Development of a large-scale synthesis of sulphostin, a dipeptidyl peptidase IV inhibitor

Article abstract of DOI:10.1021/op058000c

For the progress of the in vivo study on sulphostin, a dipeptidyl peptidase IV inhibitor, its large-scale synthetic method was investigated. The optical resolution of (3S,RS_P)-1-amino(sulfoamino)phosphinyl-3- benzyloxycarbonylamino-2-piperidinone, which was the most difficult step in the previous method, was simplified by using fractional crystallization. The use of 2 mol equiv of (1S,2R)-(+)-2-amino-1,2-diphenylethanol for optical resolution gave desired diastereomer 15 in good yield as a less soluble salt. In the present synthetic method, there were no requirements for purification using column chromatography, reaction at cryogenic temperature, and treatment using the haloalkane solvents. The total yield of the new method was 4.6%, which was an improvement of approximately 2-fold compared to the method reported previously.

Full text of DOI:10.1021/op058000c

Organic Process Research & Development 2005, 9, 570−576
Development of a Large-Scale Synthesis of Sulphostin, a Dipeptidyl Peptidase
IV Inhibitor
,
†
†
†
†
†
†
Masatoshi Abe,* Masashi Nagai, Keiichiro Yamamoto, Hiroko Yamazaki, Ichiro Koga, Yoshitaka Satoh,
‡
§
†
Yasuhiko Muraoka, Shuji Kurashige, and Yuh-ichiro Ichikawa
Research and DeVelopment DiVision, Pharmaceuticals Group, Nippon Kayaku Co. Ltd.,
3
1
3
1-12, Shimo 3-chome, Kita-ku, Tokyo 115-8588, Japan, Microbial Chemistry Research Center,
4-23, Kamiosaki 3-chome, Shinagawa-ku, Tokyo 141-0021, Japan, and Synthetic Group, NAC Co. Ltd.,
1-12, Shimo 3-chome, Kita-ku, Tokyo 115-0042, Japan
Abstract:
For the progress of the in vivo study on sulphostin, a dipeptidyl
peptidase IV inhibitor, its large-scale synthetic method was
investigated. The optical resolution of (3S,RS )-1-amino(sul-
P
foamino)phosphinyl-3-benzyloxycarbonylamino-2-piperidin-
one, which was the most difficult step in the previous method,
was simplified by using fractional crystallization. The use of 2
mol equiv of (1S,2R)-(+)-2-amino-1,2-diphenylethanol for opti-
cal resolution gave desired diastereomer 15 in good yield as a
less soluble salt. In the present synthetic method, there were
no requirements for purification using column chromatography,
reaction at cryogenic temperature, and treatment using the
haloalkane solvents. The total yield of the new method was
Figure 1.
an insulin-releasing hormone, is cleaved and inactivated by
6
DPP-IV, and DPP-IV inhibitors suppress the rise in glucose
7
,8
level in blood after a meal. The potent DPP-IV inhibitors
9
10
such as NVP-DPP728 and P32/98 are currently being
evaluated in clinical trials as therapeutic agents for type 2
diabetes. In the immune system, CD26, which is identical
to DPP-IV, has been reported to affect the proliferation and
4.6%, which was an improvement of approximately 2-fold
compared to the method reported previously.
1
1,12
activation of T cells.
It has been confirmed that two DPP-
1
3
14
IV inhibitors, Ala-boroPro and prodipine, suppress the
immune response in vivo. The DPP-IV inhibitor has been
expected to be a therapeutic agent for type 2 diabetes or
immune-related disorders. Thus, sufficient quantities of
sulphostin were a requisite for the in vivo studies.
Introduction
Sulphostin (1), a dipeptidyl peptidase IV (DPP-IV, CD26,
EC 3.4.14.5) inhibitor (IC50 ) 21 nM), which has two
asymmetric atoms at C-3 and phosphorus, was isolated from
2
However, in our previously used method, three difficult
1
the culture broth of Streptomyces sp. MK251-43F3. We
steps were involved in large-scale synthesis; i.e., separation
of the diastereomeric mixture using column chromatography,
reaction at a cryogenic temperature, and purification of the
reaction product using haloalkane solvents. Hence, to
circumvent these difficult steps, the development of a new
synthetic method was required. We evaluated the improve-
ment in yield in the available synthetic method of sulphostin
and simplification of reactions for scale-up. In this paper,
we report the large-scale synthesis of sulphostin including
synthesized four possible stereoisomers of sulphostin and
determined the configurations of the natural product by
2
comparison with the four stereoisomers. Results of the
synthetic study revealed that 3-epi sulphostin (2) also had a
strong inhibitory activity (IC50 ) 31 nM) (Figure 1).
DPP-IV is a cell surface serine exopeptidase, which
selectively cleaves the N-terminal dipeptides from polypep-
tides with proline or alanine residue at the penultimate
position.3 Glucagon-like peptide-1 (7-36 amide, GLP-1),
-5
(6) Mentlein, R.; Gallwitz, B.; Schmidt, W. E. Eur. J. Biochem. 1993, 214,
829.
*
To whom correspondence should be addressed. Telephone: +81 3 3598
(7) Balkan, B.; Kwasnik, L.; Miserendino, R.; Holst, J. J.; Li, X. Diabetologia
1999, 42, 1324.
