Synthetic optimization of rosiglitazone and related intermediates for industrial purposes

Article abstract of DOI:10.1007/s11164-015-2132-0

As an important newly Food and Drug Administration (FDA)-approved drug for treating diabetes, rosiglitazone (1) has received much attention from researchers in many areas. To search for an economical and convenient synthesis method for 1, we explored the reaction conditions and workup of a scalable five-step synthetic route by an orthogonal method to determine the best condition for each reaction step. The starting materials are commercially available, including 2-chloropyridine (2), N-methylethanolamine (3), 4-fluorobenzaldehyde (4a) or 4-hydroxybenzaldehyde (4b), and 1,3-thiazolidine-2,4-dione (5). The five sequential reaction steps are cyclization, alkylation, etherification, condensation, and reduction, having optimal yield of 90, 99, 59, 75, and 91 %, respectively. The best overall yield to synthesize rosiglitazone based on compound 2 was 40 %, being suitable for industrial purposes, using water as a green solvent and avoiding column chromatography during the last three reaction steps.

Full text of DOI:10.1007/s11164-015-2132-0

Res Chem Intermed
DOI 10.1007/s11164-015-2132-0
Synthetic optimization of rosiglitazone and related
intermediates for industrial purposes
1
1
2
•
Ge Meng
Yang Gao
Ruizhi Hu
•
Meilin Zheng
Aqun Zheng
•
•
Mengshu Dong
1
•
1
3
•
Zhenyu Li
1
Received: 22 October 2014 / Accepted: 4 June 2015
Ó Springer Science+Business Media Dordrecht 2015
Abstract As an important newly Food and Drug Administration (FDA)-approved
drug for treating diabetes, rosiglitazone (1) has received much attention from
researchers in many areas. To search for an economical and convenient synthesis
method for 1, we explored the reaction conditions and workup of a scalable five-step
synthetic route by an orthogonal method to determine the best condition for each
reaction step. The starting materials are commercially available, including
2
-chloropyridine (2), N-methylethanolamine (3), 4-fluorobenzaldehyde (4a) or
-hydroxybenzaldehyde (4b), and 1,3-thiazolidine-2,4-dione (5). The five sequential
4
reaction steps are cyclization, alkylation, etherification, condensation, and reduc-
tion, having optimal yield of 90, 99, 59, 75, and 91 %, respectively. The best overall
yield to synthesize rosiglitazone based on compound 2 was 40 %, being suitable for
industrial purposes, using water as a green solvent and avoiding column chro-
matography during the last three reaction steps.
Keywords Rosiglitazone Á Avandia Á BRL49653 Á Synthesis technology Á
Optimization Á Intermediates
&
Ge Meng
mengge@mail.xjtu.edu.cn
1
2
3
School of Pharmacy, Health Science Center, Xi’an Jiaotong University, Xi’an 710061, Shaanxi,
People’s Republic of China
School of Software Engineering, Xi’an Jiaotong University, Xi’an 710049, Shaanxi,
People’s Republic of China
School of Science, Xi’an Jiaotong University, Xi’an 710049, Shaanxi,
People’s Republic of China
123

G. Meng et al.
Introduction
Besides its well-known antihyperglycemic activity, rosiglitazone (1, Avandia, BRL
9653, by GlaxoSmithKline, Scheme 1) also displays various other biological
4
activities [1, 2], such as anti-inflammatory [3], anti-cancer [4], increasing dendritic
spine density [5], anti-Alzheimer’s disease [6], anti-ulcerative colitis [7], anti-
pancreatitis [8], anti-acanthosis nigricans [9], inducing brown adipose tissue (BAT)
recruitment [10], stimulating nitric oxide (NO) synthesis [11], and diminishing NO
synthase cofactor [12].
Retrosynthetic analysis of 1 indicated four main materials, including 2-chloropy-
ridine (2), N-methylethanolamine (3), 4-fluorobenzaldehyde (4a) or its equivalent
hydroxybenzaldehyde (4b), and 1,3-thiazolidine-2,4-dione (5) (Scheme 1).
Although many improved synthesis methods for 1 have been reported, including
a microwave (MW)-assisted approach [13], a solid-phase (SP) protocol [14], a
polymer-supported (PS) method [15], and a semiconvergent approach [16], the
linear synthesis route remains the reliable option to obtain this product efficiently
[
2], being a four-step route requiring more than 36 h to afford the final product [17].
