Synthesis and cell-based activity of a potent and selective protein tyrosine phosphatase 1B inhibitor prodrug

Article abstract of DOI:10.1021/jm061146x

Our laboratory recently reported the development of novel prodrug chemistry for the intracellular delivery of phosphotyrosine mimetics. This chemistry has now been adapted for the synthesis of a prodrug that delivers the nonhydrolyzable difluoromethylphos

Full text of DOI:10.1021/jm061146x

856
J. Med. Chem. 2007, 50, 856-864
Synthesis and Cell-Based Activity of a Potent and Selective Protein Tyrosine Phosphatase 1B
Inhibitor Prodrug
Irene G. Boutselis,† Xiao Yu,‡ Zhong-Yin Zhang,‡ and Richard F. Borch*,†
Department of Medicinal Chemistry and Molecular Pharmacology and the Cancer Center, Purdue UniVersity, West Lafayette, Indiana 47907,
Department of Biochemistry and Molecular Biology, Indiana UniVersity School of Medicine, Indianapolis, Indiana 46202
ReceiVed October 3, 2006
Our laboratory recently reported the development of novel prodrug chemistry for the intracellular delivery
of phosphotyrosine mimetics. This chemistry has now been adapted for the synthesis of a prodrug that
delivers the nonhydrolyzable difluoromethylphosphonate moiety intracellularly. Activation of the prodrug
generates a difluoromethylphosphonamidate anion that undergoes subsequent cyclization and hydrolysis
with a t1/2 ) 44 min. A highly potent and selective inhibitor of protein tyrosine phosphatase 1B (PTP1B)
with a nanomolar Ki has been reported, but this bis(difluoromethylphosphonate) lacks potential utility due
to its exceedingly low membrane permeability at physiological pH. A prodrug of this inhibitor has been
synthesized and evaluated in a cell-based assay. The prodrug exhibits nanomolar PTP1B inhibitory activity
in this assay, confirming the efficacy of intracellular phosphonate delivery using this prodrug approach.
Introduction
Protein tyrosine phosphorylation is linked to cellular signaling
processes that in turn control a host of fundamental cellular
functions, including growth, differentiation, metabolism, pro-
gression through the cell cycle, migration, apoptosis/survival,
and immune response.1 The level of protein tyrosine phospho-
rylation is regulated by the action of protein-tyrosine kinases
(PTKs) and protein-tyrosine phosphatases (PTPs). Perturbation
of the balance between these enzymes disrupts cell function
and has been implicated in a variety of disease states, including
diabetes, obesity, and cancer.2,3 Consequently, cellular pathways
regulated by tyrosine phosphorylation offer a rich source of key
targets for developing novel therapeutics.4,5 The potential of such
targeted therapeutics has been well-demonstrated by the suc-
cessful treatment of human chronic myelogenous leukemia and
gastrointestinal stromal tumors with the PTK inhibitor Gleevec,6,7
which targets Bcr/Abl or c-kit aberrantly activated in the
Figure 1. Structures of PTP1B inhibitor 1a and its prodrug 1b.
malignancies.
active site.11,12 The most useful and potent pTyr surrogate thus
far devised for PTPs is phosphonodifluoromethyl phenylalanine
(F2Pmp).13,14 Although F2Pmp likely interacts in the desired
inhibitory fashion with all PTPs, the molecular scaffolds to
which the difluorophosphonate is attached should render the
inhibitors PTP selective. Indeed, a number of potent and
selective difluorophosphonate-based PTP1B inhibitors have been
described.15-17
PTP1B is implicated as a key negative regulator of insulin
and leptin signaling. In two landmark papers, it was shown that
PTP1B deficient mice are more sensitive to insulin, have
improved glycemic control, and are resistant to diet induced
obesity.18,19 The phenotype of PTP1B knockout mice suggests
that an inhibitor of this PTP may address both obesity and
insulin resistance, thereby presenting a unique therapeutic
opportunity. Understandably, PTP1B is highly coveted by the
pharmaceutical industry and constitutes a first example of a PTP
with a physiological function that makes it an ideal drug target
for therapeutic intervention in very common human diseases,
type II diabetes and obesity.20
As observed with the PTKs, deregulation of PTP activity also
contributes to the pathogenesis of a number of human diseases.8,9
Since no formal functional linkage exists between the PTKs
and PTPs, any single PTP could catalyze the dephosphorylation
of proteins phosphorylated by more than one PTK. Thus,
inhibitors of the PTPs are expected to have therapeutic value
with unique modes of action. However, therapeutics based on
PTPs were not seriously considered until recently. Progress in
this area was initially stifled by the fact that the active site (i.e.,
pTyr binding site) is highly conserved among the large number
of PTPs, making it difficult to achieve inhibitor selectivity.
