Discovery of novel PTP1B inhibitors via pharmacophore-oriented scaffold hopping from Ertiprotafib

Article abstract of DOI:10.1016/j.bmcl.2013.10.002

An integrated molecular design strategy combining pharmacophore recognition and scaffold hopping was exploited to discover novel PTP1B inhibitors based on the known PTP1B inhibitor Ertiprotafib. A composite pharmacophore model was proposed from the intera

Full text of DOI:10.1016/j.bmcl.2013.10.002

Bioorganic & Medicinal Chemistry Letters 23 (2013) 6217–6222
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of novel PTP1B inhibitors via pharmacophore-oriented
scaffold hopping from Ertiprotafib
Jun-Zheng Liu a, , Shu-En Zhang a, , Feilin Nie a, Ying Yang a, Yan-Bo Tang a, Wenwen Yin a, Jin-Ying Tian b,
Fei Ye b, Zhiyan Xiao a,
⇑
a Beijing Key Laboratory of Active Substance Discovery and Druggability Evaluation, Institute of Materia Medica, Chinese Academy of Medical Sciences and
Peking Union Medical College, Beijing 100050, PR China
b Beijing Key Laboratory of New Drug Mechanisms and Pharmacological Evaluation Study, Institute of Materia Medica, Chinese Academy of Medical Sciences and
Peking Union Medical College, Beijing 100050, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
An integrated molecular design strategy combining pharmacophore recognition and scaffold hopping
was exploited to discover novel PTP1B inhibitors based on the known PTP1B inhibitor Ertiprotafib. A
composite pharmacophore model was proposed from the interaction mode of Ertiprotafib, and 21 diverse
molecules from five distinct structural classes were designed and synthesized accordingly. New com-
pounds with considerable inhibition against PTP1B were identified from each series, and the most active
Received 13 August 2013
Revised 29 September 2013
Accepted 1 October 2013
Available online 8 October 2013
compound 3a showed IC50 value of 1.3
cated that the new inhibitors assumed binding modes similar to that of Ertiprotafib.
l
mol LÀ1 against human recombinant PTP1B. Docking study indi-
Keywords:
PTP1B inhibitors
Pharmacophore
Scaffold hopping
Ertiprotafib
Ó 2013 Elsevier Ltd. All rights reserved.
Protein tyrosine phosphatases (PTPs) and protein tyrosine
kinases (PTKs) serve as molecular switches for a variety of func-
tional proteins by regulating their phosphorylation state.1 As a
member of the PTPs family, protein tyrosine phosphatase 1B
(PTP1B) downregulates the insulin signaling pathway by dephos-
phorylating the activated insulin receptor (IR) or insulin receptor
substrates (IRS). It is also a negative regulator in the leptin signal-
ing pathway and dephosphorylates JAK2, a kinase downstream to
the leptin receptor. Therefore, PTP1B becomes an emerging thera-
peutic target for type 2 diabetes and associated obesity.2–4
A large variety of PTP1B inhibitors have been discovered with
the aid of modern drug design techniques, such as structure-based
and fragment-based drug design techniques. Most known PTP1B
inhibitors are phosphotyrosine (pTyr) mimetics binding to the cat-
alytic site of PTP1B.3,4 However, the catalytic site of PTP1B is highly
conserved and positively charged, which makes inhibitors target-
ing this site difficult to achieve selectivity and oral availability.
Up to now, only two small-molecule PTP1B inhibitors, Ertiprotafib5
and Trodusquemine6 (Fig. 1), have reached clinical trials.
profiles via multiple mechanisms.7,8 Structurally, Ertiprotafib
incorporates hydrophobic fragments to modulate the overall polar-
ity and cellular permeability of the whole molecule, and it also
introduces a benzyl group to interact with a hydrophobic subpock-
et, which is adjacent to the active site of PTP1B and less conserva-
tive among PTPs, so as to improve affinity and possibly selectivity.
Although Ertiprotafib was discontinued at phase II clinical trials
due to insufficient efficacy and unwanted side effects,5 it pioneered
a valid approach to balance the structural requirements for high
affinity, acceptable selectivity and good oral availability. Therefore,
Ertiprotafib could serve as a chemical template for further efforts
to search for new PTP1B inhibitors. We report herein the discovery
of novel PTP1B inhibitors by exploiting an integrated molecular de-
sign strategy of pharmacophore-oriented scaffold hopping based
on the template structure of Ertiprotafib.
The interaction mode of Ertiprotafib with PTP1B was first
simulated by docking study following the previous reports on
similar structures.9,10 As illustrated in Figure 2a, the carboxylic acid
moiety of Ertiprotafib forms electrostatic interaction and hydrogen
bond networks with the catalytic site of PTP1B. The benzyl group
protrudes into a hydrophobic pocket close to the catalytic active
Ertiprotafib is a mono carboxylic acid pTyr mimetic developed
by Wyeth. It was initially discovered as a potent PTP1B inhibitor,
and was later reported to improve glycemic control and lipid
site, possibly producing
The fused aromatic rings present cation–
p
–
p
stacking interaction with Phe182.
interactions with
p
Lys120 and probably also with Lys116. Accordingly, a composite
pharmacophore model for Ertiprotafib was proposed, which fea-
tures two aromatic rings (in purple), one hydrophobic moiety (in
blue) and one carboxyl group (in red) (Fig. 2b).
⇑
Corresponding author. Tel./fax: +86 10 63189228.
E-mail address: xiaoz@imm.ac.cn (Z. Xiao).
These authors contributed equally.
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.10.002

6218
J.-Z. Liu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6217–6222
OSO3H
O
COOH
HN
OH
H
N
NH2
S
N
H
Br
Ertiprotafib
Trodusquemine
Figure 1. Structures of Ertiprotafib and Trodusquemine.
Br
S
HOOC
O
a
b
Figure 2. (a) The interaction mode of Ertiprotafib with PTP1B revealed by docking study; (b) mapping of pharmacophoric features with the structure of Ertiprotafib.
Rc
O
COOH
HN
COOH
COOH
A
III
I
II
H
H
N
HOOC
O
N
COOH
COOH
linker
B
IV
V
b
a
Figure 3. (a) General formula of target compounds, and (b) pTyr mimetic scaffolds.
Guided by the pharmacophore model, target compounds with
the general formula in Figure 3a were designed. Basically, the phar-
macophoric features in Ertiprotafib were retained, instead, the
scaffold was altered to ensure structural novelty and diversity.
Various pTyr mimetics with a carboxyl group were generated by
introducing different linkages between the phenyl ring A and the
carboxyl group. Aromatic rings, especially electron-rich aromatic
OH
O
COOCH3
O
COOH
b, c
a
Rc
6
Rc
Rc
rings, were introduced at Rc to enhance possible cation–
p interac-
7
1
tions with Lys120 and Lys116. An additional unsubstituted phenyl
ring B was kept in the neighborhood of the carboxyl group to as-
sume interaction with the hydrophobic subpocket close to the ac-
tive site. The molecular design followed the general strategy
Scheme 1. Synthesis of compound 1. Reagents and conditions: (a) (R)-2-hydroxy-
3- phenylpropanoate methyl ester, Ph3P, diethyl azodicarboxylate, THF, reflux 14 h
(for 1a) or ultrasonic 1 h (for 1b–1e); (b) KOH, C2H5OH, H2O, rt 6 h; (c) 5%
Hydrochloric acid.

J.-Z. Liu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6217–6222
6219
O
O
O
O
COOC2H5
COOC2H5
COOH
COOH
CHO
COOH
c, d
e
a
b
Rc
Rc
8
Rc
9
Rc
10
Rc
11
2a, 2b, 2e
O
O
CHO
COOH
COOH
a - e
f - h
OH
12
OH
13
Rc
2c
O
O
O
CHO
COOH
b -e
a
i, j
Rc
Rc
CH3
15
CH3
14
16
2d
Scheme 2. Synthesis of compound 2. Reagents and conditions: (a) Acetophenone, NaOH, C2H5OH, rt 12 h; (b) Diethyl malonate, KOH, C2H5OH, Ultrasonic, 1–3 h; (c) KOH,
CH3OH, H2O, rt 6 h; (d) 5% Hydrochloric acid; (e) 15% Hydrochloric acid or 10% H2SO4, reflux, 5–7 h; (f) concd H2SO4, MeOH, reflux, 5 h; (g) Benzhydrol, Ph3P, diethyl
azodicarboxylate, THF, ultrasonic, 30 min; (h) NaOH, MeOH, H2O, rt 4 h; then 5% Hydrochloric acid; (i) NBS, CCl4, hv, reflux, 7 h; (j) N-(benzo[d][1,3]dioxol- 5-ylmethyl)-4-
fluoroaniline, DMF, K2CO3, rt, 48 h.
X
X
NH2
HN COOC2H5
N
N
COOC2H5
COOH
c, d
a
b
Rc
17
Rc
18
Rc
19
Rc
3
Scheme 3. The synthesis of compound 3. Reagents and conditions: (a) Ethyl bromoacetate, K2CO3, DMF; (b) For 3a, X = COCH2, phenylacetyl chloride, Et3N, CH2Cl2; For 3b,
X = CH2, benzyl bromide, K2CO3, DMF; (c) KOH, C2H5OH, H2O; (d) 5% Hydrochloric acid.
applied in the design of Ertiprotafib, and fragments could possibly
improve cellular permeability and selectivity over PTP1B were in-
tently incorporated into the target structures. During this hit iden-
tification stage, the exploration of appropriate pTyr mimetics was
emphasized. Besides the 2-phenoxyacetic acid scaffold (I) used in
Ertiprotafib, 3-phenylpropanoic acid (II), 2-(phenyl-amino)acetic
acid (III), 2-benzamidoacetic acid (IV) and 2-(benzylamino)acetic
acid (V) (Fig. 3b) were also investigated as potential pTyr mimetics
to expand diversity and improve novelty of PTP1B inhibitors. Since
complementarity of the pTyr mimetic moieties to the active site
was the major concern at this stage, unsubstituted phenyl rings A
and B were generally used, and building blocks for Rc were mainly
selected based on the in-house availability. Optimization of substit-
uents on the aromatic rings will leave to further hit evolution.
COOH
NH
N
COOC2H5
N
COOH
O
NH
CHO
COOH
COCl
COOCH3
a
c
d, e
a, b
c, d
b
Rc
8
Rc
23
Rc
24
Rc
Rc
20
Rc
21
Rc
22
Rc
4
5
Scheme 4. The synthesis of compound 4. Reagents and conditions: (a) KOH,
Scheme 5. The synthesis of compound 5. Reagents and conditions: (a) Phenylm-
ethylamine, C2H5OH, reflux; (b) NaBH4, C2H5OH; (c) Ethyl bromoacetate, K2CO3,
DMF; (d) KOH, C2H5OH, H2O; (e) 5% Hydrochloric acid.
C2H5OH, H2O; (b) Oxalyl dichloride, CH2Cl2, DMF; (c)
(d) 5% Hydrochloric acid.
L-phenylalanine, NaOH, H2O;

6220
J.-Z. Liu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6217–6222
Table 1
Percentage inhibition of target compounds against human recombinant PTP1B
Compd
Rc
Inhibitiona (%)
Compd
Rc
Inhibitiona (%)
F
N
O
COOH
Br
Br
O
Positive control
100.0
3a
96.4
O
O
Br
F
N
N
1a
98.4
3b
44.3
O
O
F
N
N
N
1b
1c
1d
111.8
NAb
NA
4a
4b
4c
51.8
83.8
95.0
O
O
MeO
N
O
COOH
MeO
N
N
O
O
O
F
O
N
N
1e
NA
4d
53.4
N
O
O
N
N
2a
2b
2c
106.1
98.0
61.0
4e
4f
113.2
100.5
NA
MeO
N
N
O
O
O
N
O
4g
F
N
N
2d
2e
59.1
5a
5b
67.3
O
O
MeO
MeO
N
O
N
O
NA
NA
O
O
O
O
a
The percentage inhibition was measured under the concentration of 100
NA: Not active, the percentage inhibition was less than 40% under the tested concentration.
l
M.
b

J.-Z. Liu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6217–6222
6221
Table 2
that all the five scaffolds could be used as pTyr mimetics. This
hypothesis was further supported by docking studies on com-
pounds 1b, 2b, 3a and 4e, which indicated that their interaction
modes are similar to that of the template structure and the pTyr
mimetic moieties situated in the catalytic site of PTP1B as ex-
pected. The binding modes of the most active compounds 3a and
4e revealed by docking studies13 were illustrated in Figure 4. The
pTyr mimetic moieties of both 3a and 4e stretched into the active
site of PTP1B, and electrostatic interactions with Arg221 as well as
hydrogen bonding networks with the backbone amides were
observed. The phenyl ring B (cf. Fig. 3) in compounds 3a and 4e
extended into the hydrophobic subpocket near the catalytic site
as expected. Although the Rc part of both molecules presented in
PTP1B Inhibitory activity of selected compounds
Compd
IC50 M)
PC*
1a
1b
2a
2b
3a
4c
4e
(
l
0.87
40.3
10.4
10.2
10.0
1.3
23.9
3.9
*
PC: positive control.
Twenty-one compounds bearing five different pTyr mimetic
scaffolds were synthesized (Schemes 1–5). Briefly, the 2-phenoxy-
acetic acid series were prepared through Mitsunobu coupling
between phenolic derivatives and (R)-2-hydroxy-3- phenylpropan-
oate methyl ester either under reflux or ultrasonic condition
(Scheme 1). The 3-phenylpropanoic acid scaffold were generally
established via successive aldol condensation (step a), Michael
addition (step b), hydrolysis (steps c and d) and decarboxylation
(step e), although the timing and methods to introduce Rc varied
with different target compounds (Scheme 2). The synthesis of 2-
(phenylamino)acetic acid series was straightforward. Alkylation
of the aniline derivatives provided 2-(phenylamino)acetates,
which were subjected to further acylation (for 3a) or alkylation
(for 3b) to give the target compounds (Scheme 3). The 2-benzam-
idoacetic acid series were simply prepared through acylation of the
the vicinity of residues Lys120 and Lys116, cation–
with these residues were not monitored. Instead, cation–
tion between Arg45 and the benzo[d][1,3]dioxole moiety in Rc of
3a, as well as stacking interaction between Phe182 and the
p
interactions
p
interac-
p–p
benzyl group in Rc of 4e were perceived (Fig. 4). These observations
supported the molecular design to incorporate aromatic rings in Rc
area. Additional hydrogen bonding interactions were also detected
for both compounds. The carbonyl in the 2-phenylacetyl side chain
of compound 3a formed a hydrogen bond with Arg221, which at
least partly explained the activity difference between 3a and its
closely related compound 3b. The amide carbonyl in the 2-ben-
zamidoacetic acid moiety of compound 4e established a hydrogen
bond with Ala217 in the catalytic site, and it might contribute to
the relatively better activities of this compound.
free amino terminal in
a-amino acids (Scheme 4). The 2-(benzyl-
amino)acetic acid series were produced via reductive alkylation
with aldehyde and subsequent alkylation of the benzylamine
derivatives, followed by hydrolysis of the ester intermediates
(Scheme 5). The structures of the 21 target compounds were char-
acterized by their physical and spectral data.11
The target compounds were evaluated against human recombi-
nant PTP1B with protocols described previously,12 and an analogue
of Ertiprotafib9 was used as a positive control. All compounds were
initially evaluated for their percentage inhibition against PTP1B
Close examination of the biological results listed in Table 1 sug-
gested some preliminary structure–activity relationship (SAR)
clues. The data implicated that although all the five pTyr mimetic
scaffolds seemed to fit the catalytic site well, the variation in pTyr
mimetics might cause differences in inhibitory activity (1a vs 5a;
1b vs 3b). The linkage between pTyr mimetic scaffold and the phe-
nyl ring B could also affect the activity (3a vs 3b), which might be
attributed to the capability of 3a to form hydrogen bond with
Arg221, as suggested by the docking study. In addition, amide link-
ages either inside the Rc moiety (1c and 1d) or between Rc and the
phenyl ring A (1e, 2e and 4g) were obviously disfavored. As exem-
plified by compounds 1b and 1e, when the liankage between Rc
and the phenyl ring A simply changed from methylene (1b) to
carbonyl (1e), the inhibitory activity against PTP1B decreased
significantly. The discrepancy on the spacial orientation of
under the concentration of 100 lM (Table 1). Most compounds
showed apparent inhibition against PTP1B, and nine significantly
inhibited PTP1B with percentage inhibition higher than 80%. Seven
compounds were further measured for their IC50 values against
PTP1B. The two most active compounds, 3a and 4e, exhibited
IC50 values at lower micromolar level against human recombinant
PTP1B (Table 2), which is comparable to that of the positive
control.
Hit structures with considerable inhibition against PTP1B were
identified from each of the five structural classes, which suggested
Figure 4. Binding modes of compounds 3a and 4e suggested by docking studies.

6222
J.-Z. Liu et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6217–6222
(m, 4H), 7.29–7.17 (m, 4H), 7.09 (d, 2H, J = 8.4 Hz), 5.60 (s, 2H), 3.83–3.74 (m,
1H), 3.40 (d, 2H, J = 7.2 Hz), 2.77 (dd, 1H, J = 6.6, 15.9 Hz), 2.44 (dd, 1H, J = 9.6,
15.9 Hz); HR-ESI-MS: m/z 448.1901 [M+H]+ (calcd for C30H26NO3: 448.1912).
Compound 2c: Yield 68%; white powder; mp 154.2–155.6 °C; 1H NMR
(300 MHz, acetone-d6): d 7.96–6.89 (m, 19H), 6.39 (s, 1H), 3.78–3.73 (m, 1H),
3.39 (d, 2H, J = 7.2 Hz), 2.80–2.73 (dd, 1H, J = 6.6, 15.9 Hz), 2.65–2.56 (dd, 1H,
J = 8.1, 15.6 Hz); HR-ESI-MS: m/z 473.1711 [M+Na]+ (calcd for C30H26O4Na:
473.1723).
pharmacophoric features in the two compounds caused by the dif-
ferent hybridation status of the linking carbon atoms (sp3 in 1b and
sp2 in 1e) might be responsible for such activity difference. The
variation in the distance between the two aromatic rings in Rc
was generally well-accommodated (4b vs 4c, 4e), which is consis-
tent with the existence of alternative interaction modes of the aro-
matic rings as observed in 3a, 4e and Ertiprotafib. Conformational
restriction imposed on the two phenyl rings in Rc might be disfa-
vored (4a vs 4b).
In summary, pharmacophore-oriented scaffold hopping based
on the known PTP1B inhibitor Ertiprotafib has led to the discovery
of novel and potent PTP1B inhibitors with different pTyr mimetic
moieties. The results implicated the effectiveness of molecular de-
sign, and further structural optimization as well as SAR exploration
on these new hits is ongoing.
Compound 2d: Yield 41%; white powder; mp 131.1–132.6 °C; 1H NMR
(300 MHz, acetone-d6): d 10.55 (br s, 1H), 7.98–6.68 (m, 16H), 5.94 (s, 2H),
4.59 (s, 2H), 4.53 (s, 2H), 3.86–3.81 (m, 1H), 3.47–3.45 (d, 2H, J = 7.2 Hz), 2.86–
2.78 (dd, 1H, J = 6.6, 15.9 Hz), 2.72–2.64 (dd, 1H, J = 8.4, 15.9 Hz); FAB-MS: m/z:
526 [M+H]+.
Compound 2e: Yield 46%; white powder; mp 71.7–73.9 °C; 1H NMR (300 MHz,
acetone-d6): d 7.92–6.66 (m, 16H), 5.94 (s, 2H), 4.97 (s, 2H), 3.74 (dd, 1H,
J = 4.8, 14.4 Hz), 3.67 (s, 3H), 3.49–3.32 (m, 2H), 2.81–2.73 (dd, 1H, J = 6.9,
15.9 Hz), 2.64–2.56 (dd, 1H, J = 8.1, 15.9 Hz); FAB-MS: m/z: 552 [M+H]+.
Compound 3a: Yield 95%; white powder; mp 89–92 °C; 1H NMR (300 MHz,
acetone-d6): d 7.35–6.76 (m, 16H), 5.93 (s, 2H), 4.64 (s, 2H), 4.56 (s, 2H), 4.26 (s,
2H), 3.44 (s, 2H); HR-ESI-MS: m/z 527.1964 [M+H]+ (calcd for C31H28N2O5F:
527.1976).
Compound 3b: Yield 55%; white powder; mp 65.8–67.7 °C; 1H NMR (300 MHz,
acetone-d6): d 11.02 (br s, 1H), 7.33–6.64 (m, 16H), 5.94 (s, 2H), 4.65 (s, 2H),
4.50 (s, 4H), 4.20 (s, 2H); HR-ESI-MS: m/z 499.2023 [M+H]+ (calcd for
Acknowledgments
This research work was financially supported by National Nat-
ural Science Foundation of China (20972192) and Beijing Natural
Science Foundation (7102117).
C30H28N2O4F: 499.2033).
Compound 4a: Yield 61%; white solid; mp 167.7–170.1 °C; 1H NMR (300 MHz,
acetone-d6): d 8.19–7.12 (m, 17H), 5.70 (s, 2H), 4.90–4.83 (m, 1H), 3.33–3.26
(dd, 1H, J = 5.4, 14.1 Hz), 3.14–3.06 (dd, 1H, J = 9.3, 14.1 Hz); HR-ESI-MS: m/z
449.1843 [M+H]+ (calcd for C29H25N2O3: 449.1859).
Compound 4b: Yield 25%; white solid; mp 134.9–137.5 °C; 1H NMR (300 MHz,
acetone-d6): d 7.76–6.69 (m, 19H), 5.04 (s, 2H), 4.84 (m, 1H), 3.36–3.30 (dd, 1H,
J = 4.8, 13.8 Hz), 3.19–3.12 (dd, 1H, J = 8.7, 14.1 Hz); HR-ESI-MS: m/z 451.2018
[M+H]+ (calcd for C29H27N2O3: 451.2016).
References and notes
1. Barr, A. J. Future Med. Chem. 2010, 2, 1563.
2. Zhang, S.; Zhang, Z.-Y. Drug Discovery Today 2007, 12, 373.
3. Combs, A. P. J. Med. Chem. 2010, 53, 2333.
4. Thareja, S.; Aggarwal, S.; Bhardwaj, T. R.; Kumar, M. Med. Res. Rev. 2012, 32,
459.
5. Zhang, Z.-Y.; Lee, S.-Y. Expert Opin. Investig. Drugs 2003, 12, 223.
6. Lantz, K. A.; Emeigh Hart, S. G.; Planey, S. L.; Roitman, M. F.; Ruiz-White, I. A.;
Wolfe, H. R.; McLane, M. P. Obesity 2010, 18, 1516.
7. Erbe, D. V.; Wang, S.; Zhang, Y.-L.; Harding, K.; Kung, L.; Tam, M.; Stolz, L.; Xing,
Y.; Furey, S.; Qadri, A.; Klaman, L. D.; Tobin, J. F. Mol. Pharm. 2005, 67, 69.
8. Tong, Z.; Chandrasekaran, A.; Jordan, R.; Markiewicz, V.; Li, H.; Xiang, Q.; Shen,
L.; Scatina, J. Xenobiotica 2007, 37, 1.
9. Wrobel, J.; Sredy, J.; Moxham, C.; Dietrich, A.; Li, Z.; Sawicki, D. R.; Seestaller, L.;
Wu, L.; Katz, A.; Sullivan, D.; Tio, C.; Zhang, Z.-Y. J. Med. Chem. 1999, 42, 3199.
10. Wrobel, J.; Li, Z.; Sredy, J.; Sawicki, D. R.; Seestaller, L.; Sullivan, D. Bioorg. Med.
Chem. Lett. 2000, 10, 1535.
Compound 4c: Yield 53%; white powder; mp 76–78 °C; 1H NMR (300 MHz,
acetone-d6): d 7.81–7.19 (m, 19H), 4.90 (m, 1H), 3.59 (s, 2H), 3.55 (s, 4H), 3.37–
3.31 (dd, 1H, J = 8.1, 14.1 Hz), 3.20–3.13 (dd, 1H, J = 9.3, 14.1 Hz); HR-ESI-MS:
m/z 479.2328 [M+H]+ (calcd for C31H31N2O3: 479.2329).
Compound 4d: Yield 76%; white powder; mp 236.3–238.7 °C; 1H NMR
(300 MHz, DMSO-d6): d 12.75 (s, 1H), 8.63 (d, 1H, J = 8.1 Hz), 7.66–7.01 (m,
18H), 5.54 (s, 2H), 4.61–4.55 (m, 1H), 4.30 (s, 2H), 3.19–3.13 (dd, 1H, J = 4.2,
13.8 Hz), 3.06–2.98 (dd, 1H, J = 10.8, 13.8 Hz); HR-ESI-MS: m/z 490.2123
[M+H]+ (calcd for C31H28N3O3: 490.2125).
Compound 4e: Yield 35%; white powder; mp 119.8–123.1 °C; 1H NMR
(300 MHz, acetone-d6): d 7.79–6.59 (m, 19H), 4.88 (m, 1H), 4.74 (s, 2H), 4.72
(s, 2H), 3.33 (dd, 1H, J = 5.0, 13.8 Hz), 3.15 (dd, 1H, J = 9.2, 13.8 Hz); HR-ESI-MS:
m/z 465.2172 [M+H]+ (calcd for C30H29N2O3: 465.2172).
Compound 4f: Yield 42%; light yellow solid; mp 119–121 °C; 1H NMR
(300 MHz, acetone-d6): d 7.77–6.71 (m, 16H), 5.93 (s, 2H), 4.95–4.75 (m, 1H),
4.57 (s, 2H), 4.48 (s, 2H), 3.65 (s, 3H), 3.36–3.30 (dd, 1H, J = 4.8, 14.1 Hz), 3.19–
3.11 (dd, 1H, J = 9.6, 14.1 Hz); HR-ESI-MS: m/z 539.2188 [M+H]+ (calcd for
11. The physical and spectral data of the 21 target compounds are as follows.
Compound 1a: Yield 91%; white powder; mp 67.0–69.1 °C; 1H NMR (300 MHz,
acetone-d6): d 8.19–8.16 (m, 2H), 7.49–7.17 (m, 15H), 5.14–5.08 (m, 1H), 3.42–
3.28 (m, 2H); FAB-MS: m/z: 408 [M+H]+.
C32H31N2O6: 539.2176).
Compound 4g: Yield 20%; white powder; mp 84–86 °C; 1H NMR (300 MHz,
acetone-d6): d 7.90 (d, 1H, J = 8.1 Hz), 7.85–7.15 (m, 19H), 4.84 (m, 1H), 4.66 (s,
2H), 4.42 (s, 2H), 3.36–3.29 (dd, 1H, J = 4.5, 13.5 Hz), 3.18–3.10 (dd, 1H, J = 9.6,
14.1Hz); HR-ESI-MS: m/z 493.2114 [M+H]+ (calcd for C31H29N2O4: 493.2121).
Compound 5a: Yield 64%; white powder; mp 121.9–123.9 °C; 1H NMR
(300 MHz, acetone-d6): d 8.21–7.23 (m, 17H), 3.97 (s, 2H), 3.91 (s, 2H), 3.40
(s, 2H); HR-ESI-MS: m/z 421.1912 [M+H]+ (calcd for C28H25N2O2: 421.1910).
Compound 5b: Yield 84%; white powder; mp 133–136 °C; 1H NMR (300 MHz,
DMSO-d6): d 12.18 (br s, 1H), 7.31–6.69 (m, 16H), 5.98 (s, 2H), 4.94 (s, 2H), 3.66
(s, 2H), 3.64 (s, 2H), 3.60 (s, 3H), 3.09 (s, 2H); HR-ESI-MS: m/z 539.2177 [M+H]+
(calcd for C32H31N2O6: 539.2182).
Compound 1b: Yield 86%; white powder; mp 58.5–61.0 °C; 1H NMR (300 MHz,
acetone-d6): d 7.38–6.69 (m, 16H), 5.94 (s, 2H), 4.95–4.88 (dd, 1H, J = 7.8,
15.6 Hz), 4.53 (s, 2H), 4.51 (s, 2H), 3.32–3.25 (dd, 1H, J = 4.8, 14.7 Hz), 3.24–
3.16 (dd, 1H, J = 8.7, 13.5 Hz); FAB-MS: m/z: 500 [M+H]+.
Compound 1c: Yield 59%; white powder; mp 89.6–91.7 °C; 1H NMR (300 MHz,
acetone-d6): d 7.85–6.67 (m, 17H), 5.01 (s, 2H), 4.92 (dd, 1H, J = 4.8, 8.1 Hz),
3.65 (s, 3H), 3.32–3.10 (m, 2H); FAB-MS: m/z: 548 [M+Na]+.
Compound 1d: Yield 89%; white powder; mp 60.1–62.5 °C; 1H NMR (300 MHz,
acetone-d6): d 7.36–6.61 (m, 16H), 5.92 (s, 2H), 4.94 (s, 2H), 4.91–4.86 (dd, 1H,
J = 4.8, 8.1 Hz), 3.67 (s, 3H), 3.29–3.23 (dd, 1H, J = 4.8, 14.1 Hz), 3.22–3.15 (dd,
1H, J = 8.1, 14.1 Hz); FAB-MS: m/z: 526 [M+H]+.
Compound 1e: Yield 49%; light yellow solid; mp 115–117 °C; 1H NMR
(300 MHz, acetone-d6): d 7.34–6.70 (m, 16H), 5.92 (s, 2H), 5.00 (s, 2H), 4.96–
4.92 (dd, 1H, J = 4.5, 7.8 Hz), 3.30–3.24 (dd, 1H, J = 4.5, 14.7 Hz), 3.22–3.15 (dd,
1H, J = 8.1, 13.5 Hz); FAB-MS: m/z: 514 [M+H]+.
12. The measurements were performed by following the experimental protocols
described in this cited reference. Ma, Y.-M.; Tao, R.-Y.; Liu, Q.; Li, J.; Tian, J.-Y.;
Zhang, X.-L.; Xiao, Z.-Y.; Ye, F. Mol. Cell. Biochem. 2011, 357, 65.
13. The structures of Ertiprotafib and compounds 3a and 4e were generated and
molecular docking was performed with the Discovery Studio 2.5 software
package (Accelrys, San Diego, USA). The protein structure of PTP1B was
obtained from the Protein Data Bank (PDB code: 1QXK). The docking
calculation was carried out with the LigandFit protocol. Default settings were
used. All calculations were performed on a DELL Precision T5500 workstation.
The ligand interaction modes were displayed by PyMOL (DeLano, W.L. The
PyMOL Molecular Graphics System, 2002, DeLano Scientific, San Carlos, CA, USA).
Compound 2a: Yield 62%; white powder; mp 163.9–165.5 °C; 1H NMR
(300 MHz, acetone-d6): d 8.22–7.24 (m, 17H), 4.06–3.97 (m, 1H), 3.70–3.63
(dd, 1H, J = 6.3, 16.8 Hz), 3.62–3.54 (dd, 1H, J = 4.5, 16.8 Hz), 3.01–2.94 (dd, 1H,
J = 6.6, 16.2 Hz), 2.89–2.81 (dd, 1H, J = 8.1, 15.9 Hz); HR-ESI-MS: m/z 434.1738
[M+H]+ (calcd for C29H24NO3: 434.1750).
Compound 2b: Yield 32%; white powder; mp 212–215 °C; 1H NMR (300 MHz,
acetone-d6): d 8.15 (m, 2H), 7.94–7.91 (m, 2H), 7.59–7.53 (m, 3H), 7.46–7.38