Considering that MW, SP, and PS methods are not suitable for industrial scale-up,
many pharmaceutical industries still adopt linear methods involving column
chromatography to separate intermediates to provide bulk pharmaceutical chemicals
(
BPCs) [17]. Optimization for economical and convenient synthesis conditions
might provide some suggestions for companies to reduce production costs,
improving each step of the linear synthesis protocol for 1, which is still required
from the global perspective.
The yield for each step in the process to synthesize 1 varies significantly based on
various factors such as mole ratio, reaction time, and catalyst. There is still scope to
improve the productive cost efficiency by optimizing the reaction conditions of this
practical linear synthesis route based on the following five sequential reaction steps:
Intermediate 6 is first obtained via alkylation on secondary amine 3 with pyridine
chloride 2. Further etherification of alcohol 6 with 4-fluorobenzaldehyde (4a)
affords 7 [18]. Knoevenagel condensation between aldehyde 7 and 5 provides the
intermediate 8. Reduction of the double bond of compound 8 with NaBH under
4
basic condition gives the target drug 1. The main material 1,3-thiazolidine-2,4-dione
O
O
NH
N
N
O
S
1
H
N
OHC
X
OH
O
O
N
H
N
2
Cl
X = F OH etc.
,
,
S
3
4
a, 4b, etc
5
Scheme 1 Retrosynthetic analysis of rosiglitazone (1)
1
23

Synthetic optimization of rosiglitazone and related…
(5) was synthesized directly from 2-chloroacetic acid (9) and thiourea (10) via
cyclization in either hydrochloride or water using our previously improved methods
[19, 20]. The optimal overall yield of this linear synthesis route for 1 was 40 %
(Scheme 2).
Results and discussion
As an important synthon for 1 and other related glitazones, 1,3-thiazolidine-2,4-
dione (5) was earlier obtained from 2-iminothiazolidine-4-one hydrochloride salt
via cyclization between 2-chloroacetic acid (9) or its ester and thiourea (10) under
concentrated hydrochloride or sulfuric acid. The 2-iminothiazolidine-4-one HCl salt
was isolated and then neutralized with sodium hydroxide to give 2-iminothiazo-
lidine-4-one as free base, which was hydrolyzed under concentrated hydrochloride
condition to give 5. This early method was improved via a one-pot procedure
(
84 %) [19] or a green synthesis method by our group recently (79 %) [20].
Subsequent research on this cyclization reaction indicated that the yield of both
the one-pot and green methods could be enhanced by adjusting the amount and
concentration of solvent. Therefore, we designed an experiment to explore the
effects of different levels of these factors, aiming for better yield based on the well-
established orthogonal theory (Tables 1, 2).
The results showed that this cyclization process happened in both water and
hydrochloride solution with various concentrations (Table 3), the latter of which
could slightly influence the yield. However, in water solution, the yield was mainly
dependent on the concentration of the reactant. The greater the amount of water
used for a certain amount of reactant, the lower the yield, which could be explained
by the solubility of the product 6 in water. Excess amount of solvent would lead to
heavy loss during the workup and recovery difficulties.
2-(N-Methyl-N-(pyridin-2-yl)amino)ethanol (6) is an intermediate for both the
linear and convergent synthetic routes for rosiglitazone. Since hydrochloride is also
the main side-product of the same reaction, it is necessary to use an excess amount
of 2 above the theoretical mole ratio of 2 to 3 (1:1) as an acid catcher. Based on this
F
H
N
Reflux
99%
OHC
O
OH
4a
NaH, 59%
N
N
N
OH
N
Cl
N
N
2
3
CHO
NH
6
7
Piperidine, toluene
7
5%
H
N
NaBH , NaOH
O
O
4
O
O
N
N
NH
O
O
CoCl , DMG
N
O
2
S
O
5
S
9
1%
S
8
H O
90%
2
1
O
S
Cl
OH
H N NH2
2
9
10
Scheme 2 Linear synthesis route for rosiglitazone (1)
123

G. Meng et al.
Table 1 Different levels of main factors in synthesis of intermediate 5
Factor
Level
1
2
3
4
5
A: solvent
B: ratio
Hydrochloric
acid (12 mol/L)
Hydrochloric
acid (6 mol/L)
Hydrochloric
acid (4 mol/L)
Hydrochloric
acid (3 mol/L)
Water
2.00
10
6.67
4.00
3.33
(9/solvent,
mol/L)
Table 2 Experimental plan for
intermediate 5 by orthogonal
design method
Entry
Level
Factor A: solvent
Factor B: ratio (solvent/9, L/mol)
1
2
3
4
5
6
1
2
3
4
5
5
4
3
3
3
3
2
Table 3 Optimization result for synthesis of 1,3-thiazolidine-2,4-dione (5)
Entry Solvent
Time
h)
Ratio (9/solvent,
mol/L)
Yield
a
(%)
M.p.