Fortunately, PTP substrate specificity studies have shown that
pTyr alone is not sufficient for high-affinity binding, and
residues flanking the pTyr are important for PTP substrate
recognition.10 These results suggest that unique PTP sub-pockets
that border the active site may be targeted to enhance inhibitor
affinity and selectivity. Major efforts have been made to develop
nonhydrolyzable pTyr surrogates aimed to occupy the PTP
* Corresponding author. Tel.: (765) 494-1403; fax (765) 494-1414;
e-mail: borch@purdue.edu.
The most potent and selective PTP1B inhibitor reported is
compound 1a (Figure 1), which features an F2Pmp residue
linked to a second aryl difluorophosphonate group.15 At physi-
† Purdue University.
‡ Indiana University School of Medicine.
10.1021/jm061146x CCC: $37.00 © 2007 American Chemical Society
Published on Web 01/24/2007

Protein Tyrosine Phosphatase 1B Inhibitor Prodrug
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 857
Scheme 2a
Scheme 1
ologic pH, compound 1a would be expected to be highly ionized
and have exceedingly low membrane permeability. Indeed,
compound 1a is not cell permeable and exhibits no cellular
activity.21 To enhance penetration across the cell membrane,
cell permeable derivatives of 1a were prepared by coupling it
to either the cell penetrating peptide (D)-Arg8 via a disulfide
bridge22 or to a highly lipophilic fatty acid.21 While these cell
permeable analogues have enabled important validation of
PTP1B as a therapeutic target and delineation of the roles of
PTP1B in various signaling pathways,21-24 they are unlikely to
be suitable for the ultimate development of useful therapeutic
agents.
Since a number of organophosphonate-based prodrugs have
advanced to the clinic and beyond,25 we sought to apply a
prodrug strategy to improve the in vivo efficacy of compound
1a. Previous attempts to improve cellular delivery of SH2
domain targeted peptides with phosphonate ester prodrugs have
yielded very limited success.26 Our laboratory has developed a
novel prodrug approach that has been applied to the intracellular
delivery of nucleotides and aryl phosphates.27-29 Given the large
number (5) of negative charges at physiologic pH, inhibitor 1a
represents an interesting challenge for prodrug delivery. We have
extended our prodrug technology to investigate the synthesis
and activation of difluoromethyl phosphonate prodrugs and the
synthesis of prodrug 1b. Studies with 1b in a cell-based system
indicate that our prodrug strategy provides an excellent solution
to the drug delivery problem associated with organophospho-
nate-based PTP inhibitors.
a Reagents and conditions: (a) P(OEt)3, CHCl3; (b) diethylaminosulfur
trifluoride, 0 °C to rt, 20 h; (c) SOCl2, DMF (cat); (d) CH3NH(CH2)4Cl‚HCl,
DIEA, CH2Cl2, -20 °C; (e) nitrofurfuryl alcohol, LiHMDS, THF, -78
°C; (f) BnOH, n-BuLi, -78 to -40 °C, 3 h; (g) H2, Pd-C; and (h) cacodylate
buffer, pH 8.5, 37 °C.
trifluoride (DAST)31 afforded difluoromethylphosphonate 2.
