An improved and scalable synthesis of N-(5-(4-cyanophenyl)-3-hydroxypicolinoyl)glycine, a promising PHD2 inhibitor for the treatment of anemia

Article abstract of DOI:10.1016/j.tetlet.2015.07.013

Abstract An improved and scalable synthetic method of N-(5-(4-cyanophenyl)-3-hydroxypicolinoyl)glycine, a promising inhibitor of PHD2, is described. The optimized route staring from commercially available 5-bromo-3-nitropicolinonitrile takes only five chemical steps and has a dramatically increased total yield of 37.6% without chromatographic purifications.

Full text of DOI:10.1016/j.tetlet.2015.07.013

Accepted Manuscript
An improved and scalable synthesis of N-(5-(4-cyanophenyl)-3-hydroxypicoli-
noyl)glycine, a promising PHD2 inhibitor for the treatment of anemia
Yonghua Lei, Tianhan Hu, Mingyang Hu, Li Yang, Yue Wu, Hua Xiang,
Lianshan Zhang, Lingjia Shen, Qidong You, Xiaojin Zhang
PII:
S0040-4039(15)01143-0
DOI:
http://dx.doi.org/10.1016/j.tetlet.2015.07.013
TETL 46502
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 May 2015
2 July 2015
5 July 2015
Please cite this article as: Lei, Y., Hu, T., Hu, M., Yang, L., Wu, Y., Xiang, H., Zhang, L., Shen, L., You, Q., Zhang,
X., An improved and scalable synthesis of N-(5-(4-cyanophenyl)-3-hydroxypicolinoyl)glycine, a promising PHD2
inhibitor for the treatment of anemia, Tetrahedron Letters (2015), doi: http://dx.doi.org/10.1016/j.tetlet.2015.07.013
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
An improved and scalable synthesis of N-(5-(4-cyanophenyl)
-3-hydroxypicolinoyl)glycine, a promising PHD2 inhibitor
for the treatment of anemia
Leave this area blank for abstract info.
Yonghua Lei a, Tianhan Hua, Mingyang Hu a, Li Yang a, Yue Wua, Hua Xiang a,b*, Lianshan Zhang d, Lingjia Shen d,
Qidong You a,b* Xiaojin Zhang a,c*

1
Tetrahedron Letters
jour nal homepage: w ww.elsevier. com
An
improved
and
scalable
synthesis
of
N-(5-(4-cyanophenyl)-3-
hydroxypicolinoyl)glycine, a promising PHD2 inhibitor for the treatment of anemia
Yonghua Leia, Tianhan Hu a, Mingyang Hu a, Li Yang a, Yue Wu a, Hua Xiang a,b*, Lianshan Zhang d,
Lingjia Shen d, Qidong You a,b*, Xiaojin Zhang a,c*
a Jiangsu Key Laboratory of Drug Design and Optimization, China Pharmaceutical University, Nanjing, 210009, China
b State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing, 210009, China
c Department of Organic Chemistry, School of Science, China Pharmaceutical University, Nanjing, 210009, China
d National Engineering and Research Center for Target Drugs, Liangyungan, 222000, China
ARTICLE INFO
ABSTRACT
An improved and scalable synthetic method of N-(5-(4-cyanophenyl)-3-hydroxypicolinoyl)glycine, a
promising inhibitor of PHD2, is described. The optimized route staring from commercially available
5-bromo-3-nitropicolinonitrile takes only five chemical steps and has a dramatically increased total
yield of 37.6% without chromatographic purifications.
Article history:
Received
Received in revised form
Accepted
Available online
.
Keywords:
Subsituted picolinylglycine
Scalable synthesis
Suzuki reaction
PHD2 inhibitor
The hyrolyl hydroxylases domain-containing protein 2
(PHD2) is a dominant regulator of the hypoxia-inducible
transcription factor
followed by
a
Suzuki coupling reaction with 4-
cyanophenylboronic acid and then a hydrolyzation reaction. A
number of factors made this approach unsuitable as a long-term
supply route capable of delivering substantial quantities of 2. For
example, several steps suffers from suboptimal yield and require
chromatographic purifications. The seven-step synthesis provided
2 in a poor total yield of 9.0%. In addition, the microwave-
assisted high temperature reaction was not suitable for
amplification. Herein, we report an efficient and scalable
synthesis of 2 from the readily available starting material 5-
bromo-3-nitropicolinonitrile (5) in only five synthetic steps and
in 37.6% overall yield.
α
(HIFα).1-3 Inhibition of PHD2, by
preventing the hydroxylation of HIFα, has recently been
considered as a way of increasing levels of HIFα and thus
erythropoietin, which is being pursued as a promising therapy for
the treatment of anemia.4-6
A
series of N-(5-aryl-3-
hydroxypicolinoyl)glycine (1) (Fig. 1) was disclosured in 2007
by Procter & Gamble Company as potent inhibitors of PHD2.7 -9
The clinical drug AKB-6548 (phase II) is one of the family, and
its structure was presumed as N-(5-(4-cyanophenyl)-3-
hydroxypicolinoyl)glycine (2) (Fig. 1). Thus, it is of great
importance to develop an efficient and scalable route for the
industrial production of this clinically important compound.
O
O
N
OH
N
OH
N
H
N
H
O
O
OH
OH
R
Scheme 1. The reported synthesis of 2
NC
2
1
Initially, our study commenced with the substitution of 5 by
benzyl alcohol in THF with NaH as a base. Compared with
compound 3, compound 5 with nitro group exhibited a high
reaction activity. As shown in Table 1, the reaction occurred
under a relatively mild reaction condition and provided 6 in a
moderate yield of 43.2% (Table 1, entry 1). In order to access 6
Figure 1. N-(5-aryl-3-hydroxypicolinoyl)glycine
The reported route for the synthesis of 2 is depicted in
Scheme 1.7 This synthesis was carried out by preparing the key
intermediate 4 from 3, 5-dichloropicolinonitrile (3) in 5 steps,
———
*
Corresponding authors. E-mail: zxj@cpu.edu.cn (X. Zhang),
youqd@163.com (Q. You), cpuxianghua@sina.com (H. Xiang).

