Discovery of liver-directed glucokinase activator having anti-hyperglycemic effect without hypoglycemia

Article abstract of DOI:10.1016/j.ejmech.2017.03.042

Glucokinase activators (GKAs) are among the emerging drug candidates for the treatment of type 2 diabetes (T2D). Despite effective blood glucose lowering in clinical trials, many pan-GKAs “acting both in pancreas and liver” have been discontinued from cli

Full text of DOI:10.1016/j.ejmech.2017.03.042

Accepted Manuscript
Discovery of liver-directed glucokinase activator having anti-hyperglycemic effect
without hypoglycemia
Anil M. Deshpande, Debnath Bhuniya, Siddhartha De, Bhavesh Dave, Vinod P.
Vyavahare, Santosh H. Kurhade, Sachin R. Kandalkar, Keshav P. Naik, Balasaheb
S. Kobal, Rahul D. Kaduskar, Sujay Basu, Vaibhav Jain, Pratima Patil, Sandhya
Chaturvedi Joshi, Ganesh Bhat, Amol A. Raje, Satyanarayana Reddy, Jayasagar
Gundu, Vamsi Madgula, Suhas Tambe, Prasad Shitole, Dhananjay Umrani, Anita
Chugh, Venkata P. Palle, Kasim A. Mookhtiar
PII:
S0223-5234(17)30199-X
10.1016/j.ejmech.2017.03.042
EJMECH 9302
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 December 2016
Revised Date: 9 March 2017
Accepted Date: 22 March 2017
Please cite this article as: A.M. Deshpande, D. Bhuniya, S. De, B. Dave, V.P. Vyavahare, S.H. Kurhade,
S.R. Kandalkar, K.P. Naik, B.S. Kobal, R.D. Kaduskar, S. Basu, V. Jain, P. Patil, S.C. Joshi, G. Bhat,
A.A. Raje, S. Reddy, J. Gundu, V. Madgula, S. Tambe, P. Shitole, D. Umrani, A. Chugh, V.P. Palle, K.A.
Mookhtiar, Discovery of liver-directed glucokinase activator having anti-hyperglycemic effect without
hypoglycemia, European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.03.042.
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.

ACCEPTED MANUSCRIPT
GRAPHICAL ABSTRACT

ACCEPTED MANUSCRIPT
Discovery of liver-directed glucokinase activator having anti-hyperglycemic effect without
hypoglycemia
Anil M. Deshpande*, Debnath Bhuniya, Siddhartha De, Bhavesh Dave, Vinod P. Vyavahare,
Santosh H. Kurhade, Sachin R. Kandalkar, Keshav P. Naik, Balasaheb S. Kobal, Rahul D.
Kaduskar, Sujay Basu, Vaibhav Jain, Pratima Patil, Sandhya Chaturvedi Joshi, Ganesh Bhat,
Amol A. Raje, Satyanarayana Reddy, Jayasagar Gundu, Vamsi Madgula, Suhas Tambe, Prasad
Shitole, Dhananjay Umrani, Anita Chugh, Venkata P. Palle and Kasim A. Mookhtiar
Advinus Therapeutics Ltd., Drug Discovery Facility, Quantum Towers, Plot-9, Phase-I, Rajiv
Gandhi Infotech Park, Hinjewadi, Pune 411 057, India
*
Corresponding Author
Tel.: 91-20-66539637; Fax: 91-20-66539620;
E-mail: anil.deshpande@advinus.com

ACCEPTED MANUSCRIPT
Abstract: Glucokinase activators (GKAs) are among the emerging drug candidates for the
treatment of type 2 diabetes (T2D). Despite effective blood glucose lowering in clinical trials,
many pan-GKAs “acting both in pancreas and liver” have been discontinued from clinical
development mainly because of their potential to cause hypoglycemia. Pan-GKAs over sensitize
pancreatic GK, resulting in insulin secretion even at sub-normoglycemic level which might be a
possible explanation for hypoglycemia. An alternative approach to minimize the risk of
hypoglycemia is to use liver-directed GKAs, which are reported to be advancing well in clinical
development. Here, we report the discovery and structure-activity relationship (SAR) studies on
a novel 2-phenoxy-acetamide series with the aim of identifying a liver-directed GKA.
Incorporation of a carboxylic acid moiety as an active hepatocyte uptake recognizing element at
appropriate position of 2-phenoxy-acetamide core led to the identification of 26, a potent GKA
with predominant liver-directed pharmacokinetics in mice. Compound 26 on oral administration
significantly reduced blood glucose levels during an oral glucose tolerance test (oGTT)
performed in diet-induced obese (DIO) mice, while showing no sign of hypoglycemia in normal
C57 mice over a 10-fold dose range, even when dosed at fasted condition. Together, these data
demonstrate a liver-directed GKA has beneficial effect on glucose homeostasis with reduced risk
of hypoglycemia.
Keywords: Glucokinase; Activator; Liver-directed; 2-Phenoxy-acetamide; Anti-hyperglycemic
Introduction
The glucose phosphorylating enzyme Glucokinase (GK) acts as a physiological glucose
sensor in glucose-responsive tissues such as liver and pancreas, and therefore plays a pivotal role
in glucose homeostasis. The limited tissue distribution and unique kinetic properties of GK (K
m

ACCEPTED MANUSCRIPT
for glucose ~ 8 mM) allow it to maintain plasma glucose levels in the physiological range via a
dual mechanism, (i) insulin secretion in pancreatic ß-cells and (ii) increase in hepatic glucose
uptake [1]. In human, natural mutations of glucokinase are associated with loss or gain of the
enzyme function leading to phenotypes of hyper and hypoglycemia respectively [2,3]. This
etiology has provided new insights into the pathogenesis of Type-II Diabetes Mellitus (T2DM)
[
4]. It has been reported in the literature that mice deficient in GK experiences the hyperglycemia
5], whereas over expressing hepatic GK showed improved glucose tolerance [6]. GK expression
[
in the liver is progressively reduced with increasing disease severity in Zucker diabetic fatty
ZDF) rats, and normalizing liver GK expression restores plasma glucose level near to normal
(
while also normalizing hepatics glucose production [6c]. All these observations suggest that
activation of GK should be beneficial in the treatment of T2DM by reducing hyperglycemia.
Indeed, small-molecules have been discovered by several groups as allosteric GK activators
(GKAs), and these have been shown to be effective antihyperglycemic agents both in animal and
clinical settings [7]. A few representative examples of GKAs are shown in Figure 1 [8-15].
Cl
O
O
F
MeO
2
S
MeO
2
S
N
N
2
MeO S
F
3
C
N
O
S
O
S
O
S
COOH
O
S
O
N
N
N
N
N
N
N
H
H
N
H
N
H
N
H
O
O
RO-28-1675
R-1440
Piragliatin)
PSN GK1
LY-2121260
PFE-GKA2 (PF-04991532)
(
N
N
O
O
O
O
N
O
N
Me
O
N
O
O
N
N
H
HO
N
N
HN
HO
N
N
H
H
O
N
O
H
O
N
N
O
O
N
S
O
N
2
SO Et
O
O
O
MK-0941
AZD-1656
AZD-6370
AM-9074
Figure 1: Structure of representative GK activators