This compound was refluxed with thionyl chloride to generate
the labile phosphonyl dichloride, which was used without further
purification. Sequential addition of N-methyl-N-(4-chlorobutyl)-
amine to give monochloride 3 followed by reaction with the
lithium salts of the appropriate alcohols gave compounds 4a
and 4b in good overall yields.
31P NMR Kinetics Study. The intracellular activation of
prodrug 4a can be modeled in vitro by catalytic hydrogenolysis
of benzyl ester 4b. Thus, 4b was subjected to hydrogenolysis
resulting in its quantitative conversion to phosphonamidate anion
5 (as determined by 31P NMR). This anion was immediately
taken up in 0.4 M cacodylate buffer/CH3CN, and its conversion
to difluoromethylphosphonate 6 (Scheme 2) was monitored by
31P NMR at 37 °C over 2 h. The 31P triplet at -15 ppm
corresponding to phosphonamidate 5 slowly disappeared and
was replaced by a new triplet at -20 ppm corresponding to the
phosphonate 6; this conversion proceeds with a half-life (t1/2
)
of 44 min. This contrasts with t1/2 values of 2-21 min (T ) 37
°C, pH ) 7.2-7.4) for phosphoramidate analogues (X ) O or
NH, Scheme 1).28,29 It is not unexpected that the strongly
electron-withdrawing CF2 group retards the cyclization of the
difluoromethylphosphonamidate 5. The t1/2 of 44 min under
model physiologic conditions is not pharmacologically prohibi-
tive for phosphonate delivery, so our attention was directed to
the synthesis of 1b.
Target Molecule Synthesis. Construction of 1b requires the
synthesis of two prodrug building blocks 11, derived from
4-iodophenylacetic acid, and 12b, derived from N-Fmoc-(L)-4-
iodophenylalanine. Scheme 3 details installation of the difluo-
romethylphosphonate ester moiety into the corresponding
aromatic precursors. Copper-catalyzed coupling of aryl iodides
7a and 7b with the organocadmium reagent derived from diethyl
bromodifluoromethylphosphonate as described by Qabar32
provided diethyl difluoromethylphosphonates 8a and 8b in good
yields. Conversion of the phosphonate esters to the reactive
phosphonyl dichlorides was accomplished by reaction with
TMSBr and treatment of the resulting silyl phosphonate
intermediate with oxalyl chloride.33 This procedure was a vast
improvement over the prolonged reaction with thionyl chloride
utilized for the synthesis of the model compounds (Scheme 2).
Installation of the masking and delivery groups as described
for the synthesis of 4a provided the ester protected prodrugs
Results and Discussion
Chemistry. The prodrug strategy is outlined in Scheme 1.
The nitrofurfuryl delivery group in prodrug A imparts lipo-
philicity to enhance uptake into the cell. Once in the cell, the
molecule undergoes sequential enzymatic reduction of the
nitrofurfuryl group and cleavage to yield phosphonamidate B.
The nitrogen can now participate in an intramolecular cyclization
resulting in intermediate C, which hydrolyzes spontaneously
to deliver the phosphorylated drug. The efficacy of this approach
has been demonstrated for X ) O and X )NH, but the effect
of a difluoromethyl moiety (X ) CF2) on the prodrug strategy
was unknown. Therefore, a model study was carried out initially
to assess the synthetic approach and the activation process on
a representative difluoromethyl analogue.
Model Studies. Compounds 4a and 4b were synthesized as
shown in Scheme 2. Synthesis of 4a allowed development of
methods necessary to install the masking and delivery groups
onto a difluoromethylphosphonate framework. Compound 4b
would provide the necessary precursor for chemical kinetic
studies that would demonstrate the viability of the difluoro-
methyl group in prodrug activation. Arbuzov reaction of benzoyl
chloride with triethylphosphite30 followed by fluorination of the
ketophosphonate by treatment with excess diethylaminosulfur

858 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Boutselis et al.
Scheme 3a
Scheme 4a
a Reagents and conditions: (a) DBU/CH2Cl2; (b) 11, PyBOP, DIEA,
CH2Cl2; (c) LiOH/H2O, THF, 0 °C; (d) 12b + DBU/CH2Cl2, then PyBOP,
DIEA, CH2Cl2; and (e) TFA/CH2Cl2.
