Identification of the New Covalent Allosteric Binding Site of Fructose-1,6-bisphosphatase with Disulfiram Derivatives toward Glucose Reduction

Article abstract of DOI:10.1021/acs.jmedchem.0c00699

Fructose 1,6-bisphosphatase (FBPase) has attracted substantial interest as a target associated with cancer and type 2 diabetes. Herein, we found that disulfiram and its derivatives can potently inhibit FBPase by covalently binding to a new C128 allosteric

Full text of DOI:10.1021/acs.jmedchem.0c00699

Subscriber access provided by the University of Exeter
Article
Identification of the new covalent allosteric binding site of
FBPase with disulfiram derivatives toward glucose reduction
Yunyuan Huang, yixiang xu, rongrong song, shuaishuai ni, Jiaqi Liu, Yanhong Xu, Yanliang
Ren, Li Rao, Yingjie Wang, Lin Wei, lingling feng, Chen Su, Chao Peng, Jian Li, and Jian Wan
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00699 • Publication Date (Web): 07 May 2020
Downloaded from pubs.acs.org on May 7, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Identification of the new covalent allosteric binding site of
FBPase with disulfiram derivatives toward glucose reduction
Yunyuan Huang1, #, Yixiang Xu2, #, Rongrong Song1, #, Shuaishuai Ni3, #, Jiaqi Liu1,
Yanhong Xu1, Yanliang Ren1,*, Li Rao1, Yingjie Wang4, Lin Wei1, Lingling Feng1,
Chen Su5, Chao Peng5, Jian Li2, *, Jian Wan1, *
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1 Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of Education,
College of Chemistry, Central China Normal University, Wuhan 430079, China
2
Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China
University of Science and Technology, Shanghai 200237, China
3
Cancer Institute, Longhua Hospital, Shanghai University of Traditional Chinese
Medicine, Shanghai, 200032, China.
4 Shenzhen Bay Laboratory, Shenzhen 518055, Guangdong, China
5 National Facility for Protein Science in Shanghai, Zhangjiang Lab, Shanghai, 201210,
China
* Corresponding author:
Address: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China; E-mails: renyl@mail.ccnu.edu.cn (Y-L. Ren); jianwan@mail.ccnu.edu.cn (J.
Wan); Phone: +86-27-67862022. Fax: +86-27-67862022.
Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East China
University of Science and Technology, Shanghai 200237, China; E-mail:
jianli@ecust.edu.cn; Phone: +86-21-64252584. Fax: +86-21-64252584.
Keywords: disulfiram derivatives, covalent allosteric mechanism, drug repurposing,
fructose 1,6-bisphosphatase
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 35
1
2
3
4
5
ABSTRACT
6
7
8
9
Fructose 1,6-bisphosphatase (FBPase) has attracted substantial interest as a target
associated with cancer and type 2 diabetes. Herein, we found that disulfiram and its
derivatives can potently inhibit FBPase by covalently binding to a new C128 allosteric
site distinct from the original C128 site in APO FBPase. Further identification of
allosteric inhibition mechanism reveal that the covalent binding of a fragment of 214
will result in the movement of C128 and the dissociation of helix H4 (123-128), which
in turn allows S123 to more easily form new hydrogen-bonds with K71 and D74 in
helix H3 (69-72), thereby inhibiting FBPase activity. Notably, both disulfiram and 212
might moderately reduce blood glucose output in vivo. Therefore, our current findings
not only identify a new covalent allosteric site of FBPase, but also establish a structural
foundation and provide a promising way for the design of covalent allosteric drugs for
glucose reduction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
INTRODUCTION
6
7
8
9
Fructose-1,6-bisphosphatase (FBPase) acts as an important rate-limiting enzyme
central to the gluconeogenic (GNG) pathway and regulates metabolism in dysfunctional
cells or tissues, such as hepatocytes,1 renal carcinoma cells,2 soft tissue sarcomas,3
cancer stem cells,4 natural killer cells5 and hematopoietic progenitor cells.6 FBPase
inhibitors have been developed as a potential strategy for treating cancer5 and type 2
diabetes (T2D).7 Moreover, many efforts have been made to discover inhibitors that
bind to the AMP allosteric site of FBPase to treat T2D.7-9 Among these inhibitors,
MB06322 was tested in a phase II clinical trial for the treatment of T2D.10 However, it
has been suggested that a metabolite (MB05099)11 of MB06322 (Fig. S1) leads to lactic
acidosis in humans. Additionally, a previous study12, 13 reported that the AMP site was
a difficult target due to its high dependence on electrostatic binding interactions, and
lack of selectivity relative to other important enzymes involved in anabolic and
catabolic pathways. To address this concern, discovering and identifying a novel site
that can regulate the catalytic activity of FBPase is of utmost importance.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Recently, the discovery of new allosteric sites and modulators has greatly
facilitated the design of new drugs.14, 15 Discovering drugs that covalently bind target
protein residues (e.g., cysteine and serine) has become an indispensable strategy for
drug discovery due to the high biological activity, high selectivity and long action
time16, 17 of such compounds. Thus, covalent allosteric modulators, which combine the
merits of allosteric and covalent drugs, have recently emerged as a research area that
has shown great potential for drug discovery.14, 18-20 KRAS (G12C) was considered to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 35
1
2
3
4
5
6
7
8
9
be an undruggable target for cancer until the discovery of new covalent allosteric
modulators.21, 22 In the last year, AMG 510, a KRAS(G12C) allosteric covalent
modulator, was approved for clinical trials.23, 24 In our previous work,25 we found that
several nitrostyrene derivatives could potently inhibit FBPase by covalently binding to
the original C128 site,25 but the details of their covalent allosteric mechanism remain
unclear.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Facing a high risk of failure, high costs, and a long period during the research and
development (R&D) of novel antidiabetic agents, drug repurposing has been identified
as a faster and more economical strategy. Herein, disulfiram, a well-known anti-
alcoholism26 and anticancer drug,27 and its derivatives (212/214) were used to clarify
the covalent allosteric mechanism of the C128 site. It is found that the disulfiram
derivative 214 potently inhibit FBPase activity by covalently binding to a novel
allosteric binding site, which is distinct from the original C128 site in APO FBPase25
and not exploited previously. Furthermore, the covalent allosteric mechanism of this
new site has been confirmed by integrating the crystallographic structure of FBPase-
C128 modified by 214 (a disulfiram derivative), LC-MS, site-directed mutagenesis and
molecular dynamics (MD) simulations. Considering the better FBPase inhibitory
activities of disulfiram and 212, the hypoglycemic efficacies of these two compounds
in ICR and db/db mice were further evaluated.
RESULTS AND DISCUSSION
The synthesis of compounds 212/214
The synthesis of compounds 212 and 214 are shown in Scheme 1, and the
ACS Paragon Plus Environment

