Discovery of a Novel 2,6-Disubstituted Glucosamine Series of Potent and Selective Hexokinase 2 Inhibitors

Article abstract of DOI:10.1021/acsmedchemlett.5b00214

A novel series of potent and selective hexokinase 2 (HK2) inhibitors, 2,6-disubstituted glucosamines, has been identified based on HTS hits, exemplified by compound 1. Inhibitor-bound crystal structures revealed that the HK2 enzyme could adopt an induced-fit conformation. The SAR study led to the identification of potent HK2 inhibitors, such as compound 34 with greater than 100-fold selectivity over HK1. Compound 25 inhibits in situ glycolysis in a UM-UC-3 bladder tumor cell line via ¹³CNMR measurement of [3-¹³C]lactate produced from [1,6-¹³C₂]glucose added to the cell culture.

Full text of DOI:10.1021/acsmedchemlett.5b00214

Letter
pubs.acs.org/acsmedchemlett
Discovery of a Novel 2,6-Disubstituted Glucosamine Series of Potent
and Selective Hexokinase 2 Inhibitors
Hong Lin,*,† Jin Zeng,† Ren Xie,† Mark J. Schulz,† Rosanna Tedesco,† Junya Qu,† Karl F. Erhard,†
James F. Mack,† Kaushik Raha,§ Alan R. Rendina,§ Lawrence M. Szewczuk,§ Patricia M. Kratz,§
Anthony J. Jurewicz,§ Ted Cecconie,§ Stan Martens,§ Patrick J. McDevitt,§ John D. Martin,§
Stephenie B. Chen,§ Yong Jiang,§ Leng Nickels,§ Benjamin J. Schwartz,§ Angela Smallwood,§
Baoguang Zhao,§ Nino Campobasso,§ Yanqiu Qian,§ Jacques Briand,§ Cynthia M. Rominger,‡
Catherine Oleykowski,‡ Mary Ann Hardwicke,‡ and Juan I. Luengo†
†Cancer Metabolism Chemistry; ‡Cancer Metabolism Biology; and §Platform Technology & Sciences, GlaxoSmithKline, 1250 South
Collegeville Road, Collegeville, Pennsylvania 19426-0989, United States
S
* Supporting Information
ABSTRACT: A novel series of potent and selective
hexokinase 2 (HK2) inhibitors, 2,6-disubstituted glucosamines,
has been identified based on HTS hits, exemplified by
compound 1. Inhibitor-bound crystal structures revealed that
the HK2 enzyme could adopt an “induced-fit” conformation.
The SAR study led to the identification of potent HK2
inhibitors, such as compound 34 with greater than 100-fold selectivity over HK1. Compound 25 inhibits in situ glycolysis in a
UM-UC-3 bladder tumor cell line via 13CNMR measurement of [3-13C]lactate produced from [1,6-13C2]glucose added to the cell
culture.
KEYWORDS: Hexokinase 2 inhibitor, crystal structure, structure−activity relationship selectivity
nhibition of glycolysis has been considered as a therapeutic
expression of activated ERBB2/Neu (MMTV-neu).7 They
1
Iapproach for anticancer treatment for decades. As the first also showed that global Hk2 ablation in adult mice was well-
tolerated without significant physiological consequences.
While HK2 is a potential target for cancer treatment, it has
been considered intractable for the past 50 years due to its
extremely polar active site, the complexity of its protein
functions, and the uncertainty associated with finding a HK2
selective inhibitor over the housekeeping HK1 isozyme.
Encouraged by the strong rationale, we initiated research to
identify HK2 selective small-molecule inhibitors for potential
cancer treatment or in combination with existing drugs to
sensitize chemotherapy and targeted therapy.
