Discovery, synthesis and biological evaluation of novel glucokinase activators

Article abstract of DOI:10.1016/j.bmcl.2005.01.087

The identification, synthesis and SAR of a novel series of glucokinase activators is described. The interplay between lipophilicity, potency and physical properties is discussed, and compound 22 highlighted as having a suitable balance. In vivo pharmacoki

Full text of DOI:10.1016/j.bmcl.2005.01.087

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2103–2106
Discovery, synthesis and biological evaluation of
novel glucokinase activators
Darren McKerrecher,* Joanne V. Allen, Suzanne S. Bowker,
Scott Boyd, Peter W. R. Caulkett, Gordon S. Currie, Christopher D. Davies,
Mark L. Fenwick, Harold Gaskin, Emma Grange, Rod B. Hargreaves,
Barry R. Hayter, Roger James, Keith M. Johnson, Craig Johnstone,
Clifford D. Jones, Sarah Lackie, John W. Rayner and Rolf P. Walker
Cardiovascular and Gastrointestinal Research Area, AstraZeneca UK, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK
Received 12 November 2004; revised 4 January 2005; accepted 12 January 2005
Available online 22 March 2005
Abstract—The identification, synthesis and SAR of a novel series of glucokinase activators is described. The interplay between lipo-
philicity, potency and physical properties is discussed, and compound 22 highlighted as having a suitable balance. In vivo pharma-
cokinetic and acute efficacy studies on this compound are also presented.
Ó 2005 Elsevier Ltd. All rights reserved.
Type 2 Diabetes is a complex disease caused by defects
in both the action and secretion of insulin, leading to
fasting hyperglycaemia and vascular complications. It
has become the fourth leading cause of death and affects
150 million people worldwide, with its prevalence ex-
pected to double by the year 2025.1 Current therapies
do not achieve adequate glycaemic control,2 hence there
is a need for new, effective pharmacological agents.
Whilst small molecule kinase inhibitors are well prece-
dented,6 kinase activation by small molecules is rela-
tively unknown. However, a recent report by Grimsby
et al. disclosed the identification and in vivo evaluation
of the GK activator RO-28-1675 (1).7,8 The Hoffmann–
La-Roche group have also described the co-crystallisa-
tion of GK with a bound analogue.9 Banyu, in collabo-
ration with Merck, have published10 a similar crystal
structure of the bound GK activator 2 (Fig. 1).
The elevated blood glucose levels that characterise Type
2 Diabetes result from a combination of defects in hepa-
tic glucose balance and the failure of pancreatic b-cells
to secrete sufficient insulin to overcome insulin resis-
tance. Glucokinase (GK) is the rate-limiting enzyme
for glucose utilisation3 both in the liver, where its func-
tion is regulated by a 68 kDa regulatory protein,4 and in
pancreatic b-cells. Phosphorylation of glucose by GK in
the liver promotes glycogen synthesis, whilst in the b-cell
this results in glucose-sensitive insulin release.5 Activa-
tion of GK is therefore expected to improve glycaemic
control by modulating hepatic glucose balance and
decreasing the threshold for insulin secretion.
Several recent reports of GK activators have also ap-
peared in the patent literature.11 We have recently de-
scribed detailed in vitro studies with our own series of
GK activators,12 and herein describe our approach to
the identification and SAR of this novel series.13 The
importance of physical properties, and the in vivo profile
of a lead candidate, are reported for the first time.
O
N
H
S
F
N
N
S
S
N
H
O
S
N
O
N
NH2
O
1 RO-28-1675
2
Keywords: Diabetes; Glucokinase; Kinase activator.
*
Figure 1. Glucokinase activators reported by Hoffmann–La-Roche (1)
and Banyu (2).
Corresponding author. Tel.: +44 1625 518616; fax: +44 1625
516667; e-mail: darren.mckerrecher@astrazeneca.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.01.087

