Efficient synthesis of sitagliptin phosphate, a novel DPP-IV inhibitor, via a chiral aziridine intermediate

Article abstract of DOI:10.1016/j.tetlet.2013.09.136

Sitagliptin phosphate, a novel DPP-IV inhibitor of T2DM, has been synthesized via 12 linear steps, in an overall yield of 26%. The key step is the coupling reaction of 2,4,5-trifluorophenylmagnesium bromide with a chiral aziridine derivative, which was prepared from l-homo-serine by simple steps.

Full text of DOI:10.1016/j.tetlet.2013.09.136

Tetrahedron Letters 54 (2013) 6807–6809
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient synthesis of sitagliptin phosphate, a novel DPP-IV inhibitor,
via a chiral aziridine intermediate
Xianhua Pan a, Xiaojun Li a, Qingling Lu a, Wansheng Yu a, Weijin Li b, , Qunhui Zhang b, , Fei Deng b,
,
Feng Liu a,
⇑
a School of Perfume and Aroma Technology, Shanghai Institute of Technology, 100 Haiquan Rd., Shanghai 201418, PR China
b Zhejiang Hisoar Pharmaceutical Co., Ltd, 100 Waisha Branch Rd., Jiaojiang, Taizhou, Zhejiang 318000, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 May 2013
Revised 26 September 2013
Accepted 29 September 2013
Available online 3 October 2013
Sitagliptin phosphate, a novel DPP-IV inhibitor of T2DM, has been synthesized via 12 linear steps, in an
overall yield of 26%. The key step is the coupling reaction of 2,4,5-trifluorophenylmagnesium bromide
with a chiral aziridine derivative, which was prepared from
L-homo-serine by simple steps.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Sitagliptin phosphate
b-Amino acid
Grignard reaction
Aziridine
Synthesis
Type 2 diabetes mellitus (T2DM) is a rapidly growing global epi-
demic that affects millions of people. A recent study reveals that
the number of patients with diabetes has reached 350 million, dou-
bling over the past 30 years. Diabetes and its complications present
a serious threat to human health, bringing with it enormously high
economic costs, not only to individuals and families, but to global
health systems.1 It was found recently that inhibitors of the dipep-
tidyl peptidase IV (DPP IV) could form new therapeutic agents for
T2DM, by increasing GLP-1 (glucagon-like peptide-1) and GIP (glu-
cose-dependent insulinotropic peptide) levels, as well as improv-
ing glycemic control for diabetics.2–5 Sitagliptin phosphate (1)
(Fig. 1) was approved by the U.S. FDA in October 2006 as a safe
and orally active agent for the treatment of T2DM and is marketed
by Merck Co., as a monotherapy (JanuviaÒ) or in combination with
meiformin (JanumetÒ).6–8
CF3
N
N
F
F
N
(R)
N
O
NH2
F
H3PO4
Sitagliptin phosphate
1
Figure 1. Structure of sitagliptin phosphate.
ature. The core b-amino-acid fragment was then synthesized by
Arndt–Eistert homologation. However, the strict reaction condi-
tions (ꢀ78 °C) and the use of a dangerous reagent (CH2N2), had
limited this process to the laboratory and small scales.
Xiao et al.9 described another process for the synthesis of a ser-
ies of chiral beta amino acid derivatives, including sitagliptin.
S-Phenylglycine amide was used as a chiral auxiliary to obtain
Z-enamines from diketones. Finally, the title products were
obtained after PtO2 catalyzed asymmetric hydrogenation, followed
by N-deprotection. The shortcoming of this process was the use of
precious metals (Pt and Pd), and the enantiomeric excess (ee 90%)
of the reduction was less than satisfied.
Because of its unique structure and tremendous commercial va-
lue, sitagliptin phosphate (1) had captured the interest of a number
of pharmaceutical companies and synthetic teams. Considerable
efforts have been made to identify a shorter, more cost-competi-
tive, and environmentally friendly synthetic route.
Kim et al.6 synthesized sitagliptin from (R)-2-amino-3-(2,4,5-
trifluorophenyl)propanoic acid, in which the chiral center was
induced from the corresponding Schöllkopf reagent at low temper-
Dreher and Xiao10 offered alternative strategies involving the
hydrogenation of enamine using rhodium metal complexes with
chiral mono or bis phosphine ligands. However, in their processes,
the phosphine ligands of the chiral metal catalysts are very difficult
⇑
Corresponding author. Tel.: +86 21 60873006; fax: +86 21 60873134.