enzyme in the glycolytic pathway, hexokinase-2 (HK2) plays
two pivotal roles in cancer. HK2 catalyzes the first functionally
irreversible reaction in the glycolytic pathway to form glucose-
6-phosphate (G6P) from glucose and ATP. It drives the
“Warburg effect” by committing glucose to enter both the
glycolytic and pentose phosphate pathways. Second, HK2
stabilizes the Mitochondrial Outer Membrane (MOM) via
interaction with the voltage-dependent anion channel (VDAC)
to prevent apoptosis.2 The HK2 gene is amplified and HK2
RNA is overexpressed in tumors.3 Epigenetic changes such as
HK2 DNA demethylation have been reported as well. The
most profound changes in HK2 expression have been observed
in liver, kidney, endometrial, and pancreatic cancers.4 HK2
expression is regulated by many somatic stimuli: HIF1-α,
growth factors, c-MYC, and p53 mutations. Akt stimulates HK2
association with mitochondria to promote mitochondrial
integrity and prevent apoptosis. HK2 expression correlates
with tumor stage and a worse prognosis. Studies with HK2
knockdowns using siRNA or shRNA both in vitro and in vivo
have shown inhibition of tumor growth.5,6
Compound 1 was one of the glucosamine derivatives
identified from the high throughput screen (Supporting
Information) with purified HK2, and we first synthesized a
variety of C-2 amides to examine potency and HK2 vs HK1
selectivity.8 As shown in Table 1, compound 1 is weakly active
against HK2, but has no selectivity over HK1. Using a
continuous coupled assay (formation of G6P coupled to G6P
dehydrogenase), compound 1 was found to be competitive
with glucose (Ki = 2.9 0.33 μM) and noncompetitive versus
MgATP (Supporting Information). In a separate double
inhibition study (continuous coupled assay for ADP formation
with pyruvate kinase and lactate dehydrogenase), compound 1
Recently, Hay and his collaborators provided compelling
evidence for the role of HK2 in tumorigenesis and
maintenance, as well as metabolic insights via Hk2 depletion
using a genetic mouse model of nonsmall cell lung carcinoma
(NSCLC) induced by expression of activated KRAS (KRAS-
LA2) and a mouse model of breast cancer induced by
Received: May 29, 2015
Accepted: December 27, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
is equally active against HK2 and HK1 with an IC50 value of 2.0
μM.
Table 1. SAR of C-2 Substituted Glucosamines
We also examined the importance of the C1-hydroxy group
for inhibitor potency. Unfortunately, both the 1-methoxy
analog (7) and the 1-deoxy analog (8)10 were inactive.
To understand the binding of these inhibitors with HK2, we
conducted crystallography studies, and obtained the very first
ligand bound human HK2 cocrystal structure at 2.76 Å with
compound 1 and glucose-6-phosphate (G6P).11 The structure
is consistent with the double inhibition kinetic study, which
showed G6P and compound 1 could bind simultaneously to
HK2. Both N- and C-terminal domains show compound 1 at
their corresponding active sites, with the glucosamine ring
placed in the glucose-binding pocket (Figure 1), which is
consistent with the glucose competitive mode of action of these
compounds. All donor−acceptor interactions of the hydroxyl
groups of glucose with HK2 are also retained in the interaction
of compound 1 with HK2. Since the 1-hydroxyl group forms a
tight hydrogen bonding network with a water molecule and the
side chains of Gln739 and Glu742, it is not surprising that the
1-methoxy (7) and 1-deoxy (8) modifications are not tolerated.
The cocrystal structure of HK2 with compound 1 reveals
that the enzyme binding site is highly flexible. Comparison with
the HK2-glucose cocrystal structure (2NZT)12 shows that, in
both crystals (Figure 1A), the glucose ring binds tightly to the
big lobe of the enzyme via hydrogen bonds to Glu708, Gln739,
and Glu742. But for compound 1, the active-site loop (residues
616−633, colored red in Figure 1), which contains the catalytic
Lys621, is unable to fully close over the pyranose ring, as it is
displaced outward by 3 Å due to steric clash with the C-2 amide
side chain. This prevents the enzyme from phosphorylating the
C-6 hydroxyl group of compound 1. Because of such
displacement, the narrow pocket extending from the C-2
equatorial position of the pyranose ring is considerably widened
and can now easily accommodate bulkier amide substituents, a
clear example of enzyme “induced fit” by a ligand. This wider
pocket is lined with side chains from Pro605, Lys618, Lys738,
and Gln739 and is highly lipophilic, which explains the
increased potency of the biphenyl-derivative 4 (Table 1). A
hydrogen bond between the C-2 amide carbonyl and the
Thr620 side chain helps to direct the C-2 substituent into this
lipophilic pocket. The structure also suggests that the extension
of the molecule into the G6P binding site through
derivatization at the C-6 position could be a way to improve
potency of the glucosamine series. Since three oxygen atoms of
a
IC50 values given are means of at least 2 experiments.