2
Tetrahedron Letters
in a rapid and efficient manner, different conditions were
With the intermediate 6 in hand, the remained 4 steps for
synthesis of 2 were the hydrolyzation of 6, a condensation
reaction with glycine methyl ester hydrochloride, a Suzuki
coupling reaction and a deprotection reaction (Scheme 2).
examined (Table 1). An increased yield was observed when the
reaction time was prolonged from 1 h to 10 h (Table 1, entry 2).
However, a higher reaction temperature resulted in a significantly
reduced yield (Table 1, entries 3-4). It was noted that the yield
was dramatically reduced to 25.7% when the reaction carried out
in refluxing (Table 1, entry 4). Thus, the reaction was performed
at room temperature in the next condition optimization. Further
prolonging the reaction time to 24 h led to an improved yield
(Table 1, entry 5). Subsequently, when we increased the amount
of BnOH from 1.2 to 2 equiv, the yield was improved to 87.3%
(Table 1, entry 6). It was found that 2.0 equiv of the benzyl
alcohol was sufficient and further increasing its amounts
contributed little to improving the yield (Table 1, entry 7).
Therefore, the expected product was obtained optimally in the
presence of 2.0 equiv of benzyl alcohol at room temperature
using NaH as a base in 87.3% yield (Table 1, entry 6).
During the following four steps, there were still many issues
that needed to be addressed to develop a new route suitable for
further scale-up. The hydrolyzation of 6 has initially been
performed in the presence of 30% aqueous NaOH in CH3OH at
o
80 C. However, it need nearly 10 hours to make the reaction
completed. When we increased the temperature to 100 oC, the
reaction was completed in 2 h. The reaction solution was
neutralized by diluted hydrochloric acid (3 M) and then the acid
7 was conveniently obtained by simple filtration in 87.0% yield.
Subsequently, condensation between 7 and glycine methyl ester
hydrochloride gave rise to 8 in the presence of TEA in DCM at
room temperature in 64.5% yield.
Scheme 2. Reagents and conditions: a) BnOH, NaH, THF, rt, 87.3%; b) NaOH, CH3OH, H2O, reflux, 87.0%; c) GlyOMe·HCl, EDCI, HOBT,
TEA, DCM, rt, 64.5%; d) Pd(dppf)2Cl2, K3CO3, THF, 75 oC, 90.0%; e) H2, Pd/C, THF, 45 oC; then LiOH, THF, rt, 85.5%.
One of the key steps in this route is the Suzuki coupling
reaction. The reported method for this step was only in 53%
yield. To find an efficient way to access product 9, different
solvents and reagents were examined (Table 2). Firstly, the
reaction was carried out in the presence of Na2CO3 using
Pd(PPh3)4 as a catalyst under different solvents (Table 2, entries
1-3). The results showed that THF as the solvent provided a
preferable yield of 69.0% (Table 2, entry 2). Compared with
Na2CO3, K3PO4 as the base resulted in a slightly improved yield
(Table 2, entry 4). Thus, THF and K3PO4 were seem to be the
preferred combination among the tested solvents and bases.
Subsequently, with the hope of increasing the yield another three
Table 1. Optimization of substitution reaction
Entry a
BnOH (equiv)
T (oC)
t (h)
1
Yield (%) b
43.2
1
2
3
4
5
6
7
1.2
1.2
1.2
1.2
1.2
2
rt
rt
10
10
10
24
24
24
54.3
40
47.0
Reflux
rt
25.7
common
palladium
catalysts
including
Pd(dppf)2Cl2,
69.5
Pd(PPh3)2Cl2, and Pd(OAc)2 were screened under THF as the
solvent and K3PO4 as the base under reflux (Table 2, entries 5-7).
The result showed that Pd(dppf)2Cl2 provided the best reactivity
under otherwise identical conditions (Table 2, entry 5). The yield
(90.0%, Table 2, entry 5) obtained within 2 h of reaction is more
efficient than the yield obtained from the coupling reaction by
reported method. As a result, the Suzuki coupling reaction was
facile and proceeded in a high yield of 90.0%.
rt
87.3
4
rt
87.0
a The reaction was conducted with N2 protected.
b Isolated yield
Table 2. Optimization of Suzuki reaction
Entry a
Base (aq.) b
Na2CO3
Catalyst
Solvent
DMF
T (oC)
80
t (h)
10
Yield (%) c
45.5
1
2
3
Pd(PPh3)4
Pd(PPh3)4
Pd(PPh3)4
Na2CO3
THF
reflux
80
10
69.0
Na2CO3
1,4-Dioxane
10
55.3