E-mail address: liufeng@sit.edu.cn (F. Liu).
These authors contributed equally to this work.
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.09.136

6808
X. Pan et al. / Tetrahedron Letters 54 (2013) 6807–6809
to prepare and expensive. The reaction was carried out at high
pressure and temperature.
tion.18 The Br–Mg-exchange of 1-bromo-2,4,5-trifluorobenzene
was performed with i-PrMgBr at 0 °C in 1 h, monitored by GC to
the point at which 1-bromo-2,4,5-trifluorobenzene disappeared.
It was found that the reaction of 2,4,5-trifluorobromobenzene
Grignard reagents (7) and aziridine (5) could not be performed
smoothly (30%). However, several copper catalysts19,20 were
screened to improve the Grignard reaction yield (Table 1).
As indicated in Table 1, the addition of cuprous halide signifi-
cantly improved the Grignard reaction (entries 1–4), and CuBr/
Me2S gave the best result (78%, entry 5).
After the ring-opening reaction, deprotection of the hydroxyl
group was carried out by treatment with tetrabutylammonium
fluoride in MeOH, to form the amino alcohol (9) in 95% yield. After
treatment with NaClO and TEMPO, 3-R-Boc-amino-4-(2,4,5-triflu-
orophenyl)butyric acid (10) was obtained in good yield (90%).
The triazole (11)11a was then coupled to the amino acid (10) at
0 °C using HOBT-EDCI, to provide (12) in 95% yield. The Boc group
was removed by stirring (12) in the presence of concentrated HCl
and MeOH at ambient temperature. After a typical work up, the
crude product was recrystallized from toluene, and the free base
sitagliptin was isolated in 90% yield. Its phosphoric acid salt (1)
was obtained with 99.2% HPLC purity (single impurity less than
0.1%) and 99.4% ee after recrystallization from i-PrOH.
Karl and colleagues reported two different asymmetric syn-
thetic routes for sitagliptin11 in which the b-amino acid moiety
was introduced via asymmetric hydrogenation of a keto ester or
enamine amide intermediate. These two routes suffer from inade-
quate stereoselectivity and the product was contaminated with Rh,
necessitating an additional purification step at the expense of yield
to upgrade both chemical and enantiomeric purities.
Christopher et al. reported an efficient biocatalytic process to
prepare sitagliptin.12 In the presence of the enzyme, transaminase,
direct amination of prositagliptin ketone to enantiomeric pure
sitagliptin (99.95% ee) was carried out under mild conditions. This
biocatalytic route should become an important auxiliary asymmet-
ric synthesis.
As part of our continuing interest in developing efficient and
practical processes for the synthesis of active pharmaceutical
ingredients (API) and intermediates,13 we report in this work a no-
vel and efficient approach to the preparation of sitagliptin phos-
phate (1), via the nucleophilic ring-opening reaction of the
enantiomeric pure aziridine derivative (5)14–16 with Grignard re-
agent (7) as the key step (Scheme 1).
The synthesis begins with the protection of L
-homoserine17 (2)
(Scheme 2). After treated with TBSCl in the presence of DBU in
CH3CN, silyl ether was formed. This was then treated with Boc2O
to give compound (3) with 81% yield in two steps. The Boc-ami-
no-acid (3) was then condensed with N-hydroxysuccinimide to
give the corresponding ester, which was then reduced with 1 equiv
of NaBH4 at low temperature to form the diol (4) in 84% yield.
Compound 4 was then treated with methanesulfonyl chloride fol-
lowed by subsequent ring closure under basic condition to form
aziridine (5) in good yield (73%) and high enantio-purity (>99% ee).
With the chiral aziridine (5) available, the Grignard reaction,
using 2,4,5-trifluoro-phenyl magnesium bromide (7), was examined
(Scheme 3). Since the direct reaction of Mg and 1-bromo-2,4,
5-trifluorobenzene will yield benzyne instead of the anticipated
Grignard reagent, we used a bromine–magnesium-exchange reac-
In summary, we have devised a new and convergent route for
the synthesis of sitagliptin phosphate (1).The chiral b-amino group
was introduced via a ring-opening Grignard reaction of the chiral
aziridine derivative which was prepared from L-homoserine. The
use of expensive Rh or Pt catalysts has been avoided. This simple
procedure and economical process provides a novel synthesis of
sitagliptin phosphate (1).