was also shown to bind simultaneously with G6P (Supporting
Information). From structural studies with HK1, G6P is known
to bind separately from glucose in a nearby “allosteric” site with
the pyranose ring at the putative position of the ATP-bound
Mg2+ cation.9 Further modifications of the C-2 amides
indicated that bulky substitutions at the meta-positions of the
benzene ring improve HK2 potency (see 3 and 4). However,
such modifications impacted HK1 potency more; for example,
compound 4 was a very potent HK1 inhibitor with an IC50 of
40 nM. On the other hand, a bulky aliphatic amide as in
compound 5 appeared to enhance HK2 selectivity (IC50 = 16
and 160 μM, respectively, for HK2 and HK1), but suffered
from weak potency. Compound 6 with a 3,5-dinitrobenzamide
Figure 1. (A) Compound 1 in HK2 with G6P overlaid with 2NZT; (B) Compound 1 and G6P interactions with HK2 in the C-terminal catalytic
pocket.
B
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
the phosphate of G6P form H-bond interactions with Thr536,
Thr680, and Thr863, a polar group to mimic one of the H-
bond interactions might be important.
Based on this hypothesis, we set out to modify the 6-position
as depicted in Scheme 1, in which the 1-hydroxy of compound
Various functional groups at the 6-position, including
sulfonamide, amide, and urea, are tolerated (Supporting
Information Table 1). Among them, the 6-sulfonamide
(compound 18) is particularly promising with improved HK2
inhibitory activity.
Given how much has been known about HK2/HK1
substrate binding pockets and the fact that they have been
considered as intractable targets for the last 50 years, it is
remarkable that these dramatic modifications to glucose are
tolerated and result in increased inhibition of both HK2/HK1
enzymatic activities. We therefore pursued this approach
together with X-ray crystallography to understand the structural
biology of HK2.
Scheme 1. Synthesis of 2,6-Disubstituted Glucosamines
Our hypothesis to improve potency against HK2 by
modification of the 6-position of the glucosamine series was
further validated with additional 6-sulfonamide analogs. As
demonstrated in Table 2, ortho-chloride (23) improves HK2
Table 2. SAR of 2-Amido-6-benzenesulfonamide
a
glucosamines
Conditions and Reagents: (a) allyl alcohol, TsOH·H2O, 110 °C, 72 h,
93%; (b) TsCl, py, rt, 3 h; (c) NaN3, DMF, 90 °C, 17 h; 52%; (d)
Ph3P, H2O, THF, rt, 72 h, 43%; (e) benzensulfonyl chloride, py, rt, 20
h; (f) PdCl2, MeOH, rt, 4 h; 19%; (g) allyl alcohol, TsOH·H2O, 110
°C, 2 h; (h) TsCl, py, 0 °C, 6 h, 32%; (i) NaN3, DMF, 80 °C, 24 h; (j)
Ph3P, H2O, THF, rt, 24 h; (k) 2,3-dichlorobenzensulfonyl chloride, py,
rt, 24 h, 30%; (l) iodine, THF/H2O 1:1, rt, 2 h; then solid sodium
thiosulfate/sodium carbonate; (m) 4′-cyano-5-methoxy-[1,1′-biphen-
yl]-3-carboxylic acid, 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluro-
nium hexafluorophosphate (HBTU), diisopropylethylamine, CH2Cl2,
rt, 24 h, 26%; (n) PdCl2, MeOH, rt, 24 h; 46%.
4 was protected with an allyl group in a reaction catalyzed by
TsOH hydrate or, alternatively, by BF3·Et2O at lower
temperature (80 °C) and with shorter reaction time (18 h).
The 6-hydroxy of compound 9 was then converted to the
tosylate, which was treated with sodium azide in DMF to form
compound 10. Reduction of the azide to a primary amine 11
with Ph3P in THF/H2O provided a key intermediate for C6
elaboration. As an example, sulfonamide 12 was formed by slow
addition of benzenesulfonyl chloride to a solution of compound
11 in pyridine, followed by a Pd-mediated allyl deprotection in
MeOH and reverse phase HPLC purification.