With the two requisite building blocks 11 and 12b in hand,
the target molecule was then assembled. The main challenge
was to retain the somewhat labile nitrofurfuryl groups during
the two required Fmoc deprotections. Brief exposure to DBU
in CH2Cl2 at 0 °C with careful TLC monitoring effected removal
of the Fmoc group without modification of the nitrofurfuryl
group. Although solid phase synthesis can be used in the
construction of 1b, the inability to monitor Fmoc deprotection,
the strongly acidic conditions required for cleavage from the
resin, and the potential need for scaleup synthesis led us to
pursue a solution phase approach for the synthesis of acyl
dipeptide 1b.
The convergent synthesis of 1b is outlined in Scheme 4.
Protected aspartate 13 was deprotected by treatment with DBU/
CH2Cl2 at 0 °C for 25 min. The resulting aminoester was
coupled to 11 using PyBOP/DIEA to yield acyl aspartate 14a
in 65% yield. The methyl ester group of 14a was hydrolyzed
using LiOH/H2O to give the corresponding carboxylic acid 14b
in 68% yield. Compound 12b was treated with DBU/CH2Cl2
at 0 °C to remove the Fmoc group, and coupling of the resulting
amino amide to 14b was achieved in 47% yield using PyBOP
and DIEA. Finally, the t-butyl ester moiety of 15 was hydrolyzed
with TFA/CH2Cl2 to give the desired acyldipeptide prodrug 1b
in 82% yield.
Biological Evaluation. Compound 1a displays a Ki value of
2.4 nM for PTP1B and exhibits several orders of magnitude of
selectivity in favor of PTP1B against a panel of PTPs.15 The
negatively charged difluorophosphonate functional groups in
compound 1a participate in key active site and near active site
interactions that are responsible for the high PTP1B potency
and selectivity displayed by 1a.36,37 Masking the negative
charges in the difluorophosphonate groups would lead to a
decrease in inhibitory activity of compound 1a. As expected,
the prodrug 1b did not exhibit significant inhibition of the
PTP1B-catalyzed reaction even at concentrations as high as 20
µM, a nearly 4 orders of magnitude loss in activity in
comparison with the unmodified 1a. We predict that, upon entry
inside the cell, the prodrug will undergo sequential enzymatic
and spontaneous chemical transformation, regenerating the
PTP1B inhibitory species 1a.
a Reagents and conditions: (a) Cd, BrCF2P(O)(OEt)2, TMSCl, CuCl;
(b) TMSBr, CH2Cl2; (c) oxalyl chloride, CH2Cl2; (d) CH3NH(CH2)4Cl‚HCl,
DIEA, CH2Cl2, -20 °C; (e) nitrofurfuryl alcohol, LiHMDS, THF, -78
°C; (f) Me2SnOH, DCE, 80 °C; and (g) DIC, NHS, NH4HCO3, THF.
10a and 10b. Selective hydrolysis of the methyl esters proved
to be problematic. Standard lithium hydroxide hydrolysis at 0
°C resulted in decomposition. Alternative reagents such as
(NH4)2CO3, K2CO3, Ba(OH)2, and Cs2CO3 resulted either in
recovery of starting ester or in decomposition. Other ester groups
were investigated to no avail. Surprisingly, treatment of the
benzyl ester analogue corresponding to 10b with TMSI resulted
in clean removal of the nitrofurfuryl group, leaving the benzyl
ester intact. Attempts to cleave either the methyl or the benzyl
esters via enzymatic hydrolysis resulted in recovery of the
starting material. Attempted deprotection of the phenacyl ester
analogue of 10b34 resulted in decomposition as well. Reaction
of the propargyl ester analogue of 10b with tributyltin hydride
appeared to be effective but was not reproducible. Protecting
groups such as allyl and 2,4-dimethoxybenzyl did not survive
the reaction with TMSBr and could not be employed. Ultimately,
it was found that lithium hydroxide added carefully in small
increments over short reaction times led to methyl ester
hydrolysis providing the desired acids. Subsequently, it was
discovered that reaction with trimethyltin hydroxide35 in DCE
at 80 °C provided the free acids cleanly. This method is reported
to avoid epimerization sometimes experienced with LiOH and
is compatible with many commonly used protecting groups. To
confirm that chiral integrity was preserved throughout the
esterification and de-esterification reactions, methyl ester 7b was
subjected to hydrolysis by both the LiOH and the Me3SnOH
methods. The carboxylic acids obtained were analyzed on a
chiral HPLC column and were compared to authentic samples
of (L)- and (D)-N-Fmoc-iodophenylalanine. No trace of (D)-
amino acid was observed from either hydrolysis process. Finally,
the Fmoc protected amino acid was converted to the primary
amide using standard NHS-carbodiimide coupling chemistry.

