Concise preparation of a stable cyclic sulfamidate intermediate in the synthesis of a enantiopure chiral active diamine derivative

Article abstract of DOI:10.1021/op500264v

A classical resolution was studied and developed from 2-benzoyl-pyridine in order to prepare SSR504734, a novel antipsychotic derivative. The key step of this route is the substitution of a sulfamidate derivative by a benzamide anion with complete inversi

Full text of DOI:10.1021/op500264v

Article
pubs.acs.org/OPRD
Concise Preparation of a Stable Cyclic Sulfamidate Intermediate in
the Synthesis of a Enantiopure Chiral Active Diamine Derivative
Jean-Francois Rousseau,* Isaac Chekroun, Vincent Ferey, and Jean Robert Labrosse
Sanofi Recherche, Chemistry and Biotechnologies Development, 371 av Pr. J. Blayac, 34184 Montpellier, France
ABSTRACT: A classical resolution was studied and developed from 2-benzoyl-pyridine in order to prepare SSR504734, a novel
antipsychotic derivative. The key step of this route is the substitution of a sulfamidate derivative by a benzamide anion with
complete inversion of configuration. The sulfamidate is prepared in a two-step procedure by reacting erythro phenyl-piperidine-
2-yl-methanol derivative with thionyl chloride followed by oxidation with ruthenium oxide. This sulfamidate is an easily scalable
intermediate that produce a diamine intermediate with the expected configuration.
INTRODUCTION
− Azepine byproduct formation;
■
The glutamate hypothesis of the cause of schizophrenia
postulates that the positive, negative, and cognitive symptoms
of this devastating disease originate at least in part from faulty
glutamatergic signaling via NMDA receptors, This article
extends an earlier process research study in order to produce
the first kilogram of SSR504734, the lead compound discovered
by Sanofi that shows promise as a novel potential antipsychotic
agent by improving NMDA signaling.1
− Final Palladium deprotection with subsequent residual
palladium (up to 100 ppm) in the final API which
required several crystallizations to eliminate palladium.
SECOND GENERATION PROCESS
■
The intent was to prepare the amino alcohol from a cheap and
available raw material, and so the study was focused on 2-
benzoyl-pyridine 15, which could be easily hydrogenated into a
mixture of erythro 16 and threo 17 amino alcohols (Scheme 3).4
This new potential route could be very interesting if the main
erythro enantiomer could be used directly. In order to obtain
the desired diastereomer, it was necessary to perform a SN2
displacement at the carbon bearing the alcohol function. Under
these conditions, the absolute configuration (S,S) may be
obtained from an erythro amino-alcohol (R,S) 5 (Scheme 4).
This SN2 displacement becomes possible only when the
protecting group of the piperidine does not allow the formation
of an aziridinium intermediate.
During the literature searching, a very interesting article was
found where the authors demonstrated that the attack of a
nucleophile on a sulfamidate type compound 20 was
regioselective and stereoselective, with total inversion of
configuration on the carbon bearing oxygen (Scheme 5).5 So,
taking into account this regioselective and stereoselective
advantage, it was hypothesized that a cyclic sulfamidate derived
from the amino-alcohol 5 could produce the desired product
with good stereoselectivity. Thus, it was planned to use the
cyclic sulfamidate to simultaneously protect the piperidine,
activate the benzylic alcohol, and realize nucleophilic displace-
ment at the benzylic position with inversion of configuration.
Starting from 2-benzoyl-pyridine, we developed the synthesis
of SSR504734 via a sulfamidate intermediate (Scheme 6) by
performing a reduction of the ketone followed by resolution,
which was estimated to give the cheapest process.
DISCOVERY ROUTE
■
The discovery route starts with the pipecolic acid 1 derivative
(Scheme 1). The (L)-N-Boc pipecolic acid is first activated to a
Weinreb derivative in order to introduce the phenyl moiety at
low temperature via the selective addition of phenyl lithium at
−30 °C. The ketone is then selectively reduced into the
corresponding alcohol using L-selectride and, after removing
the N-Boc protecting group, the expected amino alcohol 5 is
obtained after crystallization with good enantiomeric excess (up
to 98%).2
The key step of this synthesis is the nucleophilic substitution
of the alcohol by ammonia with retention of configuration
(Scheme 2). An allyl protection is used in order to allow the
introduction of the amine function, but during this amine
introduction an azepine 9 impurity is always observed in the
reaction mixture. This ring expansion is a well-known reaction
described by Cossy et al.3 The formation of expected
compound 8 and the azepine 9 byproduct (around 10%
detected) could be explained by the formation of an aziridinium
intermediate 7. This byproduct is difficult to eliminate; at lab
scale, several chromatographic purifications must be performed
in order to eliminate it after the coupling with the aryl acid
chloride. The final deprotection is performed by using soluble
palladium catalyst.