In order to develop a robust synthetic route to modify the 2-
position of the 2,6-disubstituted glucosamine series, we
introduced pent-4-enamide at the beginning to give compound
13.13,14 It could be deprotected under mild conditions later in
the synthetic scheme. As depicted in Scheme 1, Method B,
using the same transformations, the 6-hydroxy of compound 13
was converted to amine 14. Treatment of 2,3-dichlorobenze-
nesulfonyl chloride with 14 followed by deprotection of the
pent-4-enamide with iodine afforded a second useful
intermediate 15 for 2-position modification. As an example,
compound 15 was converted to amide 16 via a coupling
reaction with 4′-cyano-5-methoxy-[1,1′-biphenyl]-3-carboxylic
acid15 in the presence of HBTU and DIEA, followed by a Pd-
mediated deprotection of the allyl group to give the final
compound 17 after reverse phase HPLC purification.
a
IC50 values given are means of at least 2 experiments
C
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
and HK1 potency by 8- to 10-fold, respectively, compared to
the parent benzenesulfonamide 12, while 2,3-dichlorobenzene-
sulfonamide 25 is the first potent HK2 inhibitor with an IC50
value less than 100 nM (HK2 IC50 = 25 nM, HK1 IC50 = 32
nM). Heterocyclic sulfonamides are also equally potent,
exemplified by 4-chloropyridine-3-sulfonamide (26), compared
to its corresponding chlorobenzenesulfonamide 23.
Table 3. HK2 Selective Glucosamine Derivatives
Some polar substitutions were introduced into the
benzensulfonamide to potentially mimic H-bond interactions
between the protein and G6P, since the 6-sulfonamides were
designed to extend into the polar G6P binding pocket. As an
example, a para-nitrile increases the potency of compound 27
against HK1 by 20-fold to 7.9 nM, while the increase of
potency against HK2 is less profound, only 6-fold.
Interestingly, a meta-carboxylic acid also increases both HK2
and HK1 potency by 16-fold and 6-fold (compound 28 versus
12), respectively. The acid−protein interactions appear to be
specific, since compound 29, bearing an ester group is far less
potent than its corresponding acid 30. Compound 30 is 63-fold
more potent against HK2 (IC50 = 10 nM), and 80-fold more
potent against HK1 (IC50 = 20 nM).
In addition, consistent with the SAR that bulky amides
increase HK1 inhibitory potency in the monosubstituted 2-
glucosamine chemical series, exemplified in Table 1 by
compound 5, it is not surprising that compound 17 is very
potent against HK1, but HK2 potency remains the same as that
of compound 25.
a
IC50 values given are means of at least 2 experiments.
Throughout the course of optimization, we have been
seeking the “clues” in SAR to improve HK2 vs HK1 selectivity
even though the sequences of these two proteins are 72%
identical and 84% similar, and the glucose binding sites are well
conserved among all hexokinases.16 Careful SAR analysis
revealed consistent trends for HK2 selectivity: (a) 2,3-
disubstituted sulfonamides at the 6-position; (b) less bulky
amides at the 2-position; and (c) sulfonamide at the 2-position.
Combination of any two of the elements could afford a
compound as selective as 500-fold against HK2 versus HK1,
such as compound 36. This has been proven to be a tractable
SAR to yield selective HK2 inhibitors exemplified by
compounds 31−36 (Table 3 and Supporting Information
Figure 5). Importantly, most selective compounds are also very
potent; for example, 2,6-bis-sulfonamide 34 has a HK2 potency
of 7.9 nM while its HK1 potency is only 1.0 μM. Therefore, it
has 125-fold selectivity against HK2 versus HK1.
conformation change induced by 2,6-disubstituted glucos-
amines.
To our surprise, some 2,6-disubstituted glucosamines,
although shown to be mutually exclusive (i.e., competitive)
with G6P in double inhibition mechanistic experiments
(Supporting Information), do not extend into the G6P pocket.
Instead, the 6-position substituents point into the opening of
the catalytic pocket to form a “folded” conformation,
exemplified by the cocrystal structure of compound 27 in
HK2,18 shown in Figure 2C. In addition to common H-bond
interactions with Glu742 and Glu708, the cyano group of
compound 27 forms a H-bond with Lys621. The overlay of the
cocrystal structures of compound 27 and 30 in HK2 (Figure
2D) shows minimal change in the protein structures, and the
catalytic pocket can accommodate both extended and folded
conformations of 2,6-disubstituted glucosamines. This also
explains the mutual exclusivity of “folded” compounds with
G6P, since the loop (aa533−538, salmon and red loops on the
left in Figure 2B) that interacts with G6P is “pushed” open,
thereby preventing G6P from binding within the pocket.