ACCEPTED MANUSCRIPT
Roche was the first company to demonstrate antihyperglycemic effect of GKA, Piragliatin
R-1440), in a clinical study [9]. Piragliatin was discontinued from the trial, mainly because of its
(
potential to cause hypoglycemia [9d]. Recently, Merck has discontinued its clinical compound,
MK-0941, after Phase IIb studies due to inability to maintain its efficacy after chronic treatment,
and also partly due to hypoglycemia observed during the trial [12a]. These results illustrate the
scope of pharmacological activation of GK in the management of T2DM. It is important to note
that majority of GKAs discontinued from clinical developments are pan-GKAs, i.e. activators of
both liver and pancreatic GK. Our hypothesis is that, activation of pancreatic GK will lead to
insulin secretion even when glucose level is below normoglycemic level. This event will
certainly have the association of hypoglycemia. In addition, chronic activation of pancreatic GK
could lead to loss of glucose sensing ability of GK in the β-cell.
Therefore, one practical strategy to avoid hypoglycemia is to activate GK selectively in liver
by a liver-directed GKA, which would help to increase glucose uptake in liver, while sparing the
pancreas from the action of insulin secretion. Expression of GK in the liver is estimated to be 20-
fold higher than in the pancreas which should offer the desired efficacy [16]. Previous studies in
transgenic mice have shown that a liver-specific increase in GK activity has an
antihyperglycemic effect without leading to hypoglycemia [6]. Additionally, liver-directed
approach might have the advantage of decreasing the stress on β-cells and protecting its mass
and function. Recently Pfefferkorn et al [14] disclosed hepatoselecetive GKA whereas
Pennington et al [15b] reported disruptors of glucokinase-glucokinase regulatory protein (GK-
GKRP) interaction as an alternative approach to minimize the risk of hypoglycemia. Consistent
with these facts, it is encouraging to note that liver-directed GKAs from Pfizer (PF-04991532)
[14], Trans Tech Pharma (TTP-399, Phase-II) [17], and Advinus (GKM-001) [18] have

ACCEPTED MANUSCRIPT
advanced into clinical trials. During phase 2 studies, PF-04991532 revealed significant lowering
of HbA1c with minimal episode of hypoglycemia. However, dose related trend for increase in
plasma triglyceride was observed [14a,b]. Recent 10 days pilot study of TTP-399 in T2DM
patients showed improved glycemic control and insulin resistant without inducing hypoglycemia
or having detrimental effect on plasma lipids [17a].
Recognizing the potential for a liver-directed approach, we aimed very early in our program
to identify liver-directed and pancreas sparing GKAs, which would provide a beneficial effect on
glucose homeostasis with reduced risk of hypoglycemia. Here, we present synthesis and
biological evaluation of novel 2-phenoxy-acetamide derivatives as small-molecule GKAs. The
structure-activity-relationship (SAR) studies were focused on GK biochemical and functional
potency, liver-directed pharmacokinetic (PK) properties, efficacy in diet induced obese (DIO)
mice, and hypoglycemic potential in normal C57 mice.
Chemistry
Compound 1 was synthesized in four steps from 2,4-difluorophenol according to the route
described in Scheme 1. O-Alkylation of 2,4-difluorophenol by ethyl-2-bromo-isobutyrate
followed by ester hydrolysis afforded an intermediate acid 4. Appropriately substituted
aminothiazole then coupled with acid 4 to provide amide derivative 5, which on ester hydrolysis
afforded the desired compounds 1 and 6.

ACCEPTED MANUSCRIPT
O
O
O H
F
OEt
O H
a
b
c
O
O
F
F
F
F
F
2
3
4
4
R
4
R
O
S
S
3
R
O
3
R
N
H
N
d
N
N
O
F
O
F
H
3
4
F
2
1: R = CH COOH; R = Cl
F
5
3
4
6
: R = H; R = 4-carboxyphenoxy
Scheme 1. Reagents and conditions: (a) ethyl 2-bromoisobutyrate, Cs CO , DMSO, rt, 15 h,
2
3
9
4
1
6%; (b) LiOH, THF/MeOH/H O (3:1:1), rt, 16 h, 77%; (c) Ethyl 2-(2-amino-5-chloro-thiazol-
2
-yl)-acetate, EDCI, HOBT, NMM, DCM, rt, 12 h, 65%; (d) LiOH, THF /MeOH/H O (3:1:1), rt,
2
2 h, 67%.
Synthesis of sulfonamide derivatives is summarized in Scheme 2. 4-mercapto phenylacetic
acid (7) was esterified to its methyl ester 8 using Fischer esterification method. Oxidative
chlorination [19] of 8 using “TMS-Cl/Nitrate salt” reagent system afforded sulfonyl chloride 9.
Subsequent bromination of 9 with N-bromosuccinimide gave α-bromo ester, which was coupled
with an appropriate amine to provide sulfonamide derivative 10. Nucleophilic substitution of 10
with 2,4-di-fluorophenol furnished compound 11b. The reaction of 11b with LDA and methyl
1
2
iodide gave an intermediate 11a where R is piperidine and R is methyl. Further ester hydrolysis
of 11 gave intermediate acid derivatives 12. Finally, coupling of 12 with an appropriately
substituted 2-amino-thiazole afforded phenoxy-acetamide derivatives 13, which on ester
hydrolysis gave desired sulfonamide derivative 14.

ACCEPTED MANUSCRIPT
a
b
c
d
7
8
9
10a,d,e
g
h
f
1b,d,e (R2 = H)
1a (R = Me)
1
1
12a,b,d,e
13a-e
14a-e
e
2
Scheme 2. General synthetic route to sulfonamide derivatives 14a-e;
Reagents and conditions: (a) H SO , MeOH, rt, 24 h, 80%; (b) KNO
h, 73%; (c) (i) NBS, benzoyl peroxide, CCl , reflux, 48 h; (ii) Amine , DCM, rt, 30 min, 50%
for two steps); (d) 2,4-difluoro phenol, Cs CO , CH CN, rt, 3 h; (e) LDA, MeI, THF, -78 °C, 2
2
4
3
, TMSCl, DCM, 50 °C, 24
4
(
2
3
3
h, 39%; (f) LiOH, THF/MeOH/ H O (3:1:1), rt, 18 hr; (g) (2-amino-5-chloro-thiazol-4-yl)-acetic
2
acid ethyl ester, EDCI, HOBT, NMM, DCM, rt, 18 h; (h) LiOH, THF/MeOH/ H O (3:1:1), rt, 18
2
h.
Synthesis of sulfone analogue 25a is delineated in Scheme 3 and the same route was
employed for 25b. Alkylation of thiophenol with cyclopropyl bromide gave an intermediate 15.
Friedel-Craft acylation of 15 obtained α-keto ester 16 which on hydrolysis gave acid 17.
Reduction of keto group in 17 obtained phenyl acetic acid derivative 18. Esterification of 18 to
obtain 19, subjected to oxidation with m-CPBA gave sulfone derivative 20. Bromination of 20
with NBS gave α-bromo ester 21, which on nucleophilic substitution by 2,4-di-fluorophenol
afforded 22. Ester hydrolysis of 22 to furnish intermediate acid 23 followed by coupling with an
amine, ethyl 4-(2-aminothiazol-5-yl)oxybenzoate, gave an amide 24. Final hydrolysis of 24 gave
the desired sulfone derivative 25a.

