Journal of Enzyme Inhibition and Medicinal Chemistry

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/ienz20

Glucokinase as an emerging anti-diabetes target

and recent progress in the development of its
agonists

Yixin Ren, Li Li, Li Wan, Yan Huang & Shuang Cao

To cite this article: Yixin Ren, Li Li, Li Wan, Yan Huang & Shuang Cao (2022) Glucokinase
as an emerging anti-diabetes target and recent progress in the development of its
agonists, Journal of Enzyme Inhibition and Medicinal Chemistry, 37:1, 606-615, DOI:
10.1080/14756366.2021.2025362

To link to this article: https://doi.org/10.1080/14756366.2021.2025362

© 2022 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 24 Jan 2022.

Submit your article to this journal

Article views: 6982

View related articles

View Crossmark data

Citing articles: 11 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ienz20
JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
2022, VOL. 37, NO. 1, 606–615
https://doi.org/10.1080/14756366.2021.2025362

REVIEW

Glucokinase as an emerging anti-diabetes target and recent progress in the

development of its agonists
Yixin Ren
, Li Li
, Li Wan, Yan Huang and Shuang Cao
Key Laboratory of Green Chemical Engineering Process of Ministry of Education, School of Chemical Engineering and Pharmacy, Wuhan
Institute of Technology, Wuhan, P. R. China

ABSTRACT ARTICLE HISTORY

Type 2 diabetes mellitus is a metabolic disorder with complicated pathogenesis, and mono-target therapy Received 13 July 2021
often fails to effectively manage the levels of blood glucose. In recent years, the anti-diabetes target glu- Revised 7 December 2021
cokinase (GK) has attracted the attention of researchers. It acts as a glucose sensor, triggering counter Accepted 29 December 2021
regulatory responses following a change in glucose levels to aid restoration of normoglycemia. Activation
KEYWORDS
of GK induces glucose metabolism and reduces glucose levels for the treatment of type 2 diabetes. GK Type 2 diabetes;
agonists (GKA) are a new class of antidiabetic drugs. Among these agents, dorzagliatin is currently being glucokinase; glucokinase
investigated in phase III clinical trials, while PB-201 and AZD-1656 have reached phase II clinical trials. This agonist; research progress;
article describes the mechanism of action of GK in diabetes and of action of GKA at the protein level, and drug target
provides a review of the research, trends, and prospects regarding the use of GKA in this setting.

1. Introduction In the human body, GK is mainly concentrated in pancreatic

cells and liver cells, as well as the hypothalamus and gastrointes-
1.1. Structure and function of GK
tinal tract. It is mainly involved in the first step of glucose metab-
GK, termed hexokinase 4, is a member of the hexokinase family1,2. olism, as the first rate-limiting enzyme. It catalyses the
It is an inducible enzyme composed of 465 amino acids with a phosphorylation of hexose (e.g. D-glucose, D-fructose, and D-man-
molecular mass of 52 kDa3. The three-dimensional structure of nitose) to hexose 6-phosphate (e.g. glucose 6-phosphate, fructose
GK can be divided into three parts: large, small, and connected 6-phosphate, and mannitose 6-phosphate) (Figure 3)8. However,
domains (Figure 1). The connected domain is composed of three GK has a distinct molecular structure and active function com-
segments of linkers and is the main active region of GK. The bind- pared with hexokinase. Its molecular weight is usually half or
ing sites of glucose and GKA are located in this region4. lower than that of other hexokinases and has a higher Km
According to the binding of endogenous substances to the kinase (Michaelis–Menten constant) value (8 mM). Its reactivity is not
connected domain, GK can be divided into three conformations: affected by the phosphorylated product glucose 6-phosphate,
closed, open, and super-open forms (Figure 2). The closed and ensuring that the blood glucose within the physiological range is
open forms correspond to the open receiving and closed process- fully phosphorylated.
Located in the beta cells of the pancreas, GK is termed “the
ing states, respectively, and are the two states that determine
glucose receptor”. Its main function is to control the release of
whether glucose is converted to glucose 6-phosphate5. The super-
insulin according to the concentration of glucose (Figure 4)9.
open form is an inactive conformation in which GK does not react
Following an increase in blood glucose levels, GK phosphorylates
with the substrate. Under the condition of low blood glucose lev-
glucose and produces a large amount of ATP through glucose
els, GK is abundant in the super-open form.
metabolism. It also inactivates the K ATP channels on the surface
The GK gene is located on the short arm of human chromo-
of the islet cells; consequently, Ca2þ influx causes the islet cells
some 7. It is a single-copy gene with a total length of to release insulin, thereby reducing the concentration of
15.3–15.1 kb. Its complementary DNA has a length of 2,439 bp, blood glucose10.
including 12 exons and 11 introns6. The cDNA of GK in different GK, located in liver cells, mainly plays a role in regulating the
parts of the human body exhibits different structures. The cDNA glycogen content in the liver. Insulin and glucagon can trigger the
of the GK gene expressed in islet cells contains a CCAAT sequence transport of GK by glucose transporters, altering the amount of
in the upstream regulatory region of the 50 end, while the corre- GK in the cytoplasm of hepatocytes and controlling the intracellu-
sponding cDNA sequence of the GK gene expressed in liver tis- lar glucose content (Figure 5)11. In the liver, GK controls blood
sues is TATTT. The different structures of these regulatory sugar by converting glucose to liver sugar; hence, the activity of
sequences determine the expression of proteins with certain dif- GK directly determines the glucose conversion rate. GK regulatory
ferences in tissue specificity and activity7. protein (GKRP) is a polypeptide (molecular weight: 68 kDa) that

CONTACT Shuang Cao [email protected] Key Laboratory of Green Chemical Engineering Process of Ministry of Education, School of Chemical
Engineering and Pharmacy, Wuhan Institute of Technology, Wuhan, 430205, P. R. China