F
F
OH
F
F
OTBS
NaClO, TEMPO
90 %
TBAF, MeOH
95 %
NHBoc
NHBoc
F
F
8
9
CF3
N
CF3
N
N
N
N
F
F
ClH HN
N
F
COOH
N
11
NHBoc
O
EDCI, HOBT, DIPEA, CH2Cl2
F
NH
F
F
95 %
Boc
10
12
CF3
CF3
N
N
N
CF3
N
N
1) HCl, MeOH
F
F
N
N
F
F
F
F
N
(R)
N
COOH
N
2) H3PO4
91 %
O
NH2
F
+
O
NHBoc
H3PO4
ClH HN
NH2
F
N
F
Sitagliptin phosphate
10
H3PO4
11
1
Sitagliptin phosphate
1
F
F
Scheme 3. Synthesis of sitagliptin phosphate.
F
OTBS
+
HO2C
OH
MgBr
N
NH2
Boc
Table 1
L-Homoserine
7
5
Influence of copper catalysts on the Grignard reactiona
2
F
F
Scheme 1. Strategy for the synthesis of sitagliptin phosphate.
F
F
5, CuBr.Me
2
S, 0ഒ
F
F
i-PrMgBr, THF
OTBS
90 %
78 %
Br
NHBoc
MgBr
F
F
6
F
8
7
Entry
Catalyst
Time (h)
Yieldb (%)
1) NHS, DCC, EtOAc
OTBS
HO2C
OH
1) TBSCl, DBU, CH3CN HO2C
2) Boc2O, Et3N, CH2Cl2
81 %
2) NaBH4, THF
84 %
1
2
3
4
5
Blank
CuCl
CuBr
CuI
8
2
2
2
2
30
60
65
71
78
NH2
NHBoc
3
L-Homoserine
2
1) MsCl, Py.
OTBS
OTBS
CuBr/Me2S
HO
2) NaH, THF
73 %
N
NHBoc
4
a
Boc
5
Reaction conditions: 7 (0.018 mol) in 40 mL THF, 5 (0.015 mol) in 30 mL THF,
catalyst (0.002 mol), 0 °C.
b
Isolated yield.
Scheme 2. Synthesis of the chiral aziridine.

X. Pan et al. / Tetrahedron Letters 54 (2013) 6807–6809
6809
20. Bornholdt, J.; Kristensen, J. L.; Felding, J. J. Org. Chem. 2010, 75, 7454–7457.
21. Kubryk, M.; Hansen, K. B. Tetrahedron: Asymmetry 2006, 17, 205–209
Spectral data:
Acknowledgments
This work was supported by the Zhejiang Hisoar Pharmaceuti-
cal Co., Ltd. and supported by the Shanghai Science and Technology
Commission (No. 12495810800).
(S)-2-((tert-Butoxycarbonyl)amino)-4-((tert-Butyl-dimethylsilyl)oxy)-butanoic
acid (3) 1H NMR (400 MHz, CDCl3) d 8.04 (s, 1H), 5.86 (d, J = 6.7 Hz, 1H), 4.48–
4.15 (m, 1H), 3.90–3.56 (m, 2H), 2.16–1.88 (m, 2H), 1.44 (d, J = 11.4 Hz, 9H),
0.90 (s, 9H), 0.06 (s, 6H); ESI-MS: 334.6 (M++1).
(S)-tert-Butyl 4-(tert-butyldimethylsilyloxy)-1-hydroxybutan-2-ylcarbamate (4)
½
a 2D0
ꢁ
ꢀ30.8 (c 1.0, CHCl3); 1H NMR (400 MHz, CDCl3) d 5.42 (d, J = 45.0 Hz,
References and notes
1H), 3.74 (t, J = 5.3 Hz, 2H), 3.70–3.58 (m, 2H), 3.51 (s, 1H), 1.83 (s, 1H), 1.73 (d,
J = 6.1 Hz, 1H), 1.44 (s, 9H), 1.26 (d, J = 2.6 Hz, 1H), 0.91 (s, 9H), 0.08 (s, 6H);
ESI-MS: 320.3(M++1).
1. See: (a) www.diabetes.org. (b) http://www.who.int.
2. International Diabetes Federation (IDF), Diabetes Atlas, 3rd ed., Dec, 2006;
http://www.iotf.org/diabetes.asp.
(S)-tert-Butyl 2-(2-(tert-butyldimethylsilyloxy)ethyl)aziridine-1-carboxylate (5)
½
a 2D0
ꢁ
ꢀ11.2 (c 1.0, CHCl3); 1H NMR (400 MHz, CDCl3)
d 3.78 (dd, J = 9.4,
3. Weber, A. J. Med. Chem. 2004, 48, 4135–4141.
3.7 Hz, 2H), 2.48 (dd, J = 6.0, 4.0 Hz, 1H), 2.28(d, J = 6.1 Hz, 1H), 1.96 (d,
J = 3.7 Hz, 1H), 1.81–1.74 (m, 1H), 1.60 (dd, J = 13.7, 6.9 Hz,1H), 1.46 (s, 9H),
0.90 (s, 9H), 0.07 (d, J = 1.9 Hz, 6H); ESI-MS:324.2 (M++Na); HRMS Calcd for:
4. Drucker, D. J. Exp. Opin. Investig. Drugs 2003, 12, 87–100.
5. Miller, S. A.; St. Onge, E. L. Ann. Pharmcother. 2006, 40, 1336.
6. Kim, D.; Wang, L.; Beconi, M.; Eiermann, G. J.; Fisher, M. H.; He, H. B.; Hickey, G.
J.; Kowalchick, J. E.; Leiting, B.; Lyons, K.; Marsilio, F.; McCann, M. E.; Patel, R.
A.; Petrov, A.; Scapin, G.; Patel, S. B.; Roy, R. S.; Wu, J. K.; Wyvratt, M. J.; Zhang,
B. B.; Zhu, L.; Thornberry, N. A.; Weber, A. E. J. Med. Chem. 2005, 48, 141–151.
7. Shultz, C. S.; Krska, S. W. Acc. Chem. Res. 2007, 40, 1320–1326.
8. Xia, L.-H. Chin. J. New Drug 2007, 16, 979–981.
C
15H31NO3SiNa (M+Na)+ requires 324.1971, found 324.1966.
(R)-tert-Butyl
ylcarbamate (8)
4-(tert-butyldimethylsilyloxy)-1-(2,4,5-trifluorophenyl)butan-2-
½ ꢁ
a 2D0 +5.2 (c 1.0, CHCl3); 1H NMR (400 MHz, CDCl3) d 7.06 (d, J = 7.0 Hz, 1H),
6.88 (d, J = 6.8 Hz, 1H), 5.24 (s, 1H), 3.97–.74 (m, 2H), 3.73–3.61 (m, 1H), 2.82
(s, 2H), 1.86–1.64 (m, 1H), 1.63–1.47 (m, 1H), 1.39 (s, 9H), 0.91 (s, 9H), 0.06 (d,
J = 4.6 Hz, 6H); ESI-MS: 456.20 (M++Na); HRMS Calcd for: C21H34F3NO3Na
(M+Na)+ requires 456.2158, found 452.2166.
9. Xiao, Y.; Armstrong, J. D.; Krska, S. W. WO 2004085378.
10. (a) Dreher, S. D.; Ikemoto, N. WO 2004085661; (b) Xiao, Y.; Sun, Y. K. WO
2005097733.
(R)-tert-Butyl-4-hydroxy-1-(2,4,5-trifluorophenyl)butan-2-ylcarbamate (9) 1H
NMR (400 MHz, CDCl3) d 7.05 (d, J = 8.3 Hz, 1H), 6.91 (d, J = 6.6 Hz, 1H), 4.56
(d, J = 9.0 Hz, 1H), 4.01 (dd, J = 25.4, 18.8 Hz, 1H), 3.68 (d, J = 6.0 Hz, 2H), 2.82–
2.70 (m, 2H), 1.86 (dd, J = 12.4, 7.7 Hz, 1H), 1.67 (d, J = 6.7 Hz, 2H), 1.45–1.37
(m, 9H); ESI-MS: 342.2 (M++Na).
11. (a) Karl, B. H.; Jaume, B.; Spencer, D.; Yi, H.; Michele, K.; Michael, P.; Nelo, R.;
Dietrich, S.; Joseph, D. A.; David, A.; Edward, J. J. G. Org. Process Res. Dev. 2005, 9,
634–639; (b) Karl, B. H.; Yi, H.; Feng, X.; Nelo, R.; Andrew, G.; Michael, K.;
Shane, K.; Thorsten, R.; Bryon, S.; Jaume, B.; Nori, I.; Yong, K. S.; Felix, S.;
Christophe, M.; Edward, J. J. G.; Joseph, D. A., III; David, A. J. Am. Chem. Soc.
2009, 9, 634–639.