ACCEPTED MANUSCRIPT
a
b
c
d
e
Thiophenol
1
5
16
17
18
g
f
h
i
1
1
9a, R = Cyclopropyl
9b, R = 4-Tetrahydropyran
2
0a-b
21a-b
22a-b
#
j
k
23a-b
24a-b
25a, R = Cyclopropyl
5b, R = 4-Tetrahydropyran
2
#
Scheme 3. Synthetic route to sulfone derivative 25a and 25b; 19b was obtained by alkylating
commercially available (4-mercapto-phenyl)-acetic acid methyl ester with 4-iodo-
tetrahydropyran.
o
Reagents and conditions: (a) Bromocyclopropane, tert-BuOK, DMSO, 60 C, 48 h, quantitative
o
yield; (b) Ethyl oxalyl chloride, AlCl
3
, DCM, 0 to 25 C, 4 h, 94%; (c) 2 molar NaOH, EtOH, 60
o
o
o
C, 2 hr, 92%; (d) N H .H O, KOH, -50 to 100 C, 20 h, 97%; (e) MeOH, SOCl , 0 C to rt, 3 h,
7%; (f) m-CPBA, DCM, 10 to 20 C, 2 h, 99%; (g) NBS, CCl , benzoyl peroxide (cat.), reflux,
2
4
2
2
o
9
4
4
o
h, 98%; (h) 2,4-Difluoro phenol, K CO3, CH CN, 25 C, 14 h, 60%; (i) LiOH,
2
3
THF/MeOH/H O (3:1:1), rt, 4 h, quantitative yield; (j) 4-(2-Amino-thiazol-5-yloxy)-benzoic acid
2
methyl ester, HATU, TEA, DCM, rt, 2.5 h, 52%; (k) LiOH, THF/MeOH(1:1), rt, 3 h, 86%.
Results and Discussion
Our initial efforts identified a tri-substituted acetamide core, a novel chemotype as GKA (Figure
2
). Early SAR studies were focused on optimizing rings A and B, wherein we evaluated different
aryl and heteroaryl rings with varied substitution patterns. Compound with 2,4-di-fluorophenyl
as ring A and 5-chlorothiazole as ring B represented a more attractive starting point based on its
potency of GK activation (EC of S reduction: 1.85 ± 0.58 µM). The subsequent priority was
5
0
0.5
to design a liver-directed GKA using the initial hit. To direct the pharmaceutical agent to the
liver, utilization of liver specific organic anion transporting polypeptides (OATPs), a membrane
transport protein, is one of the best approaches reported in the literature [20]. In general,

ACCEPTED MANUSCRIPT
carboxylic acids are preferred structural element for recognition as OATP substrates [20]. We
envisioned that incorporation of carboxylic acid into the initial hit may impart the desired liver-
directive property. Maintaining the structural features required for a GKA in our novel design,
the terminal thiazole ring was thought to be the best choice to append the carboxylic acid.
Amongst the several analogs synthesized with carboxylic acid functionality, compound 1 (Figure
2
A) was identified as an early lead from this series, with micromolar GK activation potency
(EC50: 7 ± 0.82 µM; Figure 2B & 2C) and very good metabolic stability. In order to corroborate
the idea behind designing of 2-phenoxy-acetamide series as GKA, we have performed molecular
docking to investigate the intermolecular interactions and binding pose of a few compounds with
the active site of GK. Binding mode of compound 1 reveals that both nitrogen, one from the
amino-thiazole and the other from the amide NH forms bidentate hydrogen bonds with Arg63
backbone (Figure 3), one of the most important interactions for the compound to act as GKA
[21]. In addition, di-fluoro phenyl ring of the compound occupies one of the hydrophobic
pockets and makes pi-pi interaction with Tyr214. The compound 1 was further evaluated in
mouse pharmacokinetic (PK) study following both intravenous (i.v.) and oral (p.o.) dosing along
with liver distribution study by oral dosing. It showed an excellent systemic and liver exposure,
however failed to show liver directive PK profile (Table 1).
A
B
Initial Hit
1
Phenoxy-acetamide Core
EC50: 1.85 µM
EC50: 7.0 µM
B
C

ACCEPTED MANUSCRIPT
1
00
80
60
2
00uΜ
1.50
1.35
1.20
1.05
0.90
0.75
0.60
0.45
0.30
0.15
0.00
100uΜ
30uΜ
10uΜ
3uΜ
1uM
3
00 nM
4
0
0
0
Vehicle
2
EC50 = 7 µM
Vehicle
-7
-6
-5
-4
-3
0
10
20
30
40
50
Log [Compound 1]
[
Glucose], mM
Figure 2. A) Hit compound; Evaluation of compound 1 in the in vitro enzyme assay B) Effect on
rhGK activity: The glucose requirement of rhGK is reduced by increasing concentrations of
compound 1. C) Effect on S : The EC of S reduction for rhGK is calculated by plotting the
0
.5
50
0.5
S0.5 obtained in the vehicle control and at each concentration of compound 1 against the log of
the compound 1 concentration (detailed in vitro enzyme activity assay protocol is provided in
experimental section).
Val455
Tyr215
Glu67
Ile211
2.1
2.1
Met210
Arg63
Tyr214
His218
Met235
Figure 3. Docking pose of compound 1 in the active site of GK crystal structure, PDB ID:
NO7. Ligand shown in magenta stick, important residues are shown in green lines, while
4
protein is shown in cyan cartoon representation. All distances are in angstroms.
Table 1. PK parameter and tissue concentration of compound 1 in C57BL/6J mice*
c
Concentration (µM)
Pharmacokinetic parameters
Liver
Systemic
b
b
a
a
a
C
max
AUC(0-24)
CL
Vss
T1/2
(h)
b
%
F
1 h
3 h
1 h 3 h
(µM)
(µM.h)
(mL/min/kg) L/Kg