These authors have contributed equally to this work.
ß 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits
unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
 JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 607

only exists in mammalian liver. It can competitively bind GK with applied to the design of GKA to reduce the risk of hypoglycaemia
glucose, prevent GK from catalysing the process of glucose phos- due to overactivation of GK in the hypothalamus by reducing the
phorylation to glucose 6-phosphate, and regulate GK activity in blood–brain barrier permeability for GKA.
liver cells12. Under hypoglycaemic conditions, GK forms a complex In the intestine, GK is located in endocrine K and L cells and
with GKRP (GK–GKRP complex) that aggregates in the nucleus. some pituitary cells, and its function may be related to nutrient
Following an increase in glucose levels, GKRP is replaced by glu- perception19–21.
cose, the GK–GKRP complex is dissociated, and the levels of GK in
the cytoplasm are markedly increased, thereby leading to an
1.2. Relationship between GK and diabetes mellitus
increase in GK activity13. In mammals, the content of the GK–GKRP
complex is affected by the levels of glucose, activated by fructose In a normal physiological state, blood glucose receptors respond
6-phosphate, and inhibited by fructose 1-phosphate14–16. to changes in blood glucose levels and promote the synergistic
In the hypothalamus, GK is located in glucose-sensing neurons
and prevents hypoglycaemic effects caused by overactivation of GK
in the liver and pancreas through neuromodulation17,18. Following
overexcitation of GK in the hypothalamus, the hypothalamus indu-
ces the body to decrease the secretion of adrenaline, norepineph-
rine, and glucagon, thereby resulting in hypoglycaemia. This can be

Figure 5. The mechanism of GK action in hepatocytes. The figures were drawn

using Servier Medical Art [http://www.servier.com].
Figure 1. Glucokinase protein structure (PDB ID: 1V4S). This figure was drawn
using the PyMOL program (http://www.pymol.org).

Figure 2. Three conformational transformations of glucokinase. The figures were Figure 4. Mechanism of action of GK in pancreatic cells. The figures were drawn
drawn using Servier Medical Art [http://www.servier.com]. using Servier Medical Art [http://www.servier.com].

Figure 3. Glucokinase glucose metabolism reaction formula.

608 Y. REN ET AL.

action of human organs and hormones to adjust the blood glu-

cose levels within the normal range. Each step in this regulatory
process has the potential to combat hyperglycaemia or hypogly-
caemia through blood glucose balance.
To study the targeting of GK in the liver, researchers first used
standard Cre-LoxP-based gene targeting strategies to knock out
the liver-specific expression sequence of GK in mice22. The results
showed that the mice were normal at birth, and their fasting
blood glucose levels increased with age. Six weeks later, the mice
developed hyperglycaemia, and impaired glucose tolerance was
detected. In another group of mice, overexpression of liver GK
increased the intracellular levels of glucose 6-phosphate and
glycogen, as well as the activity of L-pyruvate kinase. These find-
ings suggested that overexpression of GK could directly improve
glycolysis and glycogen synthesis in vivo23. It was also found that
a high-fat diet in the GK-overexpressing mice did not cause dia-
betes symptoms in the short term24. However, the mice showed
impaired glucose tolerance after 6 months and mild hypergly-
caemia, hyperinsulinemia, and hypertriglyceridaemia at 12 months.
Moreover, the GK-overexpressing mice gained more weight than
those in the control group, eventually leading to glucose intoler-
ance and reduced insulin sensitivity. The results of these experi-
Figure 6. GKA binding sites (PDB ID: 1V4S). This figure was drawn using the
ments indicated that the levels of GK in the liver are closely Discovery studio program.
related to the blood glucose levels, and the absence of GK in the
liver leads to the development of hyperglycaemia. Furthermore, For pancreatic beta cells, Increased GK activity eventually alters
high levels of GK expression in the liver can induce hypogly- the membrane potential, thereby promoting insulin release.
caemia in the short term; nevertheless, high levels of GK expres- In the liver, GK junction sites on the GK–GKRP complex can
sion in the long term are also associated with the risk of glucose also bind to GKA to promote complex dissociation and
intolerance. GK release27.
In a study of pancreatic GK, the researchers knocked out the
specific expression sequence of GK in islet cells in mice that died
shortly after developing severe diabetes in infancy. Studies have 2.2. Reported GKA
shown that GK in islet cells plays an indispensable role in glucose
GKA are a novel class of drugs for the treatment of diabetes,
homeostasis25. improving the sensitivity of the body to glucose through the joint
action of the pancreas and liver. At present, a number of small-
2. GKA molecule GKA have been reported28, and some of those have
entered the clinical research stage; the specific progress is shown
2.1. Molecular mechanism of GKA in Table 1. According to the structural characteristics of these
The GK junction domain is located in the connected domain of GKA, they can be divided into three types: double-conjugate
the kinase structure, which is the binding site of GKA and glucose. amides, single-conjugate amides, and others.
All GKA binding sites identified thus far are all located in the
same binding region. In this article, we focussed on human GK 2.3. Double-conjugate amide GKA
and GKA binding conformations for interaction analysis. We ana-
lysed the binding conformation of the protein 1V4S and its ori- In this section, double-conjugate amide GKA which are currently
ginal ligand of the protein (2-AMINO-4-FLUORO-5-[(1-METHYL-1H- under development are introduced in detail (Figure 7).
IMIDAZOL-2-YL)SULFANYL]-N-(1,3-THIAZOL-2-YL)BENZAMIDE). At
the binding site, GKA forms hydrogen bonds with R65 and binds
2.3.1. MK-0941
with Y214 and V455 through hydrophobic interactions (Figure 6).
MK-0941 (Figure 7) is a GKA developed by Merck Sharp & Dohme
The combination of GKA and GK can prevent the transform-
(Kenilworth, NJ, USA). A phase II trial was conducted to study the
ation from the GK conformation to the super-open form, and
dose range of MK-0941 in patients with type 2 diabetes mellitus
maintain GK in the two high-affinity conformations (i.e. closed and receiving insulin (NCT00767000)29. In the low-dose group (MK-
open forms)26. 0941 10 mg thrice daily), 80.7% of participants experienced at
Considering the secondary structure of GK, the GKA binding least one adverse event. These findings led to termination of the
site is located in the green region, while the glucose binding site study. MK-0941 was not investigated further due to its unfavour-
is located in the red region (Figure 1). Hence, there is a certain able safety profile.
distance between the two binding sites. Therefore, the binding of
GK to GKA changes the conformation of GK but does not affect
the glucose binding site. 2.3.2. AZD-6370
By controlling the conformation of GK, GKA maintains GK AZD-6370 (Figure 7), a GK activator, is being developed by
active, thereby accelerating the conversion of hexose to hexose AstraZeneca (London, UK). A dose-ranging, randomised, single-
6-phosphate. blind, placebo-controlled, crossover assignment phase I study
 JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 609

Table 1. Clinical status of glucokinase agonists (GKA).