(°C)
Yield
b
(%)
(
1
2
3
4
Hydrochloric acid
12 mol/L)
Hydrochloric acid
6.0 mol/L)
Hydrochloric acid
4.0 mol/L)
Hydrochloric acid
3.0 mol/L)
11
8
3.33
4.00
4.00
4.00
83
82
67
69
126–127 84 [19,
20]
(
126–127
126–127
126
(
8
(
8
(
5
6
a
Water
Water
8
4.00
6.67
77
90
126–127
124–125
10
Isolated yield of product (5)
b
Yield obtained from the reaction sequence used in industry
1
23

Synthetic optimization of rosiglitazone and related…
reaction mechanism and our experience, the yield of this tertiary amine 6 is
dependent on two main factors, viz. mole ratio and reaction time. Therefore, eight
main levels of these factors and experimental plans were designed according to the
orthogonal method to explore both factors efficiently at the same time, as shown in
Tables 4 and 5, respectively. The experimental results showed that the yield of 6
depended on the ratio of the materials to a large extent.
The best yield was achieved when the volume ratio of 2 to 3 (v/v) was set
between 4.0 and 5.0, and the reaction time was set between 9 and 10 h (entry 6,
Table 6). If the ratio was very low, material 3 could not be consumed completely,
increasing the amount of side-products. If the ratio was too high, recovering the
excess amount of material 2 was a difficult task requiring long distillation time,
which could lead to an increased amount of side-products. Both of the extreme
conditions led to low yield of 6. Overprolonging the reaction time could increase the
side-products and lower the yield of intermediate 6 (entry 7, Table 6). The relevant
details of these results are summarized in Table 6.
In comparison with literature [16], the improvements of this reaction step can be
summarized in the following two aspects: the workup involved no ammonium
chloride, and the reaction time was reduced from 13 to 9 h [16]. Although
compound 2 was used in excess amount, it could be recycled via distillation.
As the specific intermediate for the linear synthesis route, aldehyde 7 can be
obtained by many methods. In this work, only two main methods were investigated
for etherification between 4a and 6 (Table 7).
Method 1 using sodium hydride is the industrial approach to afford intermediate
. Dimethylformamide (DMF) was used as the solvent, with a high boiling point
7
leading to inconvenient workup. Sodium hydride is an unstable, hazardous reagent.
The yield of this step is relatively lower than other reactions of this route, so
improving this reaction was important to influence the overall yield. Method 2 using
a phase-transfer catalyst (PTC) and potassium hydroxide as the acid catcher could
be carried out under heating or microwave condition. The PTC used in a mixed
solution of toluene and water was tetrabutylammonium hydrogen sulfate (TBAHS)
or tetrabutylammonium bromide (TBAB). The yield for forming the ether bond
could be assisted by the microwave condition. Although the yield under MW
condition using TBAHS as the PTC was reported to be the best (90 %), optimization
of this method was not investigated as MW is usually not suitable for
industrialization. Only the effect of economical TBAB on this reaction was
investigated as an alternative to PTC.
The crude product 7 could be directly used in Knoevenagel condensation to
(Z)-5-[[4-[2-(methyl-2-pyridinylamino)ethoxy]phenyl]-methylene]-1,3-
provide
Table 4 Different levels of
main factors for synthesizing
intermediate 6
Factor
Level
1
2
3
4
5
6
7
8
A: ratio (2/3, v/v) 3.0 3.5 4.0 4.6 4.8 5.0 5.3 5.5
B: time (h) 10 12 15 16 17
5
6
9
123

