Novel, highly potent systemic glucokinase activators for the treatment of Type 2 Diabetes Mellitus

Article abstract of DOI:10.1016/j.bmcl.2016.10.085

Glucokinase (GK, hexokinase IV) is a unique hexokinase that plays a central role in mammalian glucose homeostasis. Glucose phosphorylation by GK in the pancreatic β-cell is the rate-limiting step that controls glucose-stimulated insulin secretion. Similarly, GK-mediated glucose phosphorylation in hepatocytes plays a major role in increasing hepatic glucose uptake and metabolism and possibly lowering hepatic glucose output. Small molecule GK activators (GKAs) have been identified that increase enzyme activity by binding to an allosteric site. GKAs offer a novel approach for the treatment of Type 2 Diabetes Mellitus (T2DM) and as such have garnered much attention. We now report the design, synthesis, and biological evaluation of a novel series of 2,5,6-trisubstituted indole derivatives that act as highly potent GKAs. Among them, Compound 1 was found to possess high in vitro potency, excellent physicochemical properties, and good pharmacokinetic profile in rodents. Oral administration of Compound 1 at doses as low as 0.03?mg/kg led to robust blood glucose lowering efficacy in 3?week high fat diet-fed mice.

Full text of DOI:10.1016/j.bmcl.2016.10.085

Accepted Manuscript
Novel, Highly Potent Systemic Glucokinase Activators for the Treatment of
Type II Diabetes Mellitus
Jiayi Xu, Songnian Lin, Robert W. Myers, George Addona, Joel P. Berger, Brian
Campbell, Hsuan-shen Chen, Zhesheng Chen, George J. Eiermann, Nadine H.
Elowe, Brian T. Farrer, Wen Feng, Qinghong Fu, Roman Kats-Kagan, Michael
Kavana, Sunita Malkani, Daniel R. McMasters, Kaushik Mitra, Michele J.
Pachanski, Xinchun Tong, Maria E. Trujillo, Libo Xu, Bei Zhang, Fengqi
Zhang, Rui Zhang, Emma R. Parmee
PII:
S0960-894X(16)31133-7
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.10.085
BMCL 24388
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
15 August 2016
26 October 2016
28 October 2016
Please cite this article as: Xu, J., Lin, S., Myers, R.W., Addona, G., Berger, J.P., Campbell, B., Chen, H-s., Chen,
Z., Eiermann, G.J., Elowe, N.H., Farrer, B.T., Feng, W., Fu, Q., Kats-Kagan, R., Kavana, M., Malkani, S.,
McMasters, D.R., Mitra, K., Pachanski, M.J., Tong, X., Trujillo, M.E., Xu, L., Zhang, B., Zhang, F., Zhang, R.,
Parmee, E.R., Novel, Highly Potent Systemic Glucokinase Activators for the Treatment of Type II Diabetes Mellitus,
Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.10.085
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.