(8) Pederson, R. A.; White, H. A.; Schlenzig, D.; Pauly, R. P.; McIntosh, C.
H.; Demuth, H.-U. Diabetes 1998, 47, 1253.
5
234. Fax: +81 3 3598 5422. E-mail: masatoshi.abe@nipponkayaku.co.jp.
†
Nippon Kayaku Co. Ltd.
Microbial Chemistry Research Center.
NAC Co. Ltd.
‡
§
(9) Ahr e´ n, B.; Simonsson, E.; Larsson, H.; Landin-Olsson, M.; Torgeirsson,
H.; Jansson, P. A.; Sandqvist, M.; Båvenholm, P.; Efendic, S.; Eriksson, J.
W.; Dickinson, S.; Holmes, D. Diabetes Care 2002, 25, 869.
(10) Sorbera, L. A.; Revel, L.; Casta n˜ er, J. Drugs Future 2001, 26, 859.
(11) Bristol, L. A.; Sakaguchi, K.; Appella, E.; Doyle, D.; Tak a´ cs, L. J. Immunol.
1992, 149, 367.
(12) Vivier, I.; Marguet, D.; Naquet, P.; Bonicel, J.; Black, D.; Li, C. X.; Bernard,
A. M.; Gorvel, J. P.; Pierres, M. J. Immunol. 1991, 147, 447.
(13) Tanaka, S.; Murakami, T.; Horikawa, H.; Sugiura, M.; Kawashima, K.;
Sugita, T. Int. J. Immunopharmacol. 1997, 19, 15.
(
1) Akiyama, T.; Abe, M.; Harada, S.; Kojima, F.; Sawa, R.; Takahashi, Y.;
Naganawa, H.; Homma, Y.; Hamada, M.; Yamaguchi, A.; Aoyagi, T.;
Muraoka, Y.; Takeuchi, T. J. Antibiot. 2001, 54, 744.
(
(
(
2) Abe, M.; Akiyama, T.; Nakamura, H.; Kojima, F.; Harada, S.; Muraoka,
Y. J. Nat. Prod. 2004, 67, 999.
3) Heins, J.; Welker, P.; Sch o¨ nlein, C.; Born, I.; Hartrodt, B.; Neubert, K.;
Tsuru, D.; Barth, A. Biochim. Biophys. Acta 1988, 954, 161.
4) Augustyns, K.; Bal, G.; Thonus, G.; Belyaev, A.; Zhang, X. M.; Bollaert,
W.; Lambeir, A. M.; Durinx, C.; Goossens, F.; Haemers, A. Curr. Med.
Chem. 1999, 6, 311.
(14) Korom, S.; De Meester, I.; Stadlbauer, T. H.; Chandraker, A.; Schaub, M.;
Sayegh, M. H.; Belyaev, A.; Haemers, A.; Scharp e´ , S.; Kupiec-Weglinski,
J. W. Transplantation 1997, 63, 1495.
(
5) De Meester, I.; Korom, S.; Van Damme, J.; Scharp e´ , S. Immunol. Today
1
999, 20, 367.
5
70
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development
10.1021/op058000c CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/12/2005

Scheme 1
Table 2. Stability of compound 4 in various pH conditions
for 1 h
run
pH
temp
4
3
1
2
3
4
5
6
7
8
6.0
8.0
9.0
10.0
10.6
11.0
11.0
13.7
rt
rt
rt
rt
0 °C
rt
0 °C
0 °C
97.7
>99
<1.0
<1.0
2.5
12.5
4.8
71.7
31.7
37.3
97.5
85.1
82.7
26.8
67.3
62.7
Table 1. Various reagents for activation of lactam amide (3)
a number of spots were observed on TLC. Next, the
1
6
activation by the treatment with chlorotrimethylsilane and
diisopropylethylamine in THF at room temperature was
examined. Diaminophosphinylation by silylation was suc-
cessful and resulted in the formation of desired compound
4 and bis((3S)-3-benzyloxycarbonylamino-2-piperidinone-1-
yl)phosphinylamide (6) along with the starting material 3.
Diaminophosphinylation in toluene also gave a similar result.
Next, the amount of phosphoryl chloride was examined, and
the employment of 4 equiv revealed that only compound 4
was produced.
run
reagents
solvents
POCl
3
temp
resultsa
1
2
3
4
NaHCO
3
THF
THF
toluene
THF
1 equiv rt
1 equiv rt
1 equiv rt
no reaction
no reaction
no reaction
i
i
Pr
Pr
2
EtN
EtN
2
NaH
1 equiv -78°C-rt multiple
products
i
5
6
7
TMS-Cl, Pr
2
2
2
EtN THF
EtN toluene
EtN toluene
1 equiv rt
1 equiv rt
4 equiv rt
3 + 4 + 6
3 + 4 + 6
3 + 4
i
TMS-Cl, Pr
i
TMS-Cl, Pr
a
Then, amidation using an aqueous ammonia solution
instead of a gaseous reagent was attempted, and a similar
result was obtained. However, it was proven that compound
The results were determined by analyzing the spots on TLC after compound
3
was treated with an activating reagent and phosphoryl chloride, followed by
bubbling with gaseous ammonia.
4
decomposed into compound 3 in the basic solution. Thus,
the practical separation method of the diastereomeric mixture
by fractional crystallization.15
we investigated the stability of compound 4 in the basic
solution in order to determine the pH of the aqueous
ammonia solution (Table 2). The results showed that
compound 4 decomposed into compound 3 in the solution
with a pH value higher than 11 even at 0 °C, and it was
comparatively stable in the solution between pH 7 and near
pH 10. On the basis of these findings, the dichlorophosphinyl
intermediate was treated with aqueous ammonia solution with
the pH value of 10.6 in an ice bath; this solution was prepared
with 14% aqueous ammonia saturated with ammonium
chloride. However, the pH value of the aqueous ammonia
solution becomes acidic since hydrogen chloride is produced
from the dichlorophosphinyl intermediate and excess phos-
phoryl chloride when the reaction mixture is added into the
ammonia solution. Therefore, the pH value of the ammonia
solution was controlled and maintained between 9 and 11
by adding 28% aqueous ammonia solution when the pH
value turned under 9.
Next, the purification procedure of compound 4 was
examined. Compound 4 and ammonium chloride exist as a
precipitate in the above-mentioned reaction mixture, and
compound 4 scarcely dissolved in the nonpolar organic
solvents with the exception of haloalkane solvents. Thus,
we compared the solubilities of compound 4 and ammonium
chloride in several solvents to examine the purification
technique not by the liquid-liquid extraction but by the
solid-liquid extraction. The solubility of compound 4 in
THF-MeOH (1:1) was approximately 10-fold higher than
Results and Discussion
The previously used synthetic route of sulphostin is shown
2
in Scheme 1. Since this synthetic route was simple and short,
we decided to modify it for large-scale synthesis. (3S)-3-
Benzyloxycarbonylamino-2-piperidinone (3) could be easily
prepared from L-ornithine (L-Orn) hydrochloride by using
2
the previous method. Therefore, we focused on the difficult
steps: diaminophosphinylation of compound 3, sulfonation
of (3S)-3-benzyloxycarbonylamino-1-diaminophosphinyl-2-
piperidinone (4), and separation of the diastereomeric
mixture.
Diaminophosphinylation of (3S)-3-Benzyloxycarbon-
2
ylamino-2-piperidinone (3). In the previous report, lactam
amide of compound 3 was activated with n-butyllithium at
-
78 °C, and liquid ammonia was added to the reaction
mixture for amidation of the dichlorophosphinyl intermediate.
Since this reaction was carried out at a cryogenic temperature,
special facilities were required and also attention was paid
to use a volatile amidation reagent. Moreover, the reaction
product was extracted with CHCl -MeOH (19:1) for
3
purification. These steps were not an option in the case of
large-scale synthesis.
First, we examined the activating reagent of lactam amide
to determine if it can be safely used at room temperature
(Table 1); treatment with sodium hydrogen carbonate and
diisopropylethylamine did not give desired compound 4. In
the case of sodium hydride, although the reaction proceeded,
(16) Murakami, M.; Sato, N.; Hashimoto, S.; Kawamura, T. JP Patent 5,4012,-
(15) The corresponding patent: WO2003051895/US20050020834.
391, 1979.
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
571

that of ammonium chloride. In contrast, the solubility of
Table 3. Sulfonation of compound 4 using various reagents
for 1 h
compound 4 in water was 2 mg/mL or lower, and that of
i
i
ammonium chloride was high. In the case of PrOH, Pr O-
2
EtOH (5:1), and some aprotic solvents, the solubility of
compound 4 was 2 mg/mL or lower. From these results,
toluene, a reaction solvent, and toluene-THF (1:1) were
chosen for the removal of some lipophilic materials from
the precipitate containing ammonium chloride and compound
resultsa
4, and then THF-MeOH (1:1) was selected for the extraction
additional
reagents
4
7
8
9
of compound 4 from the residual precipitate. Moreover, water
was selected for the removal of ammonium chloride from
run reagents
temp
(%) (%) (%) (%)
the solid residue, which was obtained by evaporation of the
1
2
3
4
5
6
7
SO
(1.2 equiv)
SO ‚py
2.0 equiv)
SO ‚py
(1.2 equiv)
SO ‚DMF 4 °C
1.2 equiv)
SO ‚DMF 4 °C
1.2 equiv)
SO ‚Me
1.2 equiv)
SO ‚Me
1.2 equiv)
3
‚py
4 °C
25.1 31.4 31.9 11.7
1.5 24.7 22.5 50.8
15.6 28.2 29.5 21.0
11.4 28.2 28.2 17.1
27.7 29.5 28.2 13.8
i
solution extracted with THF-MeOH (1:1), and then Pr
2
O-
3
4 °C
EtOH (5:1) was chosen for the removal of the trace lipophilic
materials and water.