G. Meng et al.
Table 5 Design of
experimental plan for 6 by
orthogonal method
Entry
Level
Factor A: ratio (2/3, v/v)
Factor B: reaction time
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
3
3
5
5
2
3
6
4
Table 6 Optimization results for synthesis of 6
H
N
Reflux
OH
OH
N
2
Cl
N
N
3
6
a
b
Entry
A: mole ratio (2/3, v/v)
B: reaction time (h)
Yield (%)
Yield (%)
1
2
3
4
5
6
7
8
a
3.0
3.5
4.0
4.6
4.8
5.0
5.3
5.5
9
9
84
89
99
98
97
99
90
97
77 [17]
12
12
6
9
15
10
Isolated yield of product (6)
b
Yield obtained from the reaction sequence used in industry
thiazolidine-2,4-dione (8), which could also be used for the next reduction step
without further purification. The research into this reaction showed that the yield
was high (75 %) when 1,3-thiazolidine-2,4-dione (5) was used in slight excess
(mole ratio of 7:5 = 1:1.1) above the required equivalent. Toluene was better than
methanol as an effective solvent system to enhance the yield.
The final product 1 was obtained from 8 via reduction with sodium borohydride
under basic condition containing a catalytic amount of cobalt chloride hexahydrate
and dimethylglyoxime (DMG) with yield of 91 %. The final step in the linear
synthesis of 1 was improved in this work based on the reported method in the
following three main aspects:
Firstly, lithium borohydride was reported to reduce 8 to 1 with yield of 76 %.
Reduction of the exocyclic double bond involved use of boron lithium, which is too
expensive for industrial purposes [22]. It was also necessary to add some DMF with
1
23

Synthetic optimization of rosiglitazone and related…
Table 7 Comparison of two synthesis methods for 7
a
b
Entry
Method and reagents
Time
Temperature (°C)
Yield (%)
Yield (%)
1
2
a
NaH and DMF
24 h
25
85
59
53
47 [21]
TBAB and microwave
40 min
Isolated yield of product (7)
b
Yield obtained from the reaction sequence used in industry
slight heating when using sodium borohydride for the reduction, the yield of which
is only 70 %. Our improved procedures involved use of sodium borohydride with
direct dissolution of 8 in water before adding aqueous NaOH. This might allow
complete dissolution of 8 in water to form the sodium salt of 8 without addition of
any extra organic solvent. The reaction was kept at room temperature with high
yield of 91 %.
Secondly, sodium borohydride should be dissolved in extremely dilute aqueous
sodium hydride solution before addition to the basic aqueous solution of material.
The procedures include use of the inverse addition manner to improve the
troublesome procedures. To be specific, basic solution of 8 was added with cobalt
chloride, dimethylglyoxime (DMG) before adding with sodium boron hydride
powder at room temperature, which could be manipulated more conveniently than
described in literature.
Thirdly, intermediate 8 was prepared via Knoevenagel condensation between
,3-thiazolidine-2,4-dione (5) and an unpurified intermediate of 4-[2-[N-methyl-N-
1
(
2-pyridyl)]amido]ethoxybenzaldehyde (7) prepared from 4-fluorobenzaldehyde
4a). Therefore, it is possible that unpurified material 4a might remain in the crude
(
product of the unpurified intermediate 7. Under the same basic Knoevenagel
condensation condition, 4a could condense with 1,3-thiazolidine-2,4-dione (5) also
to give 5-(4-fluorobenzylidene) 1,3-thiazolidine-2,4-dione (11) as the main side-
product (Scheme 3). The side-product 11 could be easily removed by adjusting the
pH of the reaction mixture after completing the reduction, avoiding column
chromatography. Therefore, the target compound 1 was purified easily by adjusting
pH, which is quite suitable for industrial purposes.
O
H
N
F
Piperidine, AcOH
Toluene, reflux
O
O
NH
O
OHC
S
S
F
5
4a
11
Scheme 3 Formation of 11 during the linear synthesis route
123

G. Meng et al.
Experimental
All materials were obtained from commercial suppliers and used as received.
Melting points were taken on an X-1 digital melting point apparatus and are
uncorrected. Infrared (IR) spectra were recorded on a Nicolet FT-IR 360
1
spectrophotometer. H nuclear magnetic resonance (NMR) spectra were recorded
on a Bruker AM-300 (400 MHz) spectrometer with tetramethylsilane (TMS) as
internal standard. Chemical shifts are reported in d. Mass spectra were measured on
a HP5988A instrument by direct inlet at 70 eV.