Page 5 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
procedure generally followed previous works.28 Commercially available thiazole-2-
thiol was added to a freshly prepared solution of 4-chlorophenyl hypochlorothioite in
anhydrous ethyl ether, and the reaction mixture was stirred for 12 h at room temperature
to generate compound 212. Compound 214 was obtained by the oxidation of
benzo[d]thiazole-2-thiol and ethanethiol by using DDQ as the oxidizing agent. The
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
synthetic procedures are described in the experimental section.
Scheme 1. Synthesis of compounds 212 and 214
Reagents and conditions: (a)SO2Cl2, DCM, 0 °C, 1 h; (b) Et2O, rt, 12 h; (c) DDQ, DCM, 0 °C, 1 h.
Design of Compounds 212/214.
Targeting cysteine is a practical strategy for discovering covalent inhibitors17, 29
because of their low occurrence (2.3%) in the human proteome29 and high selectivity.
Thus, several well-established drugs that could covalently modify cysteines were used
to explore their alternative functions as novel FBPase inhibitors.17 Amongst them,
disulfiram (Fig. S2) exhibited good biological activity against FBPase (IC50=1.5 μM).
The thiazole ring is very common framework in FDA-approved drugs30 and disulfiram
derivatives,31 Thus, we used a cyclization strategy to further elucidate the inhibitory
mechanism of disulfiram and its derivatives (Fig. 1). In addition, the synthesis of
asymmetric disulfide derivatives (212 and 214) helped us to effectively verify this
covalent reaction mechanism. As illustrated in Fig. 1, disulfiram derivatives 212 and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 35
1
2
3
4
5
214 also potently inhibited FBPase (IC50 = 0.54 μM for 212, and 1.6 μM for 214).
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 1. Designing strategy of asymmetric disulfide compounds 212/214
Determination and Analysis of FBPase Crystal Structures
To elaborate the allosteric inhibitory mechanism of the disulfiram derivatives,
three different FBPase structures (APO PDB:1FTA, FBP-bound PDB:5ZWK, 214b-
bound PDB:6LS5) shown in Fig. 2 were considered. As illustrated in Fig. 2A, S123--
I126 (2.9 Å) and N125--C128 (3.0 Å) hydrogen bonds could be observed in the crystal
structure of APO FBPase. These two hydrogen-bonds contribute to the formation of the
310-helix structure, named helix H4. In addition, a new hydrogen-bond between the
substrate (FBP) and S123 appeared in the crystal structure of FBP-bound FBPase.
Although the hydrogen-bonds of S123--I126 (3.3 Å) and N125--C128 (3.5 Å) could
also be observed in the crystal structure of FBP-bound FBPase (Fig. 2B), these two
hydrogen-bonds were longer than those in APO FBPase. Moreover, the helical
content32 of H4 in the FBP-bound FBPase structure was decreased compared with that
of APO FBPase. Both evidences indicate that the hydrogen-bond between S123 and
ACS Paragon Plus Environment

Page 7 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
FBP could affect the helical content of H4.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 2. Crystal Structure of FBPase. (A) Interaction networks between helix H3 (73-88) and helix
H4 (123-128) in APO FBPase (PDB: 1FTA, green). (B) Interaction networks between helix H3 and
helix H4 in FBP-bound FBPase (PDB: 5ZWK, salmon.) (C) The electron density (2Fo-Fc omit map
contoured at 0.8σ) for 214b bound to C128. (D) Interaction networks between FBPase and 214b.
(E) The pocket variation from original C128 site in APO FBPase/FBP-bound FBPase to new
allosteric binding site in 214b-bound FBPase. (F) Interaction networks between helix H3 and helix
H4 in 214b-bound FBPase (PDB: 6LS5, gray) (G) The conformation comparison between 214b-
FBPase (gray cartoon, PDB ID:6LS5), FBP-bound FBPase (salmon cartoon, PDB ID: 5ZWK) and
APO FBPase (green cartoon, PDB ID:1FTA)
The structure of the FBPase cocrystallized with 214b (a fragment of 214) was
successfully determined at 2.0 Å resolution by X-ray crystallography to identify the
allosteric inhibition mechanism of this novel C128-site for the first time. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 35
1
2
3
4
5
6
7
crystallographic data are listed in Table 1.
Table 1. Crystallographic data collection and refinement statistics
8
9
Hu-FBPase-214b (6LS5)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Data collectiona
Space group
P212121
1.06
Wavelength
Cell dimensions
a, b, c (Å)
67.44, 83.40, 276.48
90, 90, 90
42.8-2.03 (2.04-2.03)b
α, β, γ (°)
Resolution (Å)
Rmerge
11.5 (63.2)
I/σ (I)
11.3 (2.3)
99.6 (98.6)
8.1 (7.4)
Completeness (%)
Redundancy
Refinement
No.reflections
Rwork/Rfree
101,851
19.42/23.73
No.atoms
Protein
9,909
99
Ligand/ion
Water
640
All Atom
10,648
B-factors
Protein
32.8
41.9
37.5
33.1
Ligand/ion
Water
All Atom
R.m.s. deviations
Bond lengths (Å)
0.011
1.119
Bond angles (°)
aData were collected from a single crystal
bValues in parentheses are for the highest-resolution shell
A covalent linkage between C128 and 214b could be observed from the structure
of 214b in complex with FBPase (Fig. 2C). Fig. 2D shows that the C128 pocket for
214b modification involves residues L129, M172, M185, D197, H253 and V257. As
ACS Paragon Plus Environment