12. Christopher, K. S.; Jacob, M. J.; Emily, C. M.; Jeffrey, C. M.; Sarena, T.; William, R.
J.; Jeffrey, C. C.; Anke, K.; Fred, J. F.; Jos, B.; Paul, N. D.; Gjalt, W. H.; Gregory, J. H.
Science 2010, 329, 305.
13. (a) Liu, F.; Yu, W.-S.; Ou, W.-H.; Pan, X.-H. J. Chem. Res. (S) 2010, 34, 230–232;
(b) Liu, F.; Yu, W.-S.; Ou, W.-H.; Pan, X.-H. J. Chem. Res. (S) 2010, 34, 517–519;
(c) Pan, X.-H.; Ou, W.-H.; Liu, F. J. Chem. Res. (S) 2011, 35, 545–546; (d) Pan, X.-
H.; Ou, W.-H.; Liu, F. J. Chem. Res. (S) 2011, 35, 729–730; (e) Pan, X.-H.; Zhang,
Q.-H.; Liu, F. Tetrahedron Lett. 2012, 53, 5364–5366.
14. Kozikowski, A. P.; Ishida, H.; Isobe, K. J. Org. Chem. 1979, 44, 2788.
15. Alex, A. C.; Helen, M. I. O.; Joseph, S. Tetrahedron 1998, 54, 2181–2208.
16. Kim, B. M.; Bae, S. J.; So, S. M.; Yoo, H. T.; Chang, S. K.; Lee, J. H.; Kang, J. S. Org.
Lett. 2001, 3, 2349–2351.
(R)-3-(tert-Butoxycarbonyl)-4-(2,4,5-trifluorophenyl)butanoic acid (10)
½ ꢁ
a 2D0
ꢀ29.8 (c 1.0, CHCl3); mp 123–125 °C. {lit.21 a 2D0
½ ꢁ ꢀ32.2 (c 1.0, CHCl3). Mp
124–125 °C.} 1H NMR (400 MHz, CDCl3) d 7.07 (d, J = 7.8 Hz, 1H), 6.91 (d,
J = 6.8 Hz, 1H), 5.23–4.76 (m, 1H), 4.28–3.92 (m, 1H), 2.88 (d, J = 6.9 Hz, 2H),
2.62 (s, 2H), 1.36 (d, J = 23.3 Hz, 9H); ESI-MS: 356.2 (M++Na).
(R)-tert-Butyl
4-oxo-4-(3-(trifluoromethyl)-5,6-dihydro-[1,2,4]triazolo[4,3-
a]pyrazin-7(8H)-yl)-1-(2,4,5-trifluorophenyl)butan-2-ylcarbamate (12)
½ ꢁ
a 2D0
+22.1 (c 1.0, CHCl3); mp 187–191 °C. {lit.13a a 2D0
½ ꢁ +22.2 (c 1.0, CHCl3). Mp
188–191 °C.} 1H NMR (400 MHz, CDCl3) d 7.18–7.05 (m, 1H), 7.02–6.85 (m,
1H), 5.31 (s, 1H), 5.15–4.76 (m, 2H), 4.43–3.78 (m, 5H), 2.98–2.92 (m, 2H),
2.71–2.61 (m, 2H), 1.36 (s, 9H); ESI-MS: 508.0 (M++1).
Sitagliptin phosphate (1) ½a D20
ꢁ
ꢀ72.9 (c 1.0, H2O). Mp 213–216 °C. {lit.12 a 2D0
½ ꢁ
ꢀ74.4 (c 1.0, H2O). Mp 215–217 °C.} 1H NMR (400 MHz, D2O) d 7.23–7.05 (m,
1H), 7.05–6.86 (m, 1H), 4.85–4.80 (m, 2H), 4.16 (d, J = 5.5 Hz, 1H), 4.10 (dt,
J = 13.1, 6.0 Hz, 1H), 3.89–3.81 (m, 3H), 3.11–2.76 (m, 3H), 2.73 (ddd, J = 17.5,
10.1, 7.5 Hz, 1H).
17. Yu, W.-S.; Liu, F.; Ou, W.-H.; Pan, X.-H. J. Shanghai Inst. Technol. (Nat. Sci.) 2012,
12, 137–140.
18. Mohamed, A.; Florian, D.; Paul, K. Tetrahedron Lett. 1999, 40, 7449–7453.
19. Metaferia, B. B.; Fetterolf, B. J.; Syed, S. U. H. J. Med. Chem. 2007, 50, 6326–6336.