To further establish that glycolysis is inhibited by an HK2
inhibitor in a tumor cell line, the following three experiments
(details in Supporting Information) were conducted in UM-
UC-3 cells, a bladder tumor cell line with high protein
expression of HK2 and minimal expression of HK1 (Western
blot, Supporting Information Figure 6): (1) inhibition of G6P
production using a more stable 2-fluoro-2-deoxy-gluocose
(2FDG) as a surrogate for glucose; (2) reduction of glucose
consumption; and (3) reduction of downstream metabolite
lactate (Supporting Information Figure 7A−C). Compound 25
demonstrated inhibition of glycolysis: in all three experiments
with an IC50 value of 0.20 μM in the first experiment.19
Similarly, the selective compounds 33 and 36 had IC50 values of
0.47 and 0.52 μM, respectively, in the first experiment. Since
HK2 is the dominat isozyme in this cell line, these results
With various tool molecules in hand, we successfully
obtained some critical cocrystal structures to reveal the binding
modes of these 2,6-disubstituted glucosamines. As shown in
Figure 2, the cocrystal structure of compound 30 in HK2,17
a
dimer with 4 molecules of ligands in both N- and C-terminal
catalytic pockets, indeed supports our hypothesis that
substitution at the 6-position can extend into the G6P pocket
and increase potency. Similar to compound 1, compound 30
also forms H-bond interactions with Gln742 and Glu708.
However, it does not engage Lys621 because the catalytic loop
(aa616−633) is no longer in proximity, since the pocket is even
more open. The overlay (Figure 2B) of both cocrystal
structures reveals: (1) the carboxylic acid group of compound
30 mimics the phosphate of G6P to form similar H-bond
interactions with Thr680 and Thr863 but not Thr536; (2) the
catalytic loop (aa616−633, salmon and red loops on the right)
and the other loop that interacts with G6P (aa533−538, salmon
and red loops on the left) are wide open to accommodate the
D
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Figure 2. (A) Compound 30 ligand interactions in HK2 in the C-terminal catalytic pocket. (B) Overlay of compound 30 and 1 cocrystal structures
with HK2 in the C-terminal catalytic pocket. The salmon and red loops on the left are aa533−538. The salmon and red loops on the right are
aa617−633. (C) Compound 27 ligand binding interactions in HK2 in the C-terminal site. (D) Overlay of compounds 27 and 30 in HK2 in the C-
terminal catalytic pocket.
provide evidence for HK2 driven cellular activities of this novel
2,6-disubstituted glucosamine series. Although the glucose
pockets are likely very different, inhibition of glucose
transporters cannot be ruled out in these experiments and
may require further investigation.20
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.5b00214.
■
S
In summary, we have successfully optimized HK2 HTS hit
compound 1 to inhibitors with nM potency and developed a
tractable SAR to achieve a high level of HK2 versus HK1
selectivity. The 2,6-disubstituted glucosamine series was
rationally designed based on a cocrystal structure of compound
1 and G6P (Figure 1) to “link” the glucose and G6P binding
pockets. A 6-sulfonamide linker turned out to be optimal. The
hypothesis of using an acid (compound 30) to mimic the
phosphate of G6P was confirmed by the cocrystal structure of
compound 30 with HK2 (Figure 2). We also discovered a
tractable SAR via the combination of several features that lead
to selective HK2 inhibitors over HK1. Several cocrystal
structures of HK2 with inhibitors captured different con-
formations of the enzyme. These cocrystal structures reveal the
flexibility of the HK2 protein and that the catalytic site can
adopt an “induced-fit” conformation with inhibitors. Using
compound 25 as an example in a tumor cell line such as UM-
UC-3 that predominately expresses HK2, we have demon-
strated that a potent HK2 small molecule inhibitor can inhibit
glycolysis in tumor cells. To our best knowledge, compounds
such as 33, 34, and 35 are the most advanced tool molecules to
date in terms of potency and selectivity of HK2 over HK1 to
study the role of HK2 enzymatic activity in tumorigenesis
without compromising systemic homeostasis or normal
metabolic functions of HK1.
Biological assays, biological data, and experimental
procedures, and additional SAR. (PDF)
Data for 17 (PDF)
Data for 32 (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: hong.2.lin@gsk.com.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors thank Dr. Minghui Wang for NMR character-
izations of these compounds.