Page 9 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
shown in Fig. 2E, the addition of 214b to C128 results in the movement of C128 residue
and the formation of a new allosteric binding site in 214b-bound FBPase and the
disappearance of the original C128 site 25 in APO FBPase and FBP-bound FBPase. This
is most likely to be the driving force for the deactivation of FBPase. Compared with the
crystal structures of APO FBPase (Fig. 2A) and FBP-bound FBPase (Fig. 2B), the
structure of 214b-bound FBPase (Fig. 2F) lacks the hydrogen-bonds of S123--FBP,
S123--I126 and N125--C128, and helix H4 (Fig. 2F and 2G) was broken. Instead, a new
hydrogen-bonding network between S123 and D74, K71 and water was formed.
Simultaneously, the helical content of helix H3 in 214b-bound FBPase was increased
compared with those in APO FBPase and FBP-bound FBPase (Fig. 2G).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 3. The analysis of FBPase structures simulated by 100 ns molecular dynamic (MD). (A)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 35
1
2
3
4
5
6
7
8
9
Interaction networks between APO FBPase structure helix H3 and helix H4. (B) Interaction
networks between FBP-bound FBPase structure helix H3 and helix H4. (C) Interaction networks
between 214b-bound FBPase structure helix H3 and helix H4. (D) The conformation comparison
between 214b-FBPase (blue cartoon), FBP-bound FBPase (orange cartoon) and APO FBPase
(magenta cartoon). (E) Distance change (S124:O-C128:N) of three FBPase structure (APO:
magenta line, FBP-bound: orange line, 214b bound: blue line). (F) Helical content (residue 120-
129) change of three FBPase structure (APO: magenta line, FBP-bound: orange line, 214b bound:
blue line). (G) The S123 side chain population density of three FBPase structure (APO: magenta
line, FBP-bound: orange line, 214b bound: blue line) in molecular dynamics balance.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Furthermore, three FBPase structures (APO PDB, FBP-bound and 214b-bound)
were further optimized by MD simulations for 100 ns. As shown in Fig. 3A, the 310-
helix of H4 in the APO FBPase crystal (Fig. 2A) was optimized to form a more stable
H4 α-helix by MD simulations (Fig. 3A). The MD structures of FBP-bound FBPase
(Fig. 3B) and 214b-bound FBPase (Fig. 3C) are similar to those in the crystal structure
(Fig. 2B and 2F). Notably, there was only one hydrogen-bond between S123 and K71
in the MD structure of 214b-bound FBPase instead of the hydrogen-bonding network
in the 214b-bound FBPase crystal structure, indicating that the interaction between
S123 and K71 was important for the elongation of the new H3 helix (Fig. 3C and 3D).
The protein dynamics that contribute to allosteric regulation were further investigated,
and plots of the distances (S124:O-C128:N), helical contents (helix H4) and population
densities of S123-Chi are illustrated in Fig. 3E, 3F and 3G, respectively. The magenta
line represents the helical content of H4, the distance of S124:O-C128:N, and the
population density of S123-Chi in APO FBPase, while the orange line represents those
in FBP-bound FBPase, and the blue line represents those in 214b-bound FBPase. As
ACS Paragon Plus Environment

Page 11 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
shown in Fig. 3 (magenta line), in the APO FBPase structure, the S124:O-C128:N
distance was approximately 3 Å (Fig 3E), and the helical content was approximately
3.1 (Fig. 3F), both indicating a typical -helical configuration. Moreover, the
population density of S123-Chi in both swing directions (gauche- and gauche+) was
approximately 0.5 (Fig. 3G). However, when FBP was bound to FBPase (Fig. 3 orange
line), the helical content (120-129) was decreased to 0.9 and the S124:O-C128:N
distance was increased to approximately 6 Å. This indicates that the binding of FBP
into FBPase results in the partial dissociation of the 120-129 helix compared with that
in APO FBPase. As expected, the swing in/out direction and probability of the S123
side chain (Fig. 3G orange line) show that the S123 side chain in FBP-bound FBPase
has an almost gauche+ direction when forming hydrogen-bonds with FBP.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in Fig. 3 (blue line), the S124:O-C128:N distance was increased to
approximately 9 Å (Fig. 3E) and the helical content in 214b-bound FBPase was reduced
to approximately 0.3 (Fig. 3F). This demonstrates that the covalent addition of 214b to
C128 results in the dissociation of helix H4 by disrupting the hydrogen-bond between
S124 and C128. Based on the above evidence, we inferred that the hydrogen-bond (Fig.
3E) between S124 in position i and C128 in position i+4 significantly contributes to the
formation of helix H4 (Fig. 3F). Consistent with our crystal structure of 214-bound
FBPase (Fig. 2F), the addition of 214b to C128 led to significant repulsion between
C128 and the -CONH2 group of N125; thus N125 was far from C128, and the S124-
C128 hydrogen-bond was broken, which in turn led to the disruption of helix H4. As
expected, the population density of the S123 side chain increases to approximately 1.0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 35
1
2
3
4
5
6
7
8
9
in the gauche+ direction but decreases to approximately 0.3 in the gauche- direction (Fig.
3G), indicating that most S123 is pointed away from FBP, and forms hydrogen-bonds
with K71, D74 and water in helices 69-72 (Fig. 2F gray cartoon). Overall, upon addition
of 214b to C128, N125 is repelled from C128, in turn disrupting the S124-C128
hydrogen-bond and helix H4. Helix H4 is very important for the movement of S123,
and the dissociation of helix H4 allows S123 to more easily form a new hydrogen-bond
with K71 in elongated helix H3 (69-72). The loss of the S123--FBP hydrogen-bond
leads to the inactivity of FBPase. Therefore, N125 and S123 are critical residues for the
inhibition of FBPase by disulfiram derivatives.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
LC-MS and Mutation Experiments
Fig. 4. Full scan electrospray mass spectrum of the parent ion region of 214 covalent binding of WT
(wild type) FBPase and C128S. (A) The molecular weight of WT FBPase; (B) Mass weight of
FBPase + n214b (n = 2, 3); (C) Mass weight of C128S mutation; (D) Mass weight of C128S +
n214b (n = 1, 2).
Liquid chromatography-mass spectrometry (LC-MS) was performed to verify
whether these compounds covalently bind C128 in FBPase. As illustrated in Fig. 4A,
the molecular weight of FBPase is 38,009 Da. Correspondingly, the peaks at 38,129 Da
ACS Paragon Plus Environment