3
4
5
6
7
K3PO4
K3PO4
K3PO4
K3PO4
Pd(PPh3)4
THF
THF
THF
THF
reflux
reflux
reflux
reflux
8
74.3
90.0
57.9
6.0
Pd(dppf)2Cl2
Pd(PPh3)2Cl2
Pd(OAc)2
2
8
10
a The reaction was conducted with N2 protected.
b The concentration of the base is 2 M; the equivalent of the base is 2.0 eq to compound 8.
c Isolated yield.
The final chemical stage required the debenzylation and
hydrolyzation of 9. The reaction proved to be more efficient
using H2, Pd/C and 1 M aqueous LiOH in THF in one pot. When
the reaction completed, removed the organic phase by distillation.
And the concentrated mixture was neutralized by diluted
hydrochloric acid (3 M) and then the product 2 was conveniently
obtained by simple filtration in 85.5% yield.
6. Tetsuhiro T., Masaomi N., J. Pharmacol. Sci. 2009,
109, 24.
7. PCT Int. Appl. (2007), US 0299086 A1.
8. Michael H. R., J. Med. Chem. 2013, 56, 9369.
9. Yoshiki, H., Tetsuhiro, T., Masaomi, N., Expert. Opin.
Drug. Discov, 2013, 8, 965.
10. 3-(Benzyloxy)-5-bromopicolinonitrile (6). A mixture
of THF (1.8 L) and benzyl alcohol (59 mL, 114 mmol)
With the optimized route in hand, we next began to extend
the scope of this synthetic strategy to a variety of PHD2
inhibitors from the patent.7 We have tried different substrates in
the Suzuki coupling and the following deprotection to provide
five more PHD2 inhibitors with different substitutions on the
benzyl ring in 37.3-39.3% total yields (see supporting
information). This showed that the synthetic route is an efficient
way to obtain various PHD2 inhibitors with high efficiency.
o
was cooled to 0 C. NaH was added slowly to the
mixture at 0 oC. After addition, the mixture was stirred
at room temperature for 1 h and 35oC for another 0.5 h.
The resultant solution was added to the solution of 5-
bromo-3-nitropicolinonitrile (5, 100 g, 438 mmol) in
THF (1.3 L) and the mixture was stirred at room
temperature for 24 h. The reaction mixture was
quenched with H2O (200 mL). After quenched, the
mixture was concentrated to a volume of about 200 mL.
Dichloromethane (1.5 L) was added, and the higher
aqueous phase was removed. The organic phase was
wash with saturated aqueous NaHCO3 (1.0 L×1), water
(1.0 L×3), saturated aqueous NaCl (1.0 L×1). The
organic phase was concentrated, and dried under
vacuum at 55oC to give 6 (110.0 g, 87.3% yield, 99.5%
purity by HPLC) as a brown solid.
In summary, we have developed a facile and high yielding
method to synthesize 2, involving a non-chromatography five-
step synthesis in 37.6% overall yield. This synthesis route has
several advantages, including a simple experimental procedure,
mild reaction conditions and high yields of the products. The
procedure allows rapid scale-up for the large-scale synthesis of
N-(5-(4-cyanophenyl)-3-hydroxypicolinoyl)glycine (2) and other
PHD2 inhibitors that are similar in chemical structure.
11. 3-(Benzyloxy)-5-bromopicolinic
acid
(7).
A
suspension of 6 (53.3 g, 184 mmol), methanol (1.0 L)
and 30% NaOH (1.0 L) was heated to 100 oC for 2 h.
The mixture was cooled to room temperature, and the
resultant solution was concentrated to a volume of
about 1.0 L. The concentrated mixture was neutralized
by aqueous HCl (3 M). The formed precipitate was
filtered off, washed with water, and dried to give 7
(49.3 g, yield 87.0%, 98.2% purity by HPLC) as a faint
yellow solid.