(
3
50 °C py
(1.2 equiv)
Consequently, the diaminophosphinylation process was
carried out in seven continuous procedures: treatment of
compound 3 with chlorotrimethylsilane and diisopropylethyl-
amine, addition of phosphoryl chloride, pouring of the
reaction mixture into the aqueous ammonia solution with
pH 9-11, collection of the precipitate in the aqueous solution
by filtration, washing of the precipitate with toluene and
toluene-THF (1:1), extraction of compound 4 with THF-
MeOH (1:1), followed by evaporation and washing of the
3
(
3
lutidine
(
(1.2 equiv)
3
3
N
4 °C
95.5
0
0
0
(
3
3
N
rt, 24 h
79.7 3.2 3.2 0.2
(
a
The results were determined by reversed-phase HPLC analysis. Compound
was also detected as a decomposition product of compound 4.
3
resultant solid residue with water and then a mixture of
i
2
Pr O-EtOH (5:1). The desired compound 4 was obtained
by these procedures in 41% yield (purity 98%, optical purity
pound 11, as a less soluble salt, in which the configuration
at the phosphorus atom was S (Scheme 2). The optical purity
of compound 11 was 90% de, which was determined by
HPLC analysis. Additionally, the desired diastereomer 12,
in which the configuration of the phosphorus atom was R,
in the mother liquid had similar optical purity of 90% de.
Deprotection of compound 11 gave P-epi sulphostin (13) with
the optical purity of 98.5% de.
>
99% de).
Sulfonation of (3S)-3-Benzyloxycarbonylamino-1-di-
aminophosphinyl-2-piperidinone (4). In the previous re-
2
port, sulfonation of compound 4 using sulfur trioxide gave
the desired compounds in low yield. On investigating the
byproduct, it was found that the cause of the low yield was
the production of (3S)-3-benzyloxycarbonylamino-1-bis-
On the other hand, the addition of 2 equiv of ADPE
selectively gave compound 15 having the desired configu-
ration at the phosphorus atom as a less soluble salt. The
optical purity of compound 15 was 95% de. Therefore, we
selected the method using 2 equiv of ADPE for the separation
of the diastereomeric mixture. After deprotection of com-
pound 15 by hydrogenolysis in the aqueous acetic acid
solution, the desired compound was precipitated by adding
EtOH. In this reaction, ADPE was easily removed because
an acidic solution was used. Next, recrystallization of
resultant crude sulphostin from water-EtOH yielded the pure
material, of which the optical purity was >99% de. Conse-
quently, it was proven that the addition of 2 equiv of ADPE
resulted in the desired diastereomer 15 with a high optical
purity in good yield, and pure sulphostin (1) was obtained
from compound 15.
(
sulfoamino)phosphinyl-2-piperidinone (9) due to the over-
reaction. Thus, reaction temperature, the reagent, and its
quantity were carefully examined to suppress the production
of compound 9 (Table 3). The reactivity changed when a
different complex of sulfur trioxide was used (DMF >
pyridine > Me
9
reaction temperature, the quantity of the reagent, and its
complex. Pyridine/sulfur trioxide complex (1.2 equiv), which
gave the desired compound 8 in best yield, was chosen for
sulfonation.
3
N). However, the production of compound
could not be suppressed, despite reexamination of the
P
Separation of (3S,RS )-1-Amino(sulfoamino)phos-
phinyl-3-benzyloxycarbonylamino-2-piperidinone (10). In
the previous report, separation of the diastereomeric mixture
2
was performed by repeated Diaion HP-20SS column chro-
matography. In the case of the large-scale synthesis, purifica-
tion using column chromatography is not practical. Hence,
separation of the diastereomeric mixture by fractional
crystallization was examined. We hypothesized that the
diastereomeric mixture could be separated by preparing the
salt with an optically active amine. The various optically
active amines for fractional crystallization were examined,
and it was found that addition of 1 equiv of (1S,2R)-(+)-2-
amino-1,2-diphenylethanol (ADPE) selectively gave com-
Large-Scale Synthesis of Sulphostin. On the basis of
these results, the large-scale synthesis of sulphostin was
carried out by the method shown in Scheme 3. L-Orn
(
17) Ten or over kinds of optically active amines (e.g., S-(-)-1-phenylethylamine,
R-(+)-1-phenylethylamine, R-(-)-2-amino-1-butanol, R-(+)-1-(1-naphthyl)-
ethylamine, etc.) were examined; however, crystallization of resultant
sulfonic acid salts of those amines could not be performed. In the case of
17
(
1R,2S)-(-)-2-amino-1,2-diphenylethanol, the resultant salt was crystalline,
which showed no optical activity.
572
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

Scheme 2
Scheme 3
hydrochloride (16) (1250 g) in MeOH was treated with
thionyl chloride at 55 °C to give L-Orn-OMe dihydrochloride
Conclusion
The previously reported method for synthesis of sulphostin
(
17). Cyclization of compound 17 with the aqueous sodium
was reexamined in order to improve its large-scale synthesis.
The optimized procedure does not require cryogenic tem-
perature reaction, column chromatography, and the use of
haloalkane solvents. The yield of sulphostin from compound
3 was 6.1%, which was a 2-fold improvement in comparison
carbonate solution was carried out at room temperature.