ACCEPTED MANUSCRIPT
3
26
801.8
3.5
0.4
1
100
111 ± 11 104 ± 5 220 ± 18 85 ± 8
*
a
Pharmacokinetic parameters expressed as geometric mean of n = 3 animals per group; 1 mg/kg
b
c
intravenous dose; 30 mg/kg oral suspension dose; 30 mg/kg oral suspension dose in satellite
group
Impressed with PK properties of compound 1, even though non-liver directed, it was profiled
further in efficacy study. Oral administration of 1 to diet induced obese mice (DIO) followed by
an oral glucose tolerance test (oGTT) demonstrated robust and dose dependent effect on glucose
excursion starting from 10 mg/kg dose (Figure 4a). Next, hypoglycemic potential of compound 1
was investigated in normal C57 mice. Compound 1 showed hypoglycemic potential at oral dose
of 100 mg/kg as it lowered blood glucose below 60 mg/dL (Figure 4c). The observed
hypoglycemia with compound 1 was attributed to its non-liver directed PK profile, wherein
overwhelming systemic exposure indicated possibility of 1 to enter highly perfused pancreas and
resulting in insulin secretion even at sub-normoglycemic level.
a
b
4
00
00
00
00
4
0000
0000
0000
0000
-
9% -22%
-
30%
3
*
3
*
-53%
2
*
-62%
2
*
1
1
0
-
0
60 -30
0
30 60 90 120 150 180 210
TIme, min
Vehicle, 10 ml/kg, PO, n=6
Vehicle, 10 ml/kg, PO, n=6
Compound 1, 3 mg/kg, PO, n=6
Compound 1, 10 mg/kg, PO, n=6
Compound 1, 30 mg/kg, PO, n=6
Compound 1, 100 mg/kg, PO, n=6
Compound 1, 300 mg/kg, PO, n=6
Compound 1, 3 mg/kg, PO, n=6
Compound 1, 10 mg/kg, PO, n=6
Compound 1, 30 mg/kg, PO, n=6
Compound 1, 100 mg/kg, PO, n=6
Compound 1, 300 mg/kg, PO, n=6

ACCEPTED MANUSCRIPT
C
d 2
0000
150
1
20
1
5000
0000
9
0
*
1
*
6
0
3
0
5000
0
0
-
30
0
30 60 90 120 150 180 210
TIme, min
Vehicle, 10 ml/kg, PO, n=6
Vehicle, 10 ml/kg, PO, n=6
Compound 1, 3 mg/kg, PO, n=6
Compound 1, 10 mg/kg, PO, n=6
Compound 1, 30 mg/kg, PO, n=6
Compound 1, 100 mg/kg, PO, n=6
Compound 1, 300 mg/kg, PO, n=6
Compound 1, 3 mg/kg, PO, n=6
Compound 1, 10 mg/kg, PO, n=6
Compound 1, 30 mg/kg, PO, n=6
Compound 1, 100 mg/kg, PO, n=6
Compound 1, 300 mg/kg, PO, n=6
Figure 4. In vivo efficacy data of compound 1 (a) Blood glucose lowering in an acute
oGTT in DIO mice (b) AUC blood glucose during oGTT in DIO mice (c) Hypoglycemic
potential: blood glucose lowering in normoglycemic C57BL/6J mice. The dotted line in
(c) represents the hypoglycemic threshold of 60 mg/dl in mice (d) AUC blood glucose
during assessment of hypoglycemic potential in C57BL/6J mice.
Efficacy along with hypoglycemia, observed with compound 1, insisted us to look for liver-
directed and pancreas sparing GKAs. Hence, SAR efforts were simultaneously focused primarily
on incorporating liver-directed PK properties, and secondly for improving in vitro potency. In
pursuit of alternative structural element to be recognized by OATPs, one idea we investigated
was to replace 5-(chloro)-4-(carboxyacetate)thiazole of 1 by 5-(4-oxy-benzoic acid)thiazole to
obtain 6 (Figure 5), which was synthesized in a similar way as described for compound 1 in
Scheme 1.

ACCEPTED MANUSCRIPT
Modification to impart
Liver-directed PK
Modification to
improve potency
c
c
1
EC : 7.0 µM
50
6
14b
14a
EC : 6.2 µM
EC : 0.5 µM
EC : >20.0 µM
5
0
50
50
Additive SAR
1
4c
EC : 0.28 µM
50
Figure 5. SAR to impart liver-directed property and improve potency
The benzoic acid derivative 6 showed GK activation potency similar to 1 (EC : 6.2 ± 0.10
5
0
µM) and found stable in mouse and human liver microsomes. We were delighted to know the
compound 6 on oral dosing of 30 mg/kg in mice showed approximately 3 and 8 fold higher liver
concentration as compared to plasma at 1 and 3 hour time points respectively [Liver/Plasma:
2
3.3 ± 2.7/8.8 ± 2.7 µM (1 h) and 11.1 ± 2.9/1.4 ± 1.3 µM]. Compound 6 was not profiled
further due to poor potency and low exposures in liver. Hence, another approach was adopted
with goal to improve in vitro potency. Wherein, we sought to replace one of the methyl adjacent
to amide functionality in compound 1 by a phenyl sulfonamide group to obtain 14a that could
help to gain hydrophobic interactions [11, 22] (Figure 5). Unfortunately, 14a lost greater than 3-
fold potency (EC : >20 µM, 67% activation seen at 60 µM) for GK activation compared to 1. In
5
0
an order to improve the potency, we envisioned that removal of the methyl group from the chiral
carbon of 14a to yield 14b may provide scaffold similarity to potent carbon-centered GKAs

ACCEPTED MANUSCRIPT
1
reported in the literature. Indeed, compound 14b showed a significant improvement in the
potency (EC : 0.5 ± 0.06 µM). This observation is further corroborated by molecular modeling
5
0
analysis, where the docking pose and binding orientation of compounds 14a and 14b in active
site of GK differ significantly (Figure 6). In addition, compound 14b (-11.39 kcal/mol) has better
XP GlideScore in comparison to compound 14a (-6.28 kcal/mol). XP GlideScore is an extra
precision scoring function to estimate protein-ligand binding affinities. In addition to coulombic,
van der Waals and unique water desolvation energy terms, protein-ligand structural motifs
leading to enhanced binding affinity are included in this scoring function: (1) hydrophobic
enclosure where groups of lipophilic ligand atoms are enclosed on opposite faces by lipophilic
protein atoms, (2) neutral-neutral single or correlated hydrogen bonds in a hydrophobically
enclosed environment, and (3) five categories of charged-charged hydrogen bonds [23].
Compound 14a binds in a different fashion to that of compound 1, and partially occupies the GK
active site as it is unable to fit properly inside the pocket. Moreover, compound 14a also misses
molecular recognition interaction with the key residues Arg63 and Tyr214, which may be the
possible reason behind the low activity of this compound. Compound 14b mimics all the key
interactions of compound 1 and also makes additional hydrophobic interactions. However,
incorporation of 1-(benzenesulfonyl)piperidine group changes the binding orientation of
phenoxy group in comparison to 1. In detail, di-fluoro phenoxy group of 14b fits well in another
hydrophobic pocket comprises of Met210, Ile211, Tyr214 and Met235. Whereas, 1-
(benzenesulfonyl)piperidine group binds to another pocket, making pi-pi interaction with Tyr214
and hydrophobic interaction with Tyr215. Gaining these additional hydrophobic interactions
might be a possible explanation for improved GK activation potency of 14b over 1. The 2D