Acknowledgments
We are thankful for the financial support of the National
Natural Science Foundation of China (No.81302636), the Natural
Science Foundation of Jiangsu Province of China
(No.BK20130656) and the Priority Academic Program
Development of Jiangsu Higher Education Institutions (PAPD).
12. Methyl N-(5-bromo-3-hydroxypicolinoyl)glycine (8).
To a solution of 7 (49.3 g, 160 mmol) and glycine
methyl ester hydrochloride (24 g, 192 mmol) in
dichloromethane (500 mL) was added EDCI (45.9 g,
240 mmol), HOBT (32.6 g, 24 mmol) and Et3N (73
mL). The mixture was stirred at room temperature for 6
h under nitrogen atmosphere. The resultant solution was
added saturated aqueous NaHCO3 (1.0 L), and the
higher aqueous phase was removed. The organic phase
was washed with water (1.0 L×3) and saturated aqueous
NaCl (1.0 L×1), and dried with anhydrous MgSO4. The
solution was filtrated by silica gel, washed with
petroleum ether: EtOAc = 3:1 and concentrated to give
an oily product, which was recrystallized by
petroleum ether: EtOAc = 3:1, the solids were isolated
by filtration, washed with cold EtOAc and dried to give
8 (28.2 g, yield 64.5%, 99.6% purity by HPLC) of white
crystal.
Supplementary data
Supplementary data associate with this article is available.
References and notes
1. Tarhonskaya H., Chowdhury R., Leung I. K., Loik N.
D., McCullagh J. S., Claridge T. D., Schofield C. J.,
Flashman E. Biochem. J. 2014, 463, 363.
2. McDonough M. A, McNeill L. A., Tilliet M., Qiuyun
Chen, Papamicael C. A., Banerji B., Hewitson K. S.,
and Schofield C. J. J. Am. Chem. Soc. 2005, 127, 7680.
3. Berra, E., Benizri, E., Ginouves, A., Volmat, V., Roux,
D. and Pouyssegur J., EMBO J. 2003, 22, 4082.
4. Chowdhury R., Candela-Lena J. I., Chan M. C.,
Greenald D. J., Yeoh K. K., Tian Y. M., McDonough M.
A., Tumber A., Rose N. R., Conejo-Garcia A.,
Demetriades M., Mathavan S., Kawamura A., Lee M.
K., van Eeden F., Pugh C. W., Ratcliffe P. J., Schofield
C. J. Chem. Biol. 2013, 8, 1488.
13. Methyl
N-(5-(4-cyanophenyl)-3-
hydroxypicolinoyl)glycine (9). A suspension of 8 (7.26
g, 20 mmol), 4-cyanophenylboronic acid (3.0 g, 20
mmol) and THF (300 mL) was added K3PO4 (2M, 30
mL) and Pd(dppf)2Cl2 (1.0 g). The resultant solution
was stirred at 75 oC for 2 h under nitrogen atmosphere.
5. William A. D. J. Med. Chem. 2012, 55, 2943.

4
Tetrahedron
The mixture was cooled to room temperature and
concentrated. EtOAc (100 mL) was added and the
mixture was filtered off by silica gel. The filtrated
solution was concentrated to 40 mL and
recrystallization, the solids were isolated by filtration,
washed with cold EtOAc and dried to give 9 (5.62 g,
yield 90.0%, 98.3% purity by HPLC) as a white solid.
14. N-(5-(4-Cyanophenyl)-hydroxypicolinoyl)glycine
(2).The suspension of 9 (4 g, 10 mmol), and Pd/C (400
mg) in THF (120 mL) was stirred at 40oC for 5 h under
hydrogen atmosphere. After cooling to 25 oC, the Pd/C
was removed by filtration, and LiOH (1 M, 30 mL) was
added. The resultant solution was heated to 40 oC for 5
o
h. After cooling to 25 C, the reaction mixture was
concentrated. And the residue was dissolved in water
(30 mL), the solution was acidified with 10% HCl to
pH about 4-5. The formed precipitate was collected by
filtration, washed with water and dried to provide 2
(2.54 g, yield 85.5%, 99.1% purity by HPLC) as a white
solid.