Clinical
stage Amount Molecule
Phase III 1 Dorzagliatin
Phase II 13 TMG-123; AZD-1656; TTP547; PSN-010; GKM-002; GKM-001; PF-04991532; ARRY-403; MK-0941; AZD-6370; TTP-399; PB-201; piragliatin; LY2608204
Phase I 11 TTP-355; AZD-6714; MK-0599; TAK-329; BMS-820132; AZD-5658; DS-7309; RO-0281675; ZYGK1; RO-4597014
Application 1 Recombinant human glucokinase adenovirus

Figure 7. Development of double-conjugate amide glucokinase agonists (GKA).

(NCT00690287) was initiated in patients with type 2 diabetes (esti- before treatment. The placebo group exhibited a reduction of
mated n ¼ 24) in Sweden30. The study was designed to evaluate only 2%. Fasting glucose levels were reduced by 23–4,623 mg/dL,
fasting and postprandial P-glucose levels, safety, and tolerability and there was no change in the area under the curve for C-pep-
after the twice- or four times-daily oral administration of AZD- tide. C-peptide is a peptide fragment containing 31 amino acids
6370. Clinically indicated AZD-6370 resulted in dose-dependent (molecular weight: 3 kDa) that is cleaved during the transform-
reductions in plasma glucose levels by 30% versus placebo ation of proinsulin into insulin under the action of proteolytic
(p < 0.001 at 60 mg and 180 mg doses). In January 2011, research enzymes. The clinical significance of serum C-peptide is similar to
on AZD-6370 was discontinued by AstraZeneca due to undis- that of insulin. Thus, the C-peptide data showed that the hypogly-
closed reasons. Investigators suggested that the development pro- caemic mechanism of GKM-001 was independent of insulin secre-
gram for AZD-6370 may have been terminated due to the lack of tion. The trial confirmed the previous hypothesis that liver-
an advantage over the competing compound AZD-1650 and a selective GKA are effective in reducing glucose levels and do not
similar safety profile. cause hypoglycemia32,33. Impetis Biosciences (Mumbai, India)
acquired the drug discovery business of originator Advinus
2.3.3. GKM-001 Therapeutics and development of GKM-001 in 2017 and continues
GKM-001 (Figure 7) is a liver-selective GKA developed by Advinus to develop a range of GKAs34.
Therapeutics (Karnataka, India)31. Its unique liver-selective mechan-
ism enables this agent to avoid the increased insulin secretion
caused by the activation of thymic GK, thereby preventing the 2.3.4. BMS-820132
occurrence of hypoglycaemia. In 2011, Advinus Therapeutics BMS-820132 (Figure 7) is a GKA developed by Bristol-Myers
launched a phase II trial that enrolled 60 patients with type 2 dia- Squibb(New York City, [New York] USA)35. The researchers exam-
betes (CTRI/2011/04/001661). The entire clinical trial was con- ined the effects of daily administration of BMS-820132 for 1 month
ducted over a 14-day period to examine the efficacy and safety of in rats with different blood sugar levels. The results showed that
GKM-001 at incremental doses (25, 50, 200, 600, 1,000 mg). The the effective concentration 50 (EC50) in rats with hyperglycaemia
clinical trial results showed that GKM-001 was effective in decreas- (12 mM) was 39 nM. In rats with normal blood glucose levels
ing glucose levels at all examined doses. The area under the curve (5 mM), the EC50 was 73 nM35. In 2011, Bristol-Myers Squibb con-
for insulin of the subjects at 24 h was reduced in a dose-depend- ducted two phase II clinical studies (NCT01105429, NCT01290575)
ent manner (from 9% to 20%) compared with that measured to study the single/multiple-dose administration, tolerability, and
610 Y. REN ET AL.

safety of BMS-820132. Thus far, results of this trial have not placebo-controlled, phase IV, and case-crossover study which is
been published. currently ongoing.

