crystallization from isopropylether.
was produced next to two main products (Scheme 5). The struc-
tures of 19 (MIM 510, 5%) and 20 (MIM 508, 75%) were elucidated
by LC–MS and NMR. The reaction leading to 19 and 20 can be
explained by dimerization of amino ketone 14 via diimine forma-
tion, and afterwards Pd catalyzed aromatization.
With the hydroxyl amide 12 available, we turned our attention
to the key oxidation step. The choice for the mild Dess–Martin re-
agent in Medicinal Chemistry originates from the intention to
synthesize a chiral product. But with the aim to deliver a racemate,
the oxidation methods available are not limited to mild reagents.
Based on good in-house experience, our first choice was to use
the nitroxyl radical TEMPO as a catalyst in a biphasic reaction
mixture of dichloromethane and aqueous sodium hypochlorite as
co-oxidant.6 A single product was isolated in 77% yield and 95%
purity. Amazingly, analysis by LC–MS and NMR showed that the
isolated product was the hydantoin compound 16 rather than the
desired keto amide 13. We reasoned that the presence of bromide
in the reaction mixture promotes a Hoffman type rearrangement,7
giving an isocyanate as an intermediate, which in turn, is trapped
NHBoc
NHBoc
O
O
H2, Pd/C
MeOH
CbzHN
NH2
H2N
NH2
O
O
13
14
+ 14
O
NH2
NH
O
NH2
N
NHBoc
NHBoc
by the nitrogen in
(Scheme 4, (i)).
d-position leading to the hydantoin structure 16
HN
NHBoc
N
O
NHBoc
O
NH2
NH2
20
(i)
NHBoc
NHBoc
Tempo
O
O
Oxidation
CbzHN
NH2
CbzHN
NH2
OH
O
O
N
NH2
N
12
NHBoc
13
NHBOC
NHBoc
19
O
NH2
NHBoc
Scheme 5. Dimerization impurities 21 and 22.
Hofmann
Rearangement
OH
CbzHN
O
It has been observed that ketone 13 promptly equilibrate to the
hemiacetal form at 0 ꢀC in MeOH. Suspecting this behaviour to
promote dimerization, the reaction was repeated in THF, but again
a mixture of 19 and 20 was isolated. In view of these results and the
evidence that 14 is not stable and promptly dimerizes, it has been
CbzN
N
O
NH
16
O
decided to establish the
reaction (Scheme 6).
a-keto side chain after peptidic coupling
The free amine was then released one step earlier at the stage of
the -hydroxy amide 12 by conventional hydrogenation in meth-
(ii)
NHBoc
NHBoc
NHBn
a
O
O
anol. Amine 21, which isolation proved to be troublesome, was used
as such in the next step. Subsequent peptide coupling reaction was
performed according to Medicinal Chemistry prescription (EDC,
HOBt) by using diisopropylamine as base and DMF as solvent. A
satisfying 91% LC–MS purity was achieved, with as a main con-
taminant esterification product 22 (Fig. 2). Attempts to purify the
coupling product 7 revealed to be more difficult than expected
because of the different behaviour in crystallization of the two di-
astereoisomers (44:56). A first crystallization from ethyl acetate
provided 7 in 30% yield (20:80 mixture of isomers). Recovery of the
remaining product from the mother liquors was accomplished by
two additional crystallizations from ethyl acetate/diisopropylether.
By using this procedure 7 has been isolated in a disappointingly low
67% combined yield. When the reaction was performed at large
scale, the purity of 7 measured by LC–MS dropped to 71%, con-
taminated by three main impurities 22, 23 and 24 (Fig. 2). Impu-
rities 23 and 24 are the result of a polymerization mechanism, by
addition of acid monomers 3 to sulfonamide functionality.
Tempo
Oxidation
CbzHN
CbzHN
NHBn
OH
O
17
18
Scheme 4. Formation of hydantoin 18dproposed mechanism.
In order to confirm the validity of this mechanism, the same
sequence was carried out with benzylamine (Scheme 4, (ii)) since
Hoffmann rearrangement do not take place with substituted am-
ides. In that case no rearrangement was observed, leading to the
expected oxidation product 18 in 81% yield. To the best of our
knowledge, this is the first example in the literature of a chiral
hydantoin synthesis via a TEMPO-catalyzed mechanism.
Among other tested oxidation conditions, Ru-oxidation by using
tetrapropylammonium perruthenate (TPAP) reagent8 has resulted
mainly in degradation. The best result was obtained with a Moffat-
type oxidation. When a combination of DMSO, N-(3-dimethyl-
aminopropyl)-N0-ethylcarbodiimine (EDC) and catalytic amount of
dichloroacetic acid (DCAA)9 have been used, 13 was isolated in 95%
yield and >90% purity. The low solubility of hydroxyl amide 12 was
in addition nicely circumvented by using DMSO as reagent.
In order to release the free amine for coupling reaction with
building block 3, keto amide 13 was submitted to hydrogenolysis
conditions (H2, Pd/C) in methanol. Only 11% of the desired amine 14
To circumvent the formation of these impurities, other peptidic
coupling conditions were screened. It turned out that when the
reaction was performed in N-methyl-2-pyrrolidinone (NMP), to-
gether with 4-methylmorpholine (NMM) as a base, the reaction
took place without side products formation. Reproducibility was
demonstrated upon scaling-up: compound 7 was isolated in 91%

2692
S. Wilmouth et al. / Tetrahedron 65 (2009) 2689–2694
NHBoc
NHBoc
NHBoc
a
b
O
O
O
O
O
+
S
O
N
NH2
CbzHN
OMe
CbzHN
N
H
OH
H2N
NH2
OH
OH
O
O
OH
12
21
3
11
NHBoc
NHBoc
NH2
O
O
O
c
d
O
S
O
N
O
O
S
N
NH2
N
H
N
N
H
N
H
H
O
OH
O
O
8
7
Scheme 6. Selected alternative route. Reagents and condition: (a) NH3, AlMe3, CH2Cl2/THF, recrystallization from i-Pr2O, 75%; (b) H2, Pd/C, 98%; (c) EDCI, HOBt, NMM, NMP, 91%; (d)
DMSO, EDC, dichloroacetic acid, THF, recrystallization from AcOEt, 74%.
yield with a purity of 94% and thus engaged without any purifica-
tion into the next step. By using such a ‘one-pot procedure’, loss of
material due to difference in crystallization behaviour of the di-
astereoisomers could be avoided.
Oxidation of hydroxy amide 7 by using a catalytic TEMPO, so-
dium hypochlorite and potassium bromide system10 gave a lot of
degradation. With a catalytic ruthenium oxide/sodium periodate
system11 or by using Swern oxidation,12 no reaction took place.
Again, the Moffat oxidation was the most effective.13 To optimize
yield and impurity profile, a screening of the parameters, which
may influence the oxidation reaction was undertaken. Results are
summarized in (Table 1).
THF is the solvent of choice as can be seen from (entries 2 and 3)
and EDC the best reagent (entries 3 and 6). Dichloroacetic acid
(DCAA) is the best additive but to achieve complete conversion,
0.75 equiv is required (entries 1 and 3). Phosphoric acid (H3PO4)
was identified as a potential alternative catalyst, though the quality
of the isolated product was lower (entry 7). As a last observation,
10 equiv of DMSO is required to consume all starting material
(entries 3 and 5).
On large scale, the Moffat oxidation was complete within 3 h at
20 ꢀC. The crude reaction mixture was crystallized from ethyl ace-
tate, and 8 isolated as an off-white solid in 80% yield and >95%
purity. The content of precursor 7 in keto amide 8 is a critical
quality attribute as there is no way to purge it by the end of the
synthesis. Hence, stringent specification was set at this stage.
The process developed thus far was reproduced in our kilo-lab
facilities starting from 465 g of building block 11. The synthesis
proved to be robust and reproducible as no major difference with
the optimized process was noted. The four steps sequence starting
from 11 was performed in an overall 51% yield delivering 328 g of
keto amide 8 with the desired high quality level.
To finalize the synthesis of 1, Boc-deprotection and racemization
remains to be done (Scheme 7).14 In order to change costly and
cumbersome reverse phase purification into normal phase purifi-
cation, racemization of 8 was envisioned prior to Boc-deprotection.
Unfortunately, any attempts to racemize 8 were not successful,
presumably due to the poor solubility of 8 in water and in most
organic solvents. By using organic bases in DMF or DMSO, degra-
dation takes place while racemization was partly observed when
basic buffers were used. Compound 8 was thus deprotected with
TFA in dichloromethane. The TFA salt of the deprotected amine was
isolated quantitatively after precipitation from tert-butyl methyl
ether in 94% HPLC purity. Racemization was investigated on the TFA
salt with phosphate buffers at different pH, controlling the enan-
tiomeric excess by capillary electrophoresis. Complete racemiza-
tion was observed between pH¼7.5 and pH¼8.0. At higher pH,
degradation occurs. The process was reproduced at 300 g scale and
resulted in a clean and complete racemization in 24 h.
NHBoc
O
O
O
S
O
N
O
N
H
NH2
O
N
H
O
O
N
O
N
H
S
O
22 (MIM 911)
NHBoc
Purification of the resulting racemate was accomplished suc-
cessfully by reverse phase chromatography. Optimization of the
Medicinal Chemistry procedure led to a streamlined approach
where desalting of the phosphate solution and purification were
O
O
O
N
N
NH2
N
H
HN
S
N
S
O
O
O
OH
O
Table 1
O
O
Oxidation of
a-hydroxy amide 7ddesign of experiment
23 (MIM 911)
Entry
Reagentsa
Catalystb
Solvent
7e
8e
NHBoc
1
2
3
4
5
6
7
10 equiv/10 equiv
30 equiv/10 equiv
10 equiv/10 equiv
7.5 equiv/10 equiv
5 equiv/10 equiv
10 equiv/10 equivd
10 equiv/10 equiv
0.5 equiv
0.5 equiv
0.75 equiv
0.75 equiv
0.75 equiv
0.5 equiv
0.5 equivc
THF
CH2Cl2
THF
THF
THF
16%
14%
0%
0.4%
3%
54%
14%
89%
86%
83%
33%
64%
O
O
O
O
N
N
N
NH2
N
H
HN
S
N
S
N
S
THF
THF
4%
2%
O
O
O
O
OH
O
O
O
O
O
a
Reagents¼DMSO/EDC.
Catalyst¼DCAA.
b
c
24 (MIM 1229)
Catalyst¼H3PO4.
d
e
Reagent¼DMSO/DCC.
Yield determined by LC–MS.
Figure 2. Peptidic coupling reactiondstructure of main impurities.

S. Wilmouth et al. / Tetrahedron 65 (2009) 2689–2694
2693
NHBOC
NH2
O
O
O
O
O
O
S
O
Deprotection
S
O
N
N
NH2
NH2
N
H
N
H
N
H
N
H
O
O
O
O
8
(S)-1.TFA
NH2
O
NH2
O
O
O
Ion exchange
Racemization
Purification by RPC
O
O
S
O
S
O
N
N
N
H
N
H
NH2
N
N
H
NH2
H
O
O
O
O
(+/-)-1.TFA
Scheme 7. Racemization and purification process of 1.
(+/-)-1.HCl
Table 2
ionization (ESI) mass spectra were obtained using an API 165 Sciex
mass spectrometer. All the reactions were performed under a
positive pressure of nitrogen. Commercial grade anhydrous
solvents and reagents were used without further purification. HPLC
was conducted with an Agilent Technologies HP 1100 on a
Ion exchange in purification process
Small scale
Large scale
HPLC purity
TFA anion
Chlorine anion
>98.0% a/a
<0.1% w/w
6.6% w/w
>95.0% a/a
11.6% w/w
2.7% w/w
Symmetry C18 (5
250 mmꢂ4.6 mm) and
m
m, 150 mmꢂ2.1 mm) an Hypersil-Phenyl (5
m
m
m,
m,
a
Supelcosil LC-18-DB column (5
both accomplished in a single run by using acetonitrile and an
aqueous HCl buffer solution as eluent.