2104
D. McKerrecher et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2103–2106
CO2H
Further SAR studies demonstrated that although the 2-
monobenzyl analogue 10 was less active, a single substi-
tuent at R3 (11) gave a significant increase in potency.
Given the disubstituted nature of the original lead, we
investigated the effect of adding additional substitution
within this series. Whilst 2,3-dibenzyloxy substitution
(12) led to a drop in potency, 3,5-dialkoxy substitution
(13) gave a modest improvement. 3,5-Dibenzyloxy (14)
and the corresponding ortho-functionalised analogues
(15) also showed increased potency, with the latter dem-
onstrating an EC50 < 100 nM for the first time (Table 2).
S
N
O
3
Figure 2. Structure of HTS lead compound.
A high throughput screen (HTS) of the company com-
pound collection was carried out, with hits selected
based upon their activation of recombinant human pan-
creatic enzyme.14 Of the hits identified, compound 3 was
of particular interest, mediating GK activation with an
EC50 of 3.2 lM14 (Fig. 2).
Despite promising potency, compounds such as 14 and
15 suffered from poor aqueous solubility and very high
plasma protein binding (>99.9% bound). We therefore
sought to decrease lipophilicity in order to achieve im-
proved physical properties.
We were aware that compounds such as 3 were potential
Michael acceptors, and therefore decided to investigate
alternatives to the unsaturated linker group. Preliminary
studies indicated that whilst saturated alkyl (4), ether
(5), ÔreversedÕ ether (6) and ÔreversedÕ amide (7, 8) linkers
were not tolerated, amide linkers (9) retained activity.
As a result, compounds such as 9 represented a more
attractive startpoint based on their anticipated chemical
stability and synthetic accessibility (Table 1).
Our initial attempts focused on replacing the benzyl
groups with more polar substituents. Whilst such
changes allowed us to demonstrate the expected
improvements in both solubility and unbound drug,
the data for the tetrahydropyran analogue 16 illustrates
that such compounds were significantly less potent
(Table 3).
We therefore decided to undertake an iterative
approach, seeking to build on the balance of moderate
potency and better physical properties exhibited by com-
pounds such as the isopropyl derivative 13. These stud-
ies demonstrated that cyclopentylmethyloxy (17) and
cyclopentylethyloxy (18) substituents were potent, but
failed to show an improvement in physical properties.
Polar heterocyclic terminal groups such as 3- or 4-
pyridyl (19, 20) showed a dramatic reduction in potency
despite exhibiting good solubility and unbound levels. In
contrast, phenyl (21) and, in particular, less polar het-
erocylic terminal groups such as thiophene (22) demon-
strated excellent potency whilst retaining accept-
able physical properties. Attempts to achieve further
improvements by chain length modification (23, 24) or
aromatic ring substitution (25) proved unfruitful due
to the deleterious effects on potency and physical prop-
erties, respectively (Table 3).
Table 1. Identification of lead series and investigation of linker
SAR
CO2H
R5
X
Y
N
R2
R3
R3
Compd R2
R5
X–Y
EC50
(lM)14
3
4
5
6
7
8
9
–OCH2Ph H
–SCH3
–OCH2Ph –OCH2Ph –CH2CH2– >10
–CH@CH–
3.20
H
H
H
H
–OCH3
–OCH3
–OCH2–
>10
>10
>10
>10
–OCH2Ph –OCH2Ph –CH2O–
–OCH2Ph –OCH2Ph –NHCO–
–OCH2Ph H
–OCH2Ph H
–SCH3
–SCH3
–NHCO–
–CONH–
Benzamides 13 and 16–25 were synthesised in five steps
from methyl 3,5-dihydroxybenzoate according to the
4.22
Table 2. Identification of phenyl ring substitution pattern
CO2H
O
R5
N
N
H
R2
R3
Compd
R2
R3
H
R5
EC50 (lM)14
10
11
12
13
14
15
–OCH2Ph
H
–OCH2–o-Cl–Ph
H
>10
0.91
>10
0.57
0.40
0.09
–OCH2–o-Cl–Ph
–OCH2–o-Cl–Ph
–OCH(CH3)2
–OCH2Ph
H
H
H
H
H
–OCH2CH(CH3)2
–OCH2Ph
–OCH2–o-F–Ph
–OCH2–o-F–Ph

D. McKerrecher et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2103–2106
2105
Table 3. Balancing potency and physical properties
CO2H
7.5
7
6.5
6
(a)
(b)
O
O
7
N
N
H
5.5
5
OR
6.5
6
Compd
R
Solubilitya % Freeb EC50
(lM)
(rat)
(lM)14
13
16
17
18
19
20
21
22
23
24
25
–CH2CH(CH3)2
–CH2CH2–4-THP
–CH2–cPent
198
2380
44
0.53
7.38
0.15
0.11
5.11
6.35
0.26
0.26
0.20
0.13
0.14
0.57
1.33
0.65
0.17
1.26
1.78
0.13
0.09
0.29
0.42
0.10
4.5
4
–CH2CH2–cPent
–CH2CH2–3-pyridyl
–CH2CH2–4-pyridyl
–CH2CH2Ph
3
5.5
3.5
3080
NDc
29
3
3.5
4
4.5
5
5.5
3
3.5
4
4.5
5
5.5
clogP
clogP
–CH2CH2–3-thiophene
–CH2Ph
48
15
Figure 3. Correlation of lipophilicity with (a) potency, and (b) plasma
protein binding.
–CH2CH2CH2Ph
–CH2–o-F–Ph
15
9
a Solubility measured in 0.1 M phosphate buffer at pH 7.4.
b Protein binding measured by equilibrium dialysis.
c Not determined.
represented by the first apparent binding constant
K1app, Fig. 3b), the challenge faced to simultaneously
optimise both potency and free drug levels is evident.
Compound 22 was considered to have a suitable balance
of potency and physical properties, and was selected for
further evaluation. In vivo pharmacokinetic studies were
carried out in female Han-Wistar rats (Charles River
Laboratories), which demonstrated that the compound
had low clearance and good bioavailability (Table 4).
route highlighted in Scheme 1. A key feature of the syn-
thesis is the use of sequential alkylation and/or Mitsun-
obu reactions to install the ether side chains. Benzoate
cleavage, followed by amide coupling (via the acid
chloride due to the relatively low nucleophilicity of
methyl 6-aminonicotinate17) and hydrolysis of the
methyl ester yielded the final compounds. The chemistry
described in Scheme 1 proved to be a quick and robust
route to a diverse set of analogues, and as such consti-
tuted an attractive strategy for lead optimisation.
The acute in vivo efficacy of compound 22 was also evalu-
ated in an Oral Glucose Tolerance Test (OGTT) in fe-
male Zucker rats fed with a high fat diet to render
them insulin resistant.16 Administration of a 30 mg/kg
oral dose of compound 22 delivered a statistically signi-
ficant reduction in plasma glucose levels following the
oral glucose challenge (Fig. 4a), resulting in a 14%
reduction in the area under the glucose curve16 (Fig. 4b).
Obtaining a suitable balance between potency and phys-
ical properties is a common problem in medicinal chem-
istry,15 especially when optimising a lipophilic lead. This
is emphasised by Figure 3a, which demonstrates the
relationship observed for this series between the enzyme
pEC50 and clogP. Since we also observed a good correl-
ation between clogP and plasma protein binding (as
In summary, we have described the discovery, synthesis
and detailed SAR studies of a novel series of glucoki-
nase activators, and highlighted the interdependence of
CO2Me
HO
CO2Me
O
CO2Me
O
(i)
(ii)
OH
OR
OH
CO2Me
O
(iii)
(iv)
CO2H
O
O
N
N
H
OR
OR
CO2H
O
O
(v)
N
N
H
OR
Scheme 1. General synthetic route to benzamides 13 and 16–25. Reagents and conditions: (i) (CH3)2 CHBr, K2CO3, DMF (35%); (ii) Ph3P, DEAD,
ROH, THF or RBr, K2CO3, DMF (60–80%); (iii) 1 M NaOH, THF/H2O (90%); (iv) (COCl)2, CH2Cl2, DMF (cat.) followed by methyl 6-
aminonicotinate, THF, pyridine (70%); (v) 1 M NaOH, THF/H2O (90%).

2106
D. McKerrecher et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2103–2106
Table 4. Pharmacokinetic parameters of compound 22a
4. Van Schaftingen, E.; Veigha da Cunha, M. In Frontiers in
Diabetes; Matschinsky, F. M., Magnuson, M. A., Eds.;
Karger: Basel, 2004; Vol. 16, pp 193–207.
CO2H
O
5. Matschinsky, F. M. Diabetes 2002, 51, S394.
6. Cohen, P. Nature Rev. Drug. Discov. 2002, 1, 309.
7. Grimsby, J.; Sarabu, R.; Corbett, W. L.; Haynes, N. E.;
Bizzarro, F. T.; Coffey, J. W.; Guertin, K. R.; Hilliard, D.
W.; Kester, R. F.; Mahaney, P. E.; Marcus, L.; Qi, L.;
Spence, C. L.; Tengi, J.; Magnuson, M. A.; Chu, C. A.;
Dvorozniak, M. T.; Matschinsky, F. M.; Grippo, J. F.
Science 2003, 301, 370.
8. Grimsby, J.; Matschinsky, F. M.; Magnuson, M. A. In
Frontiers in Diabetes; Matschinsky, F. M., Grippo, J. F.,
Eds.; Karger: Basel, 2004; Vol. 16, pp 360–378.
9. Dunten, P.; Swain, A.; Kammlott, U.; Crowther, R.;
Lukacs, C. M.; Levin, W.; Reik, L.; Grimsby, J.; Corbett,
W. L.; Magnuson, M. A.; Matschinsky, F. M.; Grippo, J.
F. In Frontiers in Diabetes; Matschinsky, F. M., Magnu-
son, M. A., Eds.; Karger: Basel, 2004; Vol. 16, pp 145–
154.
O
N
H
N
O
S
Volume of
distribution (L/kg)
Clearance
(mL/min/kg)
Half life
(h)
Bioavailability
(%)
0.3
3.8
0.9
100
a Doses 1 mg/kg (iv) and 10 mg/kg (po).
10
9.0
(a)
(b)
9
10. Kamata, K.; Mitsuya, M.; Nishimura, T.; Eiki, J.-i.;
Nagata, Y. Structure 2004, 12, 429.
8.5
8.0
7.5
7.0
6.5
6.0
5.5
5.0
8
7
6
5
11. GK activators reported in the recent patent literature
include: (a) Fyfe, M. C. T.; Gardner, L. S.; Nawano, M.;
Procter, M. J.; Rasamison, C. M.; Schofield, K. L.; Shah,
V. K.; Yasuda, K. PCT Int. Appl. WO2004072031, 2004;
Chem. Abstr. 2004, 141, 225496; (b) Weichert, A. G.;
Barrett, D. G.; Heuser, S.; Riedl, R.; Tebbe, M. J.; Zaliani,
A. PCT Int. Appl. WO2004063179, 2004; Chem. Abstr.
2004, 141, 157108; (c) Bebernitz, G. R. PCT Int. Appl.
WO2004050645, 2004; Chem. Abstr. 2004, 141, 54350; (d)
Polisetti, D. R.; Kodra, J. T., Lau; J.; Bloch, P.; Valcarce-
Lopez, M. C.; Blume, N.; Guzel, M.; Santhosh, K. C.;
Mjalli, A. M. M.; Andrews, R. C.; Subramanian, G.;
Ankersen, M.; Vedsø, P.; Murray, A.; Jeppesen, L. PCT
Int. Appl. WO2004002481, 2004; Chem. Abstr. 2004, 140,
77140.
Vehicle
4
Cpd 22
3
Pre
Cpd
0
20
40 60
Vehicle Cpd 22
Time (min)
Figure 4. In vivo OGTT efficacy data for compound 22 (30 mg/kg po).
12. Brocklehurst, K. J.; Payne, V. A.; Davies, R. A.; Carroll,
D.; Vertigan, H. L.; Wightman, H. J.; Aiston, S.; Waddell,
I. D.; Leighton, B.; Coghlan, M. P.; Agius, L. Diabetes
2004, 53, 535.
potency, lipophilicity and physical properties within the
series. Thiophene 22 was identified as having a suitable
balance of potency and physical properties, and has
been shown to demonstrate low clearance, good bio-
availability and promising in vivo efficacy in an acute
rat model. Further studies, including our attempts to
uncouple the relationship between potency and plasma
protein binding, will be communicated in due course.
13. Preliminary aspects of this work have previously been
presented in poster form: McKerrecher, D.; Allen, J. V.;
Block, M. H.; Bowker, S. S.; Boyd, S.; Caulkett, P. W. R.;
Currie, G. S.; Davies, C. D.; Fenwick, M. L.; Gaskin, H.;
Hargreaves, R. B.; Hayter, B. R.; James, R.; Johnson, K.
M.; Johnstone, C.; Jones, C. D.; Rayner, J. W.; Walker, R.
P. Abstracts of Papers, 227th National Meeting of the
American Chemical Society, Anaheim, CA, Mar 28–Apr
1, 2004; American Chemical Society: Washington, DC,
2004; MEDI 341.
Acknowledgements
14. EC50 values were determined at 10 mmol/L glucose, and
are quoted as the geometric mean of at least three
independent experiments. Assay to assay variability was
within 2-fold based on the results of a standard
compound. All compounds in this series investigated
exhibited a negligible effect on Vmax. Detailed in vitro
investigations and assay conditions are described in Ref.
12.
The authors would like to acknowledge the contribu-
tions of colleagues in bioscience, physical chemistry
and pharmacokinetics in generating the data reported
in this communication. Drs. K. G. Pike and M. J. War-
ing are also thanked for helpful discussions and assis-
tance with preparation of the manuscript.
15. (a) Van de Waterbeemd, H.; Smith, D. A.; Beaumont, K.;
Walker, D. K. J. Med. Chem. 2001, 44, 1313; (b) Van de
Waterbeemd, H.; Smith, D. A.; Jones, B. C. J. Comput.-
Aided Mol. Des. 2001, 15, 273.
16. Further details of the high fat fed female Zucker OGTT
model will be described in a future publication.
17. Aranapakam, V.; Albright, J. D.; Grosu, G. T.; Delos
Santos, E. G.; Chan, P. S.; Coupet, J.; Ru, X.; Saunders,
T.; Mazandarani, H. Bioorg. Med. Chem. Lett. 1999, 9,
1737.
References and notes
1. Tadayyon, M.; Smith, S. A. Expert Opin. Investig. Drugs
2003, 12, 307.
2. Turner, R. C.; Cull, C. A.; Frighi, V.; Holmann, R. J. Am.
Med. Assoc. 1999, 281, 2005.
3. Matschinsky, F. M.; Magnuson, M. A. In Frontiers in
Diabetes; Matschinsky, F. M., Magnuson, M. A., Eds.;
Karger: Basel, 2004; Vol. 16, pp 1–17.