2.3.5. Pf-04937319/PB-201 2.3.6. AZD-1656

Pf-04937319 (Figure 7) is a partial GKA developed by Pfizer (New AZD-1656 (Figure 7) is a GKA developed by AstraZeneca (London,
York City, [New York] USA) that acts simultaneously on the liver UK). In 2010, a randomised, double-blind, placebo-controlled,
and pancreas36. However, As Pf-04937319 reduces the Vmax of GK phase II clinical study (NCT01152385) was conducted in Japan. In
while activating GK, the ability of excitation is reduced, resulting that trial, 224 patients with type 2 diabetes received treatment
with high-(daily titration 200 mg), medium-(daily titration 140 mg),
in a limited hypoglycaemic effect on type 2 diabetes.
or low-dose (daily titration 80 mg) AZD-1656 or placebo to
In 2011, Pfizer conducted a randomised, double-blind, phase II
observe changes in HbA1c levels within 4 months and evaluate
clinical trial (NCT01475461) to evaluate the safety and efficacy of
drug safety and tolerability. The results showed that, after
PF-04937319 on glycemic control in adult patients with type 2
2 months of treatment, the levels of HbA1c in the patients were
diabetes mellitus inadequately controlled on metformin. At the
decreased by 0.3–0.8% compared with those recorded at baseline;
same time, glimepiride and sitagliptin were also studied in paral- of note, the levels of HbA1c in the placebo group were decreased
lel. The results showed that the reduction rate of haemoglobin by only 0.1%. At 4 months, the reduction in HbA1c in the
A1c (HbA1c) in the PF-04937319 (100 mg), sitagliptin, and glime- 40–200 mg group was similar to that observed in the placebo
piride groups was 0.47, 0.43, and 0.83%, respectively. In group. This study showed that AZD-1656 significantly reduced the
terms of side effects, 5.1, 2.5, 1.8, and 34.4% of subjects developed levels of glycosylated haemoglobin after short-term treatment;
hypoglycaemia in the PF-04937319 (100 mg), placebo, sitagliptin, however, the efficacy diminished over time39,40. In 2011,
and glimepiride groups, respectively. Therefore, the addition of AstraZeneca terminated the AZD-1656 program due to unsatisfac-
PF-04937319 (100 mg) to treatment with metformin was shown to tory results.
be effective and well tolerated. However, the incidence of adverse
events caused by PF-04937319 was 3.5-fold higher than that of
metformin36–39. 2.4. Single-conjugate amide GKA
In 2016, pegbio (Suzhou, China) purchased the PF-04937319 Piragliatin, developed by F. Hoffmann-La Roche AG (Basel,
compound and renamed it PB-201. In 2019, PB-201 was investi- Switzerland), was the first single-conjugate amide GKA to enter
gated in a phase I clinical study involving Chinese patients with the clinical stage. Also, HMS5552 developed by Huazhong
type 2 diabetes. The safety, tolerability, pharmacokinetics, and Pharmaceutical co., Ltd (Guangdong, China). is the most advanced
pharmacodynamics of PB-201 at three doses (50, 100, and and has entered Phase III trials. In this section, some single-conju-
150 mg) were investigated in a randomised, double-blind, gate amide GKA are described in detail (Figure 8).

Figure 8. Development of single-conjugate amide GKA.

 JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 611

2.4.1. RO-281675 randomised, double-blind, phase II clinical trial (NCT02405260) to

In 2003, F. Hoffmann-La Roche AG first reported that GK could be evaluate the safety, tolerability, pharmacokinetic characteristics,
excited by small molecules (RO-281675) (Figure 8). The evidence and pharmacodynamic effect of TTP399 as adjuvant treatment for
demonstrated that GK could be a potential target for diabetes, diabetes. In this 6-month study, TTP399 (800 mg/day) was directly
and these small molecules could be developed into oral drugs41. associated with a decrease in HbA1c levels, with an average 0.9%
At a concentration of 3 mM, RO-281675 induced a 1.5-fold increase reduction from baseline compared with placebo (p < 0.01). TTP399
in the maximum metabolic rate (Vmax) of GK. In addition, the sub- (800 mg/day) also increased the levels of HDL cholesterol (3.2 mg/
strate concentration [S]0.5 (glucose) at half of the Vmax was dL) compared with placebo. It also reduced the levels of fasting
decreased from 86 mM to 20 mmol/L41. plasma glucagon by 20 pg/mL (p < 0.05), and induced weight loss
RO-281675 is a mixed agonist. An insulin release experiment of 3.4 kg in patients with body weight of 100 kg (p < 0.05)48,49.
was conducted on freshly isolated fused rat islets. These cells
were treated with 0.3, 1, 3, and 10 mM RO-281675. To observe the
lowest glucose concentration at which insulin is released, the glu- 2.4.5. PF-04991532
cose concentration was gradually increased from 0 to 20 mM, at a PF-04991532 (Figure 8) is a liver-selective GKA developed by Pfizer
rate of 1 mM every 20 min. Finally, the glucose sensitivity thresh- (New York City, NY, USA). A randomised, double-blind, placebo-
old of islet cells was decreased from 7 to 3 mM, when islets were controlled, dose-varying phase II clinical study (NCT01336738)
treated with 0 to 10 mM RO-28167542. However, due to the poten- showed that the levels of HbA1c in patients were decreased by
tial cardiovascular hazards, the compound did not enter the clin- 0.49%, compared with those recorded at baseline (PF-04991532:
ical stage. Although this compound did not progress to clinical 750 mg once daily). Research on PF-04991532 was discontinued
development, it paved the way for pharmaceutical companies to due to poor clinical activity. Notably, treatment with sitagliptin
develop efficient and safe GKA. (100 mg once daily) decreased the levels of HbA1c in patients by
0.79% compared with baseline.

2.4.2. Piragliatin
F. Hoffmann-La Roche AG avoided the cardiovascular hazards 2.4.6. LY2608204
associated with the first-generation GKA through a series of struc- LY2608204 (Figure 8) is a GKA developed by Eli Lilly (Indianapolis,
tural modifications of the compound RO-281675 and, subse- USA). A phase II clinical trial (NCT01247363) recruited 20 eligible
quently, produced the second-generation GKA piragliatin43. patients who received treatment orally with LY2608204 capsules
Piragliatin, also termed RO4389620 (Figure 8), is a mixed GKA once daily. The study examined the safety and tolerability of
which can simultaneously stimulate the GK in both the pancreas LY2608204 in patients with type 2 diabetes receiving increasing
and liver44. Piragliatin (110 mM) increases the maximum metabolic dosages of medicine. The starting dose was 160 mg, which was
rate of GK from 10.6 to 17.9 mM43,45. Piragliatin is the first GKA to gradually increased to 240, 320, and 400 mg every 7 days of treat-
enter the clinical study phase. ment (total duration of treatment: 28 days). The results showed
However, the drug has a cyclopentanone structure that is that the drug exerted a good hypoglycaemic effect without ser-
metabolised to produce cyclopentyl alcohol, which is toxic to liver ious adverse events; the incidence rate of other adverse events
cells. In 4-week toxicity studies (40, 80, and 120 mg/kg, orally, was 47.37%. Common adverse events included gastrointestinal
once daily), rats fed with any dose of piragliatin metabolite devel-
diseases, site reactions, headache, and hypertriglyceridaemia
oped hepatic lipidosis. A phase II clinical study (NCT00266240)
caused by high dosages. The maximum hypoglycaemic effect on
also found that long-term use of this product can cause greater
the glucose area under the curve was 42% compared with the
liver load. Research on piragliatin was terminated because of its
control group at high doses (30 mg/kg). Interpolation analysis
potential liver-poisoning effects in patients who regularly use the
showed that, when the average concentration of LY2608204 in
drug for the treatment of chronic conditions, such as type 2
plasma was 99 ng/mL (179 nM) (equivalent to 6.9 mg/kg
diabetes42,44,45.
LY2608204), the glucose area under the curve concentration
decreased by 20%.
2.4.3. RO-4597014 In 2019, a parallel, randomised, double-blind, phase II clinical
RO-4597014 (Figure 8) is the second GKA, produced by F. trial (CRT20192351) was conducted in China to assess the safety,
Hoffmann-La Roche AG (Basel, Switzerland), to enter clinical devel- efficacy, and tolerability of LY2608204 in patients with type 2 dia-
opment. The company prioritised RO-4597014 as back-up com- betes (Active with enrolment of 200 volunteers).
pound of piragliatin. In April 2009, research on this drug was
terminated because its cyclopentane structure may result in the
accumulation of metabolites. 2.4.7. HMS5552
F. Hoffmann-La Roche AG developed the fourth-generation GKA
HMS5552, also termed Dorzagliatin (Figure 8), to address the
2.4.4. TTP399 problems linked to the use of second-generation GKA (liver cell
TTP399 (Figure 8) is an oral GKA designed by vTv Therapeutics damage or inflammation in human metabolism)50. This is a mixed
LLC (High Point, USA). In animal studies, TTP399 did not activate agonist, which can simultaneously activate GK in the pancreas and
GK in pancreatic cells, affect insulin secretion, or cause insulin- liver, promote pancreatic insulin secretion and liver glucose trans-
related hypoglycemia39,46,47. The researchers examined the effects formation in patients with type 2 diabetes, and exert multiple
of TTP399 on patients with different blood glucose levels. At hypoglycaemic effects by activating intestinal GK to regulate the
hyperglycaemic levels (15 mM), the EC50 was 304 nM. At normal secretion of GLP-1. The most striking feature of this compound is
blood glucose levels (5 mM), the EC50 was 762 nM48. The drug is that it increased GK activity at low levels of blood sugar and does
designed to activate GK only in the liver for optimal control of not enhance insulin secretion, thus avoiding the development of
glucose levels48. The company conducted a multicenter, hypoglycaemic symptoms.
612 Y. REN ET AL.

In 2017, the results of a phase II clinical trial (NCT02561338) of metformin (33–35 patients per group) to evaluate the dose-
HMS5552 were published. The data showed a distinct curative response relationship between ARRY-403 and placebo. A signifi-
effect, as well as an excellent tolerability and safety profile with cant linear dose-response relationship between efficacy and dose
low risk of hypoglycaemia and absence of the most common side was observed in subjects who received 50, 100, and 200 mg
effects associated with the pharmacological treatment of diabetes. ARRY-403 orally twice daily (p ¼ 0.004). After 7 days of treatment,
In this multicenter, randomised, double-blind, placebo-controlled, fasting plasma glucose levels were reduced by 1.38 mM. The
multi-dose study, 258 patients with diabetes were divided into the reduction remained constant until the last day of administration.
placebo group and four groups of patients who received different However, the incidence of adverse events associated with ARRY-
doses of GKA. After 12 weeks, the mean change in HbA1c least- 403 was 1.5-fold higher than that linked to metformin, mainly due
squares was 0.35% in the placebo group, 0.39% in the 75 mg to elevated serum triglyceride levels which increased by 17–25%
once-daily group, 0.65% in the 100 mg once-daily group, from baseline; however, there was no evidence of dose depend-
0.79% in the 50 mg twice-daily group, and 1.12% in the 75 mg ence57–60. Amgen eventually discontinued the research on
twice-daily group. Patients in the 50 mg or 75 mg twice-daily this compound.
groups exhibited more significant changes in HbA1c than those in
the placebo group51. The incidence of adverse events was similar
between the treatment groups and placebo group, and there 2.5.2. TMG-123
were no reports of drug-related serious adverse events or severe TMG-123 was developed by Teijin Limited (Chiyoda, Japan); the
hypoglycaemia. Treatment with HMS5552 showed significant definite chemical structure of this compound has not been
advantages over similar GKA, no significant changes in liver reported61. At glucose levels of 5 nM, the EC50 value for the
enzymes, blood lipids, or other laboratory parameters compared human liver and pancreas was 0.35 and 0.32 mM respectively, indi-
with placebo, and no abnormalities in physical examination, vital cating that TMG-123 could effectively activate GK in both organs.
signs, or electrocardiogram findings52,53. Following the administration of 30 nM of TMG-123, the [S]0.5 was
In 2018, a multicenter, randomised, double-blind, placebo-con- 0.5 nM and there was no change in Vmax61. TMG-123 may exert a
trolled, 24-week treatment, phase III clinical study (NCT03141073) favourable hypoglycaemic effect on type 2 diabetes.
was conducted to evaluate the efficacy and safety of GKA At present, TMG-123 has entered the phase II clinical research
HMS5552 in combination with metformin in patients with type 2 (JapicCTI173698). In this study, involving 90 patients with type 2
diabetes; the study also add-on to metformin with additional 28- diabetes (age: 20–74 years), the researchers investigated the effi-
week open-label treatment. By July 2019, 718 patients were cacy, safety, and dose-response of TMG-123 for 12 weeks versus
enrolled in the trial. The primary clinical outcome was achieved placebo. The patients received TMG-123 orally daily for 24 weeks
within the first 24 weeks of the trial. At 24 weeks, patients receiv- in a fasting state at a gradient range from 1 to 160 mg. TMG-123
ing the combination therapy had a 1.07% decrease in HbA1c lev- did not cause serious adverse events or hypoglycaemia.
els compared with baseline (8.35%). Of note, those receiving the Electrocardiography, vital signs, and laboratory testing did not
placebo had a 0.50% decrease in HbA1c levels compared with show abnormality. When the blood glucose concentration was
baseline (8.37%). The reduction in HbA1c noted in the combin- less than 200 ng/mL, TMG-123 exerted a significant dose-depend-
ation group was statistically significant compared with that ent hypoglycaemic effect. At 30 min after receiving TMG-123
observed in the placebo group (p < .001)54,55. (80 mg), the serum insulin levels were significantly increased. The
results showed that the oral administration of the liver GKA TMG-
123 was well tolerated, and the agent could be rapidly absorbed
2.5. Other GKA on an empty stomach without being affected by eating. The use
2.5.1. ARRY-403 of TMG-123 was safe and effective in healthy individuals.
ARRY-403, also termed AMG-151 (Figure 9), is a mixed agonist
that simultaneously activates GK in the pancreas and liver56.
Amgen conducted a phase II clinical study (NCT01464437) to 2.5.3. ZYGK1
evaluate the dose-effect relationship of AMG 151 compared with ZYGK1 was developed by Zydus Cadila (Ahmedabad, India); its
placebo, on fasting plasma glucose in subjects with type 2 dia- specific chemical structure has not been reported. Studies using a
betes treated with metformin57. Fasting plasma glucose levels variety of preclinical models have shown that ZYGK1 is safe and
were assessed in 236 patients with type 2 diabetes treated with effective in controlling fasting and non-fasting blood glucose lev-
els. ZYGK1 reduces blood sugar levels by enhancing the ability of
the pancreas to sense the concentration of glucose, thereby
increasing insulin production and improving the net uptake of
blood sugar by the liver. In 2011, the company conducted a rand-
omised, double-blind, placebo-controlled, phase I clinical trial
(NCT01472809). The trial recruited 96 healthy volunteers and
patients with type 2 diabetes from India to evaluate the safety,
pharmacokinetics, and pharmacodynamics of ZYGK1. However, the
clinical trial was suspended because the results in the 4 mg single-
dose group did not meet the researchers’ expectations.

3. Conclusion
Typically, the treatment of diabetes requires long-term administra-
tion of medications, resulting in great burden to patients and soci-
Figure 9. Structure of ARRY-403. ety. Long-term use of metformin – a first-line antidiabetic drug –
 JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 613

is associated with the development of drug resistance. In addition, diabetes of the young, permanent neonatal diabetes, and
the traditional oral hypoglycaemic drugs (e.g. sulfonylureas, acar- hyperinsulinemic hypoglycemia. Human Mutation 2009;30:
bose, and pyrazolidines) have many disadvantages (e.g. poor effi- 1512–26.
cacy) and are linked to severe side effects. Therefore, the 8. Lenzen S. A fresh view of glycolysis and glucokinase regula-
development of new antidiabetic drugs is urgently warranted. tion: history and current status. J Biol Chem 2014;289:
In this article, we reviewed the GKA that have been investi- 12189–94.
gated in clinical studies thus far. The currently available clinical 9. Iynedjian PB, Mobius G, Seitz HJ, et al. Tissue-specific expres-
data suggest that GKA can effectively control the levels of HbA1c sion of glucokinase: identification of the gene product in
in patients with type 2 diabetes and can be used to treat this dis- liver and pancreatic islets. Proc Natl Acad Sci USA 1986;83:
ease. In addition, GKA were well tolerated by patients and associ- 1998–2001.
ated with a low risk of developing minor adverse 10. Del Guerra S, Lupi R, Marselli L, et al. Functional and
effects (p > 0.05). molecular defects of pancreatic islets in human type 2 dia-
There are two major challenges in the development of GKA, betes. Diabetes 2005;54:727–35.
namely overcoming the rapid development of drug resistance and 11. Brown KS, Kalinowski SS, Megill JR, et al. Glucokinase regula-
reducing the occurrence of treatment-related adverse effects (e.g. tory protein may interact with glucokinase in the hepato-
hypoglycaemia and abnormal serum triglyceride levels). cyte nucleus. Diabetes 1997;46:179–86.
Both types of parent ring structures of GKA described herein 12. de la Iglesia N, Mukhtar M, Seoane J, et al. The role of the
have good research potential. Double-conjugated amide PB-201 regulatory protein of glucokinase in the glucose sensory
and single-conjugated amides TTP 399 and HMS 5552 are charac- mechanism of the hepatocyte. J Biol Chem 2000;275:
terised by high activity and low toxicity. However, the usefulness 10597–603.
of MK-0941, AZD-6370, and AZD-1656 is limited by the rapid 13. Kaminski MT, Schultz J, Waterstradt R, et al. Glucose-induced
development of resistance, as a result of the introduction of pro- dissociation of glucokinase from its regulatory protein in the
pane-1, 2-diol. Treatment with LY 2608204 and AMG-151 has been nucleus of hepatocytes prior to nuclear export. Biochim
linked to elevation in the levels of serum triglycerides. Biophys Acta 2014;1843:554–64.
Owing to their favourable safety profile, GKA can be used for 14. Pautsch A, Stadler N, Lohle A, et al. Crystal structure of glu-
the routine treatment of diabetes and as a temporary hypogly-
cokinase regulatory protein. Biochemistry 2013;52:3523–31.
caemic agent for diabetic patients with other diseases. As a new
15. Agius L. Glucokinase and molecular aspects of liver glycogen
target, this type of therapeutic agents is suitable for combination
metabolism. Biochem J 2008;414:1–18.
with other hypoglycaemic drugs to achieve a better effect in the
16. Lloyd DJ, St Jean DJ, Jr., ; Kurzeja RJ, et al. Antidiabetic
treatment of diabetes.
effects of glucokinase regulatory protein small-molecule dis-
ruptors. Nature 2013;504:437–40.
Disclosure statement 17. Salgado M, Tarifeno-Saldivia E, Ordenes P, et al. Dynamic
localization of glucokinase and its regulatory protein in
No potential conflict of interest was reported by the author(s). hypothalamic tanycytes. PLOS One 2014;9:e94035.
18. Elizondo-Vega R, Cortes-Campos C, Barahona MJ, et al. The
role of tanycytes in hypothalamic glucosensing. J Cell Mol
Funding
Med 2015;19:1471–82.
The author(s) reported there is no funding associated with the 19. Magnuson MA. Tissue-specific regulation of glucokinase
work featured in this article. gene expression. J Cell Biochem 1992;48:115–21.
20. Moates JM, Nanda S, Cissell MA, et al. BETA2 activates tran-
scription from the upstream glucokinase gene promoter in
References islet beta-cells and gut endocrine cells. Diabetes 2003;52:
1. Wilson JE. Hexokinases. Rev Physiol Biochem Pharmacol 403–8.
1995;126:65–198. 21. Speck M, Cho YM, Asadi A, et al. Duodenal-jejunal bypass
2. Grossbard L, Schimke RT. Multiple hexokinases of rat tissues: protects GK rats from fbetag-cell loss and aggravation of
purification and comparison of soluble forms. J Biol Chem hyperglycemia and increases enteroendocrine cells coex-
1966;241:3546–60. pressing GIP and GLP-1. Am J Physiol Endocrinol Metab
3. Sarabu R, Taub R, Grimsby J. Glucokinase activation–a strat- 2011;300:E923–E932.
egy for T2D therapy: recent developments. Drug Discov 22. Terauchi Y, Sakura H, Yasuda K, et al. Pancreatic beta-cell-
Today 2007;4:111–5. specific targeted disruption of glucokinase gene. Diabetes
4. Matschinsky FM. Assessing the potential of glucokinase acti- mellitus due to defective insulin secretion to glucose(*). J
vators in diabetes therapy. Nat Rev Drug Discov 2009;8: Biol Chem 1995;270:30253–6.
399–416. 23. Ferre T, Riu E, Bosch F, Valera A. Evidence from transgenic
5. Choi JM, Seo MH, Kyeong HH, et al. Molecular basis for the mice that glucokinase is rate limiting for glucose utilization
role of glucokinase regulatory protein as the allosteric in the liver. Faseb J 1996;10:1213–8.
switch for glucokinase. Proc Natl Acad Sci USA 2013;110: 24. Shiota M, Postic C, Fujimoto Y, et al. Glucokinase gene locus
10171–6. transgenic mice are resistant to the development of obesity-
6. Matschinsky FM. Regulation of pancreatic beta-cell glucoki- induced type 2 diabetes. Diabetes 2001;50:622–9.
nase: from basics to therapeutics. Diabetes 2002;51(Suppl 3): 25. Postic C, Shiota M, Niswender KD, et al. Dual roles for gluco-
S394–S404. kinase in glucose homeostasis as determined by liver and
7. Osbak KK, Colclough K, Saint-Martin C, et al. Update on pancreatic beta cell-specific gene knock-outs using Cre
mutations in glucokinase (GCK), which cause maturity-onset recombinase. J Biol Chem 1999;274:305–15.
614 Y. REN ET AL.

26. Kamata K, Mitsuya M, Nishimura T, et al. Structural basis for 41. Grimsby J, Sarabu R, Corbett WL, et al. Allosteric activators
allosteric regulation of the monomeric allosteric enzyme of glucokinase: potential role in diabetes therapy. Science
human glucokinase. Structure 2004;12:429–38. 2003;301:370–3.
27. Dunten P, Swain A, Kammlott U, et al. Crystal structure of 42. Georgy A, Zhai S, Liang Z, et al. Lack of potential pharmaco-
human liver glucokinase bound to a small molecule allo- kinetic and pharmacodynamic interactions between piraglia-
steric activator. Front Diabet 2004;16:145–54. tin, a glucokinase activator, and simvastatin in patients
28. Spasov AA, Lobasenko VS, Kosolapov VA, et al. Synthesis with type 2 diabetes mellitus. J Clin Pharmacol 2016;56:
and pharmacological activity of 3-phenoxybenzoic acid 675–82.
derivatives. Pharma Chem J 2020;54:229–35. 43. Bonadonna RC, Heise T, Arbet-Engels C, et al. Piragliatin
29. Meininger GE, Scott R, Alba M, et al. Effects of MK-0941, a (RO4389620), a novel glucokinase activator, lowers plasma
novel glucokinase activator, on glycemic control in insulin- glucose both in the postabsorptive state and after a glucose
treated patients with type 2 diabetes. Diabetes Care 2011; challenge in patients with type 2 diabetes mellitus: a mech-
34:2560–6. anistic study. J Clin Endocrinol Metab 2010;95:5028–36.
30. Ericsson H, Sjo€berg F, Heijer M, et al. The glucokinase activa- 44. Sarabu R, Bizzarro FT, Corbett WL, et al. Discovery of pira-
gliatin-first glucokinase activator studied in type 2 diabetic
tor AZD6370 decreases fasting and postprandial glucose in
patients. J Med Chem 2012;55:7021–36.
type 2 diabetes mellitus patients with effects influenced by
45. Zhi J, Zhai S. Effects of piragliatin, a glucokinase activator,
dosing regimen and food. Diabet Res Clin Pract 2012;98:
on fasting and postprandial plasma glucose in patients
436–44.
with type 2 diabetes mellitus. J Clin Pharmacol 2016;56:
31. Ramanathan V, Vachharajani N, Patel R, Barbhaiya R. GKM-
231–8.
001, a liver-directed/pancreas-sparing glucokinase modulator 46. Raimondo A, Rees MG, Gloyn AL. Glucokinase regulatory
(GKM), lowers fasting and post-prandial glucose without protein: complexity at the crossroads of triglyceride and glu-
hypoglycemia in type 2 diabetic (T2D) patients. Diabetes cose metabolism. Curr Opin Lipidol 2015;26:88–95.
2012;61:A76. 47. Egan A, Vella A. TTP399: an investigational liver-selective
32. Bunkholt Elstrand M, Dong HP, Ødegaard E, et al. glucokinase (GK) activator as a potential treatment for type
Mammalian target of rapamycin is a biomarker of poor sur- 2 diabetes. Drug Eval. 2019;28:741–7.
vival in metastatic serous ovarian carcinoma. Human Pathol 48. Vella A, Freeman JLR, Dunn I, et al. Targeting hepatic gluco-
2010;41:794–804. kinase to treat diabetes with TTP399, a hepatoselective glu-
33. Bertelsen BI, Steine SJ, Sandvei R, et al. Molecular analysis of cokinase activator. Sci Transl Med 2019;11(475).
the PI3K-AKT pathway in uterine cervical neoplasia: frequent 49. Buse JB, Valcarce C, Freeman JL, et al. Simplici-T1—first clin-
PIK3CA amplification and AKT phosphorylation. Int J Cancer ical trial to test activation of glucokinase as an adjunctive
2006;118:1877–83. treatment for type 1 diabetes. Am Diabetes Assoc 2018;67:
34. Itamochi H, Kigawa J. Clinical trials and future potential of 126-LB.
targeted therapy for ovarian cancer. Int J Clin Oncol 2012; 50. Zhu XX, Zhu DL, Li XY, et al. Dorzagliatin (HMS5552), a novel
17:430–40. dual-acting glucokinase activator, improves glycaemic con-
35. Tirmenstein M, Horvath J, Graziano M, et al. Utilization of trol and pancreatic b-cell function in patients with type 2
the Zucker diabetic fatty (ZDF) rat model for investigating diabetes: a 28-day treatment study using biomarker-guided
hypoglycemia-related toxicities. Toxicol Pathol 2015;43: patient selection. Diabetes Obes Metab 2018;20:2113–20.
825–37. 51. Zhu D, Gan S, Liu Y, et al. Dorzagliatin monotherapy in
36. Amin NB, Aggarwal N, Pall D, et al. Two dose-ranging stud- Chinese patients with type 2 diabetes: a dose-ranging, rand-
ies with PF-04937319, a systemic partial activator of glucoki- omised, double-blind, placebo-controlled, phase 2 study.
nase, as add-on therapy to metformin in adults with type 2 Lancet Diabetes Endocrinol 2018;6:627–36.
diabetes. Diabetes Obes Metab 2015;17:751–9. 52. Xu H, Sheng L, Chen W, et al. Safety, tolerability, pharmaco-
37. Denney WS, Denham DS, Riggs MR, Amin NB. Glycemic kinetics, and pharmacodynamics of novel glucokinase activa-
effect and safety of a systemic, partial glucokinase activator, tor HMS5552: results from a first-in-human single ascending
dose study. Drug Design Develop Ther 2016;10:1619.
PF-04937319, in patients with type 2 diabetes mellitus inad-
53. Wang P, Liu H, Chen L, et al. Effects of a novel glucokinase
equately controlled on metformin-a randomized, crossover,
activator, HMS5552, on glucose metabolism in a rat model
active-controlled study. Clin Pharmacol Drug Dev 2016;5:
of type 2 diabetes mellitus. J Diabetes Res 2017;2017:1–9.
517–27.
54. Zhu D, Zhang Y, Chen L. 182-OR: a novel dual-acting gluco-
38. Kamimura H, Ito S, Chijiwa H, et al. Simulation of human
kinase activator (GKA) dorzagliatin (HMS5552) achieved pri-
plasma concentration-time profiles of the partial glucokinase mary efficacy endpoint with good safety profiles in T2DM
activator PF-04937319 and its disproportionate N-demethy- patients after 24 weeks of treatment in a phase III mono-
lated metabolite using humanized chimeric mice and semi- therapy trial. Am Diabetes Assoc 2020;69:182-OR.
physiological pharmacokinetic modeling. Xenobiotica 2017; 55. Zhu D, Zhao Y, TANG C, HMM0201 STUDY GROUP, et al.
47:382–93. Pharmacodynamics post-hoc analysis of glucose kinase acti-
39. Hale C, Lloyd DJ, Pellacani A, Veniant MM. Molecular target- vator dorzagliatin (HMS5552)—twelve weeks treatment in
ing of the GK-GKRP pathway in diabetes. Expert Opin Ther T2D patients in China. Am Diabetes Assoc 2018;
Targets 2015;19:129–39. 67(Supplement_1):1201-P.
40. Kiyosue A, Hayashi N, Komori H, et al. Dose-ranging study 56. Katz L, Manamley N, Snyder W, et al. AMG 151 (ARRY-403), a
with the glucokinase activator AZD1656 as monotherapy in novel glucokinase activator, decreases fasting and postpran-
Japanese patients with type 2 diabetes mellitus. Diabetes dial glycaemia in patients with type 2 diabetes. Diabetes
Obes Metab 2013;15:923–30. Obes Metab 2016;18:191–5.
 JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY 615