General procedures for synthesis of target molecule and intermediates
Synthesis of 1,3-thiazolidine-2,4-dione (5)
2
-Chloroacetic acid (9, 39.6 g, 420 mmol) and thiourea (10, 31.9 g, 420 mmol)
were mixed together in a flask, to which water (60.0 mL) was added. The reaction
mixture was heated to reflux for 10 h, then allowed to cool down to room
temperature. Large amounts of solid appeared from the mixture, which were
collected by filtering over vacuum and washed with a small amount of ice water
(
10.0 mL). The filter cake was dried under vacuum to give the product 5 (44.3 g,
-
1
9
0 %), m.p. 126–127 °C. lit m.p. 125–126 °C [19], IR (KBr, cm ): 3130 (mN–H),
821 (mC–H), 1670 (mC=O).
2
Synthesis of 2-(N-methyl-N-(pyridin-2-yl)amino)ethanol (6)
2
-Chloropyridine (3, 16.1 mL, 19.3 g, 169.7 mmol, d = 1.20 g/mL, b.p. 166.0 °C)
and N-methylethanolamine (2, 80.5 mL, 75.7 g, 1.01 mol, d = 0.94 g/mL, b.p.
58.1 °C) were added into a clean flask. The reaction mixture was heated to reflux
1
for 9 h. The reaction process was monitored by thin-layer chromatography (TLC).
After 9 h of heating, the reaction mixture was condensed under vacuum. The excess
amount of 2 was recovered. The residue was allowed to cool down to room
temperature and was then added with some ice water (100 mL). The mixture was
extracted with dichloromethane (150 mL 9 3). The organic layers were combined
and washed with saturated brine solution (150 mL), and dried over sodium sulfate.
The final product 6 was obtained after filtration and concentration under vacuum as
1
clear yellow oil (25.6 g, 99 %) (b.p. 112–115 °C at 0.8 Torr [23]). H NMR
(
400 MHz, CDCl ) d: 8.01 (m, 1H, pyridine-6-H), 7.44 (m, 1H, pyridine-4-H), 6.54
3
(
m, 1H, pyridine-3-H), 6.51 (m, 1H, pyridine-5-H), 3.80 (t, 2H, CH OH), 3.67 (t,
1
2
2
2
the next step without further purification.
H, R R NCH CH OH), 3.03 (s, 3H, CH N). Compound 6 could be used directly in
2 2 3
Synthesis of 4-[2-[N-methyl-N-(2-pyridyl)]amido]ethoxybenzaldehyde (7)
Method 1 Using sodium hydride: 2-[N-methyl-N-(2-pyridyl)]amidoethanol (6,
4
.42 mL, 5.00 g, 32.9 mmol, d = 1.13 g/mL, b.p. 281 °C) and anhydrous DMF
1
23

Synthetic optimization of rosiglitazone and related…
(
(
60.0 mL) were mixed together in a three-necked flask, to which sodium hydride
70 %, 2.30 g, 95.8 mmol) was added portionwise. Nitrogen protection was applied
after the addition of sodium hydride, and the speed of hydrogen gas evolution was
abated. 4-Fluorobenzaldehyde (4a, 7.76 mL, 9.00 g, 72.5 mmol, d = 1.16 g/mL,
b.p. 181 °C) was then added dropwise to the mixture, keeping the temperature
below 50 °C. The reaction mixture was kept at r.t. for 24 h while monitoring with
TLC, then concentrated by evaporating DMF partially. The mixture became turbid
after diluting with cold water (30.0 mL), which was adjusted with aqueous
hydrochloride (2.0 mol/L) to pH = 3. The mixture was extracted with dichlor-
omethane (30.0 mL 9 3) to remove the unused material 6. Aqueous sodium
hydroxide (5.0 mol/L) was added to the aqueous layer to modulate the pH to 11
again, followed by extraction with dichloromethane (30.0 mL 9 3). The combined
organic phase was washed with brine (30.0 mL 9 2), dried over anhydrous sodium
sulfate, filtered, concentrated, and purified by column chromatography (silica gel,
eluent: chloroform/methanol, v/v = 10:1) to provide the crude product (5.00 g,
5
9 %), which was recrystallized with ethyl acetate/petroleum ether (v/v = 1:1) to
1
afford white crystal 7 (4.20 g, 50 %), m.p. 66–67 °C [13]. H NMR (400 MHz,
CDCl ): d 9.88 (s, 1H, CHO), 8.17 (d, 1H, J = 4.0 Hz, pyridine-6-H), 7.83 (d, 2H,
3
J = 8.0 Hz, Ar–H), 7.48 (t, 1H, J = 12.0 Hz, pyridine-4-H), 7.02 (d, 2H,
J = 8.0 Hz, Ar–H), 6.58 (m, 1H, pyridine-3-H), 6.53 (d, 1H, pyridine-5-H), 4.30
(
t, 2H, OCH CH N), 4.03 (t, 2H, OCH CH N), 3.15 (s, 3H, CH N).
2 2 2 2 3
Method 2 Using tetrabutylammonium bromide (TBAB): 2-[N-methyl-N-(2-
pyridyl)]amidoethanol (6, 2.30 mL, 2.60 g, 16.8 mmol), 4-fluorobenzaldehyde (4b,
.81 mL, 2.10 g, 16.8 mmol), potassium hydroxide (2.80 g, 50.4 mmol), tetrabuty-
lammonium bromide (TBAB, 0.60 g, 1.70 mmol), water (0.50 mL), and toluene
10.0 mL) were put into a three-necked flask. The reaction mixture was put into a
1
(
microwave reactor with power of 800 W. Heating was kept under 80 °C for 40 min
to complete the reaction. The residue was extracted with dichloromethane
(
50 mL 9 3). The organic layers were combined and washed with water, then
dried over anhydrous sodium sulfate. The filtrate was concentrated before separating
via stepwise column chromatography (eluent: petroleum/ethyl acetate ester,
v/v = from 10:1 to 7:1) to give the crude product 7 (2.30 g, 53 %), which was
recrystallized from ethyl acetate ester/petroleum (v/v = 1:1) to give needle-like
1
white crystals as pure 7 (1.80 g, 41 %), m.p. 65–66 °C, lit. viscous liquid [16]. H
NMR (400 MHz, CDCl ) d: 9.88 (s, 1H, CHO), 8.17 (d, 1H, J = 4.0 Hz, pyridine-
3
6
-H), 7.83 (d, 2H, J = 8.0 Hz, Ar–H), 7.48 (t, 1H, J = 12.0 Hz, pyridine-4-H), 7.02
d, 2H, J = 8.0 Hz, Ar–H), 6.58 (m, 1H, pyridine-3-H), 6.53 (d, 1H, pyridine-5-H),
.30 (t, 2H, OCH CH N), 4.03 (t, 2H, OCH CH N), 3.15 (s, 3H, CH N).
(
4
2
2
2
2
3
Synthesis of (Z)-5-[[4-[2-(methyl-2-pyridinylamino)ethoxy]phenyl]-methylene]-1,3-
thiazolidine-2,4-dione (8)
4
-[2-[N-Methyl-N-(2-pyridyl)]amido]ethoxybenzaldehyde (7, 9.00 g, 35.1 mmol),
,3-thiazolidine-2,4-dione (5, 6.20 g, 52.7 mmol), and toluene (60.0 mL) were put
1
into a three-necked flask and stirred until dissolving, to which was added a few
drops of piperidine (about 0.5 mL) under nitrogen atmosphere. The mixture was
123

G. Meng et al.
heated to reflux for 6 h until reaction completion. Large amounts of yellow solid
precipitated from the solution after cooling down to room temperature. Methanol
(10.0 mL) was added to the reaction mixture with stirring for 10 min before
filtering. The filter cake was washed with a small amount of cold methanol and dried
under vacuum to give yellow powder as the crude product 8 (9.40 g, 75 %), m.p.
1
96–200 °C. Yellow crystals were obtained after recrystallization with ethanol as
1
the pure product (8.50 g, 80 %), m.p. 196–197 °C, lit. m.p. 197–198 °C [13]. H
NMR (400 MHz, DMSO) d: 8.0 (d, 1H, pyridine-6-H), 7.7 (s, 1H, Ar–CH=), 7.5 (m,
3
H, pyridine-4-H & O–Ar–H), 7.1 (m, 2H, O–Ar–H), 6.6 (m, 1H, pyridine-3-H), 6.5
m, 1H, pyridine-5-H), 4.2 (t, 2H, OCH CH N), 3.9 (t, 2H, OCH CH N), 3.0 (s,
(
2
?