Continuous addition of sodium hydrogen carbonate and
benzyl chloroformate into the cyclization mixture gave
compound 3 (purity >99%, optical purity >99% ee) in 75%
yield from compound 16. Reaction of compound 3 (600 g)
with chlorotrimethylsilane and diisopropylethylamine at room
temperature, followed by treatment with phosphoryl chloride,
gave a phosphinyl chloride intermediate, which was poured
into an aqueous ammonia solution with the pH value of 9-11
to give compound 4 (purity >99%, optical purity >99% ee)
in 44% yield. After treatment of compound 4 (294 g) with
2
with that of our previously reported method. The required
time for the synthesis of sulphostin had been drastically
reduced because of the discovery of fractional crystallization
using 2 equiv of ADPE. Additionally, the large-scale
synthesis of 3-epi sulphostin (2), which also had a strong
DPP-IV inhibitory activity, from D-Orn hydrochloride could
be possible using a similar method, which employed 1 equiv
of (1R,2S)-(-)-2-amino-1,2-diphenylethanol as an optically
active amine reagent for the separation of the diastereomeric
mixture.
1
.2 equiv of pyridine/sulfur trioxide complex at 0-5 °C,
addition of 2.5 equiv of ADPE to the reaction mixture
resulted in sulfonic acid 2 ADPE salt (15) (optical purity
92% de). Deprotection of compound 15 was carried out by
hydrogenolysis using palladium-black catalyst in the aqueous
acetic acid solution to give crude sulphostin (purity 98.5%,
optical purity 95% de) in 20% yield from compound 4. The
crude material was recrystallized from water-EtOH to give
pure sulphostin (1) (purity >99%, optical purity >99% de)
in 74% yield. The physicochemical and biological properties
of the pure material were identical with those of previously
Experimental Section
General Experimental Procedures. Optical rotations
were measured with a Perkin-Elmer Model 241 polar meter.
1
13
31
H NMR, C NMR, and P NMR spectra were recorded
with a Varian Gemini 200, Bruker Avance 400, or JEOL
JNM-ECA600 spectrometer. TMS or 3-(trimethylsilyl)-
propionic-2,2,3,3-d
standard substance for H and C NMR spectra, and 85%
PO P
was used as an external standard substance for 31
4
acid sodium salt was used as an internal
2
1
13
reported sulphostin. Finally, we examined the recovery of
ADPE because it is an expensive reagent and 76% of ADPE
was recovered from the optical resolution and deprotection
procedure.
H
3
4
NMR spectra. HR-ESIMS spectra were measured with a
JEOL JMS-T100LC.
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
573

(
3S,SR
P
)-1-Amino(sulfoamino)phosphinyl-3-benzyloxy-
were combined, and EtOH (2.0 L) was slowly added to the
combined aqueous solution. The precipitate was collected
by filtration to give compound 13 (24.5 g, 98.5% de) in 71%
yield. To prepare the optical analytical specimen, compound
13 (20 mg) was reacted with benzyl chloroformate (0.025
carbonylamino-2-piperidinone Sodium Salt (10). To a
solution of compound 4 (10.00 g, 30.6 mmol) in DMF (80
mL) was added SO
at 0 °C, and the mixture was stirred for 12 h at 6-8 °C. To
the mixture were added water (100 mL) and NaHCO (6.72
3
/pyridine complex (6.35 g, 40.0 mmol)
3
mL) and NaHCO (33 mg) in water (10 mL) and THF (5
3
g, 80.0 mmol), and the mixture was stirred for 5 min at room
temperature. The reaction mixture was diluted with water
mL) at room temperature. After 30 min, the aqueous solution
of the reaction mixture was measured by reversed-phase
HPLC analysis. The optical purity was obtained from a rate
(400 mL) and was purified by Diaion HP-20SS column
2
0
chromatography (water-30% MeOH(aq)) to give compound
of diastereomer in this reaction mixture. [R]
+43.8° (c
D
1
1
1
1
3
2
0 (5.80 g) in 45% yield. H NMR (D O, 600 MHz) δ:
0.5, H O); H NMR (400 MHz, D O) δ: 1.85-2.12 (3H,
2
2
.73-2.07 (3H, m, H-4a, H-5), 2.21 (1H, m, H-4b), 3.56-
.82 (2H, m, H-6), 4.23 and 4.31 (1H, m, H-3), 5.13-5.23
m, H-4a and H-5), 2.41 (1H, m, H-4b), 3.63-3.74 (2H, m,
1
3
H-6), 4.13 (1H, dd, J ) 6.5,7.0 Hz, H-3); C NMR (D
2
O,
31
(2H, m, OCH
2
Ph), 7.42-7.50 (5H, m, Ph); P NMR (D
43 MHz) δ: 4.87 and 5.31.
Separation of (3S,RS )-1-Amino(sulfoamino)phos-
2
O,
1
00 MHz) δ: 23.1 (d, C-5), 26.7 (C-4), 47.8 (C-6), 53.4 (d,
2
C-3), 174.5 (C-2).