Novel, Highly Potent Systemic Glucokinase Activators for the Treatment of Type
II Diabetes Mellitus
Jiayi Xu*, Songnian Lin*, Robert W. Myers, George Addona, Joel P. Berger, Brian Campbell, Hsuan-shen
Chen, Zhesheng Chen, George J. Eiermann, Nadine H. Elowe, Brian T. Farrer, Wen Feng, Qinghong Fu,
Roman Kats-Kagan, Michael Kavana, Sunita Malkani, Daniel R. McMasters, Kaushik Mitra, Michele J.
Pachanski, Xinchun Tong, Maria E. Trujillo, Libo Xu, Bei Zhang, Fengqi Zhang, Rui Zhang, and Emma R.
Parmee
Discovery, Preclinical and Early Development, Merck & Co., Inc., 2015 Galloping Hill Rd, Kenilworth, NJ
USA
*Corresponding authors. Tel.: +1 732 594 1438; fax: +1 732 594 2840; Email: Jiayi_xu@merck.com (J.
Xu). Tel.: +1 908 740 0585; fax: +1 908 740 3126; Email: songnian_lin@merck.com (S. Lin)
Abstract
Glucokinase (GK, hexokinase IV) is a unique hexokinase that plays a central role in mammalian glucose
homeostasis. Glucose phosphorylation by GK in the pancreatic β-cell is the rate-limiting step that
controls glucose-stimulated insulin secretion. Similarly, GK-mediated glucose phosphorylation in
hepatocytes plays a major role in increasing hepatic glucose uptake and metabolism and possibly
lowering hepatic glucose output. Small molecule GK activators (GKAs) have been identified that increase
enzyme activity by binding to an allosteric site. GKAs offer a novel approach for the treatment of Type 2
Diabetes Mellitus (T2DM) and as such have garnered much attention. We now report the design,
synthesis, and biological evaluation of a novel series of 2,5,6-trisubstituted indole derivatives that act as
highly potent GKAs. Among them, Compound 1 was found to possess high in vitro potency, excellent
physicochemical properties, and good pharmacokinetic profile in rodents. Oral administration of
Compound 1 at doses as low as 0.03 mg/kg led to robust blood glucose lowering efficacy in 3 week high
fat diet-fed mice.

Keywords
Glucokinase, Glucokinase activator (GKA), hexokinase IV, glucose phosphorylation, glucose homeostasis,
Type 2 Diabetes Mellitus, diabetes, antidiabetic, allosteric activator, 2,5,6-trisubstituted indole
Glucose homeostasis is essential to mammalian life and as such it is carefully maintained through a
complex network of factors that control both glucose uptake and utilization.1 Imbalance in these factors
can lead to serious disease, including Type 2 Diabetes Mellitus (T2DM). T2DM is characterized by
hyperglycemia (elevated blood glucose) due to a combination of insulin resistance and the failure of
pancreatic beta-cells to produce sufficient insulin.2 The long-term complications of uncontrolled T2DM
include cardiovascular morbidity and mortality, peripheral neuropathy, retinopathy, and chronic kidney
disease.2 T2DM has become one of the most devastating diseases of mankind, affecting hundreds
millions of people worldwide. The estimated cost of diagnosed diabetes in the U.S. alone was $245
billion in 2012, representing a substantial economic burden to society.3 Treatment of T2DM aims to
restore and maintain blood glucose level in the normal range by the use of antidiabetic drugs that have
4
varying mechanisms of action (MOA) which impact different pathways in glucose metabolism.2,
Currently available antihyperglycemic agents include insulin, insulin secretagogues (e.g. sulfonylureas),
metformin, insulin sensitizers (PPAR- agonists), GLP-1 analogs and the related DPP-4 inhibitors, and
SGLT2 inhibitors.4 Despite these options, many patients are still unable to safely and effectively maintain
normal blood glucose levels. Therefore, the search for novel antidiabetic drugs continues.
Glucokinase (GK, hexokinase IV) plays a unique role in glucose homeostasis.5, 6 GK is a monomeric
enzyme that catalyzes the ATP-dependent phosphorylation of D-glucose to D-glucose-6-phosphate,
which is the initial, as well as rate-limiting step, in cellular glucose uptake and metabolism. GK exhibits
positive cooperativity kinetics with respect to glucose, with a S0.5 (substrate required for 50% of maximal
activity) for D-glucose of 8 mM, which approximates the physiological blood glucose concentration, and
a Hill coefficient (nH) for glucose dose response cooperativity of 1.75.7-9 GK has limited expression in
mammals. In pancreatic beta-cells, GK functions as a “glucose sensor” leading to glucose-dependent
insulin release.10,11 Small molecule activators of GK (GKA) have been identified that increase insulin
secretion from pancreatic beta-cells at any given concentration of glucose by lowering the S0.5 and/or
increasing the maximal reaction velocity (Vmax) of GK.6, 12, 13 In hepatocytes, GKAs act to increase D-
glucose uptake (and likely decrease hepatic glucose production) by trapping it as glucose-6-phosphate,
which is then incorporated into glycogen and/or metabolized by the glycolytic pathway.14 The activity of
hepatic GK is also regulated by glucokinase regulatory protein (GKRP), a GK inhibitory protein present in
hepatocytes but not found in pancreatic -cells.15 In the past two decades, several direct-acting GKAs, as
well as GK-GKRP disruptors, have been identified that lower blood glucose levels in diabetic rodent
models.15-17
As part of our efforts to identify structurally diverse GKAs,18-21 we discovered 2,5,6-trisubstituted indole
compounds (such a Compound 1) as a novel series of highly potent GKAs. The general synthetic route to