■
REFERENCES
■
(1) Pelicano, H.; Martin, D. S.; Xu, R.-H.; Huang, P. Glycolysis
inhibition for anticancer treatment. Oncogene 2006, 25, 4633−4646.
(2) Mathupala, S. P.; Ko, Y. H.; Pedersen, P. L. Hexokinase-2 bound
to mitochondria: Cancer’s stygian link to the “Warburg effect” and a
pivotal target for effective therapy. Semin. Cancer Biol. 2009, 19, 17−
24.
(3) Mathupala, S. P.; Rempel, A.; Pedersen, P. L. Aberrant Glycolytic
Metabolism of Cancer Cells: A Remarkable Coordination of Genetic,
Transcriptional, Post-translational, and Mutational Events That Lead
E
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
to a Critical Role for Type II Hexokinase. J. Bioenerg. Biomembr. 1997,
29, 339−343.
(4) Pedersen, P. L.; Mathupala, S.; Rempel, A.; Geschwind, J. F.; Ko,
Y. H. Mitochondrial bound type II hexohinase: a key player in the
growth and survival of many cancers and an ideal prospect for the
therapeutic intervention. Biochim. Biophys. Acta, Bioenerg. 2002, 1555,
14.
(5) Ahn, K. J.; Hwang, H. S.; Park, J. H.; Bang, S. H.; Kang, W. J.;
Yun, M.; Lee, J. D. Evaluation of the Role of hexokinase type II
incellular proliferation and apoptosis using human hepatocellular
carcinoma cell lines. J. Nucl. Med. 2009, 50 (9), 1525−1532.
(6) Peng, Q.; Zhou, Q.; Zhou, J.; Zhong, D.; Pan, F.; Liang, H. Stable
RNA interference of hexokinase II gene inhibits human colon cancer
LoVo cell growth in vitro and in vivo. Cancer Biol. Ther. 2008, 7 (7),
1128−1135.
(7) Patra, K. C.; Wang, Q.; Bhaskar, P. T.; Miller, L.; Wang, Z.;
Wheaton, W.; Chandel, N.; Laakso, M.; Muller, W. J.; Allen, E. L.; Jha,
A. K.; Smolen, G. A.; Clasquin, M. F.; Robey, R. B.; Hay, N.
Hexokinase 2 Is Required for Tumor Initiation and Maintenance and
Its Systemic Deletion Is Therapeutic in Mouse Models of Cancer.
Cancer Cell 2013, 24, 213−228.
(8) Lin, H.; Luengo, J. I.; Schulz, M.; Xie, R.; Zeng, J. PCT Int. Appl.
(2012), WO2012083145A1, Example 1.
(9) Mulichak, A. M.; Wilson, J. E.; Padmanabhan, K.; Garavito, R. M.
The mammalian hexokinase-1. Nat. Struct. Biol. 1998, 5, 555−560.
(10) Kocienski, P.; Pant, C. A convenient preparation of some
2,3,4,6-tetra-O-acetyl-1,5-anhydro-D-hexitols. Carbohydr. Res. 1982,
110 (2), 330−332.
(11) Pdb code 5HGI.
(12) DOI: 10.2210/pdb2nzt/pdb.
(13) Madsen, R.; Roberts, C.; Fraser-Reid, B. The Pent-4-enoyl
Group: A Novel Amine-Protecting Group That is Readily Cleaved
under Mild Conditions. J. Org. Chem. 1995, 60, 7920−7926.
(14) Ref 8, intermediate 1.
(15) Ref 8, intermediate 6a.
(16) Deeb, S. S.; Malkki, M.; Laakso, M. Human Hexokinase II:
Sequence and Homology to Other Hexokinases. Biochem. Biophys. Res.
Commun. 1993, 197 (1), 68−74.
(17) Pdb code 5HEX.
(18) Pdb code 5HFU.
(19) Although not selective for HK2 vs HK1, compound 25 is a very
weak inhibitor of HK4 (glucokinase IC50 = 3.5 μM; 120-fold selective).
(20) Deng, D.; Xu, C.; Sun, P.; Wu, J.; Yan, C.; Hu, M.; Yan, N.
Crystal structure of the human glucose transporter GLUT1. Nature
2014, 510, 121−125.
F
DOI: 10.1021/acsmedchemlett.5b00214
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX