Y-shaped bis-arylethenesulfonic acid esters: Potential potent and membrane permeable protein tyrosine phosphatase 1B inhibitors

Article abstract of DOI:10.1016/j.bmcl.2017.03.060

Known PTP1B inhibitors with bis-anionic moieties exhibit potent inhibitory activity, good selectivity, however, they are incapable of penetrating cellular membranes. Based upon our finding of a new pharmacophoric group in inhibition of PTP1B and the struc

Full text of DOI:10.1016/j.bmcl.2017.03.060

Bioorganic & Medicinal Chemistry Letters 27 (2017) 2166–2170
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Y-shaped bis-arylethenesulfonic acid esters: Potential potent and
membrane permeable protein tyrosine phosphatase 1B inhibitors
Fengzhi Yang a,b, Fangzhou Xie , Ying Zhang , Yu Xia , Wenlu Liu , Faqin Jiang , Celine Lam , Yixue Qiao ,
a
a
a,c
a
a
a
a
a,
⇑
b,
⇑
a,
⇑
Dongsheng Xie , Jianqi Li , Lei Fu
a
School of Pharmacy, Shanghai Jiao Tong University, Shanghai 200240, PR China
China State Institute of Pharmaceutical Industry, Novel Technology Center of Pharmaceutical Chemistry, Shanghai Institute of Pharmaceutical Industry, Shanghai 201203, PR China
Viva Biotech (Shanghai) Limited, Shanghai 201203, PR China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Known PTP1B inhibitors with bis-anionic moieties exhibit potent inhibitory activity, good selectivity,
however, they are incapable of penetrating cellular membranes. Based upon our finding of a new phar-
macophoric group in inhibition of PTP1B and the structural characteristics of the binding pocket of PTP1B,
a series of bis-arylethenesulfonic acid ester derivatives were designed and synthesized. These novel
molecules, particularly Y-shaped bis-arylethenesulfonic acid ester derivatives, exhibited high PTP1B inhi-
Received 7 December 2016
Revised 7 March 2017
Accepted 22 March 2017
Available online 23 March 2017
bitory activity, moderate selectivity, and great potential in penetrating cellular membranes (compound
Keywords:
Type 2 diabetes
PTP1B inhibitors
Phosphotyrosine mimics
Arylethenesulfonic acid esters
-6
7p, CLogP = 9.73, Papp = 9.6 ꢀ 10 cm/s; IC50 = 140, 1290 and 920 nM on PTP1B, TCPTP and SHP2, respec-
tively). Docking simulations suggested that these Y-shaped inhibitors might interact with multiple sec-
ondary binding sites in addition to the catalytic site of PTP1B.
Ó 2017 Published by Elsevier Ltd.
Substantial evidence indicated that protein tyrosine phos-
phatase 1B (PTP1B), the first isolated member of the PTP family,
negatively regulate insulin signal transduction by directly or indi-
rectly dephosphorylating the phosphotyrosine of many related sig-
nal proteins, such as insulin receptor (IR), insulin receptor kinase
membrane permeability, efforts were focused on developing
uncharged or nonionic PTyr mimics.
Some new scaffolds of inhibitors emerged in response to this
7
demand, such as thioxothiazolidinone derivative (Fig. 1, IV) and
8
sulfonamide derivatives (Fig. 1, V & VI) despite the fact that phe-
(
IRK), and insulin receptor substrates (IRS) etc. PTP1B is a potential
nolic hydroxyl and diimide moieties could be ionized in basic
water solution. Moreover, our previous research indicated that
arylethenesulfonic acid ester derivatives (Fig. 1, VII & VIII) , not
being hydrolyzed and ionized at physiological pH, could also exhi-
bit comparable activity to those ionic PTP1B inhibitors. This sug-
gested that sulfonic acid ester is an effective bioisostere of
phosphoric acid and carboxylic acid moiety in the inhibition of
PTP1B. These results prompted us to further develop more potent
PTP1B inhibitors with improved membrane permeability.
1
drug target for the treatment of type 2 diabetes and obesity.
Due to promising application for type 2 diabetes, PTP1B inhibi-
tors have become an attractive research area over the past several
decades. Zhang and coworkers reported, a phosphotyrosine (PTyr)
mimic in 1997, and it was the first dimeric type of PTP1B inhibitor
with low molecular weight but high affinity (BPPM, Fig. 1, I).
number of PTyr mimics with bis-anionic moieties were later dis-
covered, including bis- -difluoromethylphosphonic acid (DFMP,
Fig. 1, II) and bis-benzoylformic acid (Fig. 1, III). They exhibited
more potent activity and better selectivity than their analogues
with single anionic moiety. Nevertheless, these anionic inhibitors
which matched well with the positively charged catalytic site of
PTP1B, suffered from an apparent drawback that they are incapable
9
2
3
A
a,a
4
5
In addition, the catalytic site of PTP1B, where PTyr residues of
the IRK activation peptide are dephosphorylated, is a highly con-
served site of all PTP family members, thus making it difficult to
1
0
develop selective inhibitors. However, complex crystals and
molecular docking simulations of some inhibitors with branched
linkers revealed multiple secondary binding sites (B, C, D and/or
6
of penetrating cellular membranes. In order to improve molecular
1
1
E site) around the catalytic site (A site) of PTP1B. Therefore, a fea-
sible method to design potent and selective PTP inhibitors with
improved membrane permeability is to engage both the active site
and peripheral binding sites thorough tethering appropriately
⇑
Corresponding authors.
E-mail addresses: dshxie@sjtu.edu.cn (D. Xie), li.jianqi@sipi.com.cn (J. Li),
1
2
functionalized moieties to a nonionic PTyr mimic.
leifu@sjtu.edu.cn (L. Fu).
http://dx.doi.org/10.1016/j.bmcl.2017.03.060
0
960-894X/Ó 2017 Published by Elsevier Ltd.

F. Yang et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 2166–2170
2167
O
HO
F
O
P
O
P
O
OH
O
O
F
CO2H
O
O
H
N
NH2
N
H
O
O
O
O
O
O
F
HO
O
O
O
O
P
O
N
H
Ph
O
P
F
OH
CO2H
OH
BPPM
I
DFMP
II
III
Km = 16 µM
K = 2.4 nM (PTP1B)
i
IC50 = 220 nM (PTP1B)
IC50 = 6.5 µM (TCPTP)
K = 26 nM (TCPTP)
i
O
N
N
H
OMe
OH
O
O
N
S
O
N
S
H
O
Cl
F
S
O
S
NO2
O
O
IV
NH
HN
O
O
NH
S
S
O
IC50 = 4.1 µM
CF3
F3C
VI
V
O
IC50 = 7.7 µM
IC50 = 18 nM
Ph
O
O
O
O
N
S
S
O
N
S
O
N
HN
H
N
S
OMe
O
VII
IC50 = 1.9 µM
VIII
IC50 = 3.5 µM
Fig. 1. Representative PTP1B inhibitors.
In this report, we continued to design, synthesize and evaluate a
series of nonionic PTyr mimics with bis-arylethenesulfonic acid
ester moiety. Preliminary results indicated that Y-shaped bis-
arylethenesulfonic acid esters are potent, membrane permeable
and moderate selective PTP1B inhibitors, which could interact with
multiple peripheral binding sites (probably, C and E site) as well as
the catalytic site of PTP1B.
ner reaction by treating with ethyl (diethoxyphosphoryl) methane-
sulfonate (4) in the presence of NaH in THF at room temperature,
affording the target compound 5a in 94.0% yield. Other bis-
arylethenesulfonic acid esters (5b–5i) were prepared from 4-
hydroxybenzaldehyde (3) via two steps. Firstly, two molecules of
4-hydroxybenzaldehyde were coupled using dihalohydrocarbons
or dimethanesulfonates as linking reagents, affording bis-ben-
zaldehydes intermediates (2b–2i). Then 2b–2i underwent the
same reaction as 2a to provide bis-arylethenesulfonic acid ester
derivatives (5b–5i) with high yields (84.7%–95.6%). Subsequently,
bis-arylethenesulfonic acid ester derivatives (7a–7r) were pre-
pared from the above synthesized 5i via two steps (Scheme 2). 5i
was hydrolyzed with LiOH to give the benzoic acid intermediate
1
4
Strategies for synthesis of bis-arylethenesulfonic acid ester
derivatives (5a–5i and 7a–7r) are outlined in Schemes 1 and 2.
0
Compound 5a was prepared from 4,4 -oxydibenzoic acid (1)
through three step reactions. Thus, 1 was first reduced by LiAlH
to a diol intermediate, which was then oxidized by PCC to afford
4
0
13
4
,4 -oxydibenzaldehyde (2a). 2a underwent a simple Wittig-Hor-
O
CO H
2
O
O
i
S
O
O
O
P
O
O
O
O
=
R'
S
O
O
O
HO
2
C
2
a
1
4
O
ii
R'
O
O
i'
S
O
O
O
O
5
a-i
R
= R'
HO
O
3
2b-i
0
Scheme 1. The synthesis of bis-arylethenesulfonic acid ester derivatives 5a–5i. Reagents and conditions: (i) a: LiAlH
dihalohydrocarbons, K CO , ACN, refluxed, 16 h; or NaOH, DMSO, 100 °C, 8 h; (ii) NaH, THF, rt, 6 h.
4
, THF, 60 °C, 2 h; b: PCC, DCM, rt, 3 h; (i )
2
3

2
168
F. Yang et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 2166–2170
O
O
S
O
O
S
O O
S
O
O
O
O
O
O
O
i
ii
R"
CO
2
Me
2
CO H
O
O
O
O
O
O
S
O
S
O
S
O
O
O O
5
i
6
7a-r
Scheme 2. The synthesis of bis-arylethenesulfonic acid ester derivatives 7a–r. Reagents and conditions: (i) 1 N LiOH/THF, rt, 16 h; (ii) arylamines, HATU, DIPEA, DMF, rt, 6 h.
(
6) and the latter condensed with various arylamines in the pres-
ence of HATU to form bis-arylethenesulfonic acid ester derivatives
a–7r.
caused by linkers will reduce bis-arylethenesulfonic acid esters’
affinity.
7
Based on compound 5h which is a standard dimeric form, we
designed a set of branched bis-arylethenesulfonic acid ester
derivatives in order to discover additional binding interactions
with PTP1B. As shown in Table 2, short branch-chains such as
methoxycarbonyl (5i), carboxyl (6) and anilinocarbonyl groups
(7a–7d) couldn’t improve the inhibitory activity of bis-arylethene-
sulfonic acid esters. Further expansion of branch-chain with aryl
groups enhanced the PTP1B inhibitory activity of compound 5h
for 3–9 times, such as compounds 7j–7m and 7p. This suggests
additional interactions between these Y-shaped bis-arylethenesul-
fonic acid esters with multiple peripheral binding sites of PTP1B.
The design of PTP1B inhibitors and their SARs
Due to the higher potency exhibited by known PTP1B inhibitors
with bis-anionic moieties, several unbranched bis-arylethenesul-
fonic acid esters (5a–5h) were initially designed, which used dif-
ferent length and flexibility of alkylenedioxyl bridges (diether
linkages) as linker. As shown in Table 1, and with the exception
of compound 5e, all the other tested bis-arylethenesulfonic acid
ester derivatives showed moderate to potent PTP1B inhibitory
activity (IC50 = 1–40 lM). This proved true regardless of whether
the alkylenedioxyl bridges were linear alkylidenes, cyclohexyli-
dene or aralkylidenes. Increasing the length of linear alkylidene
from C0 to C2 slightly enhanced PTP1B inhibitory activity of bis-
arylethenesulfonic acid esters (5a/5b/5c). Further increasing the
length from C2 to C4 resulted in a rapid decline of molecular activ-
ity (5c/5d/5e). This seems to indicate that the length of alkylene-
dioxyl bridges exert a critical influence over the inhibitory
activity of these derivatives. However, 1,4-cyclohexylidene, with
similar length but different flexibility to 1,4-butylidene, endued
corresponding derivative (5f) with moderate activity. Further, lin-
ear aralkylidenes derivatives, particularly p-xylene derivatives (5h)
which have longer chain length, exhibited the most potent activity
The selectivity and membrane permeability of PTP1B inhibitors
For those Y-shaped bis-arylethenesulfonic acid esters with IC50
values below 500 nM on PTP1B, we further tested their inhibitory
activity on other PTPs (TCPTP and SHP2) and evaluated their mem-
brane permeability. As shown in Table 3, most of these Y-shaped
bis-arylethenesulfonic acid esters showed moderate selective inhi-
bitory activity for PTP1B over TCPTP and SHP2. Specifically, 5–9-
fold selectivity for PTP1B over TCPTP was observed in compounds
7j, 7k, 7l, 7m and 7p, and 5–6-fold selectivity for PTP1B over SHP2
was observed in compounds 7j, 7l, 7m and 7p. Furthermore, these
Y-shaped bis-arylethenesulfonic acid esters have significantly
higher CLogP values than that of bis-anionic PTP1B inhibitors such
as BBPM and DFMP (CLogP = 0.20 & 1.20 respectively). And as we
expected, sufficient membrane permeabilities of this kind of
inhibitors are demonstrated by parallel artificial membrane
(
IC50 = 1.2 lM). These results strongly suggested that the PTP1B
inhibitory activity of chain bis-arylethenesulfonic acid esters
depend largely on their spatial configurations rather than the
lengths of alkylenedioxyl bridges. Excess of molecular flexibility
Table 1
PTP1B inhibitory activities of bis-arylethenesulfonic acid esters.a
O
O
S
O O
S
O
O
R'
0
0
Cmpd
R
IC50
(lM)
Cmpd
R
IC50 (lM)
5
5
5
a
b
c
O
O
O
11.8
5e
O
O
O
4
>100
29.2
18.4
O
1
O
2
8.7
3.0
5f
O
5g
O
O
O
O
5
d
O
O
37.3
5h
1.2
3
a
Values are means of triplicates, repeated two times.

F. Yang et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 2166–2170
2169
Table 2
PTP1B inhibitory activities of branched bis-arylethenesulfonic acid ester derivatives.
a
O
O
S
O
O
O
R"
O
O
S
O
O
0
0
00
R
Cmpd
R
IC50
3.4
(lM)
Cmpd
IC50 (lM)
5
i
OCH
3
7i
H
N
0.87
3
F C
O
6
OH
1.3
1.4
2.6
7j
H
N
0.35
0.40
0.16
O
CF
Cl
3
7
7
a
H
N
7k
7l
H
N
O
b
H
N
H
N
Cl
Br
O
O
OMe
7
7
c
H
N
1.3
1.1
7m
7n
H
N
0.19
0.75
Et
d
H
N
H
N
i-Pr
O
F
7
7
7
7
e
f
H
N
0.86
1.3
7o
7p
7q
7r
H
N
0.91
0.14
1.1
t-Bu
Ph
O
O
O
O
H
N
H
N
F
O
F
g
h
H
N
0.85
0.82
H
N
O
Ph
H
N
H
N
0.79
CF
3
O
O
a
Values are means of triplicates, repeated two times.
permeation assay (PAMPA).15 Compared with DFMP which is one
of the most potent PTP1B inhibitors so far, compound 7p, despite
relatively weak potency (IC50 = 140 nM), not only exhibit similar
selectivity for PTP1B but have a great potential in penetrating
cellular membranes.
active site (A site), forming several hydrogen bonds and hydropho-
bic interactions with key residues including Ala215, Arg221,
Gly218, Ile219, Gly220 and the backbone NHs of Ser216. The phe-
nyl ring (A1 cycle) interacts by
p-p stacking interaction with the
phenyl ring of Tyr46, and the phenyl ring (A2 cycle) interacts with
the phenyl ring of Phe182. The sulfonic acid ester group forms
hydrogen bond with the backbone NHs of Lys36 in the second sul-
fonic acid ester-binding site (C site). Additionally, a hydrophobic
interaction may be established between the benzyl ring (E1 and
E2 ring) of the ligands and Thr263 and Ala264 in E site. These mul-
tiple interactions contribute to improvements of the inhibitory
activity and selectivity of compound 7p.
Molecular docking simulation
Molecular docking of the most potent and selective compound
7
1
p for PTP1B was performed on GOLD 5.0.1 and presented in PyMol
.1r1. As showed in Fig. 2, compound 7p extended deep into the

2
170
F. Yang et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 2166–2170
Table 3
The inhibitory activities and membrane permeability of partial Y-shaped bis-arylethenesulfonic acid esters on PTP1B, TCPTP and SHP2.
PTP1B IC50 (nM)a
TCPTP IC50 (nM)a
SHP2 IC50 (nM)a
CLogPb
P
app (10 cm/s)c
ꢁ6
Cmpd
Na VO
BBPM
DFMP
3
4
46
16000
2.4
14
NT
NT
NT
4700
1700
940
1050
1100
920
–
ꢁ0.92
0.20
1.20
5.58
8.73
8.56
9.19
8.87
9.73
–
ND
ND
ND
ND
6.9
9.1
12.6
8.5
9.7
1.5
18.7
d
NT
e
c
26
5
7
7
7
7
7
h
j
k
l
m
p
1200
350
400
160
190
140
–
3390
1620
2970
1430
1320
1290
–
f
Atenolol
Propranololf
–
–
–
–
a
b
c
d
e
f
Values are means of triplicates, repeated two to three times.
Predicted by ChemBioDraw Ultra 13.0.
Values of parallel artificial membrane permeation assay (PAMPA).
K
K
m
value, reported by literature.3
values, reported by literature.4
i
Controls in PAMPA.
(
b)
O
S
O
O
Lys36
O
E2
Ala264
Tyr46
Ile219
E1
O
O
N
A2
Thr263
A1
O
H
Gly218
Gly220
Phe182
O
O
S
O
Arg221
Ser216 Ala215
Fig. 2. Docking of compound 7p to PTP1B (PDB code 1G1H). (a) Binding conformation of compound 7p in PTP1B. (b) Schematic representation of the interactions between
compound 7p and PTP1B. Hydrogen bonds are depicted with red arrows, and
p-
p
interactions or hydrophobic interactions are depicted with brown discs.
In summary, the design and synthesis of a series of bis-ethene-
sulfonic acid ester derivatives as potential PTP1B inhibitors, pos-
sessing high potency and excellent membrane permeability, was
carried out. Preliminary research demonstrated that bis-ethenesul-
fonic acid esters, particularly Y-shaped bis-ethenesulfonic acid
ester derivatives are potent and selective PTP1B inhibitors (com-
pound 7p, IC50 = 140, 1290 and 920 nM on PTP1B, TCPTP and
SHP2, respectively). More importantly, they have a great advantage
in penetrating cellular membranes. This provides useful informa-
tion for further design and discovery of new PTP1B inhibitors with
high potency, selectivity and good membrane permeability.
2. (a) Andrew PC. J Med Chem. 2010;53:2333;
(
b) Joo-Leng L, Christina LLC, Shao QY. Antioxid Redox Signal. 2014;20:2225.
3.
Puius YA, Zhao Y, Sullivan M, Lawrence DS, Almo SC, Zhang ZY. Proc Natl Acad
Sci USA. 1997;94:13420.
4. Shen K, Keng YF, Wu L, Guo XL, Lawrence DS, Zhang ZY.
001;276:47311.
(a) Anthony BC, David AC, Rebecca P, Christopher TS.
010;53:6768;
J
Biol Chem.
2
5.
J Med Chem.
2
(b) Jian X, Christopher TS. Bioorg Med Chem. 2005;13:2981.
(a) Barr AJ. Future Med Chem. 2010;2:1563;
6
.
.
(
b) Iversen LF, Møller KB, Pedersen AK, et al. J Biol Chem. 2002;277:19982.
7
Stuible M, Zhao L, Aubry I, et al. Chem Biol Chem. 2007;8:179.
8. (a) Huang P, Ramphal J, Wei J, et al. Bioorg Med Chem. 2003;11:1835;
(
(
b) Combs AP, Zhu W, Crawley ML, et al. J Med Chem. 2006;49:3774;
c) Sparks RB, Polam P, Zhu W, et al. Bioorg Med Chem Lett. 2007;17:736.
9.
(a) Liu J, Jiang F, Jin Y, et al. Eur J Med Chem. 2012;57:10;
(b) Deng X, Liu J, Jin Y, et al. Pharmazie. 2015;70:777;
Acknowledgements
(
c) Liu J, Deng X, Jin Y, et al. Pharmazie. 2015;70:446.
1
1
0. Barr AJ, Ugochukwu E, Lee WH, et al. Cell. 2009;136:352.
1. Ala PJ, Gonneville L, Hillman M, et al. J Biol Chem. 2006;281:38013.
We gratefully acknowledge Dr. Wei Fu of Fudan University
China) for Molecular Docking Simulation assistance.
(
12. (a) Liu P, Du Y, Song L, Shen J, Li Q. Bioorg Med Chem. 2015;23:7079;
(b) Du Y, Ling H, Zhang M, Shen J, Li Q. Bioorg Med Chem. 2015;23:4891;
(c) Liu P, Du Y, Song L, Shen J, Li Q. Eur J Med Chem. 2016;118:27;
(d) Zeng L, Zhang R, Yu Z, et al. J Med Chem. 2014;57:6594.
A. Supplementary data
1
1
3. Gurung SK, Kim SB, Park H. Arch Pharm Res. 2010;33:1919.
4. (a) Lan Y, Li S, Qin J, et al. Inorg Chem. 2008;47:10600;
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.03.
(
b) Jiang W, Nowosinski K, Loew NL, et al. J Am Chem Soc. 2012;134:1860.
1
5. (a) Basu S, Prathipati P, Thorat S, et al. Bioorg Med Chem. 2017;25:67;
0
60.
(b) Chen X, Murawski A, Patel K, Crespi CL, Balimane PV. Pharm Res.
2008;25:1511.
References
1
.
(a) Elchebly M, Payette P, Michaliszyn E, et al. Science. 1999;283:1544;
b) Klaman LD, Boss O, Peroni OD, et al. Mol Cell Biol. 2000;20:5479.
(