Potent, selective inhibitors of protein tyrosine phosphatase 1B

Article abstract of DOI:10.1016/S0960-894X(03)00302-0

We have previously reported a novel series of oxalyl-aryl-amino benzoic acid-based, catalytic site-directed, competitive, reversible protein tyrosine phosphatase 1B (PTP1B) inhibitors. With readily access to key intermediates, we utilized a solution phase

Full text of DOI:10.1016/S0960-894X(03)00302-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1887–1890
Potent, Selective Inhibitors of Protein Tyrosine Phosphatase 1B
Zhili Xin,a,* Thorsten K. Oost,b Cele Abad-Zapatero,b Philip J. Hajduk,b Zhonghua Pei,a
Bruce G. Szczepankiewicz,a Charles W. Hutchins,b Steve J. Ballaron,a Mike A. Stashko,a
Tom Lubben,a James M. Trevillyan,a Mike R. Jirouseka,y and Gang Liua,*
aMetabolicDiseaseResearch, GlobalPharmaceuticalResearchandDevelopment,AbbottLaboratories,AbbottPark,IL60064-6098, USA
bAdvanced Technology, Global Pharmaceutical Research and Development, Abbott Laboratories, Abbott Park, IL 60064-6098, USA
Received 27 December 2002; accepted 24 February 2003
Abstract—We have previously reported a novel series of oxalyl-aryl-amino benzoic acid-based, catalytic site-directed, competitive,
reversible protein tyrosine phosphatase 1B(PTP1B) inhibitors. With readily access to key intermediates, we utilized a solution
phase parallel synthesis approach and rapidly identified a highly potent PTP1Binhibitor ( 19, Ki=76 nM) with moderate selectivity
(5-fold) over T-cell PTPase (TCPTP) through interacting with a second phosphotyrosine binding site (site 2) in the close proximity
to the catalytic site.
# 2003 Elsevier Science Ltd. All rights reserved.
Protein tyrosine phosphatases (PTPases) that function
as negative regulators of the insulin signaling cascade
have been identified as novel targets for the therapeutic
enhancement of insulin action in insulin-resistant dis-
ease states.1 Recent studies have provided compelling
evidence that one of the main functions of the intra-
cellular enzyme PTPase 1B(PTP1B) is to suppress
insulin action.2 Reducing PTP1Babundance in mice
not only enhances insulin sensitivity and improves glu-
cose metabolism, but also protects against obesity
induced by high-fat feeding.3 Inhibition of PTP1Bin
insulin sensitive tissues using novel antisense oligonu-
cleotides has shown enhanced insulin signaling and
glucose tolerance in preclinical models.4 Small molecule
PTP1Binhibitors may find an important clinical role as
novel insulin sensitizers in the treatment of type II
diabetes.5
so far have not addressed the issue of achieving selec-
tivity over TCPTP. Zhang and co-workers have pro-
posed a new paradigm for designing highly potent and
selective PTP1Binhibitors, based on the presence of a
second phosphotyrosine-binding site (site 2) positioned
adjacent to the active site, which lies within a region
that is not conserved among PTPases.7 Recently pub-
lished results from the same group have confirmed the
feasibility of such approach.8
Compound 1, an oxalyl-aryl-amino benzoic acid, was
discovered by a NMR-based fragment screening as a
novel PTP1Binhibitor with a Ki value of 93 mM from
this laboratory.9,10 Subsequent structure-based drug
design approach facilitated the discovery of a relatively
potent inhibitors of PTP1B, as represented by com-
pound 2 (Ki=0.99 mM).9 X-ray crystal structure of the
complex between PTP1Band 2 demonstrated that the
di-carboxylic acid portion of the molecule binds in the
catalytic site of PTP1Bas a novel phosphotyrosine
mimetic.9 These compounds exhibited good selectivity
for PTP1Bover most of the phosphatases in our selec-
tivity panel, such as SHP-2, LAR and CD45. However,
they do not discriminate between PTP1Band TCPTP.
Analogous to Zhang’s approach, we have expanded
beyond the di-carboxylic acid pharmacophore anchored
in the catalytic site to interact with site 2, and identified
PTP1Binhibitors with greater affinity and moderate
selectivity over TCPTP.