2
2
2
3
H, CH N); EI MS m/z (%): 355.0 ([M] , 3), 221.0 (35), 150.1 (100).
3
Synthesis of 5-[[4-[2-(methyl-2-pyridinylamino)ethoxy]phenyl]-methyl]-1,3-
thiazolidine-2,4-dione (1)
(
Z)-5-[[4-[2-(Methyl-2-pyridinylamino)ethoxy]phenyl]-methylene]-1,3-thiazo-
lidine-2,4-dione (8, 3.60 g, 10.1 mmol) was mixed in water (450 mL) with
stirring, to which was added aqueous sodium hydroxide (0.5 mol/L) to pH = 11
until the material 8 dissolved completely. The solution was added with cobalt
chloride hexahydrate (0.20 g, 0.84 mmol), DMG (0.20 g, 1.72 mmol), and
sodium boron hydride (4.50 g, 120 mmol). The mixture was kept at room
temperature for 24 h until the reaction was complete. The insoluble residue was
removed by filtration. The filtrate was adjusted with hydrochloride to pH = 3,
then a small amount of solids appeared. The filtrate from the second filtration was
adjusted to pH 6–7 with aqueous sodium hydroxide, then a large amount of solids
appeared from the solution and was collected by filtration. The filter cake was
washed with a small amount of water and dried under vacuum to give the final
product 1 (0.80 g), m.p. 154–155 °C. The filtrate was extracted with dichlor-
omethane (50.0 mL 9 3), washed with saturated brine (50.0 mL 9 2), and dried
over anhydrous sodium sulfate. The solid was obtained via filtration and
concentration as the second portion of 1 (2.50 g, total amount: 3.30 g, crude
yield: 91 %). Pale-white crystals were obtained via recrystallization from ethanol
as pure 1 (1.80 g, 50 %), m.p. 154–156 °C, lit. m.p. 152–153 °C [16], m.p.
1
1
54–155 °C [13]. H NMR (400 MHz, DMSO) d: 12.00 (s, 1H, NH), 8.07 (d, 1H,
J = 8.0 Hz, pyridine-6-H), 7.48 (t, 1H, J = 8.0 Hz, pyridine-4-H), 7.12 (d, 2H,
J = 8.0 Hz, Ar–H), 6.86 (d, 2H, J = 8.0 Hz, Ar–H), 6.63 (d, 1H, J = 8.0 Hz,
pyridine-3-H), 6.55 (m, 1H, pyridine-5-H), 4.84 (m, 1H, thiazolidine-5-H), 4.09
(
CH N); EI MS m/z (%): 342.0 (3), 107.1 (100), 223.1 (5).
t, 2H, OCH CH N), 3.88 (t, 2H, OCH CHN), 3.05 (m, 2H, CH Ar), 2.49 (s, 3H,
2 2 2 2
3
Conclusions
Rosiglitazone was synthesized from commercially available materials via a five-step
linear synthetic route with improved overall yield of 40 % (based on 2).
Optimization for each reaction step was designed and explored using an orthogonal
1
23

Synthetic optimization of rosiglitazone and related…
method on the linear synthesis route to determine the best conditions. The yield for
each reaction step was 90, 99, 59, 75, and 91 %, respectively. The three main
aspects of the improvement in the final step offered a practical and reliable
technological protocol to synthesize 1 efficiently. The newly improved workup of
the final reaction step offers an alternative method to afford the purified product
without column chromatography, including purification of the final target compound
1, intermediate 6, and intermediate 8. All the improved manipulations in the related
procedures are conveniently suitable for industrial purposes.
Acknowledgments The authors would like to thank the Returned Overseas Chinese Scholars, State
Education Ministry (SRF for ROCS, SEM), P. R. China (2011) and Technology Research and
Development Program of China, International Cooperation (2013KW31-04) for financial support. The
project was also sponsored by the Cooperation Project between Xi’an Jiaotong University and Weifang
Keli Chemical Limited Company, P. R. China (2013).
References
1
2
3
4
5
6
7
8
9
. B.B. Lohray, V. Bhushan, B.P. Rao, G.R. Madhavan, N. Murali, K.N. Rao, A.K. Reddy, B.M.