Compound 1 and its analogues is illustrated in Scheme 1. Zinc iodide mediated homo-Reformatsky
reaction22 of 2-chloro-5-nitrobenzaldehyde and (1-ethoxycyclopropoxy)trimethylsilane, followed by acid
catalyzed desilylation-cyclization, afforded lactone 1c in 70% yield. The lactone ring was opened by DBU
in methanol to give -hydroxy ester 1d. Mitsunobu reaction23 of 1d with diphenyl phosphorazidate
(DPPA) afforded azide 1e in good yield (83%). The azide group was reduced to an amino group by
tributylphosphine in THF/water at room temperature and the product was heated with triethylamine in
methanol at 50oC to give lactam 1f in 71% yield. Methylation of 1f, followed by substitution of the chloro
group with 6-(methylsulfonyl)pyridin-3-ol, gave compound 1h. The nitro group was reduced by
iron/ammonium chloride to give aniline 1i. Iodination at the ortho-position of the aniline and protection
as the ethyl carbamate afforded intermediate 1k. Sonogashira coupling24 with 2-ethynylpyridine and
base mediated cyclization (Larock indole synthesis25) gave target Compound 1 in high yield. By varying
the reagent used in the displacement of the chloro group in 1g, and the alkyne used in the Larock indole
synthesis step, we were able to prepare a series of analogues for biological testing.
Once synthesized, compounds were tested for their ability to activate recombinant human GK in the
presence of glucose at both 2.5 mM and 10 mM. The GK activation EC50 and maximum percent velocity
increase at saturating GKA values were used to define the in vitro potency of the GKAs.26 Compounds of
interest were also evaluated in the GKACA (Glucokinase Activator Cell Activity) assay using both rat INS-1
Clone-13 insulinoma cells and freshly isolated, primary rat hepatocytes in order to determine their
ability to activate GK in relevant cell types. In those assays, the ability of a GKA to increase the
production of intracellular D-2-deoxyglucose-6-phosphate from exogenous D-2-deoxyglucose was
measured by monitoring product formation using HILIC LC-MS.27 Representative results from these
assays are summarized in Table 1.
The 2,5,6-trisubstituted indole derivatives in Table I are potent, low nM GKAs that significantly increase
GK activity at both hypoglycemic and hyperglycemic glucose levels. Cellular potencies in the low nM
range were commonly observed in this GKA series. Similar to previously reported GKA’s series19, the dual
hydrogen-donor/acceptor binding motif between the GKA indole-NH and the N-atom in the adjacent R1
heterocycle with Arg 63 of GK is presumed to be crucial for good potency. A wide variety of 5- or 6-
membered R1 heterocycles with a N-atom at the 2’-position, including substituted pyridine, thiazole and
pyridazine, were well tolerated (compare Compounds 1, 2 and 3). Using pyridine as an example, the 4’-
position of the ring can be substituted with various groups of differing polarity and steric bulk, such as
dimethyl-hydroxymethyl (Compound 4), -CF3 (Compound 5) and methyl ester (Compound 6).
The chirality at the pyrrolidinone ring has variable impact on enzymatic potency across the series; often
one enantiomer is more potent than the other (Compound 3, 4 and 5 compared to their corresponding
enantiomer). Significant differences between enantiomers were also observed in their cellular activities.
For example, the enantiomers of Compound 1 have similar EC50 values in the enzyme assay (EC50
=
3.7/2.4 nM vs. 4.2/0.8 nM), but one enantiomer is significantly more potent in the cell assays (EC50 =
0.9/7.5 nM vs. 33/93 nM). The reason(s) for this are unclear.
SAR of the aryl ether moiety reveal that good enzymatic potency can be retained with a variety of
substituted phenyl or pyridyl groups. However, methyl sulfonylpyridine appears to be a preferred group

at that position. Compared to Compound 5, phenyl analogue Compound 7 is equipotent in the enzyme
assays, but 10-fold less potent in hepatocyte assay. Similarly, Compound 8, with a replacement of the
methyl sulfonylpyridine of 7 with an ethyl carboxylic ester, is also less potent in the cell assays. Its acid
analogue, Compound 9, which has good solubility at pH 7 [SOLY (pH2, 7) = 5, 113 µM], also has weak
activity in cells, despite being reasonably potent against the enzyme. Presumably, this difference
between the enzymatic and cellular activities is due to the much lower membrane permeability of the
carboxylic acid salt of Compound 9 at physiological pH.
Scheme 1. a. ZnI2, (1- ethoxycyclopropoxy)trimethylsilane, DCM, rt, 24 h; b. TFA, DCM/H2O (100:1), rt, 2
o
d, 70% two steps; c. DBU, MeOH, rt, 2 h, 88%; d. PPh3, DEAD, DPPA, THF, 0 C to rt, 83%; e. PBu3, H2O,
THF, rt, 30 min; then, TEA, methanol, 50 oC, 24 h, 71%; f. NaH, MeI, THF/DMF (1/1), rt, 30 min, 99%; g.
K2CO3, 6-(methylsulfonyl)pyridin-3-ol, DMA, 120 oC, 16 h, 66%; h. Fe, NH4Cl, isopropanol/water (1/1), 60
o
oC, 40 min, 99%; i. KI, KIO3, HCl (5 M), MeOH/dioxane/water (4/3/3), 50 C, 15 min, 99%; j. Ethyl
chloroformate, pyridine, rt, 16 h, 99%; Chiral separation by HPLC, 36%; k. 2-Ethynylpyridine,
bis(triphenylphosphine) palladium(II) dichloride, CuI, TEA, THF, 55 oC, 30 min, 99%; l. TBAF, THF, 55 oC, 1
h, 97%.
Table 1. Structure activity relationships of the 2,5,6-trisubstituted indole series GKAs

Maximum
activity
(vs. DMSO)
Maximum
activity
(vs. DMSO)
GKACA assay
Rat INS-1 Cells
EC50 (nM)
GKACA assay
Rat Hepatocytes
EC50 (nM)
EC (nM) @
EC (nM) @
R1
R2
50
50
Compound
2.5mM glucose
10mM glucose
1
1-en
2
3.7
784%
628%
620%
482%
713%
577%
739%
581%
764%
583%
739%
602%
452%
683%
2.4
193%
158%
168%
132%
188%
155%
173%
132%
174%
141%
184%
158%
123%
179%
0.9
33
7.5
93
4.2
4.5
5.5
4.5
48
0.8
3.8
1.0
4.2
18
21
16
2-en
3
37
77
7.2
247
3.9
38
23
3-en
4
1510
5.2
34
4.8
31
2.1
9.8
<0.85
31
4-en
5
2.4
86
2.5
375
7.8
1.7
22
2.7
199
20
5-en
6-rac
7
8.2
2.0
4.8
24
3.4
1.4
9.2
9.6
174
232
728
8
9
2111
Data are from a single titration.
Compound 1 was analyzed for its solubility, liver microsomal stability, plasma protein binding and cell
permeability. The results are summarized in Table 2. The compound has excellent solubility at pH 2 and
7 (194 and 142 M, respectively). It has reasonable stability upon incubation with human and rat liver
microsome preparations (75% and 79% remaining after 30 min, respectively). A significant fraction of
Compound 1 remains unbound in 100% human and C57BL/6 mouse plasma (8.7% and 7.2%,
respectively). Compound 1 is also highly cell permeable, with a Papp = 40 x10-6 cm/s. These characteristics
suggested that Compound 1 might have an attractive in vivo pharmacokinetic profile. Indeed,
Compound 1 was found to have low clearance (18 mL/min/kg) in the rat, resulting in an acceptable half-
life of 2.8 h. The oral bioavailability (%F) of Compound 1 was 30% (Table 3).
Table 2. Selected properties of Compound 1