250 mmꢂ4.6 mm), with purities being determined by peak area % at
the UV detector wavelength of 210 nm. The enantiomeric purities
were determined by capillary electrophoresis with a HP3DCE
TFA displacement by the HCl buffer used for purification was
demonstrated on small scale. Unfortunately this was not reproducible
on large scale as only 40% of the ion exchange occurred (Table 2).
The ion exchange experiment was performed in batch with two
different resins. By using the low loaded Q Sepharose resin (0.2 equiv
Clꢁ/g) 0.25% w/w of TFA salt remained, but when the higher loaded
resin, AG 1X8 (1.2 equiv Clꢁ/g) was used, the exchange was complete.
On large scale, the purification of crude (þ/ꢁ)-1$TFA, was per-
formed on a 110 mm column filled with Delta-Pack C18. After ion
exchange chromatography on AG 1X8 resin, the solution was fil-
tered dust free and lyophilized to give 140 g of (þ/ꢁ)-1$HCl. The
isolated off-white amorphous powder was of suitable quality for
toxicological and first into human evaluation. The measured HPLC
purity was >97% without single impurity above the qualification
level as defined by the ICH guidelines.
equipment (Agilent Technologies) using sulphopropyl-b-cyclo-
dextrine as chiral selector at the UV detector wavelength of 200 nm.
4.2. N-((1S)-1-{4-[(tert-Butoxy)carbonylamino]butyl}-2-
carbamoyl-2-hydroxyethyl)(phenylmethoxy)-carboxamide (12)
To CH2Cl2 (3 L) cooled at ꢁ30 ꢀC was added NH3 (500 g,
27.5 mol) under gentle stirring. AlMe3 (0.7 L, 1.4 mol) was slowly
added and the solution aged for 1 h at ꢁ30 ꢀC, then warmed to 0 ꢀC.
A solution of 11 (465 g,1.1 mol) in THF (3.5 L) was added slowly. The
mixture was aged 5 h at 40 ꢀC and 16 h at rt before quenching with
0.01 M aqueous HCl (12 L). After concentration of the volatiles in
vacuo, the mixture was diluted with water and the pH was adjusted
to 7 by using 0.01 M aqueous HCl. The precipitate was filtered off
and washed with heptane. The white solid was poured in hot
methanol (10 L) and stirred for 1 h. The precipitate containing in-
soluble aluminium salts was filtered through a pad of Celite. The
filtrate was concentrated in vacuo at 40 ꢀC to give a white solid,
which was recrystallized from diisopropylether (8 L). The white
crystals were filtered and dried in vacuo to yield 12 (334 g, 75%). MS
(ESI) m/z 410 (MþþH); HPLC (Symmetry C18) 98.4% pure; 1H NMR
(DMSO-d6) (assignments with asterisk are related to the signals of
3. Conclusion
In summary, we have developed an efficient process for the
preparation of the anti-hypertensive compound 1, in high purity
(>97%) with an improved overall yield of 23% versus 11% in the earlier
route. The new process is amenable to large-scale preparation and
successfully overcomes the drawbacks of a troublesome aminolysis
and a Dess–Martin oxidation. The improved robustness of the new
route was demonstrated by the synthesis of a batch of 140 g of com-
pound1. Aparticularattentionhasbeendedicatedtotheidentification
of impurities in order to get better knowledge and understanding of
the processes to help in the decision making to develop alternative
pathway. Additionally, the process developed in the end, for the pu-
rification of compound 1, has contributed to a much lower cost of the
overall synthesis even though a chromatography is still required.
a second diastereoisomer): d w1.35 (m, 2H), 1.53–1.05 (m, 2H), 1.42
(m, 2H), 1.36 (s, 9H), 2.86 (m, 2H), w3.78 (m, 1H), 3.90 (dd, 1H),
3.82) (dd, 1H), 5.01 (m, 2H), 5.56 (d, 1H, J¼5.7 Hz), 5.41) (d, 1H,
J¼6.2 Hz), 6.92 (d, 1H, J¼8.5 Hz), 6.57) (d, 1H, J¼9.4 Hz), 6.73 (t, 1H),
7.23–7.15 (2br s, 2H), 7.39–7.28 (m, 5H).
4.3. N-((5S)-5-Amino-6-carbamoyl-6-hydroxyhexyl)-
(tert-butoxy)carboxamide (21)
4. Experimental
4.1. General
A mixture of 12 (330 g, 0.81 mol) and 5% Pd–/C (33 g,10 mol %) in
methanol (7 L) was introduced in an autoclave and put under
hydrogen pressure (4 bar) at RT for 3 h. The reaction mass was fil-
tered over a layer of Celite and the catalyst was washed with
methanol (4 L). The combined filtrate was concentrated. The re-
sidual solid was submitted to three azeotropic distillations with
The NMR spectra were recorded with a Brucker DRX-500 in-
strument at 400 MHz for 1H and 100 MHz for 13C. The electronspray

2694
S. Wilmouth et al. / Tetrahedron 65 (2009) 2689–2694
toluene at 50 ꢀC, to remove all trace of water and methanol to give 21
(217 g, 98%). MS (ESI) m/z 276 (MþþH); 1H NMR (DMSO-d6) (as-
signments with asterisk are related to the signals of a second di-
The white solid was filtered off and washed two times with TBME
(since the compound is very hygroscopic, it was not filtered to
dryness). The wet crystals still containing residual TBME were
dried in vacuo to give (S)-1$TFA. To a stirred solution of (S)-1$TFA
in water (34 L), phosphate buffer (8 L, 0.2 M, pH 7.9) was added
dropwise to adjust the pH to 7.5 and the reaction mixture was
stirred at rt for 24 h at which point the racemization was com-
plete. The pH was then adjusted to 3 with a concentrated HCl and
filtered ‘dust free’. The solution of (þ/ꢁ)-1$TFA was purified on
astereoisomer): d 1.29–1.16 (m, 2H),1.34 (m, 2H),1.37 (s, 9H),1.41 (m,
2H), 1.85 (br s, 2H), 2.79 (m, 1H), 2.88 (m, 2H), 3.71 (d, 1H, J¼3.8 Hz),
3.64) (d,1H, J¼3.1 Hz), 5.41 (brs,1H), 6.74 (t,1H), 7.19/7.11 (2br s, 2H).
4.4. N-((1S)-1-{4-[(tert-Butoxy)carbonylamino]butyl}-2-
carbamoyl-2-hydroxyethyl)-2-(6-methyl-2-oxo-3-
{[benzylsulfonyl]amino}hydropyridyl)acetamide (7)
reverse phase silica (Deltapack C18 15 mm) at a flow rate of
0.5 L/min by using ACN/H2O pH 3 as eluent. The fractions with
purity higher than 97% were collected and ACN was concentrated
in vacuo. The resulting water solution was freeze-dried to give
pure (þ/ꢁ)-1$TFA (186 g). The solid was dissolved in water
(1.86 L) and charged in a glass reactor containing an ion exchange
resin (Biorad AG 1X8) conditioned at pH 3.4. The solution was
aged for 1 h, filtered and freeze-dried to give (þ/ꢁ)-1$HCl
(132.6 g, 45%). MS (ESI) m/z 492 (MþþH); HPLC (Supelcosil) 97.8%
To a stirred suspension of 21 (230 g, 0.79 mol) in NMP (8.5 L) at
10 ꢀC was added 3 (280 g, 0.79 mol). The mixture was aged for
30 min to reach rt at which HOBt (118 g, 0.87 mol), EDCI (167 g,
0.87 mol) and NMM (88 g, 0.87 mol) were added successively. The
resulting mixture was aged at rt for 16 h until all starting material
has been consumed. The reaction mixture was poured into water at
a temperature below 25 ꢀC. The aqueous layer was extracted three
times with ethyl acetate, and the combined organic layer was
concentrated in vacuo to give 7 as a white-brown solid (468 g, 91%).
MS (ESI) m/z 594 (MþþH); HPLC (Hypersil-Phenyl) 99.3% pure; 1H
NMR (DMSO-d6) (assignments with asterisk are related to the sig-
purity; CE enantiomeric ratio 1.06; 1H NMR (D2O):
d 1.33 (m, 1H),
1.48 (m, 2H), 1.64 (m, 2H), 1.74 (m, 1H), 2.28 (s, 3H), 2.93 (m, 2H),
4.16 (dd, 1H, hydrate form), 4.57 (dd, 2H), 4.84 (dd, 2H), 5.17 (dd,
1H, keto form), 6.25 (dd, 1H), 7.40–7.29 (m, 6H).
nals of a second diastereoisomer):
d 1.37 (s, 9H), 1.59–1.08 (m, 6H),
2.23 (s, 3H), 2.20) (s, 3H), 2.87 (m, 2H), 3.88 (dd, 1H), 3.84) (dd, 1H),
4.08 (m, 1H), 4.02) (m, 1H), 4.51 (s, 2H), 4.52)/4.49) (d/d, 2H), 4.77
(s, 2H), 4.83)/4.66) (d/d, 2H), 5.68 (d, 1H, J¼5.6 Hz), 5.56) (d, 1H,
J¼5.6 Hz), 6.08 (d, 1H, J¼6.6 Hz), 6.07) (d, 1H, J¼6.6 Hz), 6.74 (t, 1H),
6.73) (t, 1H), 7.12 (wd, 1H), 7.21/7.15 (s/s, 2H), 7.39–7.27 (m, 5H),
Acknowledgements
We thank Schering-Plough for allowing us to publish the above
described results. Furthermore, we gratefully acknowledge Y. Ber-
net, I. Bourgeois, J. Bretogne, A. Dreau, O. Kostelitz, J. C. Laigle, F.
Lombard, for all their efforts in turning the original laboratory
procedure into a solid process.
8.02 (d, 1H, J¼8.8 Hz), 7.79 (d, 1H, J¼8.8 Hz), 8.54 (s, 1H).
*
4.5. N-((1S)-1-{4-[(tert-Butoxy)carbonylamino]butyl}-2-
carbamoyl-2-oxoethyl)-2-(6-methyl-2-oxo-3-{[benzyl-
sulfonyl]amino}hydropyridyl)acetamide (8)
References and notes
To a stirred solution of 7 (444 g, 0.74 mol) in THF (5 L) and DMSO
(531 mL, 7.4 mol) at rt, was added EDCI (716 g, 3.7 mol). The
resulting heterogeneous mixture was aged for 20 min and
dichloroacetic acid (46 mL, 0.56 mol) was added. The reaction was
monitored by HPLC and complete conversion was observed after
5 h at rt. About 90% of the solvent was concentrated in vacuo and
water was added to the suspension. The solid was filtered, washed
three times with water and heptane and dried in vacuo. The crude
product was recrystallized from ethyl acetate (1.5 L). The white
precipitate was filtered and washed once with ethyl acetate at 0 ꢀC
then dried in vacuo to give 8 (328 g, 74%). MS (ESI) m/z 592
(MþþH); HPLC (Hypersil-Phenyl) 95.2% pure; 1H NMR (DMSO-d6):
1. (a) Alexander, J. H.; Singh, K. P. Am. J. Cardiovasc. Drugs 2005, 5, 279–290; (b)
Walenga, J. M.; Jeske, W. P.; Hoppensteadt, D.; Fareed, J. Curr. Opin. Invest. Drugs
2003, 4, 272–281; (c) Kunitada, S.; Nagahara, T.; Hara, T. Handb. Exp. Pharmacol.
1999, 132, 397–420.
2. Adang, A. E. P. WO 9850420, 1998.
3. Adang, A. E. P.; Van Boeckel, C. A. A.; Grootenhuis, P. D. J.; Peters, J. A. M. WO
9717363, 1997.
4. Sanderson, P. E. J.; Lyle, T. A.; Cutrona, K. J.; Dyer, D. L.; Dorsey, B. D.; McDo-
nough, C. M.; Naylor-Olsen, A. M.; Chen, I.-W.; Chen, Z.; Cook, J. J.; Cooper, C. M.;
Gardell, S. J.; Hare, T. R.; Krueger, J. A.; Lewis, S. D.; Lin, J. H.; Lucas, B. J.; Lyle,
E. A.; Lynch, J. J.; Stranieri, M. T.; Vastag, K.; Yan, Y.; Shafer, J. A.; Vacca, J. P. J.
Med. Chem. 1998, 41, 4466–4474.
5. Acherki, H.; Alvarez-Ibarra, C.; Dios, A.-de; Quiroga, M. L. Tetrahedron 2002, 16,
3217–3228.
6. (a) Bobbitt, M. Z. J. Org. Chem. 1991, 56, 6110–6114; (b) Harbeson, S. L.; Abelleira,
S. M.; Akiyama, A.; Barrett, R.; Carroll, R. M. J. Med. Chem. 1994, 18, 2918–2929;
(c) Zhao, M.; Eiichi Mano, J. L.; Song, Z.; Tschaen, D. M.; Grabowski, E. J. J.;
Reider, P. J. J. Org. Chem. 1999, 64, 2564–2566; (d) Palomo, C.; Oiarbide, M.;
Landa, A. J. Org. Chem. 2000, 65, 41–46.
7. For a review, see: Wallis, E. S.; Lane, J. F. Org. React. 1946, 3, 267–306.
8. Ley, S. V. Synthesis 1994, 7, 639–666.
9. Yuan, W. Y.; Munoz, B.; Wong, C.-H. J. Med. Chem. 1993, 36, 211–222.
10. See Ref. 3.
d
1.35 (m, 2H), 1.37 (s, 9H), 1.38 (m, 2H), 1.75/1.50 (m/m, 2H), 2.22 (s,
3H), 2.89 (m, 2H), 4.51 (s, 2H), 4.90/4.73 (d/d, 2H), 5.00 (m,1H), 6.07
(d, 1H, J¼7.7 Hz), 6.77 (t, 1H), 7.11 (d, 1H, J¼7.7 Hz), 7.39–7.28 (m,
5H), 8.02/7.77 (s/s, 2H), 8.56 (s, 1H), 8.67 (d, 1H, J¼6.9 Hz).
4.6. N-[1-(4-Aminobutyl)-2-carbamoyl-2-oxoethyl]-2-(6-
methyl-2-oxo-3-{[benzylsulfonyl]amino}hydropyridyl)-
acetamide (D/L)-1$HCl
11. Narukawa, Y.; Nishi, K.; Onoue, H. Tetrahedron 1997, 2, 539–556.
12. Burkhart, J. P.; Peet, N. P.; Bey, P. Tetrahedron Lett. 1990, 31, 1385–1388.
13. (a) See Ref. 9; (b) Chandrasekaran, S.; Kluge, A. F.; Edwards, J. A. J. Org. Chem.
1977, 42, 3972–3974; (c) Semple, J. E.; Owens, T. D.; Nguyen, K.; Levy, O. E. Org.
Lett. 2000, 18, 2769–2772.
To a stirred solution of 11 (300 g, 0.507 mol) in dichloro-
methane (0.6 L) at rt, TFA (2.4 L) was added and the mixture aged
for 2 h. TBME (18 L) was added at 10 ꢀC to crystallize the TFA salt.
14. Analysis of a sample of 8 by 1H NMR using Pirckle’s alcohol as chiral chelating
agent, showed an enantiomeric purity of 80% meaning that the compound
partly racemizes over the four steps sequence.