Protein Tyrosine Phosphatase 1B Inhibitor Prodrug
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 859
Figure 2. Concentration dependence of prodrug 1b on insulin signaling in HepG2 cells. Subconfluent HepG2 cells were treated with a range of
concentrations of prodrug 1b (0, 5, 20, and 100 nM) or 200 nM of the cell permeable fatty acid conjugated derivative of 1a (compound II21) for
1 h, which was followed by a 5 min treatment with (lanes 2, 4, 6, and 8) or without (lanes 1, 3, 5, 7, and 9) 100 nM insulin. Cell lysates were
subjected to SDS-PAGE, and the resolved proteins were transferred to nitrocellulose membranes, which were blocked with anti-phospho-IRâ and
anti-phospho-ERK1/2 antibodies. â-Actin was used as a loading control.
Insulin mediated activation of the insulin receptor (IR)
produces a wide array of cellular responses, including enhanced
phosphorylation of the insulin receptor â (IRâ) subunit and the
extracellular signal regulated protein kinase 1 and 2 (ERK1/
2).38,39 Previous studies have demonstrated that PTP1B serves
as a negative regulator of the insulin activated signaling
pathways by catalyzing the dephosphorylation of IRâ. Conse-
quently, inhibition of PTP1B activity should enhance insulin
signaling. For example, IRâ phosphorylation in the liver of the
PTP1B knockout mice is prolonged 2-fold.18,19 The antisense
mediated reduction of PTP1B in rat hepatoma cells generates a
2.3-fold increase in the phosphorylated IRâ level.40 In addition,
significant enhancements in IRâ phosphotyrosine levels have
been reported with a fatty acid conjugated derivative of 1a
(identified as compound II in Figure 2).21
To assess the efficacy of the prodrug delivery strategy, we
determined the cellular effects of compound 1b on insulin
signaling in human hepatoma HepG2 cells. The cells were
exposed to prodrug 1b over a selected range of concentrations
for 1 h and subsequently treated either with or without insulin
(100 nM) for 5 min. Cell lysates were resolved by SDS-PAGE
and the resolved proteins were transferred to nitrocellulose
membranes and probed with both anti-phospho-IRâ (pY1162/
pY1163) and anti-phospho-ERK1/2 antibodies. As shown in
Figure 2, prodrug 1b can increase the insulin induced activation
of both IRâ and its downstream targets ERK1/2. The maximal
effect of 1b in HepG2 cells appears to peak around 20 nM. At
20 nM (∼8× the Ki of 1a for PTP1B), prodrug 1b enhanced
the phosphorylation levels of IRâ and ERK1/2 by 1.9- and 1.3-
fold, respectively. A somewhat more pronounced effect was
observed at 100 nM 1b (2.1- and 1.6-fold increase for IRâ and
ERK1/2, respectively). As a positive control, treatment of the
HepG2 cells with a known cell permeable fatty acid conjugated
1a derivative (compound II in Figure 2) (Ki ) 26 nM for
PTP1B) led to similar increases in IRâ and ERK1/2 phospho-
rylation (2- and 1.6-fold, respectively) at a concentration of 200
nM (∼8× the Ki) (Figure 2). These results suggest that the
prodrug strategy is very effective in promoting cellular uptake
of difluoromethyl phosphonates and that the liberated PTP1B
inhibitor has ready access to the cytoplasmically oriented protein
target.