Rajesh, P.G. Reddy, R. Chakrabarti, J. Med. Chem. 41(10), 1619–1630 (1998)
. B.C. Cantello, M.A. Cawthorne, D. Haigh, R.M. Hindley, S.A. Smith, P.L. Thurlby, Bioorg. Med.
Chem. Lett. 4(10), 1181–1184 (1994)
. P. Mohanty, A. Aljada, H. Ghanim, D. Hofmeyer, D. Tripathy, T. Syed, W. Al-Haddad, S. Dhindsa,
P. Dandona, J. Clin. Endocr. Metab. 89(6), 2728–2735 (2004)
. S.S. Palakurthi, H. Aktas, L.M. Grubissich, R.M. Mortensen, J.A. Halperin, Cancer Res. 61(16),
6213–6218 (2001)
. J. Brodbeck, M.E. Balestra, A.M. Saunders, A.D. Roses, R.W. Mahley, Y. Huang, Proc. Natl. Acad.
Sci. 105(4), 1343–1346 (2008)
. M. Risner, A. Saunders, J. Altman, G. Ormandy, S. Craft, I. Foley, M. Zvartau-Hind, D. Hosford, A.
Roses, Pharmacogenomics J. 6(4), 246–254 (2006)
. J.D. Lewis, G.R. Lichtenstein, J.J. Deren, B.E. Sands, S.B. Hanauer, J.A. Katz, B. Lashner, D.H.
Present, S. Chuai, J.H. Ellenberg, Gastroenterology 134(3), 688–695 (2008)
. C. Chen, S. Xu, W.-X. Wang, Y.-M. Ding, K.-H. Yu, B. Wang, X.-Y. Chen, Arch. Med. Res. 40(2),
79–88 (2009)
. A. Romo, S. Benavides, Ann. Pharmacother. 42(7–8), 1090–1094 (2008)
10. W.T. Festuccia, P.-G. Blanchard, D. Richard, Y. Deshaies, Am. J. Physiol. Regul. Integr. Comparat.
Physiol. 299(1), R159–R167 (2010)
11. J.G. Boyle, P.J. Logan, M.-A. Ewart, J.A. Reihill, S.A. Ritchie, J.M. Connell, S.J. Cleland, I.P. Salt, J.
Biol. Chem. 283(17), 11210–11217 (2008)
1
1
1
1
1
1
2. P. Linscheid, U. Keller, N. Blau, D.J. Schaer, B. M u¨ ller, Biochem. Pharmacol. 65(4), 593–598 (2003)
3. S.L. Gaonkar, H. Shimizu, Tetrahedron 66(18), 3314–3317 (2010)
4. K.M. Brummond, J. Lu, J. Org. Chem. 64(5), 1723–1726 (1999)
5. X. Li, C. Abell, B.H. Warrington, M. Ladlow, Org. Biomol. Chem. 1(24), 4392–4395 (2003)
6. D.V. Jawale, U.R. Pratap, R.A. Mane, Bioorg. Med. Chem. Lett. 22(2), 924–928 (2012)
7. B.C. Cantello, M.A. Cawthorne, G.P. Cottam, P.T. Duff, D. Haigh, R.M. Hindley, C.A. Lister, S.A.
Smith, P.L. Thurlby, J. Med. Chem. 37(23), 3977–3985 (1994)
18. S. Raza, S.P. Srivastava, D.S. Srivastava, A.K. Srivastava, W. Haq, S. Katti, Eur. J. Med. Chem. 63,
611–620 (2013)
1
2
2
2
9. G. Meng, Z.-Y. Li, M.-L. Zheng, Org. Prep. Proced. Int. 40(6), 572–574 (2008)
0. G. Meng, Y. Gao, M.-L. Zheng, Org. Prep. Proced. Int. 43(3), 312–313 (2011)
1. T. Sohda, K. Mizuno, Y. Momose, H. Ikeda, T. Fujita, K. Meguro, J. Med. Chem. 35(14), 2617–2626 (1992)
2. R.G. Giles, N.J. Lewis, J.K. Quick, M.J. Sasse, M.W.J. Urquhart, L. Youssef, Tetrahedron 56,
4531–4537 (2000)
23. M.K. Chmielewski, V. Marcha, J. Cieslak, A. Grajkowski, V. Livengood, U. Munch, A. Wilk, S.L.
Beaucage, J. Org. Chem. 68(26), 10003–10012 (2003)
123