57. Katz L, Manamley N, Snyder WJ, et al. AMG 151 (ARRY-403), formulation and clinical development for a compound with
a novel glucokinase activator, decreases fasting and post- pH-dependent solubility. J Pharm Sci 2015;104:1522–32.
prandial glycaemia in patients with type 2 diabetes. 60. Aicher TD, Anderson D, Boyd SA, et al. ARRY-403, a novel
Diabetes Obes Metab 2016;18:191–5. glucokinase activator with potent glucose-dependent anti-
58. McVean M, Aicher TD, Boyd SA, et al. In combination ther- hyperglycemic activity in animal models of type 2 diabetes
apy of ARRY-403 with metformin, sitagliptin or pioglitazone mellitus. Keystone Symposium: Type.
results in additive glucose lowering in female ZDF rats. 61. Tsumura Y, Tsushima Y, Tamura A, et al. TMG-123, a novel
Keystone Symposium: Type 2 Diabetes and Insulin glucokinase activator, exerts durable effects on hypergly-
Resistance. Banff, AB, Canada, 20–25 January; 2009. cemia without increasing triglyceride in diabetic animal
59. Chung J, Alvarez-Nunez F, Chow V, et al. Utilizing models. PLOS One 2017;12:e0172252.
physiologically based pharmacokinetic modeling to inform