ACCEPTED MANUSCRIPT
interaction diagram for the docking pose of the compounds is depicted in Figure S1
(Supplementary Information).
Compound 14a
Val455
Compound 14b
Tyr215
Glu67
Ile211
2.6
1.9
Arg63
Met210
Tyr214
His218
Met235
Figure 6. Docking pose compound 14a (salmon color) and 14b (magenta color) in the active site
of GK crystal structure, PDB ID: 4NO7. Ligands are shown as sticks; important residues are
shown in green lines, while protein is shown in cyan cartoon representation. All distances are in
angstroms.
This encouraging result prompted us to embrace additive SAR strategy to prepare a series of
compounds. Towards that end, first we attempted synthesis of few sulfonamide derivatives (14c-
e) followed by sulfone analogs 25a and 25b as exemplified in Table 2. Subsequently, compounds
were evaluated for in vitro potency and metabolic stability. Interestingly, all sulfonamide and
sulfone analogs (Table 2) were found ~20 fold more potent GK activator than early lead
compound 1. Compound 14c showed poor metabolic stability in mouse and human liver
microsomes. Attempts of lowering Log D value by approximately 0.5 unit for remaining analogs
compare to 14c (CLogD = 2.54) proved fruitful to improve the metabolic stability [24].
Subsequently all compounds were evaluated in a functional assay for glycogen synthesis in rat
primary hepatocytes. In hepatocyte GK is rate limiting enzyme controlling glucose utilization; its
activation promotes glycogen synthesis [25]. Hence evaluation of potency in this functional
assay will give compounds ability to activate endogenous GK. All sulfonamide derivatives

ACCEPTED MANUSCRIPT
showed approximately 3- fold increase in glycogen synthesis, compared to vehicle control. We
were gratified to observe that sulfone derivative 25a demonstrated 11.36 fold increases in
glycogen synthesis and appeared 3 fold better in functional activity, compared to all other
analogs.
Table 2. Sulfonamide and sulfone derivatives: In vitro potency, metabolic stability and
glycogen synthesis data
a
b
Compound
rhGK Activation
EC of S
MR
(nmol/min/mg) glycogen synthesis
at 10
Fold increase in c
1
R
50
0.5
reduction (
µM)
µM
0
0
.15 (MLM)
.13 (HLM)
1
4c
0.28 ± 0.02
0.33 ± 0.02
3.40 ± 0.29
3.00 ± 0.36
0
0
.08 (MLM)
.03 (HLM)
1
4d
4e
5a
5b
0
0
0
0
.10 (MLM)
.06 (HLM)
.01 (MLM)
.05 (HLM)
1
0.24 ± 0.05
0.19 ± 0.05
3.76 ± 0.38
11.36 ± 1.7
2
0
0
.04 (MLM)
.02 (HLM)
2
0.31 ± 0.05
3.04 ± 0.27
a
b
GK activation potency reported as average (n
≥ 3). In vitro microsomal stability expressed as MR
(
metabolic rate) detailed experimental reported in supporting information, MR < 0.1 considered
c
metabolically stable. MLM: Mouse liver microsomes, HLM: Human liver microsomes. Detailed
glycogen synthesis assay protocol is provided in experimental section.
Next, we explored structural variation of 25a by replacing 4-oxy-benzoic acid at position-5
on thiazole ring with various substituents bearing acid functionality (25c-h) as summarized in
Table 3.
Table 3. Sulfone derivatives: In vitro potency, metabolic stability and glycogen synthesis data

ACCEPTED MANUSCRIPT
4
a
b
Compound
R
rhGK Activation
MR
EC of S reduction ( µM) (nmol/min/mg)
5
0
0.5
<
<
0.04 (MLM)
0.04 (HLM)
2
5c
2.73 ± 0.15
<
<
0.04 (MLM)
0.04 (HLM)
2
5d
0.40 ± 0.03
0
.04 (MLM)
2
5e
5f
0.45 ± 0.12
>10
<
0.04 (HLM)
0.09 (MLM)
2
0
.04 (HLM)
<
<
0.04 (MLM)
0.04 (HLM)
2
5g
>10
<
<
0.04 (MLM)
0.04 (HLM)
2
5h
2.3 ± 0.40
a
GK
activation potency reported as average (n
b
≥
3). In vitro microsomal stability expressed as MR (metabolic
rate) detailed experimental reported in supporting information, MR < 0.1 considered metabolically
stable. MLM: Mouse liver microsomes, HLM: Human liver microsomes.
We assumed that the modification at this position might have an impact on PK properties and
to further understand structural activity trends in this series. Compounds 25c-h were synthesized
via similar route used for compound 25a using appropriately susbstituted 2-aminothiazoles
(Scheme 3). Disappointingly, 25c, 25f-h resulted in significant loss of potency. Though,
compounds wherein homologation of carboxylate moiety (25a to 25d, EC : 0.40 ± 0.03µM) and
50
replacement of oxygen by methylene (25a to 25e, EC : 0.45 ± 0.12µM)) were reasonably
5
0

ACCEPTED MANUSCRIPT
tolerated but found to be ~2-fold less potent compared to 25a, rendering 25a as a compound of
choice for further profiling.
Since 25a is a mixture of enantiomer, we used preparative HPLC with a chiral column to
separate racemate into (+) and (‒) enantiomers 26 and 27 respectively. The absolute
stereochemistry of 26 or 27 have not been determined; still a tentative assignment is given as
(
R)-isomer for 26 based on in-house single crystal X-ray data of analogous compound form the
same series. Enantiomers, 26 and 27 were evaluated in enzyme assay for their GK activation
potency and results are captured in Table 4. When comparing a pair of enantiomers, 26 (EC50
.034 ± 0.01 µM) was found to be approximately 10-fold more potent than 27 (EC50: 0.69 ± 0.28
µM). Docking pose for the more active (+)-(R)-isomer compound 26 is illustrated in Figure S2
Supplementary information). Upon subsequent evaluation in a functional assay for glycogen
:
0
(
synthesis in rat primary hepatocytes, compound 26 displayed 12.2-fold increase in glycogen
synthesis compared to vehicle control. Both enantiomers have shown good metabolic stability in
mouse and human liver microsomes.
Table 4. Enantiomers of 25a: In vitro potency, metabolic stability and glycogen deposition data
rhGK
Activation
a
b
Fold increase in c
glycogen synthesis
at 10 µM
MR
(nmol/min/mg)
Compound
EC50 of S0.5
reduction ( µM)
0
.01 (MLM)
.08 (HLM)
0.09 (MLM)
2
6, (+)-(R)-enantiomer
0.034 ± 0.01
0.69 ± 0.28
12.2 ± 0.6
ND
0
2
7 , (-)-(S)-enantiomer

ACCEPTED MANUSCRIPT
0.03 (HLM)
a
b
ND = Not determined. GK activation potency reported as average (n
≥ 3). In vitro microsomal stability
expressed as MR (metabolic rate) detailed experimental reported in supporting information, MR < 0.1
considered metabolically stable. MLM: Mouse liver microsomes, HLM: Human liver microsomes.
c
Detailed glycogen synthesis assay protocol is provided in experimental section
Based on GK activation potency in enzyme and functional assay as well as warranted by
metabolic stability, compound 26 was progressed into mouse PK study. It showed an excellent
PK profile and exhibited 35- and 68- fold higher liver concentrations at 1 and 3 h time point,
respectively, compared with plasma (Table 5).
Table 5: PK parameter and tissue concentration of compound 26 in C57 BL/6J mice
c
Concentration (µM)
b
b
a
a
a
b
C
(
max AUC(0-t)
CL
Vss T1/2
F
%
Liver
Systemic
3 h
215 ± 57 184 ± 34 6 ± 2.9 2.7 ± 0.1
µM) (µM.h) (ml/min/kg) L/Kg (h)
1
h
3h
1 h
1
3.3
22
32.6
1.67 2.16
86
*
a
Pharmacokinetic parameters expressed as geometric mean of n = 3 animals per group; 3mg/kg
intravenous dose; 30 mg/kg oral suspension dose; 30 mg/kg oral suspension dose in satellite group.
b
c
To characterize the active transport mechanism responsible for the selective liver distribution;
we performed the in vitro hepatocyte uptake study [26] for compounds 1 and 26 in fresh mouse
primary hepatocytes. On incubating test compounds in suspended mouse hepatocytes at 37 °C
for 5 minutes; approximately 30%-40% of 26 was taken up into the hepatocytes by active uptake
in the concentration range of 1-100 µM. Approximately <5% of compound 1 was taken into the
hepatocytes with respect to initial concentration. Conducting similar experiments at 4°C for 5
minutes; <5% of compound 26 was transported into the hepatocytes and approximately 5% of 1
was transported in to the hepatocytes. In addition active uptake of compound 26 was inhibited
(90% inhibition) in presence of rifamycin SV (a pan inhibitor of OATP transport pathway) and

ACCEPTED MANUSCRIPT
no inhibition was seen with compound 1. Above results suggest that active uptake into the
hepatocytes is the predominant pathway for 26.
Compound 26 was emerged as a viable lead due to alignment of desired in vitro potency,
metabolic stability and liver directed pancreas-sparing PK properties. Based on its favorable
profile, compound 26 was tested in acute in vivo efficacy study. In an oGTT in DIO mice, 26
demonstrated 21% and 34% significant AUCglucose reduction at oral dose of 30 and 100 mg/kg,
respectively, as shown in Figure 7.
a
b
4
00
00
00
00
3
5000
28000
1000
4000
000
-
7% -21%
-34%
3
*
*
2
2
1
1
7
0
-
0
60 -30
0
30 60 90 120 150 180 210
Time, min
Vehicle, 10 ml/kg, PO, n=6
Vehicle, 10 ml/kg, PO, n=6
Compound 26, 3 mg/kg, PO, n=6
Compound 26, 30 mg/kg, PO, n=6
Compound 26, 100 mg/kg, PO, n=6
Compound 26, 3 mg/kg, PO, n=6
Compound 26, 30 mg/kg, PO, n=6
Compound 26, 100 mg/kg, PO, n=6
Figure 7. In vivo efficacy of compound 26 in an acute oGTT in DIO mice (a)
Blood glucose lowering and (b) Reduction in AUCglucose
To access the hypoglycemic potential of 26, compound was further evaluated in
normoglycemic C57 mice. Compound 26 decreased blood glucose significantly but no
hypoglycemia (i.e. plasma glucose level < 60 mg/dl) was observed up to an oral dose of 300
mg/kg as shown in Figure 8. A therapeutic window for a lead compound 26 was observed greater

ACCEPTED MANUSCRIPT
a 1
80
50
20
b
2
5000
0000
5000
0000
000
-
16%
1
-16%
2
1
*
*
9
0
1
6
0
1
3
0
5
0
-
30
0
30 60 90 120 150 180 210
Time, min
0
Vehicle, 10 ml/kg, PO, n=6
Vehicle, 10 ml/kg, PO, n=6
Compound 26, 30 mg/kg, PO, n=6
Compound 26, 100 mg/kg, PO, n=6
Compound 26, 300 mg/kg, PO, n=6
Compound 26, 30 mg/kg, PO, n=6
Compound 26, 100 mg/kg, PO, n=6
Compound 26, 300 mg/kg, PO, n=6
Figure 8. Hypoglycemic potential of compound 26 in normoglycemic C57
mice (a) Blood glucose lowering and (b) Reduction in AUCglucose. The dotted
line in (a) represents the hypoglycemic threshold of 60 mg/dl in mice
than 10-fold dose range i.e. efficacy at 30 mg/kg and no hypoglycemia up to 300 mg/kg dose.
Indeed, with improved in vitro potency and liver directed tissue distribution, compound 26
demonstrated in vivo efficacy without hypoglycemia. Thus, compound 26 validated our
hypothesis that a liver-directed and pancreas sparing GKA offers beneficial effect on glucose
homeostasis while minimizing the risk of hypoglycemia. Further optimization efforts of this
series will be reported in due course.
Conclusions
In summary, we identified a novel class of 2-phenoxy-acetamide derivatives as glucokinase
activators. Subsequent SAR studies of this class of compounds were explored with the aim to
incorporate liver-selective tissue distribution. Inclusion of carboxylic acid moiety enabled an
active hepatocyte uptake of compound 26 by liver specific organic anion transporting
polypeptides (OATPs), as demonstrated by liver directed PK profile. Compound 26 showed a
dose dependent improvement in glucose excursion in oGTT in a DIO mouse model of T2D. The
predominant liver-directed tissue distribution of compound 26 ensured a good safety window

ACCEPTED MANUSCRIPT
with respect to hypoglycemia over a 10-fold dose range. Together, these results demonstrated
our hypothesis that a liver-directed GKA has beneficial effect on glucose homeostasis with
reduced risk of hypoglycemia.

ACCEPTED MANUSCRIPT
Experimental Section
Biological Assays
Definition of EC S reduction
5
0
0.5
The enzymatic properties of glucokinase can be described in terms of its velocity (i.e. its rate of
converting glucose to glucose-6-phosphate (G6P)) and its S0.5 for glucose (i.e. the apparent
concentration of glucose at which GK converts glucose to G6P at half of its maximal enzyme
velocity). The S0.5 of glucose, in an in vitro assay using recombinant human GK, is
approximately 8 mM. GKAs induce increased conversion of glucose to G6P by GK by either
decreasing the S0.5 of GK for glucose, increasing its Vmax, or by combination of both. The ∆S0.5
was calculated by plotting the S0.5 obtained in the vehicle control and at each concentration of
GKAs against the log of the GKAs concentration. As the concentration of GKA is increased, the
S0.5 of GK for glucose decreases, while the Vmax increases.
In vitro GK enzyme assay
Recombinant human GK activity was measured by a coupled enzymatic assay. Briefly, glucose
is converted by GK to glucose-6-phosphate, which in turn is converted to 6-phosphogluconate by
glucose-6-phosphate dehydrogenase (G6PD). This second enzymatic step requires NAD, which
gets converted to NADH by the action of G6PD. The progress of the enzymatic reaction can
therefore be monitored by measuring the accumulation of NADH at 340 nm. Recombinant
human GK was assayed in a 96-well plate with different concentrations of GKA ranging from
1
00 nM to 60 µM, in buffer containing 25 mM Hepes, pH 7.2, 100 mM KCl, 10 mM MgCl , 5
2
mM ATP, 0.5 mM NAD, 2 mM DTT and 1 U/mL G6PD. The enzymatic reaction was initiated
by the addition of glucose at final concentrations ranging from 0.5 mM to 50 mM. The progress
of the enzymatic reaction was monitored by measuring the continuous accumulation of NADH at

ACCEPTED MANUSCRIPT
3
40 nm in a Tecan Safire plate reader, and converting this accumulation into rate of increase in
absorbance per hour. Kinetic data is plotted using GraphPad Prism 5.0. The equation for the non-
linear fit to the data is a modified version of the Michaelis-Menten equation, wherein a Hill
coefficient of 1.7 is introduced to reflect the non-Michaelis-Menten kinetic properties of GK.
Molecular Modeling
Enzyme and Ligand Structure Preparation
The X-ray crystallographic structure of human GK in complex with a nanomolar activator (PDB
ID: 4NO7, resolution 1.7 Å) was considered for the purpose of molecular docking [27]. Among
the several other crystal structures in the Protein Data Bank (PDB), this structure was
particularly selected due to its high resolution, better overall B-factor and structural similarity of
co-crystallized ligand with the compounds to be docked. GK enzyme structure was prepared by
using the Protein Preparation Wizard tool implemented in Maestro interface (Maestro 10.0,
Schrodinger, LLC, New York 2014) using the default parameters and standard protocol. Ligands
were prepared using LigPrep module of Maestro interface by generating all possible low energy
ionization and tautomeric states within the pH range 7.0 ± 2.0. For the minimization purpose in
both the preparation wizards OPLS_2005 force field was implemented.
GLIDE Docking
An interaction grid was generated for an enzyme by using the receptor grid generation wizard of
GLIDE 6.5 by considering bound ligand ((2R)-2-[3-chloro-4-(methylsulfonyl)phenyl]- 3-[(1R)-
3
-oxocyclopentyl]-N-(pyrazin-2-yl)propanamide) as the reference structure for defining the
binding site. The centroid (XYZ coordinates: -9.07, 3.19, 20.29) of workspace ligand was used to
generate the grid box. The set of prepared ligands were docked into the GK ligand binding site as
defined above using Grid box by using GLIDE 6.5 at XP (Extra Precision) mode. The docking

ACCEPTED MANUSCRIPT
and scoring algorithms of GLIDE have been fully described elsewhere [28]. All docking
calculations were performed using the OPLS_2005, while keeping all other parameters at their
default value. The best docking pose was selected manually for each molecule on the basis of
intermolecular interactions with the key residues and XP Glide docking score. Docking analysis
showed that (R)- isomer of the synthesized compounds have better intermolecular interactions,
XP GlideScore as well as Emodel score as compared to (S)- isomer. Therefore, in all the docking
pose representation we have depicted (R)- isomer of the compound, unless otherwise mentioned
explicitly.
Animal care, tissue isolation and in vivo protocols:
Animal care, hepatocyte isolation, and all in vivo experimental procedures (please refer
supporting information) followed were according to approved Institutional Animal Ethics
Committee (IAEC) protocols, based on Committee for the Purpose of Control and Supervision of
Experiments on Animals (CPCSEA) guidelines.
Glycogen synthesis in rat primary hepatocytes
Rat primary hepatocytes (200,000 cells/well) were plated in minimal essential medium (MEM)
containing 10% fetal bovine serum (FBS), 100 nM insulin and 10 nM dexamethasone (Dex).
After 4 hours, the medium was changed to MEM containing 10% FBS and 10 nM Dex without
insulin and maintained overnight. Hepatocytes were starved in MEM without FBS for 2 h. The
1
4
cells were then treated with medium containing 10.5 mM glucose and 2 µCi/mL of [U C]-D-
glucose with 1 nM insulin in the presence or absence of GKM-001 at various concentrations for
another 3 hours. The hepatocytes were washed twice with 150 mM NaCl and lysed with 0.1 N
NaOH. Lysates were used for protein estimation using a Bradford protein estimation kit (Thermo
Scientific). Cellular proteins were then precipitated with 8% (w/v) trichloroacetic acid containing

ACCEPTED MANUSCRIPT
1
mg/well glycogen carrier (Sigma), followed by precipitation of glycogen with 66% (v/v)
ethanol. The precipitates were dried overnight at 37 ºC, dissolved in water, and transferred to a
6-well plate. Optiphase supermix scintillation fluid (PerkinElmer) was added to the wells, and
9
the plate was read in a Microbeta TriLux luminescence counter (PerkinElmer).
General Methods for Chemistry:
All reagents, starting materials, and solvents (including dry solvents) were obtained from
commercial suppliers and used as such without further purification. Reactions were carried out in
oven-dried glassware under a positive pressure of argon/nitrogen unless otherwise mentioned.
Column chromatography was performed on silica gel (Rankem, 60-120 mesh). Deuterated
solvents (Cambridge Isotope Laboratories) for NMR spectroscopic analysis were used as
received. All NMR spectra were recorded on Varian 400 MHz spectrometer. Coupling constants
are measured in Hertz. All chemical shifts are quoted in ppm, relative to tetramethylsilane, using
the residual solvent peak as a reference standard. Melting points were determined using Büchi B-
5
45 digital melting point apparatus and are uncorrected. HRMS (ESI+ ionization mode) were
recorded on a JEOL-AccuTOF JMS-T100LC Mass spectrometer having a DART (Direct
Analysis in Real Time) source (Sophisticated Analytical Instrument Facility, Central Drug
Research Institute, Luknow India). LC-Mass spectra were recorded on an Agilent MSD/VL with
ESI ionization.
Purity of all final analogs for biological testing were confirmed to be >95% as determined by
HPLC analysis. HPLC analysis was carried on Agilent 1200 series instrument equipped with
Diode Array Detector, with chromatography performed on Agilent Eclipse XDB C-18, 5µM,
2
50mm X 4.6mm column with mobile phase 0.05% formic acid in water and acetonitrile with a
flow rate of 1.5 ml/min, detected by UV at 220 nm and retention time (t ) expressed in minutes.
R

ACCEPTED MANUSCRIPT
Ethyl 2-(2,4-difluorophenoxy)-2-methyl-propanoate (3): To a stirred solution of 2,4-
difluorophenol 2 (5 g, 38.43 mmol) in DMSO (50 mL) was added Cs CO (31.3 g, 96.08 mmol)
2
3
and sodium iodide (2.88 g, 19.21 mmol) under nitrogen atmosphere at room temperature. Then
ethyl 2-bromoisobutyrate (11.28 mL, 76.86 mmol) was added drop wise at the same temperature
and the reaction mixture was stirred for 16 h. The reaction mixture was filtered through celite to
afford a clear filtrate. Water was added to the filtrate and the solution was extracted with ethyl
acetate, the combined organic solution was washed with water, brine and dried over sodium
sulfate. The organic solvent was evaporated in vacuo. The residue was purified by silica gel
column chromatography (5 % ethyl acetate in hexanes) to afford 3 as a solid (9.0 g, 96%).
1
H NMR (400 MHz, CDCl ): δ 1.30 (t, J = 7.1 Hz, 3H), 1.55 (s, 6H), 4.25 (q, J = 7.1 Hz, 2H),
3
6
2
3
.70-6.78 (m, 1H) 6.83 (ddd, J = 3.05, 8.4 and 11.0 Hz, 1H), 7.02 (dt, J = 5.6 and 9.3 Hz, 1H).
-(2,4-difluorophenoxy)-2-methyl-propanoic acid (4): To a stirred solution of ester 3 (9.0 g,
6.88 mmol) in THF/MeOH/ H O (60 mL, 3:1:1) was added lithium hydroxide monohydrate
2
(7.74 g, 184.4 mmol) and the reaction mixture was stirred for 16 h at room temperature. The
reaction mixture was concentrated to remove organic solvents. The aqueous layer was extracted
with diethyl ether and then acidified to pH 2.0 with 1N HCl. The product was extracted with
ethyl acetate. The organic layer was washed with water, brine, dried over sodium sulfate filtered,
and evaporated in vacuo to afford compound 4 as a solid (6.15 g, 77 %).
1
H NMR (400 MHz, CDCl ): 1.57 (s, 6H), 6.78-6.83 (m, 1H), 6.86-6.91 (m, 1H), 7.11 (dt, J =
3
5
.6 and 9.1 Hz, 1H); LC-MS (m/z): 217.0 [M+1]
Ethyl 2-[5-chloro-2-[[2-(2,4-difluorophenoxy)-2-methyl- propanoyl]amino] thiazol-4-
yl]acetate (5): To a stirred solution of 4 (4.0 g, 18.5 mmol) in DMF was added EDCI (5.3 g,
2
7.75 mmol), HOBt (3.74 g, 27.75 mmol), NMM (2.80 g, 27.75 mmol) and ethyl 2-(2-amino-5-

ACCEPTED MANUSCRIPT
chloro-thiazol-4-yl)acetate (6.12 g, 27.75 mmol) at 0 °C. Then the reaction mixture was stirred
for 16 h at room temperature. Water was added to the reaction mixture and extracted with ethyl
acetate. Organic layer was washed with 1N HCl, water, brine, dried over sodium sulfate and
filtered. The organic solvent was evaporated in vacuo to yield a crude product, which was
purified by silica gel column chromatography ( 10 % ethyl acetate in hexanes) to obtain pure
1
compound 5 as a solid (5.0 g, 65%). H NMR (400 MHz, CDCl ): δ 1.28 (t, J = 7.08 Hz, 3H),
3
1
7
2
.54 (s, 6H), 3.71 (s, 2H), 4.20 (q, J = 7.08 Hz, 2H), 6.80-6.87 (m, 1H), 6.88-6.96 (m, 1H), 7.00-
3
5
.08 (m, 1H), 10.01 (bs, 1H); LC-MS (m/z): 419.0 [M( Cl) +1]
-[5-chloro-2-[[2-(2,4-difluorophenoxy)-2-methyl-propanoyl]amino]thiazol-4-yl]acetic acid
(1): To a stirred solution of 5 (5.0 g, 11.95 mmol) in THF (24 mL) and methanol (8 mL) was
added a solution of lithium hydroxide monohydrate (3.0 g, 71.62 mmol, in 8 mL water). The
resulting solution was stirred at room temperature for 12 h. After evaporation of organic
solvents, the residue was extracted with diethyl ether and the aqueous layer was acidified to pH
2
.0 with 1N HCl. The aqueous layer was extracted with ethyl acetate, washed with water, brine,
dried over sodium sulfate, filtered and evaporated in vacuo to obtain crude product. The crude
product was purified by silica gel column chromatography (2% methanol in DCM) to afford 1 as
1
a white solid (3.1 g, 67%). mp 154-155 °C; H NMR (400 MHz, DMSO-d ): δ 1.52 (s, 6H), 3.61
6
(s, 2H), 6.96-7.15 (m, 2H), 7.33 (ddd, J = 3.0, 9.0 and 11.7 Hz, 1H) 12.63 (s, 2H).
1
3
C NMR (100 MHz, DMSO-d6)
δ 24.03 (2C), 34.10, 81.72, 105.09 (dd, J = 23.6 and 26.7 Hz),
1
11.21 (dd, J = 3.1 and 22.4 Hz), 115.31, 124.14 (d, J = 10 Hz), 138.18 (dd, J = 3.5 and 11.2
Hz), 141.12, 154.04, 154.67 (dd, J = 13.2 and 247 Hz), 157.53 (dd, J = 10.8 and 240.7 Hz),
3
5
37
1
70.44, 172.60; LC-MS (m/z): 390.8 [M ( Cl)+1], 392.9 [M ( Cl)+1]; HPLC t : 9.016 min.,
R