The 10 step discovery route used to prepare SSR504734
presents numerous drawbacks related to industrial scale up,
especially when considering that the targeted production
should be several metric tons per year:
− Availability and price of the (L)-N-Boc pipecolic raw
material at metric ton scale;
Received: August 18, 2014
− Low temperature required in two different steps;
© XXXX American Chemical Society
A
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 1
Scheme 2
is only 2.5 volumes with regard to the mass of reagent. In
addition, the 2.5% w/w platinum oxide necessary for the
hydrogenation does not suffer from scale-up conditions. Only 2
h were required to hydrogenate 10 kg of 2-benzoylpyridine.
RESULTS AND DISCUSSION
■
The 2-benzoylpyridine 15 was hydrogenated in acetic acid
using 2.5% platinum oxide at 3.5 bar in 2 h (Scheme 6). This
step is very productive because the quantity of acetic acid used
B
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 3
Scheme 4
combination of DMAP and thionyl chloride forms a much
more reactive intermediate, resulting in a better selectivity
without formation of numerous byproducts.
Scheme 5
A study of various bases used for the sulfamidite 22
substitution allowed replacement of triethylamine by the
combination of pyridine/DMAP, which gave better results.
The choice of the mixture of these two bases arose from two
results:
− When pyridine/DMAP was used, the product could be
isolated as a white solid with a good yield (90%) and a
very good purity (95%).
− When pyridine alone was used, the product was isolated
as a brownish solid with a good yield (90%) and a good
purity (90%), but with the concomitant presence of tar.
The expected phenyl-piperidine-2-yl-methanol was isolated
as a mixture of diastereomers (erythro/threo: 80/20) with 95%
yield and 99.4% purity by HPLC. The product contains
approximately 5% of a fully hydrogenated impurity 25 (Scheme
6), which was measured only by LC-MS analysis.
The phenyl-piperidine-2-yl-methanol isomer required for the
synthesis of SSR504734 via the “sulfamidate” process is the
erythro diastereomer (R,S) 5. To obtain this product, the threo
diastereoisomer must be eliminated and the erythro racemic
mixture resolved.
The slow addition of thionyl chloride allowed control of the
exothermicity, enabling a total conversion directly at the end of
addition.
Finally, the synthesis of the erythro sulfamidite (R,S) 22
could be realized by the addition of thionyl chloride in the
presence of pyridine and DMAP (10% w/w) in dichloro-
methane at −5 °C. The product was isolated as a mixture of
two diastereomers (due to the presence of chiral sulfoxide) with
a 92% yield and 98.7% purity. Furthermore, various trials
showed that this sulfamidite could be synthesized in no more
than five volumes of dichloromethane and in a temperature
range between 5 and 10 °C.
The oxidation into sulfamidate (R,S) 23 was initially realized
by the combination of RuCl3/NaIO4 in an acetonitrile/water
mixture.8 This effective method was used during the synthesis
of the first batch of SSR504734 and allowed the isolation of the
erythro sulfamidate (R,S) with a 96% yield and 97.3% purity.
Other oxidizing systems are effective.9 The replacement of
sodium periodate by NaOBr3 or NaOCl in the presence of
ruthenium trichloride in an acetonitrile/water mixture also
permits the conversion into the sulfamidate with good yield and
purity. Acetonitrile can be replaced by dichloromethane to
avoid a further change of solvent. Finally, oxidation was also
possible using bleach only, without ruthenium, but during scale
up, the percentage of various impurities was significantly higher.
Therefore, further work focused on the combination RuCl3/
NaOCl in dichloromethane. After several trials at different
temperatures and with different amounts of RuCl3 and NaOCl,
An important property of this amino-alcohol (16/17)
allowed the separation of both diastereomers: the hydro-
chloride erythro salt is less soluble than the threo salt. Therefore,
a crystallization of the hydrochloride salt as a mixture of
erythro/threo 80:20 in hydrochloric isopropanol/isopropyl ether
produced the erythro amino-alcohol (16) with a good yield but
a moderate diastereomeric purity (2.5% of threo). After
displacement of the hydrochloride salt, a direct separation of
the enantiomers of the enriched erythro phenyl-piperidine-2-yl-
methanol 16 was tried. Crystallization of the erythro base 16 in
a mixture of ethanol/water (80/20) in the presence of 0.25
equiv of (+)-(D)-di-p-toluoyl tartaric acid was successful. After
displacement of the tartaric salt, the amino-alcohol erythro (R,S)
enantiomer 5 was isolated in 40% yield and 99% e.e.
The synthesis of sulfamidate 23 was first envisaged from the
amino-alcohol in the presence of sulfuryl chloride which might
avoid an oxidation step.6 Unfortunately, the trials were not
satisfactory due to the fact that low yield and numerous
degradation products were observed.
The next attempt was to generate the sulfamidite 22
derivative by thionyl chloride addition followed by triethyl-
amine in dichloromethane.7 However, this reaction was not
reproducible and product was isolated as an oil with a deep
purple color and a considerable amount of tar. Morever, the
C
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 6
the sulfamidate was obtained in good yield (up to 95%) and
with good purity (up to 97%).
batch. The NaOCl oxidation need further study due to its easier
industrial handling and lower price compared to NaIO4.
Afterwards, the sulfamidate 23 was not isolated but used
directly as a solution in NMP, after solvent exchange with
dichloromethane, for the subsequent step.
The 2-chloro-3-trifluorobenzamide 24 is obtained from the
2-chloro-3-trifluoro benzoic acid 10 via the corresponding acid
chloride with a 84% yield and 99.3% purity.10
Even after trying numerous bases, solvents and temperatures,
SN2 displacement of sulfamidite 22 was unsuccessful, so it was
mandatory to use the more reactive sulfamidate 23. In the
literature,11 the coupling of sulfamidates with nucleophiles is
mainly performed in the presence of NaH in DMF. These
conditions are not scalable due to safety issues. Thus, the
influence of various bases, solvents and temperatures were
studied instead. Some trials were performed under phase
transfer conditions.
The anion of 2-chloro-3-trifluorobenzamide 24 is obtained
by reaction of sodium tert-amylate in THF. The sodium tert-
amylate could be used as a solid or in solution in THF with no
impact on yield. The anion of compound 24 was slowly added
to the sulfamidate 23 solution in NMP. After hydrolysis of the
sulfamic acid intermediate by sulfuric acid, the 2-chloro-N-[(S)-
phenyl[-(2S piperidine-2-yl]-methyl]-3-trifluoromethyl) benza-
mide, SSR504734, was isolated with 85% crude yield. A
recrystallization in 20 volumes of methyl cyclohexane and 0.2
An important point remained to study: the elimination of the
ruthenium tetroxide and dioxide as byproducts of the oxidation.
The presence of RuO2 darkens the organic phase black. Diverse
trials to wash the crude with reducing aqueous media such as
sodium thiosulfate or sodium metabisulfite, followed by several
aqueous washes, gave no reproducible results. Indeed, the
treated organic phase often returned to a black color, meaning
that some residual ruthenium tetroxide is still present in the
organic layer. The alternative was to reduce completely the
ruthenium tetroxide to ruthenium dioxide. When studying the
reaction in dichloromethane in order to solve this problem, a
stoichiometric amount (0.17 equiv) of isopropanol with respect
to the amount of ruthenium was added. The metal oxidizes
isopropanol into acetone and is reduced into ruthenium
dioxide. After filtration, a clear and colorless organic phase
was obtained with less than 10 ppm of residual ruthenium
(Scheme 7).
It is therefore possible to perform the oxidation of
sulfamidite 22 into sulfamidate 23 by using 0.75% w/w of
RuCl3 in the presence of 3 equiv of NaOCl (1 M) in 5 volumes
of dichloromethane at room temperature followed by treatment
with isopropanol. But because the NaOCl oxidation process
was not reproducible, NaIO4 was used to scale up the first
D
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 7
vol of N-methylpyrrolidinone produced a final SSR504734 in
high purity (up to 98%).
Finally, SSR504734 was obtained starting from 2-benzoyl-
pyridine through the synthesis of an original sulfamidate 23
intermediate as shown in Scheme 8.
wavelength, 220 nm; mobile phase CH3CN/phosphate buffer
pH = 3, gradient 20/80 increasing to 80/20 over a 20 min
1
period. H and 13C NMR spectra were recorded on a Varian
NMR 400 MHz spectrometer operating at 399.9 MHz in
1
DMSO-d6 for H NMR.
This robust and scalable process resulted in the expected
SSR504734 in seven steps with 15% overall yield, compared
with 10% via the discovery route. The process was optimized
later on in order to produce kilogram quantities of SS504734
for first GMP batches.
The chemical shifts, δ, were recorded relative to
tetramethylsilane as an internal standard; all coupling constants,
J, are reported in Hz.
Preparation of Phenyl-piperidine-2-yl-methanol (16/
17). A solution of 10.061 kg (54.915 mol) of 2-benzoyl-
pyridine (15) in 20 L (2 volumes) of acetic acid was loaded in a
40 L hydrogenator followed by the addition of 250 g (2.5 wt %)
platinum oxide (PtO2) suspended in 5 L (0.5 volumes) of
acetic acid. The reaction media was shaken at 20 °C under 50
psi (3.5 bar) of hydrogen. Beginning with the agitation under
pressure, the internal temperature increases to 45 °C. The flow
of hydrogen is then controlled so as to maintain the
temperature around 35 °C. After reacting for approximately
3−4 h, the addition of hydrogen was stopped and the reaction
worked up.
The acetic acid solution was filtered on Celite (Clarcel) to
eliminate the catalyst and the cake is washed with 2 × 8 L (0.8
volumes) of water. The filtrate was then concentrated, 40 L (4
volumes) of water was added, and pH of the mixture was
adjusted to 14 using concentrated aqueous NaOH (30%). The
precipitate was filtered, washed using 2 × 10 L (1 volume) of
water, and then dried at 50 °C under vacuum. Phenyl-
piperidine-2-yl-methanol (16/17) was isolated as a racemic
mixture of diastereomers (erythro/threo: 80/20) with a yield of
95% (9 970 g), 99.4% purity, and 100% assay. The product
EXPERIMENTAL SECTION
■
General Considerations. All reagents were purchased
from commercial sources and used without any additional
purification. All experiments were carried out under a nitrogen
atmosphere to maintain anhydrous conditions. The reaction
profiles were followed quantitatively by HPLC on a Varian
series 9000 machine used for Method 1: Modulo-Cart QS
Uptisphere 120A 5 μm NEC 250 × 4.6, column at 30 °C; flow
rate, 0.8 mL/min; injection volume, 10 μL; detection
wavelength, 215 nm; mobile phase CH3CN/sodium heptane-
sulfonate buffer pH = 3, gradient 20/80 increasing to 80/20
over a 25 min period. Method 2: Chiral method used for
phenyl-piperidine-2-yl-methanol: AGP 150 × 4.0 chiral AGP 5
μm column at 30 °C; flow rate, 0.8 mL/min; injection volume,
5 μL; detection wavelength, 210 nm; mobile phase MeOH/
KH2PO4 buffer pH = 6.5, 5/95:7.7 min: erythro SR, 8.7 min:
erythro RS; 9.7 min: threo RR; 10.5 min: threo SS. Method 3:
YMC Pack ODS AQ, 150 × 4.6 mm, 5 mm, 12 mm column at
30 °C; flow rate, 1 mL/min; injection volume, 5 μL; detection
E
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 8
contained approximately 5% of a fully saturated impurity (25),
which was not detected in HPLC but by GCMS.
-CH2-), 2.4−3.0 (3H, m, -NCH2-, -NCH-), 4.5 (1H, d, J = 5.1
Hz, -CHOH-), 7.1−7.3 (5H, m, -CH- phenyl).
HPLC (Method 1): RT, 12.6 min; RT, 20.6 min (2-benzoyl-
pyridine). TLC: dichloromethane/methanol 90/10, Rf = 0.2;
Preparation of erythro (R,S) Phenyl-piperidine-2-yl-
methanol (5). A solution was prepared by dissolving 298 g
(1.6 mol) of erythro phenyl-piperidin-2-yl-methanol (16) in
1.14 L (4 volumes) of an ethanol/water (80/20) mixture. To
this solution was added 143.3 g (0.4 mol) (+)-(D)-di-p-toluoyl
tartaric acid previously dissolved in 274 mL (2 volumes) of an
ethanol/water (80/20) mixture. The reaction media was stirred
at room temperature for 1 h and the precipitate was filtered,
washed with 1 L of absolute ethanol, and dried under vacuum.
The erythro (R,S) phenyl-piperidine-2-yl-methanol (5) was
obtained as a (+)-(D)-di-p-toluoyl taratric acid salt in a ratio of
2:1 with a 40% yield (241 g). HPLC (Method 1): RT, 12.6
min. HPLC (Method 2): Diastereomeric purity: erythro 99%,
threo 1%. Enantiomeric purity: erythro (R,S) 98.2%, (S,R) 0.8%,
and threo (S,S) 0.8%.
1
mp 132.2 °C. H NMR (CDCl3) δ 1.0−1.8 (6H, m, -CH2-),
2.2−3.0 (3H, m, -NCH2-, -NCH-), 4.3 (0.2H, d, J = 6.3 Hz,
-CHOH- threo), 4.5 (0.8H, d, J = 5.1 Hz, -CHOH- erythro),
7.1−7.4 (5H, m, -CH- phenyl).
Purification of erythro Phenyl-piperidine-2-yl-meth-
anol (16). In 1 L (2 volumes) of absolute ethanol, 500 g (2.61
mol) of the previous phenyl-piperidine-2-yl-methanol (16/17)
erythro/threo racemic mixture 80:20 was dissolved and cooled
at 0−5 °C followed by the addition of 363 mL (3.14 mol of
HCl) of hydrochloric ethanol 8.66 M. The initially heteroge-
neous reaction media became homogeneous and then
reprecipitated. Diethyl ether (6 L) was added, and the reaction
medium was shaken for 4 h. The precipitate was filtered,
washed with 2 L of diethyl ether, and dried under vacuum. The
erythro amino-alcohol (16) was obtained as a hydrochloride salt
with a 64% yield (382 g). Analysis: purity diastereomeric
erythro 97.5%; threo 2.5%.
The salt was suspended in 1.25 L (5 volumes) of water and
treated with 30% NaOH. The heterogeneous media was shaken
for 16 h. After filtration, washing with 1 L of water, and drying
under vacuum conditions, the enantiomer (R,S) of the erythro
form (5) was isolated as a white powder with a 90% yield (108
g). The overall yield for these steps was 36%. HPLC (Method
1): RT, 12.6 min. HPLC (Method 2): diastereomeric purity,
erythro 100%; enantiomeric purity, erythro (R,S) 100%, fully
hydrogenated impurity 25 < 0.1% (GC-MS); mp 127.6 °C
Preparation of Sulfamidite (3R,3aS)-3-Phenyl-hexahy-
dro-[1,2,3]oxathiazolo[3,4-a]pyridine 1-Oxide (22). To a
solution of 381 g (1.99 mol) of erythro (R,S) phenyl-piperidine-
2-yl-methanol (5), 480 mL (6 mol) of pyridine, and 38 g (10%
w/w) of DMAP in 3 L (9.5 volumes) of dichloromethane, were
The hydrochloride was finally suspended in 3.4 L (9
volumes) of water and basified with concentrated aqueous
30% NaOH. The heterogeneous reaction media was shaken for
16 h. After filtration, washing with 3 L of water, and drying
under vacuum, the diastereoisomer erythro (16) was isolated as
a white powder with a 93% yield (298 g). Overall yield for these
steps was 59%. HPLC (Method 1): RT, 12.6 min; HPLC
(Method 2): Diastereomeric purity: erythro 99%, threo 1,0%,
HPLC purity 99,7%, fully hydrogenated impurity 25: 0.4%
(GC-MS); mp 139.2 °C. 1H NMR (CDCl3) δ 1.0−1.8 (6H, m,
F
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
added 174 mL (2.39 mol) of thionyl chloride dissolved 160 mL
(0.5 volumes) of dichloromethane over a period of 3 h. The
temperature was maintained at −5 °C during the entire
addition. The initial heterogeneous reaction medium became
homogeneous, and then again heterogeneous. After 15 min at
−5 °C, the reaction was complete and the reaction medium was
returned to room temperature and hydrolyzed by 4 L of water.
The organic phase was washed with 4 L of 1 M hydrochloric
acid and by 3 × 4 L of water until the pH was neutral.
After concentration in vacuo, the sulfamidite erythro (R,S)
(22) was isolated as a white solid and a mixture of two
diastereomers (due to the sulfoxide) with a 92% yield (436 g), a
98.7% purity, and a 100% assay. HPLC (Method 3): RT, 17.7
min. TLC: pure dichloromethane, Rf1 = 0.66, Rf2 = 0.46; mp
117.5 °C. 1H NMR (CDCl3) δ 0.5 (0.7H, qd, J = 11.4/2.7 Hz,
-CH2-), 0.8−1.8 (5.3H, m, -CH2-), 2.6 (0.7H, td, J = 11.4/2.9
Hz, -NCH2-, -NCH-), 2.8−3.2 (0.3H, m, -NCH2-, -NCH-),
3.3−3.8 (2H, m, -NCH2-, -NCH-), 4.3 (0.3H, d, Dia1, J = 7.1
Hz, -CHOSO-), 5.8 (0.7H, d, Dia2, J = 6,3 Hz, -CHOSO-),
7.0−7.4 (5H, m, -CH- phenyl).
380 g (1.5 mol) of sulfamidate (R,S) (3R,3aS)-3-phenyl-
hexahydro-[1,2,3] oxathiazolo[3,4-a]pyridine 1,1-dioxide (23)
in 1.9 L (5 volumes) of NMP was added over 2 h. At the end of
the addition, the temperature of the reaction medium was
maintained under reflux another 2 h. After cooling to 5 °C, 3.8
L of 20% H2SO4 solution and 6 L of MTBE were added to the
reaction medium, which was stirred overnight at room
temperature. After decantation, the aqueous phase was washed
by 3 × 2 L of MTBE and then the pH was adjusted to 14 with
concentrated (30%) NaOH (2.7 L) in the presence of 3 L of
toluene. The reaction media was diluted with 15 L with water
to solubilize all the sodium sulfate. The organic phase was then
washed until neutral pH using 3 × 6 L of water.
After concentration under vacuum of this organic phase, the
threo SSR504734 (S,S) (14) was isolated as an off-white solid
with a 85% yield (507 g). A recrystallization in 20 vol of methyl
cyclohexane and 0.2 vol of N-methylpyrrolidinone produced a
final SSR504734 in high purity (min. 98%). HPLC (Method
3): RT SSR504734, 12.2 min, 98% purity, enantiomeric purity
e.e. = 99.9%. TLC: dichloromethane/methanol 95/5, Rf = 0.17;
1
mp 146 °C. H NMR (CDCl3) δ 1.2−1.8 (6H, m, -CH2-), 2.2
Preparation of Sulfamidate (3R,3aS)-3-Phenyl-hexa-
hydro-[1,2,3]oxathiazolo[3,4-a]pyridine 1,1-Dioxide
(23). On a solution of 584 g (2.75 mol) of sodium periodate
and 1.1 g (0.25% w/w/sulfamidite) of ruthenium trichloride in
4.3 L (10 volumes) of water and 1.3 L (3 volumes) of
acetonitrile, are added, in 1 h, 435 g (1.83 mol) of erythro
sulfamidite (R,S) (22) in a solution in 3 L (7 volumes) of
acetonitrile. The temperature was maintained around +5 °C
during the entire addition. After 15 min, the reaction was
complete. The reaction media was filtered and the cake wass
washed with 5 L of ethyl acetate. The organic phase was washed
using 5 L of a 0.6 M sodium thiosulfate solution then by 2 × 5
L of water until neutral pH. After concentration in vacuo, the
sulfamidate erythro (R,S) (23) was isolated as a white solid with
a 96% yield (445 g), 97.3% purity, and 98.6% assay. HPLC
(Method 3): the sulfamidate (23) and the sulfamidite (22)
have the same retention time in the conditions used. TLC, pure
dichloromethane, Rf = 0.60. Revelation with phosphomolybdic
acid gave a black color for sulfamidate instead of the sulfamidite
(1H, td, J = 11.4/2.8 Hz, -NCH-), 2.8 (1H, m, -NCH2-), 5.1
(1H, dd, J = 7.2/1.7 Hz, -CHOH-), 7.1−7.8 (9H, m, -NHCO-,
-CH- phenyl).
AUTHOR INFORMATION
■
Corresponding Author
*Telephone: +33 4 99 77 47 75. Fax: +33 4 99 77 68 03. E-
mail: jean-francois.rousseau@sanofi.com.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS
■
API=active pharmaceutical ingredient
DMAP=dimethylaminopyridin
HPLC=high pressure liquid chromatography
LC-MS=liquid chromatography−mass spectrometry
NMDA=N-methyl-D-aspartate
1
which turned brown; mp: 135.2 °C. H NMR (CDCl3) δ 0.7
TLC=thin layer chromatography
(1H, qd, J = 11,4/4,6 Hz, -CH2-), 1.1−1.8 (5H, m, -CH2-), 2.6
(1H, td, J = 13.2/3.0 Hz, -NCH-), 3.5−3.8 (2H, m, -NCH2-),
5.6 (0.7H, d, Dia2, J = 6.3 Hz, -CHOSO2-), 7.3 (5H, m, -CH-
phenyl).
REFERENCES
■
(1) (a) Lowe, J. A., III Expert Opin. Ther. Pat. 2005, 15, 1657.
(b) Ronan, D.; Gihad, D.; Genevieve, E.-B.; Annick, C.; Christophe,
L.; Christophe, D.; Martine, P.; Michel, H.; Vincent, S.; Michel, D.;
Annie, C.; Carolle, V.; Denis, B.; Jean, P. T.; Jeanne, S.; Pierre, R.;
Benoit, M.; Mireille, S.; Xavier, V.; Bruno Biton, R. S.; Dominique, F.;
Richard, A.; Patrick, A.; Florence, O.-D.; Ghislaine, P.; Guy, G.; Pascal,
G.; Philippe, S.; Bernard, S. Neuropsychopharmacology 2005, 30, 1963.
(c) Coyle, J. T.; Balu, D.; Benneyworth, M.; Basu, A.; Roseman, A.
Dialogues Clin. Neurosci. 2010, 12, 359−382.
Preparation of 2-Chloro-3-trifluoromethyl-benzamide
̈
(24). A solution of 898 g (4 mol) of 2-chloro-3-trifluorobenzoic
acid (10) in 4.5 L (5 volumes) of toluene was warmed to 80 °C
before adding, over 2 h, 584 mL (8 mol) of thionyl chloride.
The reaction medium was heated under reflux for 2 h and then
concentrated. The obtained oil was added to 5 L of 20% w/w
aqueous ammonia. After stirring for 16 h, the precipitate was
filtered, washed with 1 L of water and dried room under
vacuum. The 2-chloro-3-trifluoromethyl benzamide (24) was
obtained as a white powder with a 84% yield (749 g) and a
HPLC purity of 99.3%; HPLC (Method 3): RT, 9.9 min; mp
(2) (a) Dargazanli, G. WO 2005/037792 and Aletru, M.; et al. PCT
Int. Appl. 2009010660, 22 Jan 2009. (b) A similar procedure
described by Sekiguchi, Y.; et al. PCT Int. Appl. 2008018639, 14 Feb
2008.
(3) Cossy, J.; Dumas, C.; Michel, P.; Gomez Pardo, D. Tetrahedron
Lett. 1995, 36, 549−552.
(4) (a) Solladie-Cavallo, A.; Roje, M.; Baram, A.; Sunjic, V.
Tetrahedron Lett. 2003, 44, 8501. (b) Aerberli, P.; Houlihan, W. J.;
Takesue, E. I. J. Med. Chem. 1969, 12, 51. (c) Villani, F. J.; King, M. S.;
Villani, F. J. J. Med. Chem. 1963, 6, 51. (c) Rhone Poulenc S. A. GB
843,070; priority date Aug 1, 1957.
1
118.9 °C. H NMR (CDCl3) δ 7.2−8.2 (5H, m, CONH2 and
phenyl CH-).
Preparation of SSR504734: 2-Chloro-3-methyl-N-((S)-
phenyl-(S)-piperidin-2-yl-trifluoromethyl)-benzamide
(14). To a solution of 502 g (2.25 mol) 2-chloro-3-
trifluoromethyl-benzamide (24) in 2.5 L (5 volumes) of THF
was added 252.6 g (2.175 mol) of sodium tert-amylate. The
reaction medium was heated to reflux. After 1 h, a solution of
(5) (a) Cohen, S. B.; Halcomb, R. L. Org. Lett. 2001, 3, 405.
(b) Williams, A. J.; Chakthong, S.; Gray, D.; Lawrence, R. M.;
G
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Gallagher, T. Org. Lett. 2003, 5, 811. (c) Melendez, R. E.; Lubell, W.
D. Tetrahedron 2003, 59, 2581.
(6) Pilkington, M.; Wallis, J. D. J. Chem. Soc., Chem. Commun. 1993,
24, 1857−1858.
(7) (a) Posakony, J. J.; Grierson, J. R.; Tewson, T. J. J. Org. Chem.
2002, 67, 5164−5169. (c) Pound, M. K.; Davies, D. L.; Pilkington, M.;
Sousa, M.; Wallis, J. D. Tetrahedron: Asymmetry 1990, 1, 881−884.
Boulton, L. T.; Stock, H. T.; Raphy, J.; Horwell, D. C. J. Chem. Soc.,
Perkin Trans. 1 1999, 11, 1421−1430. Atfani, M.; Wei, L.; Lubell, W.
D. Org. Lett. 2001, 3, 405.
(8) (a) White, G. J.; Garst, M. E. J. Org. Chem. 1991, 56, 3177−3178.
Posakony, J. J.; Tewson, T. J. Synthesis 2002, 766−770. Posakony, J. J.;
Tewson, T. J. Synthesis 2002, 859−864. RuCl3−3H2O and NaIO4
(b) Wei, L.; Lubell, W. D. Can. J. Chem. 2001, 79, 94−104. Nikolai, A.
K.; Manfred, H. Helv. Chim. Acta 2003, 86, 2028−2057. Dolence, E.
K.; Mayer, G.; Kelly, B. D. Tetrahedron: Asymmetry 2005, 16, 1583−
1594. Peters, R.; Fischer, D. F Org. Lett. 2005, 7, 4137−4140 RuCl3
and NaIO4.
(9) (a) Zubovics, Z.; Toldy, L.; Varro, A.; Rabloczky, G.; Kurthy, M.;
Dvortsak, P.; Jerkovich, G.; Tomori, E. Eur. J. Med. Chem. 1986, 21,
370. KMnO4 and AcOH (b) Andersen, Kenneth K.; Bray, Diana D.;
Chumpradit, Sumalee; Clark, M. E.; Habgood, G. J.; Hubbard, C. D.;
Young, K. M. J. Org. Chem. 1991, 56, 3177−3178.
mCpBA
(c) Roennholm, P.; Soedergren, M.; Hilmersson, G. Org. Lett. 2007, 9,
3781−3783. RuCl3, NaIO4, and SiO2 (d) Zanatta, N.; Borchhardt, D.
M.; Flores, D. C.; Coelho, H. S.; Marchi, T. M.; Flores, A. F. C.;
Bonacorso, H. G.; Martins, M. A. P. J. Heterocycl. Chem. 2008, 45,
335−341 RuCl3 and KIO4.
(10) Fuller, R. W.; Molloy, B. B.; Day, W. A.; Roush, B. W.; Marsh,
M. M. J. Med. Chem. 1973, 16, 101.
(11) For general nucleophilic displacements of 1,2- and 1,3-cyclic
sulfamidates, see (a) Meunier, N.; Veith, U.; Jager, V. Chem. Commun.
1996, 331−332. (b) Pound, M. K.; Davies, D. L.; Pilkington, M.;
Sousa, M.; Wallis, J. D. Tetrahedron Lett. 2002, 43, 1915−1918.
Synthesis of fluoroamines: (c) Posakony, J. J.; Tewson, T. J. Synthesis
2002, 766−770. (d) Posakony, J. J.; Tewson, T. J. Synthesis 2002,
859−864. (e) Ok, D.; Fisher, M. H.; Wyvratt, M. J.; Meinke, P. T.
Tetrahedron Lett. 1999, 40, 3831−3834. (f) Lyle, T. A.; Magill, C. A.;
Pitzenberger, S. M. J. Am. Chem. Soc. 1987, 109, 7890−7891.
(g) Thompson, W. J.; Anderson, P. S.; Britcher, S. F.; Lyle, T. A.;
Thies, J. E.; Magill, C. A.; Varga, S. L.; Schwering, J. E.; Lyle, P. A.;
Christy, M. E.; Evans, B. E.; Colton, C. D.; Holloway, M. K.; Springer,
J. P.; Hirshfield, J. M.; Ball, R. G.; Amato, J. S.; Larsen, R. D.; Wong, E.
H. F.; Kemp, J. A.; Tricklebank, M. D.; Singh, L.; Oles, R.; Priestly, T.;
Marshall, G. R.; Knight, A. R.; Middlemiss, D. N.; Woodruff, G. N.;
Iversen, L. L. J. Med. Chem. 1990, 33, 789−808. (h) Van Dort, M. E.;
Jung, Y.-W.; Sherman, P. S.; Kilbourn, M. R.; Wieland, D. M. J. Med.
Chem. 1995, 38, 810−815. Synthesis of chiral ether ligands:
(i) Okuda, M.; Tomioka, K. Tetrahedron Lett. 1994, 35, 4585−4586.
In manipulating carbohydrates: (j) Aguilera, B.; Fernan
A. Chem. Commun. 1996, 127−128. (k) Aguilera, B.; Fernan
́
dez-Mayoralas,
dez-
́
Mayoralas, A.; Jaramillo, C. Tetrahedron 1997, 53, 5863−5876.
(l) Aguilera, B.; Fernandez-Mayoralas, A. J. Org. Chem. 1998, 63,
2719−2723. Manipulation of R-amino acids, including serine
derivatives: (m) Baldwin, J. E.; Spivey, A. C.; Schofield, C. J.
Tetrahedron: Asymmetry 1990, 1, 881−884. (n) Boulton, L. T.; Stock,
H. T.; Raphy, J.; Horwell, D. C. J. Chem. Soc., Perkin Trans. 1 1999,
1421−1430. (o) Wei, L.; Lubell, W. D. Org. Lett. 2000, 2, 2595−2598.
(p) Wei, L.; Lubell, W. D. Can. J. Chem. 2001, 79, 94−104. (q) Atfani,
M.; Wei, L.; Lubell, W. D. Org. Lett. 2001, 3, 2965−2968. (r) Cohen,
S. B.; Halcomb, R. L. Org. Lett. 2001, 3, 405−407. (s) Cohen, S. B.;
Halcomb, R. L. J. Am. Chem. Soc. 2002, 124, 2534−2543. Synthesis of
2-substituted pyrrolidines: (t) Cooper, G. F.; McCarthy, K. E.; Martin,
M. G. Tetrahedron Lett. 1992, 33, 5895−5896.
H
DOI: 10.1021/op500264v
Org. Process Res. Dev. XXXX, XXX, XXX−XXX