Page 13 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(FBPase+2*214b) and 38,189 Da (FBPase+3*214b) in Fig. 4B match the molecular
weight of FBPase covalently bound to two and three 214b fragments of 214, which
indicates that 214b could covalently bind to multiple cysteines in FBPase. The peak of
37,993 Da in Fig. 4C corresponds to the molecular weight of the C128S mutation (when
the C128 residue in FBPase was mutated to serine). Notably, Fig. 4D shows peaks of
38,053 Da (C128S+214b) and 38,113 Da (C128S+2*214b), which indicate that 214b
could covalently bind with cysteines in FBPase other than C128. The peaks in the LC-
MS spectrum of wild type FBPase (Fig. 4B) show that one equivalent of the 214b
fragment was lost in the peaks of C128S (Fig. 4D), which implied that 214b was most
likely covalently bound to C128. In fact, the crystallographic data mentioned above
have already lent a solid credit for this finding. In addition, the possible reaction
pathway between FBPase and 214 was illustrated in Fig. S3. In addition, the possible
reaction mechanism between FBPase and compound 212 was also predicted by using
XO methods at ωB97X-D/6-31G*//ωB97X-D/6-311-G** theoretical level, as shown in
Fig. S4.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To identify the central and important role of cysteine in covalent inhibitory
regulation, when FBPase was covalently modified by disulfiram and its derivatives, the
IC50 values of these three compounds against cysteine mutants were measured
systematically.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 35
1
2
3
4
Table 2. The IC50 of Disulfiram/212/214 against cysteine mutation and WT FBPase
5
6
WT
C38S
C92S
C116S
C128S
C179S
C183S
C281S
7
8
kcat (s-1)
2.5±0.1
0.80±0.06 0.99±0.06 0.99±0.12
1.7±0.1
1.4±0.1
2.3±0.1
2.2±0.1
9
Disulfiram
IC50 (μM)
1.5±0.1
1
0.60±0.03
0.4
1.2±0.1
0.8
1.3±0.06
0.9
> 1000a
> 667
1.9±0.2
1.3
1.5±0.2
1
1.1± 0.1
0.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Disulfiram
(ICM50 / ICW
)
50
212 IC50 (μM)
212 (ICM50 / ICW
214 IC50 (μM)
0.54±0.06
2.1±0.7
3.8
0.64±0.12 0.85±0.08
> 1000a
> 1800
> 1000a
> 625
0.67±0.06
1.2
1.1± 0.1
2.1
1.3± 0.1
2.5
)
)
1
1.6±0.2
1
1.2
0.98±0.26
0.6
1.6
1.9±0.3
1.2
50
6.6±1.0
4.1
1.8±0.2
1.1
1.2± 0.3
0.75
1.5± 0.2
0.93
214 (ICM50 / ICW
50
a1000 μM inhibition rate is less than 50%
As shown in Table 2, the IC50 values of 212, 214 and disulfiram against the C38S,
C92S, C116S, C179S, C183S and C281S mutants were almost the same as their IC50
values against the wild type (WT) protein, and only the C128S mutation resulted in a
significant increase (>600-fold) in the IC50 values compared to that of the WT protein.
The above mutant assays and crystallographic data show that C128 plays a unique and
crucial role in the covalent allosteric inhibition of FBPase when disulfiram, 212, and
214 covalently bind with FBPase.
In addition, a series of Ala mutation assays of residues S123, S124 and N125 in
helix H4 were carried out, and the corresponding IC50 values of disulfiram, 212 and 214
against these mutants were determined as well. The enzymatic activities of S123A,
S124A and N125A were determined as described in our previous report25. Compared
with the WT (kcat=2.5 s-1), S124A (kcat=3.1 s-1) and N125A (kcat=0.93 s-1) retained
comparable catalytic activities, while S123A (kcat=0.04 s-1) lost almost all of its
catalytic activity25. Thus, these results demonstrated that among the above three
ACS Paragon Plus Environment

Page 15 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
residues, only S123 was crucial for the catalytic activity of the enzyme, which was
consistent with the crystallographic structure of FBP-bound (PDB:5ZWK), showing a
direct interaction between S123 and the substrate. As listed in Table S1, the IC50 values
of disulfiram, 212 and 214 against the S124A mutant of FBPase showed almost no
changes compared to those of WT FBPase, however, the N125A mutation led to 320-,
900- and 313-fold increases in the IC50 values of disulfiram, 212 and 214, respectively,
indicating that N125 but not S124 played an important role in the covalent allosteric
inhibitory regulation of helix H4 when C128 was covalently bound with these
disulfiram derivatives. Notably, the results of this mutation assay are consistent with
the abovementioned MD results and helpful for further understanding the inhibitory
regulation mechanism of new covalent allosteric binding sites through the dissociation
of helix H4 in FBPase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Glucose Reduction in ICR and db/db Mice
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 35
1
2
3
4
5
6
7
8
9
Fig. 5. Effect of a single administration of disulfiram or compound 212 on blood glucose in 7-9
weeks old ICR mice or db/db mice (n = 5-6 in each group). (A) and (B) Blood glucose changes and
AUC of blood glucose between 0 and 8 h in 12 h-fasted ICR mice after oral administration of
disulfiram and compound 212. (C) and (D) Blood glucose changes and AUC of blood glucose
between 0 and 8 h in 12 h-fasted ICR mice after intraperitoneal administration of compound 212.
(E) and (F) Blood glucose responses to an oral glucose challenge (2 g/kg) and AUC of blood glucose
between 0 and 2 h in 12 h-fasted ICR mice. Compound 212 was intraperitoneally administered 1.5
h prior to oral glucose challenge. (G) and (H) Blood glucose changes and AUC of blood glucose
between 0 and 6 h in freely feeding db/db mice after intraperitoneal administration of compound
212. (I) and (J) Blood glucose responses to an oral glucose challenge (2 g/kg) and AUC of blood
glucose between 0 and 2 h in 12 h-fasted db/db mice. Compound 212 was administered 1.5 h prior
to oral glucose challenge. Data are presented as the mean ± SEM. (*) P < 0.05, (**) P < 0.01, (***)
P < 0.001 vs vehicle. Student’s t test.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Considering the better FBPase inhibitory activities of disulfiram and 212, it is
imperative to further evaluate the hypoglycemic efficacies of these two compounds in
ICR and db/db mice. As shown in Fig. 5A, in an oral administration model, disulfiram
at doses of 100 mg/kg significantly reduced blood glucose at 4 h (p < 0.01 vs vehicle),
and the area under the curve (AUC)0-8h of the 100 mg/kg disulfiram treated group was
reduced by 12.8% compared to that of the vehicle group (Fig. 5B). However, compound
212 at oral doses of 100 mg/kg failed to display hypoglycemic effect, most likely due
to its poor oral bioavailability. Interestingly, we found that compound 212 at an
intraperitoneal dose of 50 mg/kg could remarkably reduce blood glucose in mice during
2 h to 6 h compared to that of the vehicle group at 50 mg/kg (Fig. 5C). AUC0-8h analysis
suggested that the 50 mg/kg 212 treated group exhibited a 36.3% reduction in blood
glucose (Fig. 5D). Unfortunately, disulfiram cannot be administered intraperitoneally
ACS Paragon Plus Environment

Page 17 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
due to its poor water solubility. Therefore, 212 was selected for further
pharmacodynamics studies. To better evaluate glucose tolerance in response to
treatment with 212, an oral glucose tolerance test (OGTT) was carried out in ICR mice.
Glucose challenge (2 g/kg) dramatically increased the blood glucose levels in the
vehicle group, whereas the 212 administered groups exhibited much smaller increases
in blood glucose, especially at 30 min after glucose administration (Fig. 5E). The AUC0-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
analysis suggested that 50 mg/kg 212 observably increased glucose tolerance by
2h
reducing the AUC0-2h by 61.7% relative to that of the vehicle group, and the
pharmaceutical effect was superior to that of 25 mg/kg 212 (ΔAUC0−2h = 19.8%) (Fig.
5F). These results clearly demonstrated that disulfiram and 212 have hypoglycemic
effects in ICR mice.
As an important pathway for glucose homeostasis, GNG can convert lactate and
other carbon substrates into glucose (Fig. S5A). Therefore, the inhibition of FBPase
likely blocks the GNG pathway and leads to decreased glucose and lactate
concentrations. To preliminarily determine whether compound 212 could inhibit
FBPase in vivo, we measured several parameters in ICR mice, including the
concentrations of blood glucose, blood lactate and the FBPase activity. As illustrated
in Fig. S5B, compound 212 at a dose of 50 mg/kg had hypoglycemic effects (P < 0.05
vs vehicle) 2 h after i.p. administration. At the same time, the blood lactate
concentration of the 212-treated group (4.1 mM) was elevated relative to that of the
33
vehicle group (2.0 mM) (Fig. S5C) but lactic acidosis was not observed (>5 mM ).
Moreover, 212 significantly inhibited the activity of FBPase at 50 mg/kg (Fig. S5D).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 35
1
2
3
4
5
6
7
These results further indicate that compound 212 could block GNG by inhibiting
FBPase.
8
9
In light of the favorable hypoglycemic effect of 212 in ICR mice, the ability of
212 to lower glucose was further evaluated in diabetic db/db mice. Low doses (25
mg/kg and 50 mg/kg) of 212 had hypoglycemic effects at 1-4 h compared to the vehicle
(Fig. 5G), and remarkably reduced blood glucose AUC0-6h by 21.7% and 25.5%,
respectively (Fig. 5H). An OGTT was also performed to evaluate the glucose tolerance
ability of 212. AUC0-2h analysis suggested that compound 212 at a dose of 25 mg/kg
observably increased glucose tolerance by reducing the AUC0-2h by 21.8% relative to
the vehicle, which was similar to the effect of 212 at 50 mg/kg (ΔAUC0−2h = 19.7%)
(Fig. 5I and 5J). These experiments suggested that 212 could also improve glucose
homeostasis in diabetic db/db mice.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSIONS
In this study, a novel covalent allosteric binding site, which is distinct from the
original C128 site in APO FBPase and not exploited by previous works, was discovered
based on the cocrystal structure of FBPase-214b. Disulfiram, a well-known anti-
alcoholism drug, and its derivatives (212 and 214) were found to potently inhibit
FBPase by covalently binding to this new C128 site. Crystallographic and MD studies
revealed that the addition of 214b to C128 would repel N125 away from C128, and
disrupt the S124--C128 hydrogen-bond and helix H4 (123-128). The dissociation of
helix H4 allows S123 to more easily move away from FBP and form a new hydrogen-
bond with K71, which in turn results in the extension of helix H3 (69-72). The
ACS Paragon Plus Environment

Page 19 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hydrogen-bond of S123--FBP is essential for the catalytic activity of FBPase, and the
addition of 214b to C128 could disrupt the S123--FBP hydrogen bond and thereby
inhibiting FBPase. To the best of our knowledge, this is the first report that disulfiram
or its derivatives act as covalent allosteric modulators of FBPase and exhibit
hypoglycemic effects. In particular, disulfiram and 212 moderately reduced blood
glucose in an ICR mouse model and a db/db mouse model. Our current findings identify
a new covalent allosteric binding site of FBPase, and establish a structural foundation
for the discovery of new covalent allosteric drugs that inhibit FBPase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Chemistry. The synthetic starting materials, reagents, and solvents were obtained
from a commercial supplier, such as Energy Chemical, J&K, and TCI at the highest
commercial quality and were used without further purification. The reaction progress
was monitored using analytical thin layer chromatography (TLC), HSGF 254 (150-200
μm thickness; Yantai Huiyou Co., China), UV light (254 nm) and I2 were used to
visualize the components. 1H NMR and 13C NMR spectra were recorded on a Bruker
AMX 400 and AMX 600 spectrometer in DMSO-d6, or CDCl3 with TMS as an internal
standard. Chemical shifts are reported in parts per million (ppm, δ) downfield from
tetramethylsilane. Proton coupling patterns are described as singlet (s), doublet (d),
triplet (t), quartet (q), and multiplet (m). High-resolution mass spectrometry (HRMS)
data were obtained by electron ionization (EI) using a Waters GCT Premier. All final
compounds were purified to > 95% purity, as determined by high-performance liquid
chromatography (HPLC) on an Agilent 1100 instrument with a quaternary pump and a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 35
1
2
3
4
5
6
7
diode array detector (DAD), MeOH:H2O=95:5 (v:v) as the eluent, and a flow rate of
0.5 mL/min.
8
9
2-((4-Chlorophenyl)disulfanyl)thiazole (212). To a stirred solution of 4-
chlorobenzenethiol (1.4 g, 10 mmol) in 20 mL of anhydrous dichloromethane was
added dropwise sulfuryl chloride (810 μL, 10 mmol) at 0 °C. The reaction mixture was
stirred for 1 h at the same temperature and then concentrated under reduced pressure to
afford 4-chlorophenyl hypochlorothioite as a brownish yellow liquid, which was used
in the next step without purification. A solution of freshly prepared 4-chlorophenyl
hypochlorothioite (5 mmol) in 10 mL of anhydrous ethyl ether was added dropwise to
a solution of thiazole-2-thiol (5 mmol) in 10 mL of anhydrous ethyl ether at room
temperature. The reaction mixture was stirred for 12 h, and removal of the solvents
produced a liquid that was purified using column chromatography, to afford compound
212. Yellow liquid. 75% yield. 1H NMR (400 MHz, CDCl3) δ 7.73 (d, J = 3.3 Hz, 1H),
7.57–7.50 (m, 2H), 7.34 –7.29 (m, 3H). 13C NMR (151 MHz, DMSO) δ 144.93, 130.91,
130.03, 129.92, 129.62, 124.03. HRMS (EI): m/z calcd C9H6ClNS3 (M+) 258.9351,
found 258.9352. HPLC purity: 99.2%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-(Ethyldisulfanyl)benzo[d]thiazole (214). 2-Mercaptobenzothiazole (1.0 mmol)
and ethanethiol (1.5 mmol) were dissolved in 5 mL of DCM at 0 °C. DDQ (1.0 mmol)
was then slowly added in portions to the solution, and the reaction mixture was stirred
for 0.5−1 h on an ice bath. Upon completion of the reaction, the solvents were
evaporated under vacuum, and the crude compound was purified via column
chromatography to give compound 214. Colorless oil. 85% yield. 1H NMR (400 MHz,
ACS Paragon Plus Environment

Page 21 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CDCl3) δ 7.87 (d, J = 8.1 Hz, 1H), 7.80 (dd, J = 8.0, 0.5 Hz, 1H), 7.47–7.40 (m, 1H),
7.36–7.30 (m, 1H), 2.98 (q, J = 7.3 Hz, 2H), 1.42 (t, J = 7.3 Hz, 3H). 13C NMR (150
MHz, DMSO-d6) δ 172.79, 155.13, 135.72, 126.99, 125.25, 122.46, 122.18, 33.26,
14.62. HRMS (EI) m/z calcd C9H9NS3 [M]+ 226.9897, found 226.9894. HPLC purity:
99.9%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Drugs and Chemicals. Mercaptopurine, furazolidone, disulfiram, carmofur,
lipoic acid, afatinib, phosphonomycin, eflornithine, warfarin and phenoxybenzamine
were obtained from Sigma-Aldrich.
Enzyme Activity Assays and Mutagenesis. Full-length WT human liver FBPase
(GenBank: D26055.1) was expressed in E.coli BL21 (DE3) and purified with a
HisTrap_FF_5 mL [Global] column following the standard ÄKTA™ pure system, and
this process was performed as described in our previous work.34 Enzymatic reactions
were performed with purified enzyme (11.2 μg/mL), and FBP (0.4 mM/L) in 30 μL of
buffer (50 mM/L Tris, 0.8 mM/L Mg2+). After reacting for 5 min at 37 ℃, 15 μL of 1
M perchloric acid was used to stop the reaction, and then malachite green (0.35%
polyvinyl alcohol and 0.0035% malachite green) was added to quantify the inorganic
phosphate. The A620 was measured after these additions.25 The kinetic constants values
or half maximal inhibitory concentration (IC50) were identified using logistic equations
(nonlinear analysis) or Hill kinetic equations in the Origin 7.5 program package.
Mutations were performed by introducing specific base changes into a double-
stranded DNA plasmid. The mutant constructs were generated using the two-step PCR
method. DNA encoding WT human liver FBPase cloned into pET-21b was used as a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 35
1
2
3
4
5
6
7
8
9
template for mutagenesis. The parental methylated and hemimethylated DNA were
digested by Nde I and Xhol1 restriction enzymes. Then, the mutant constructs were
ligated into a previous plasmid. The plasmids of the recombinant mutant were
transformed into DH5α competent cells. All of the mutations were confirmed by DNA
sequencing. The verified plasmids with mutations were transformed into strain E.coli
BL21 (DE3) cells. The mutant FBPase proteins were purified in the same manner as
WT FBPase.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Crystallization, Data Collection and Structure Determination. APO FBPase
crystals were obtained after 5-7 days at 18 ℃ by the hanging drop vapor diffusion
method against 500 μL of well solution using 24-well crystallization plates. The
crystallization drops contained 1 μL of protein solution (10 mg/mL FBPase, 0.1 M NaCl,
1 mM AMP and 10 mM Mg2+) mixed with 1 μL of reservoir solution containing 0.05
M Tris (pH 6.8) and 10% (v/v) polyethylene glycol 8000. 214 was soaked into the drop
of APO FBPase crystals by adding 0.2 μL of the compound (5 mM) in the same
reservoir solution for 10 h.
Crystals were cryoprotected with 20% (v/v) glycerol and flash-cooled in liquid
nitrogen. The data were acquired at the Shanghai Synchrotron Radiation Facility on
beamline BL19U. The data were processed with XDS packages.35 Furthermore, the
structure was solved by molecular replacement using the CCP436 program and a
previous human FBPase structure (PDB ID: 5ZWK) was used as the search model.
Model analysis, building and refinement were performed with the Coot37 and Phenix38
programs.
ACS Paragon Plus Environment

Page 23 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Animals. Male ICR mice (6-7 weeks) and male BKS. Cg-
+Leprdb/+Leprdb/JclSlac mice (6-8 weeks) were purchased from Shanghai JSJ Lab
Animal, Ltd. and Shanghai SLAC Laboratory Animal Co., respectively. The animals
were housed under specific pathogen-free conditions with a 12-h light-dark cycle, at a
temperature of 25 °C, humidity of 55%, and free access to water and food. The animal
study was approved by the Animal Care and Use Committee of Central China Normal
University
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Glucose Reduction in ICR Mice. ICR mice that had been fasted for 12 h (n = 6
in each group) were administered vehicle (10% Ricinus oil in water), 212 (100 mg/kg,
p.o.), disulfiram (100 mg/kg p.o.) or compound 212 (25 mg/kg and 50 mg/kg, i.p.).
Food was withheld throughout the study. Blood samples were obtained from the tail
vein at 0, 2, 4, 6 and 8 h and were analyzed using a glucometer (Sanicare).
Glucose Reduction in db/db Mice. Freely fed db/db mice (n = 6 in each group)
were intraperitoneally administered compound 212 (25 mg/kg and 50 mg/kg),
metformin (250 mg/kg), or vehicle (10% Ricinus oil in water). Food was withheld after
the administration of the test compounds. Blood samples were obtained from the tail
vein at 0, 2, 4 and 6 h and were analyzed using a glucometer (Sanicare).
OGTT in ICR Mice and db/db Mice. ICR mice or db/db mice that had been
fasted for 12 h (n = 6 in each group) were intraperitoneally administered compound 212
(25 mg/kg and 50 mg/kg), metformin (250 mg/kg), or vehicle (10% Ricinus oil in water)
1.5 h before glucose challenge. Glucose (2 g/kg) was orally administered at 0 h, and
blood samples were drawn from the tail vein at 0, 0.5, 1, 2, 3, and 4.5 h after glucose
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 35
1
2
3
4
5
6
7
administration. Plasma glucose was measured using a glucometer (Sanicare). Food was
withheld throughout the study.
8
9
Inhibition of FBPase ex vivo. ICR mice received the test compound via i.p.
administration and were euthanized by decapitation after 2 h. The liver was removed,
weighed, and homogenized in double distilled water. Blood samples were obtained
from the tail vein at 0 h and 2 h. The activities of liver FBPase were tested as previously
described.1 Blood glucose was measured using a glucometer, and blood lactate was
measured using a Lactic Acid LD Assay Kit (Nanjing Jiancheng Bioengineering
Institute).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mass Spectrometry. WT FBPase (50 μM) or C128S FBPase (50 μM) was
incubated with 500 μM disulfiram and 212 for 2 h. Then, the samples were analyzed
using an Agilent 6550 quadrupole-time-of-flight (QTOF) mass spectrometer (Santa
Clara, CA) coupled with an Agilent 1260 high-performance liquid chromatograph
(HPLC; Santa Clara, CA).
The samples were separated on a Phenomenex Jupiter C4 300 Å LC column
(2×150 mm, 5 μm) over 15 min using a gradient from 5% to 100% acetonitrile
containing 0.1% formic acid at a flow rate of 0.5 mL/min. Data were acquired in
positive-ion mode with Dual Agilent Jet Stream electrospray voltage using a capillary
temperature of 250 °C, a fragmentor voltage of 175 V, a capillary voltage of 3000 V,
and a m/z 600−3200 mass window. Mass deconvolution was then performed using
Agilent Mass Hunter Qualitative Analysis B.06.00 software.
Molecular Dynamic Simulations. The crystal structures of the apo state (1FTA), FBP-
ACS Paragon Plus Environment

Page 25 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
bound state (5ZWK) and FBP/214b-bound state obtained in this study were used as
templates to set up the system for MD simulations. The CHARMM force field39 and
CGENFF force field40 were used for the protein and the ligands (including FBP and
214b), respectively. The whole system was surrounded by TIP3P water molecules41 in
a truncated octahedral box 10 Å from any solute atoms. The system was neutralized
with K+ counterions. MD simulations of 100 ns were performed at 300 K. LINCS42 was
used to constrain bonds involving hydrogen atoms, and the time step was 2.0 fs. The
nonbonded cutoff was set to 10 Å. The long-range electrostatics were calculated by the
particle mesh Ewald (PME) algorithm.43 Parallel simulations were performed
simultaneously using GROMACS 5.1.4.44 To characterize the changes in the secondary
structures at helix E, i.e., residues 120-129, the collective variable ALPHARMSD32
was used to probe the number of six-residue segments in the α-helical configuration.
ASSOCIATED CONTENT
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting information.
The Supporting Information is available free of charge at (http://pubs.acs.org).
1
Molecular formula strings (CSV); molecular dynamic results (PDB); H NMR, 13C
NMR, and high-resolution mass spectra (HRMS) data; HPLC spectra data for 212/214;
WT and mutant enzyme data and proposed reaction mechanism calculation for 212
(PDF).
Accession code.
The atomic coordinates and structure factors have been deposited into the RCSB
Protein Data Bank with accession numbers 6LS5. Authors will release the atomic
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 35
1
2
3
4
5
6
7
coordinates and experimental data upon article publication.
AUTHOR INFORMATION
8
9
Corresponding Authors
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Jian Wan: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China; Email: jianwan@mail.ccnu.edu.cn; Phone: +86-27-67862022. Fax: +86-27-
67862022
Yanliang Ren: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China; E-mail: renyl@mail.ccnu.edu.cn; Phone: +86-27-67862022. Fax: +86-27-
67862022
Jian Li: Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East
China University of Science and Technology, Shanghai 200237, China; E-mail:
jianli@ecust.edu.cn; Phone: +86-21-64252584. Fax: +86-21-64252584.
Authors
Yunyuan Huang: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry
of Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
Yixiang Xu: Shanghai Key Laboratory of New Drug Design, School of Pharmacy, East
China University of Science and Technology, Shanghai 200237, China
Rongrong Song: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry
of Education, College of Chemistry, Central China Normal University, Wuhan 430079,
ACS Paragon Plus Environment

Page 27 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
China
6
7
8
9
Shuaishuai Ni: Cancer Institute, Longhua Hospital, Shanghai University of
Traditional Chinese Medicine, Shanghai, 200032, China
Jiaqi Liu: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yanhong Xu: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
Li Rao: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
Yingjie Wang: Shenzhen Bay Laboratory, Shenzhen 518055, Guangdong, China
Lin Wei: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
Lingling Feng: Key Laboratory of Pesticide & Chemical Biology (CCNU), Ministry of
Education, College of Chemistry, Central China Normal University, Wuhan 430079,
China
Chen Su: National Facility for Protein Science in Shanghai, Zhangjiang Lab, Shanghai,
201210, China
Chao Peng: National Facility for Protein Science in Shanghai, Zhangjiang Lab,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 35
1
2
3
4
5
Shanghai, 201210, China
6
7
Author Contributions
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
All authors contributed to the writing of the manuscript and have approved the
#
final version of the manuscript. These authors contributed equally. (Match
statement to author names with a symbol)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the Natural Science Foundation of China (Nos.
21877046, 21873035, 21572077 and 21472061), the Program for the PCSIRT
(No. IRT0953), and the self-determined research funds of CCNU from the
colleges’ basic research and operation of MOE (Nos. CCNU19TS011,
CCNU18TS010, 2018YBZZ019 and CCNU16A02041), the support from the
Program of Introducing Talents of Discipline to Universities of China (111
Program, B17019) is also appreciated. We thank the staff of the BL19U1
beamline of the NCPSS at the Shanghai Synchrotron Radiation Facility for
assistance during data collection.
Abbreviations
FBPase: fructose-1,6-bisphosphatase; FBP: fructose-1,6-biphosphate; F6P: fructose-6-
biphosphate; IC50: half maximal inhibitory concentration; T2D: Type 2 diabetes; AMP:
adenosine monophosphate; ICR: Institute of Cancer Research; OGTT: oral glucose
ACS Paragon Plus Environment

Page 29 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
tolerance test; LC-MS: liquid chromatography−mass spectrometry; GNG:
gluconeogenesis;
8
9
REFERENCES
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1) Hunter, R. W.; Hughey, C. C.; Lantier, L.; Sundelin, E. I.; Peggie, M.; Zeqiraj, E.; Sicheri, F.; Jessen,
N.; Wasserman, D. H.; Sakamoto, K. Metformin reduces liver glucose production by inhibition of
fructose-1-6-bisphosphatase. Nat. Med. 2018, 24, 1395-1406.
(2) Li, B.; Qiu, B.; Lee, D. S.; Walton, Z. E.; Ochocki, J. D.; Mathew, L. K.; Mancuso, A.; Gade, T. P.;
Keith, B.; Nissim, I.; Simon, M. C. Fructose-1,6-bisphosphatase opposes renal carcinoma progression.
Nature 2014, 513, 251-255.
(3) Huangyang, P.; Li, F.; Lee, P.; Nissim, I.; Weljie, A. M.; Mancuso, A.; Li, B.; Keith, B.; Yoon, S.
S.; Simon, M. C. Fructose-1,6-Bisphosphatase 2 Inhibits Sarcoma Progression by Restraining
Mitochondrial Biogenesis. Cell Metab. 2020, 31, 174-188.
(4) Dong, C.; Yuan, T.; Wu, Y.; Wang, Y.; Fan, T. W.; Miriyala, S.; Lin, Y.; Yao, J.; Shi, J.; Kang, T.;
Lorkiewicz, P.; St Clair, D.; Hung, M. C.; Evers, B. M.; Zhou, B. P. Loss of FBP1 by Snail-mediated
repression provides metabolic advantages in basal-like breast cancer. Cancer Cell 2013, 23, 316-331.
(5) Cong, J.; Wang, X.; Zheng, X.; Wang, D.; Fu, B.; Sun, R.; Tian, Z.; Wei, H. Dysfunction of Natural
Killer Cells by FBP1-Induced Inhibition of Glycolysis during Lung Cancer Progression. Cell Metab.
2018, 28, 243-255.
(6) Guo, B.; Huang, X.; Lee, M. R.; Lee, S. A.; Broxmeyer, H. E. Antagonism of PPAR-gamma
signaling expands human hematopoietic stem and progenitor cells by enhancing glycolysis. Nat. Med.
2018, 24, 360-367.
(7) Rines, A. K.; Sharabi, K.; Tavares, C. D. J.; Puigserver, P. Targeting hepatic glucose metabolism in
ACS Paragon Plus Environment