Liver Microsomal Stability
% remaining @ 1 M after 30 min
Plasma Protein Binding
% Unbound
LLC-PK1 MDCK
Papp (10-6cm/s)
Solubility (M)
pH 2
194
pH 7
142
Human
75
Rat
79
Human
8.7
Mouse
7.2
40
Table 3. Pharmacokinetics of Compound 1 in the rat
AUCN
(M.h.kg/mg)
Cmax/dose
(M.kg/mg)
CL
Vdss
(L/kg)
T1/2
(h)
MRT
(h)
F
(%)
Compound
(mL/min/kg)
1
18
3.05
2.8
2.85
0.59
0.235
30
AUCN = normalized area under the curve; Vdss = volume of distribution; MRT = mean residence time
Based on its promising properties, Compound 1 was evaluated for its acute glucose lowering efficacy in
vivo using a 3 week high fat diet (HFD)-fed C57BL/6 mouse model.27 In this study, the compound was
administered orally at 0.01, 0.03 and 0.1 mg/kg doses to mice having a baseline blood glucose level of
~200 mg/dl. Compound 1 was extremely efficacious in this model, with highly significant blood glucose
lowering observed as early as 30 min post administration at doses as low as 0.03 mg/kg (Figure 1). The
effect persisted throughout the 4 h experimental period, resulting in a large reduction in blood glucose
area under the curve (AUC) at the two highest doses. As expected, statistically significant increases in
plasma insulin were observed at the 0.1 mg/kg dose. There was a trend towards increased plasma
insulin levels at the 0.03 mg/kg dose. The observed glucose lowering efficacy in the absence of
statistically significant insulin elevation at this dose may be due solely to the hepatic effects of GKAs to
lower glucose. This data suggests that despite the high potency of Compound 1 as a GKA, a clinically
useful therapeutic window might exist between its ability to meaningfully lower glucose while avoiding
hypoglycemia. The plasma levels of Compound 1 (total drug) at 4 hours post dosing were 3.8 –18 nM
and roughly dose-proportional (Table 4).
Figure 1. Acute effect of varying doses of orally administered Compound 1 on blood glucose and plasma
insulin levels in HFD-fed mice.

N = 7 mice for each treatment group.
Table 4. Total Compound 1 drug levels in HFD mouse plasma at 4 h post dosing
Dose
Total [Compound 1] in plasma (M)
0.01 mg/kg
0.03 mg/kg
0.10 mg/kg
0.0038
0.012
0.018
In conclusion, a series of 2,5,6-trisubstituted indole compounds have been prepared and demonstrated
to be novel, highly potent GK activators both in vitro and in vivo. SAR studies revealed excellent
tractability in this series of GKAs. Optimization efforts led to the discovery of Compound 1, which
activates GK in vitro both in the enzymatic assays and in pancreatic -cells and hepatocytes with high
potency. Compound 1 exhibited good oral bioavailability and PK. Oral administration of Compound 1 at
doses as low as 0.03 mg/kg led to robust blood glucose lowering in 3 week HFD-fed mice. Additional
studies of this series of GKAs will be reported in due course.
References:

1. Szablewski, L. Glucose Homeostasis and Insulin Resistance, Bentham Science Publishers, 2011.
2. Goldstein, B. J. and Muller-Wieland, D. Type 2 Diabetes: Principles and Practice, 2nd edition,
Informa Healthcare, 2008.
3. American Diabetes Association. Diabetes Care 2013, 36, 1033.
4. DeFronzo, R. A. Am. J. Med. 2010, 123, S38.
5. Matschinsky, F. M. Curr. Diab. Rep. 2005, 5, 171.
6. Matschinsky, F. M.; Zelent, B.; Doliba, N. M. et. al. “Research and Development of Glucokinase
Activators for Diabetes Therapy: Theoretical and Practical Aspects”, in Handbook of
Experimental Pharmacology 203: Diabetes - Perspectives in Drug Therapy, Schwanstecher, M.
Ed.; Springer-Verlag Berlin Heidelberg, 2011, pp 357.
7. Xu, L. Z.; Weber, I. T.; Harrison R. W.; Gidh-Jain, M.; Pilkis, S. J. Biochemistry 1995, 34, 6083.
8. Cárdenas, M. L. "Glucokinase": Its Regulation and Role in Liver Metabolism, R G Landes
Publishers, Austin, TX, 1995, pp 21.
9. Matschinsky, F. M. Diabetes, 1996, 45, 223.
10. Matschinsky, F. M.; Glaser, B.; Magnuson, M. A. Diabetes 1998, 47, 307.
11. Schuit,F. C.; Huypens, P.; Heimberg, H. and Pipeleers, D. G. Diabetes 2001, 50, 1.
12. Fyfe, M.C.T.; Procter, M.J. Drugs Future. 2009, 34, 641.
13. Kamata,K.; Mitsuya, M.; Nishimura, T.; Eiki, J. and Nagata, Y. Structure 2004, 12, 429.
14. Agius, L. Biochem J. 2008, 414, 1.
15. Lloyd, D. J.; St. Jean Jr, D. J.; Kurzeja, R. J. et. al. Nature 2013, 504, 437.
16. Grewal, A. S.; Sekhon, B. S.; Lather, V. Mini Rev. Med. Chem. 2014, 14, 585.
17. Filipski, K. J.; Pfefferkorn, J. A. Expert Opin. Ther. Patents 2014, 24, 875.
18. Nishimura, T.; Iino, T.; Mitsuya, M.; Bamba, M. et. al. Bioorg. Med. Chem. Lett. 2009, 19, 1357.
19. Mitsuya, M.; Kamata,K.; Bamba, M. et. al. Bioorg. Med. Chem. Lett. 2009, 19, 2718.
20. Ishikawa, M.; Nonoshita, K.; Ogino, Y. et. al. Bioorg. Med. Chem. Lett. 2009, 19, 4450.
21. Iino, T.; Sasaki, Y.; Bamba, M.; Mitsuya, M.; Ohno, A. et. al. Bioorg. Med. Chem. Lett. 2009, 19,
5531.
22. Oshino, H.; Nakamura, E.; Kuwajima, I. J. Org. Chem. 1985, 50, 2802.
23. Mitsunobu, O.; Yamada, Y. Bull. Chem. Soc. Jap. 1967, 40, 2380.
24. Sonogashira, K. J. Organomet. Chem. 2002, 653, 46.
25. Larock, R. C.; Yum, E. K. J. Am. Chem. Soc. 1991, 113, 6689.
26. Glucokinase activity was measured by determining the ATP-dependent production of D-glucose-
6-phosphate from D-glucose at 2.5 and 10 mM using a continuous glucose-6-phosphate
dehydrogenase coupled spectrophotometric assay. All GKA were titrated using a final assay
concentration of 1% DMSO, which served as the control. EC50 values are defined as
concentration of GKA which gave half of maximal increase in glucokinase-catalyzed product
formation at either 2.5 or 10 mM glucose.
27. Detailed experimental procedures will be published separately.

Graphical abstract