P
P
Separation of (3S,RS )-1-Amino(sulfoamino)phos-
phinyl-3-benzyloxycarbonylamino-2-piperidinone Sodium
Salt(10)Using1equivof(1S,2R)-(+)-2-Amino-1,2-diphenyl-
ethanol (ADPE). To a solution of compound 10 (1.05 g,
phinyl-3-benzyloxycarbonylamino-2-piperidinone Sodium
Salt (10) Using 2 equiv of ADPE. To a solution of
compound 10 (6.94 g, 16.2 mmol) in water (200 mL) and
EtOH (350 mL) were added ADPE (7.96 g, 37.3 mmol) and
2
2
.45 mmol) in water (50 mL) were added ADPE (0.522 g,
.45 mmol) and 1 N HCl(aq) (2.45 mL, 2.45 mmol). The
1
N HCl(aq) (16.2 mL, 16.2 mmol). The suspension was
suspension was dissolved by heating to 90 °C and was
allowed to cool to room temperature. The precipitate was
collected by filtration to give (3S,S )-1-amino(sulfoamino)-
P
phosphinyl-3-benzyloxycarbonylamino-2-piperidinone ADPE
dissolved by heating to internal temperature at 55 °C and
was allowed to cool to room temperature. The precipitate
was collected by filtration to give (3S,R )-1-amino(sulfoami-
P
no)phosphinyl-3-benzyloxycarbonylamino-2-piperidinone 2
salt (11) (0.66 g) in 42% yield. The optical purity of
ADPE salt (15) (5.40 g) in 40% yield. The optical purity of
compound 11 was 90% de, and that of (3S,R
P
)-compound
compound 15 was 95% de, and that of (3S,S
4 in the filtrate was 80% de.
P
)-compound
12 in the filtrate was also 90% de. Each optical purity was
1
measured by reversed-phase HPLC (conditions; column,
Senshu Pak PEGASIL ODS 4.6 mm × 250 mm (rt); eluent,
MeCN/1% phosphoric acid(aq) ) 14:86; flow rate, 1.0 mL/
min; detection, UV 210 nm). The retention times of
compounds 11, 12, and ADPE were 27, 29, and 17 min,
respectively.
1
Physicochemical Data of Compound 15: H NMR
18
(
3
CD OD, 600 MHz) δ: 1.74-1.84 (2H, m, H-4a and H-5a),
2.07 (1H, m, H-5b), 2.14 (1H, m, H-4b), 3.57 (1H, m, H-6a),
3
.86 (1H, m, H-6b), 4.27 (1H, m, H-3), 4.28 (2H, d, J ) 4.8
Hz, CH(NH
2
)CH(OH) × 2), 5.04 (2H, d, J ) 4.8 Hz, CH-
(
NH
2
)CH(OH) × 2), 5.09 (2H, s, OCH
2
Ph), 7.12-7.37 (25H,
Physicochemical Data of Compound 11: 1H NMR
13
18
m, Ph × 5); C NMR (CD
3
OD, 150 MHz) δ: 21.9 (C-5),
(
(
(
CD
3
OD, 600 MHz)18 δ: 1.77 (1H, m, H-4a), 1.85-2.00
2H, m, H-5), 2.21 (1H, m, H-4b), 3.63 (1H, m, H-6a), 3.79
1H, m, H-6b), 4.23 (1H, m, H-3), 4.41 (1H, d, J ) 4.8 Hz,
2
(
1
1
7.7 (C-4), 45.6 (C-6), 53.4 (C-3), 62.1 (CHNH
OCH Ph), 76.6 (CHOH), 127.8 (Ph), 128.7 (Ph), 128.8 (Ph),
28.9 (Ph), 129.0 (Ph), 129.1 (Ph), 129.4 (Ph), 129.5 (Ph),
38.2 (Ph), 142.0 (Ph), 158.8 (NHCOO), 175.4 (C-2); 31
OD, 243 MHz) δ: 5.31; HR-ESIMS m/z: calcd
2
), 67.7
2
CH(NH
.8 Hz, CH(NH
6H, m, Ph), 7.23-7.30 (4H, m, Ph), 7.31-7.36 (3H, m,
2 2
)CH(OH)), 5.07 (2H, s, OCH Ph), 5.16 (1H, d, J )
P
4
2
)CH(OH)), 7.09 (2H, m, Ph), 7.17-7.22
NMR (CD
3
(
-
18 4 7
for C13H N O PS (M-2ADPE-H) 405.0634, found
13
18
Ph); C NMR (CD
C-4), 45.9 (C-6), 53.4 (C-3), 61.8 (CHNH
Ph), 75.1 (CHOH), 127.6 (Ph), 128.8 (Ph), 128.9 (Ph), 129.0
3
OD, 150 MHz) δ: 22.4 (C-5), 28.1
4
05.0644.
Small-scale Synthesis of (3S,R
sulfoamino)phosphinyl-2-piperidinone: Sulphostin (1).
(
2
), 67.6 (OCH
2
-
P
)-3-Amino-1-amino-
(
(
(
(
Ph), 129.1 (Ph), 129.2 (Ph), 129.6 (Ph), 129.8 (Ph), 138.2
To a suspension of compound 15 (1.00 g, 1.20 mmol, 95%
de) in acetic acid (2 mL) and water (5 mL) was added Pd-
black (0.050 g), and the suspension was stirred for 2 h at
room temperature under a hydrogen atmosphere. The reaction
mixture was filtrated to remove Pd-black, and the residue
was washed with water (6 mL). The filtrate and the washing
solution were combined, and then EtOH (21 mL) was slowly
added. The precipitate was collected by filtration to give
crude sulphostin (0.28 g, 98.6% de) in 81% yield. Recrys-
tallization of crude sulphostin from water-EtOH gave pure
Ph), 142.0 (Ph), 158.5 (NHCOO), 175.7 (C-2); 3 P NMR
1
CD
3
OD, 243 MHz) δ: 6.04; HR-ESIMS m/z: calcd for
-
C
13 18 4 7
H N O PS (M-ADPE-H) 405.0634, found 405.0627.