T-cell PTPase (TCPTP), the most homologous phos-
phatase to PTP1Bknown to date, is implicated in reg-
ulating T-cell activation.6 Therefore, selective inhibition
of PTP1Bwithout antagonizing TCPTP is highly desir-
able for an anti-diabetic agent. With one exception,
most of the small molecule PTP1Binhibitors reported
*Corresponding author. Tel.: +1-847-935-1224; fax: +1-847-938-
1674; e-mail: gang.liu@abbott.com
yCurrent address: Pfizer Global R&D, La Jolla Laboratories, 10770
Science Center Drive, San Diego, CA 92121-1187, USA.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00302-0

1888
Z. Xin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1887–1890
X-ray crystal structure 2 also revealed two hydrogen-
bonding interactions with Asp48 of PTP1Bthat are cri-
tical for the enhanced affinity.9 These interactions also
orientated the terminus of the pentyl amide of 2 within
site 2. Therefore, we developed a solution phase parallel
synthesis route to rapidly screen ligands for site 2 of
PTP1Busing amide-coupling chemistry. Benzyl 2-acet-
amidoacrylate was coupled via Heck reaction with 4-
bromo-2-ethylaniline to give dehydroamino ester 3
(Scheme 1). The double bond was saturated and the
benzyl ester of 3 was cleaved via hydrogenation to give
the p-aminophenylalanine derivative, which was then
protected as its allyl ester 4. Condensation of 4 with
DPIC generated benzoic acid 5.11 The diphenylaniline
was then acylated to provide the t-butyloxalyl ester 6,
and the benzoic acid moiety of 6 was protected as its
benzhydryl ester.12 The allyl ester was then removed to
liberate the acid core 7 for coupling with different
amines. Final deprotection of the two acid-sensitive
esters with trifluoroacetic acid provided diacid 8.13
Coupling of acid 7 with amino pentanoic acid derivative
9 followed by TBAF treatment of the resulting amide
generated acid 10 for the library synthesis.
Scheme 1. Reagents and conditions: (a) BnBr, K2CO3, DMF, rt,
100%; (b) 4-bromo-2-ethylaniline, Pd(OAc)2, (o-Tol)3P, Et3N, MeCN,
reflux, 58%; (c) 10% Pd/C, 4 atm H2, rt, 100%; (d) allyl bromide,
Cs2CO3, rt, 62%; (e) diphenyliodonium carboxylate, Cu(OAc)2,
i-PrOH, 80 ꢀC, 85%; (f) ClCOCOt-Bu, i-Pr2NEt, CH2Cl2, 0 ꢀC–rt,
72%; (g) diphenyldiazomethane, acetone, rt, 92%; (h) Pd(Ph3P)4,
morpholine, CH2Cl2, 100%; (i) TBTU, H2NR, i-Pr2NEt, DMF, 72–
94%; (j) TFA, anisole, CH2Cl2, 58–88%; (k) TBAF, THF, 94%.
We explored extensively the SAR of the pentyl amide of
2 with the objective to find better linkers for accessing
site 2. Among over 300 compounds synthesized, only
benzyl amines with para electron withdrawing sub-
stituents, such as nitro (8a) or chlorine (8b), appeared to
be a good replacement of the amyl amine, while electron
donating methoxy group yielded less potent analogue 8c
(Table 1). However, further SAR on benzyl amides
failed to provide any improvement on PTP1Binhibitory
potency and selectivity against TCPTP (data not
shown). In the ensuing efforts of searching for site 2
ligands, it was decided to use the more flexible amino-
pentanoic acid (8d) as the linker, since the initial phe-
nethyl amide 8e demonstrated reasonable potency
against PTP1B. More importantly, a key hydrogen-
bond between the carbonyl of the phenethyl amide
of 8e and Gln262 of PTP1Bwas observed by X-ray
crystallography.
decreased potency than compound 2. Phenylalanine-
based 12 with an extra acid functionality gained 24-fold
in potency compared to 8e. This critical observation
prompted us to screen a library of amino acids. With
different l-amino acids coupled to the core template,
most of the compounds retained submicromolar
potency, suggesting a rather tolerant site 2 (13–20). It
appears that amino acids with the more lipophilic side
chains (17–19) were slightly more favored than the ones
with hydrophilic side chains (15–16, 20). Remarkably,
compound 18 and 19 not only showed improved binding
affinity, but also for the first time, exhibited some moder-
ate selectivity against TCPTP (5-fold). Again, the
importance of the acid functionality for site 2 ligands is
confirmed with ester 21, which is 7 times less potent than
the corresponding acid 19. Additionally, the stereo-
chemistry of the amino acid played an important role in
anchoring the acid to the right position: the R-isomer
(22) is about 20-fold less potent than its S-isomer.
Neither 21 nor 22 is selective for PTP1Bover TCPTP.
In order to rapidly identify suitable site 2 ligands, dif-
ferent amines were coupled to the readily available acid
core 10 in a parallel fashion. The presence of two argi-
nine residues (Arg24 and Arg254) in the vicinity of the
phenethyl amide portion of 8e suggested that a nega-
tively charged acid functionality might be preferred as
part of site 2 ligands. Such preference is clearly reflected
by the analogues listed in Table 2. Analogues with neu-
tral groups, such as 8e and 8f, exhibited slightly

Z. Xin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1887–1890
Table 1. SAR on modification of the pentyl amide of 2 Table 2. SAR on site 2 ligand exploration
1889
Ki (mMb)
(PTP1B)
Ki (mMb)
(TCPTP)
Compd
R0
Ki (mM) (PTP1B) Ki (mM) (TCPTP)
Compd
R
2a
1.2 (Æ0.3)
0.54 (Æ0.14)
1.2
0.9 (Æ0.1)
0.45 (Æ0.14)
1.5
8e
3.4 (Æ0.9)
1.3 (Æ0.4)
1.4 (Æ0.3)
0.7 (Æ0.2)
8aa
8ba
8ca
8da
8f
12a
13a
0.14 (Æ0.02)
0.25 (Æ0.10)
0.16 (Æ0.02)
0.29 (Æ0.03)
4.7 (Æ1.6)
2.5 (Æ0.2)
3.4 (Æ0.9)
3.6
2.2 (Æ0.3)
1.4 (Æ0.3)
14a
15a
0.43 (Æ0.2)
0.40 (Æ0.1)
8ea
0.33 (Æ0.16)
0.27 (Æ0.09)
a1:1 racemic mixture.
bThe kinetic analysis was conducted using pNPP as the small molecule
substrate in a continuously-monitored colorimetry assay. Values are
means of more than two experiments with the number in the par-
entheses being the standard deviation.
16a
17a
18a
19a
20a
21a
22a
0.71 (Æ0.28)
0.12 (Æ0.07)
0.13 (Æ0.01)
0.076 (Æ0.015)
0.47 (Æ0.19)
0.54 (Æ0.19)
1.7 (Æ0.7)
0.82 (Æ0.22)
0.24 (Æ0.07)
0.51 (Æ0.15)
0.38 (Æ0.078)
0.37 (Æ0.08)
0.67
The selectivity of 19 against other phosphatases in our
selectivity panel was also confirmed, including SHP-2,
LAR, CD45, and cdc25C (Table 3). In general, 19 is at
least 100-fold selective for these less homologous phos-
phatases.
The X-ray crystal structure of PTP1Bcomplexed with
1914 demonstrated unambiguously that extends the full
span of catalytic site and site 2 (Fig. 1). With 19 bound
to PTP1B, the conformation of the WPD loop of
PTP1Bremains open, unlike most of the other complex
structures reported.15 The oxamic acid and benzoic acid
of 19 bind in the catalytic site as a novel phosphotyro-
sine mimetic.,The diamides of the alanine linker hydro-
gen bond with Asp48, leading the methionine to site 2,
where the carboxylic acid of methionine in compound
19 interacts with Arg24 and the carbonyl group of
adjacent amide interacts with Gln262. The lipophilic
1.0 (Æ0.3)
a1:1 mixture of diastereomers.
Table 3. Phosphatase selectivity profile of compound 19
Ki (mM) (PTP1B)
0.076 (Æ0.015)
Ki (mM) (TCPTP)
0.38 (Æ0.078)
Ki (mM) (SHP-2)
Ki (mM) (LAR)
8.75 (Æ8.84)
Ki (mM) (CD45)
Ki (mM) (cdc25C)
>300
73.17
>300

1890
Z. Xin et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1887–1890
1544. (b) Klaman, L. D.; Boss, O.; Peroni, O. D.; Kim, J. K.;
Martino, J. L.; Zabotny, J. M.; Moghal, N.; Lubkin, M.; Kim,
Y.-B.; Sharpe, A. H.; Stricker-Krongrad, A.; Shulman, G. I.;
Neel, B. G.; Kahn, B. B. Mol. Cell. Biol. 2000, 20, 5479.
4. Zinker, B. A.; Rondinone, C. M.; Trevillyan, J. M.; Gum,
R. J.; Clampit, J. E.; Waring, J. F.; Xie, N.; Wilcox, D.;
Jacobson, P.; Frost, L.; Kroeger, P. E.; Reilly, R. M.;
Koterski, S.; Opgenorth, T. J.; Ulrich, R. G.; Crosby, S.;
Butler, M.; Murray, S. F.; Mckay, R. A.; Bhanot, S.; Monia,
B. P.; Jirousek, M. R. Proc. Natl. Acad. Sci. U.S.A. 2002, 99,
11357.
5. Blaskovich, M. A.; Kim, H.-O. Exp. Opin. Ther. Pat. 2002,
12, 871.
´
6. You-Ten, K. E.; Muise, E. S.; Itie, A.; Michaliszyn, E.;
Wagner, J.; Jothy, S.; Lapp, W. S.; Tremblay, M. L. J. Exp.
Med. 1997, 186, 683.
7. Puius, Y. A.; Zhao, Y.; Sullivan, M.; Lawrence, D. S.;
Almo, S. C.; Zhang, Z.-Y. Proc. Natl. Acad. Sci. U.S.A. 1997,
94, 13420.
8. Shen, K.; Keng, Y. F.; Wu, L.; Guo, X. L.; Lawrence, D. S.;
Zhang, Z.-Y. J. Biol. Chem. 2001, 276, 47311.
˚
Figure 1. X-ray structure (2.4 A resolution) of PTP1Bcomplexed with
19. Color scheme: carbons in orange and green for compound 19,
oxygen in red, nitrogen in blue, and sulfur in yellow.
side chain of the methionine lies on top of a largely
hydrophobic region of the protein, consisted of Met258,
Phe52. The origin of TCPTP selectivity for 19 is not
very clear, since 19 did not interact with any residues
that are different between PTP1Band TCPTP, although
the interaction with Arg24 and Met258 appear to be
critical for achieving such selectivity.
9. Liu, G.; Szczepankiewicz, B. G.; Pei, Z.; Janowick, D.; Xin,
Z.; Liang, H.; Hadjuk, P. J.; Abad-Zapatero, C.; Hutchins, C.
W.; Fesik, S. W.; Ballaron, S. J.; Stashko, M. A.; Lubben, T.;
Mika, A. K.; Zinker, B. A.; Trevillyan, J. M.; Jirousek, M. R.
J. Med. Chem. In press.
10. For in vitro assay protocols, see: Lubben, T.; Clampit, J.;
Stashko, M.; Trevillyan, J.; Jirousek, M.R. In Current Proto-
cols in Pharmacology; Enna, S. J., Williams, M., Ferkany, J.
W., Kenakin, T., Porsolt, R. D., Sullivan, J. P. Eds.; Wiley:
New York, 2001; p 3.8.1.
11. Scherrer, R. A.; Beatty, H. R. J. Org. Chem. 1980, 45,
2127.
12. Skelakatos, G. C.; Paganou, A.; Zervas, L. J. Chem. Soc.
C 1966, 1191.
13. The purity of the compounds tested exceeds 90% as
determined by analytical HPLC.
14. Refined crystallographic coordinates for the structure of
PTP1Bcomplexed with compounds 19 have been deposited
with the Protein Data Bank (www.rcsb.org) with entry codes
1NZ7.
15. (a) Andersen, H. S.; Olsen, O. H.; Iversen, L. F.; Sor-
ensen, A. L. P.; Mortensen, S. B.; Christensen, M. S.; Branner,
S.; Hansen, T. K.; Lau, J. F.; Jeppesen, L.; Moran, E. J.; Su,
J.; Bakir, F.; Judge, L.; Shahbaz, M.; Collins, T.; Vo, T.;
Newman, M. J.; Ripka, W. C.; Moller, N. P. H. J. Med.
Chem. 2002, 45, 4443. (b) Larsen, S. D.; Barf, T.; Liljebris, C.;
May, P. D.; Ogg, D.; O’Sullivan, T. J.; Palazuk, B. J.; Schos-
tarez, H. J.; Stevens, F. C.; Bleasdale, J. E. J. Med. Chem.
2002, 45, 598.
16. Co-crystalization of TCPTP with inhibitor has failed due
to a multimerization process. See: Iversen, L. F.; Moller, K. B.;
Pedersen, A. K.; Peters, G. H.; Petersen, A. S.; Andersen,
H. S.; Branner, S.; Mortensen, S. B.; Moller, N. P. H. J. Biol.
Chem. 2002, 277, 19982.
In summary, we have discovered the first series of
potent, non-phosphonic acid-containing PTP1Binhibi-
tors with moderate specificity for PTP1Bover TCPTP,
the most homologous phosphatase to PTP1B. We uti-
lized the strategy of occupying both the catalytic site
and the nearby, less homologous, non-catalytic phos-
photyrosyl binding site. This series of PTP1Binhibitors
identified the residues important for achieving TCPTP
specificity in the absence of any structural informa-
tion,16 and provided us with opportunities for designing
more potent and selective PTP1Binhibitors with
improved physical properties.
References and Notes
1. (a) Saltiel, A. R.; Kahn, C. R. Nature 2001, 414, 799. (b)
Ukkola, O.; Santaniemi, M. J. Intern. Med. 2002, 251, 467.
2. Cheng, A.; Dube, N.; Gu, F.; Tremblay, M. L. Eur. J.
Biochem. 2002, 269, 1050.
3. (a) Elchebly, M.; Payette, P.; Michaliszyn, E.; Cromlish,
W.; Collins, S.; Loy, A. L.; Normandin, D.; Cheng, A.;
Himms-Hagen, J.; Chan, C. C.; Ramachandran, C.; Gresser,
M. J.; Tremblay, M. L.; Kennedy, B. P. Science 1999, 283,