P
(3S,S )-3-Amino-1-amino(sulfoamino)phosphinyl-2-pi-
peridinone: P-epi Sulphostin (13). To a suspension of
compound 11 (73.56 g, 116 mmol, 88.4% de) in acetic acid
(150 mL) and water (375 mL) was added Pd-black (3.65 g),
and the suspension was stirred for 24 h at room temperature
under a hydrogen atmosphere. The reaction mixture was
filtrated to remove Pd-black, and the residue was washed
with water (500 mL). The filtrate and the washing solution
2
0
1
compound 1 (>99% de): [R]
NMR (400 MHz, D O) δ: 1.85-2.02 (2H, m, H-4a and
D
2
-21.8° (c 5.0, H O); H
2
18) As an internal standard substance for 1H and 13C NMR spectra, a solvent
signal (3.30 and 49.0 ppm) was used, respectively.
H-5a), 2.11 (1H, m, H-5b), 2.40 (1H, m, H-4b), 3.65 (1H,
m, H-6a), 3.79 (1H, m, H-6b), 4.15 (1H, dd, J ) 11.9, 6.9
(
574
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

1
3
Hz, H-3); C NMR (100 MHz, D
C-4), 47.5 (C-6), 53.4 (d, C-3), 174.5 (C-2).
Large-Scale Synthesis of Sulphostin. (3S)-3-Benzyl-
2
O) δ: 22.6 (d, C-5), 26.3
of 28% ammonia(aq) in an ice bath. The precipitate was
collected by filtration and was washed with toluene (6.0 L)
and a solution (4.5 L) of toluene-THF (1:1), followed by
extraction with a solution (10 L) of THF-MeOH (1:1). After
evaporation of the extract under reduced pressure, the solid
(
oxycarbonylamino-2-piperidinone (3). To a suspension of
L-ornithine hydrochloride (16) (1250 g, 7.40 mol) in MeOH
(
3.7 L) was slowly added thionyl chloride (568 mL, 7.78
residue was washed with water (6.0 L) and then with a
i
mol) keeping the internal temperature below 0 °C, and the
mixture was stirred for 3 h at 50-55 °C. The reaction
mixture was evaporated under reduced pressure, and the
solution (10 L) of EtOH- Pr
2
O (1:5). The precipitate was
dried at 40 °C under reduced pressure to give compound 4
(323 g, purity 98.4%, optical purity >99% ee) in 41% yield.
The optical purity was measured by HPLC analysis using a
chiral column (conditions were similar to those described
for compound 3). The retention time of compound 4 was
i
residue was crystallized by trituration in Pr
2
O (4.0 L). The
crystals were collected by filtration and dried at room
temperature and then at 40 °C under reduced pressure to
give L-ornithine methyl ester dihydrochloride (17). The
accurate quantity of compound 17 could not be measured
due to being hydroscopic. To a solution of compound 17 in
18.9 min. For reference, that of 3-epimer of compound 4
1
was 15.7 min. H NMR (DMSO-d
6
, 200 MHz) δ: 1.62 (1H,
m, H-4a), 1.73-1.78 (2H, m, H-5), 2.01 (1H, m, H-4b), 3.46
water (3.6 L) was slowly added a solution of Na
2
CO
3
(1202
(1H, m, H-6a), 3.58 (1H, m, H-6b), 4.09 (1H, m, H-3), 4.14
g, 11.36 mol) in water (6.0 L) keeping the internal temper-
ature below 5 °C. After the mixture was stirred at the same
temperature for 2 h, stirring of the reaction was continued
for 10 h at room temperature. To the reaction mixture were
(2H, br s, PNH
Ph), 7.28-7.41 (6H, m, Ph and NHCOO).
(3S,R )-1-Amino(sulfoamino)phosphinyl-3-benzyloxy-
carbonylamino-2-piperidinone 2 ADPE Salt (15). A
suspension of compound 4 (294 g, 0.90 mol) in DMF (3.0
L) was dissolved by heating internal temperature at 60 °C
and was allowed to cool in an ice bath. To the solution was
2 2 2
), 4.19 (2H, br s, PNH ), 5.00 (2H, s, OCH -
P
3
added NaHCO (954 g, 11.36 mol) and THF (2.1 L).
Continuously, benzyl chloroformate (1136 mL, 7.96 mol)
was slowly added with keeping the internal temperature
below 10 °C. Stirring of the reaction was continued for 1 h
at room temperature. The reaction mixture was separated,
and the aqueous layer was extracted twice with EtOAc (6.0
and 3.0 L). The organic layers were combined and were
3
added SO /pyridine complex (172 g, 1.08 mol) keeping
internal temperature at 0-5 °C, and the mixture was stirred
for 1 h. After addition of water (100 mL), the mixture was
allowed to warm to room temperature. To the reaction
mixture were added MeOH (1.0 L) and a half of the ADPE
(480 g, 2.25 mol) solution in MeOH (4.0 L), and the mixture
was heated at 50 °C. The remainder of the ADPE solution
was added, and the mixture was stirred for 30 min. The
mixture was filtrated to remove the precipitate, and the filtrate
was evaporated to remove MeOH under reduced pressure.
The residue was diluted with EtOH (15.0 L) and water (6.0
L), and the solution was stirred for 16 h at room temperature
and then was stirred for 1 h below 10 °C. The precipitate
was collected by filtration and was dried at 40 °C under
reduced pressure to give compound 15 (257 g, 92% de).
washed with 5% citric acid(aq) (4.0 L), 5% NaHCO
4.0 L), and saturated NaCl(aq) (4.0 L × 2). The organic
solution was dried over anhydrous Na SO and was evapo-
rated under reduced pressure until its volume approximately
3
(aq)
(
2
4
i
became 3.0 L. To the solution was slowly added Pr
2
O (2.0
L). The crystals were collected by filtration, washed with
i
Pr
2
O (0.5 L), and dried at room temperature and then at 40
°
C under reduced pressure to give compound 3 (1383 g,
purity >99%, optical purity >99% ee) in 75% yield from
L-ornithine hydrochloride (16). The optical purity was
measured by HPLC analysis using a chiral column (condi-
tions: column, CHIRALPAK AS (4.6 mm × 250 mm);
eluent, n-hexane/EtOH ) 1:1; flow rate, 0.4 mL/min;
detection, UV 215 mm). The retention time of compound 3
P
Crude (3S,R )-3-Amino-1-amino(sulfoamino)phos-
phinyl-2-piperidinone: Crude Sulphostin (1). To a sus-
pension of compound 15 (255 g) in acetic acid (400 mL)
and water (2.0 L) was added Pd-black (11 g), and the
suspension was stirred for 2 h at room temperature under
10 atm of pressure of hydrogen. The reaction mixture was
filtrated to remove Pd-black, and the filtrate was diluted with
EtOH (5.0 L). The solution was stirred for 2 h keeping
internal temperature below 5 °C. The crystals were collected
by filtration, washed with EtOH (0.5 L), and dried at 40 °C
under reduced pressure to give crude sulphostin (51.6 g,
purity 98.6%, optical purity 95.3% de) in 20% yield from
compound 4. The optical purity was measured by the method
described for compound 13.
Purification of Crude Sulphostin (1). A solution of
crude sulphostin (150 g, 0.52 mol, 97.2% de) in water (1.80
L) was filtrated by the membrane filter. The filtrate was
diluted with EtOH (0.90 L) keeping the internal temperature
at 55-60 °C. The solution was stirred for 16 h at room
temperature. The crystals were collected by filtration, washed
was 25.4 min. For reference, that of 3-epimer of compound
1
3
was 47.2 min. H NMR (CDCl
3
, 200 MHz) δ: 1.61 (1H,
tdd, J ) 11.7, 5.4, 12.7 Hz, H-4a), 1.82-1.95 (2H, m, H-5),
.53 (1H, m, H-4b), 3.27-3.33 (2H, m, H-6), 4.09 (1H, m,
H-3), 5.11 (2H, s, OCH Ph), 5.79 (1H, br s, NHCOO), 6.36
1H, br s, H-1), 7.28-7.38 (5H, m, Ph).
3S)-3-Benzyloxycarbonylamino-1-diaminophosphinyl-
-piperidinone (4). To a suspension of compound 3 (600
2
2
(
(
2
g, 2.52 mol) in toluene (6.0 L) were added chlorotrimeth-
ylsilane (612 mL, 4.80 mol) and diisopropylethylamine (823
mL, 4.80 mol), and the mixture was stirred for 24 h at room
temperature. To the mixture was added phosphoryl chloride
(
894 mL, 9.60 mol), and the mixture was stirred for 2 days
at room temperature. The reaction mixture was poured into
the aqueous ammonia solution prepared from NH Cl (1596
4
g), 28% ammonia(aq) (3420 mL), and water (2850 mL)
keeping the pH of the ammonia solution at 9-11 by addition
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
575

with a solution (400 mL) of water-EtOH (1:2), and dried
at 40 °C under reduced pressure to give pure sulphostin (111
g, purity >99%, optical purity >99% de) in 74% yield. This
synthetic sulphostin was identical with the authentic sample
in physicochemical and biochemical properties.
Recovery of ADPE (480 g × 3 lots). The precipitate and
the aqueous solution containing ADPE were combined and
were evaporated under reduced pressure to remove the
organic solvents, followed by addition of 2 equiv of 3 N
NaOH(aq) against the amount of ADPE calculated by
reversed-phase HPLC analysis. The aqueous solution was
stirred overnight, and then the crystals were collected by
filtration. The crystals were recrystallized from MeOH-
water to recover ADPE (1100 g) in 76% yield.
2
Acknowledgment
We thank Dr. R. Sawa and Ms. Y. Kubota (Microbial
Chemistry Research Center) for helpful advice and measure-
ment of various spectral data.
Received for review February 26, 2005.
OP058000C
576
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development