To investigate how fast it would take the prodrug to exert its
biological effect, HepG2 cells pretreated with 20 nM prodrug
1b for various time intervals were stimulated with or without
100 nM insulin for 5 min. As shown in Figure 3, a 35% increase
in the insulin induced IRâ phosphorylation was observed when
the cells were pretreated with the prodrug for only 5 min.
Preincubation of the cells with the prodrug for 20 and 60 min
generated 1.7- and 2-fold increases in the insulin induced IRâ
phosphorylation, while the ERK1/2 phosphorylation was in-
creased by 25 and 50%, respectively. Prolonged incubation (2
h) with 20 nM prodrug only led to a modest increase in cellular
activity (a 2.1-fold increase in IRâ phosphorylation and no
further increase in ERK1/2 phosphorylation) (data not shown).
These results are in general agreement with the observed
conversion half-life (t1/2) of 44 min for model prodrug 4b in
solution. The fact that the biological effects produced by the
prodrug at a concentration only a few times over its Ki are
identical to those observed in PTP1B deficient mice, cells treated
with PTP1B antisense oligo, or a cell permeable PTP1B inhibitor

860 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Boutselis et al.
Figure 3. Time course of prodrug 1b delivery. Subconfluent HepG2 cells were preincubated with 20 nM prodrug 1b for various amounts of time.
Subsequently, the cells were stimulated without (lanes 1, 3, 5, and 7) or with (lanes 2, 4, 6, and 8) 100 nM insulin for 5 min. Cell lysates were
subjected to SDS-PAGE, and the resolved proteins were transferred to nitrocellulose membranes, which were blocked with anti-phospho-IRâ and
anti-phospho-ERK1/2 antibodies. â-Actin was used as a loading control.
from the mass spectrometry laboratory at Purdue University, West
Lafayette, IN.
indicates that the conversion of the prodrug 1b to PTP1B
inhibitor 1a is nearly complete in the cellular environment.
All anhydrous reactions were carried out under an atmosphere
of argon. All organic solvents were distilled prior to use unless
otherwise specified. Flash chromatography using silica gel grade
60 (230-400 mesh) was carried out for all chromatographic
separations. Thin layer chromatography was performed using
Analtech glass plates precoated with silica gel (250 nm). Visualiza-
tion of the plates was accomplished using UV and/or Hanessian’s
Stain (5.0 g of Ce(SO4)2, 25.0 g of (NH4)Mo7)24‚4H2O, 450 mL of
H2O, and 50 mL of H2SO4) followed by heating.
Antibodies and Chemicals. Polyclonal insulin receptor (IR) â
subunit dual-phospho-specific (pYpY1162/1163) antibody was
purchased from BioSource International (Camarillo, CA); polyclonal
anti-phospho-ERK1/2 antibody, secondary anti-mouse, as well as
anti-rabbit IgG(H + L) conjugated with HRP were obtained from
Cell Signaling Technology. Human insulin, sodium orthovanadate,
iodoacetic acid, and Nonidet P-40 were from Sigma (St. Louis,
MO).
Conclusion
We have extended our prodrug technology to investigate the
synthesis and activation of difluoromethyl phosphonates and
the synthesis of PTP1B prodrug 1b. Studies with 1b in human
HepG2 cells indicate that our prodrug strategy provides an
excellent solution to the drug delivery problem associated with
organophosphonate-based PTP inhibitors. The cell-based studies
in this paper show that prodrug 1b can potentiate the insulin
signaling pathway. Future studies will assess the utility of
prodrug 1b for the treatment of diabetes and obesity.
Experimental Procedures
Materials and Methods. All NMR spectra were recorded using
a 300 MHz Bruker spectrometer equipped with a 5 mm multinuclear
probe unless otherwise specified. H chemical shifts are reported
1
Cell Lines and Tissue Culture. The human hepatoma cells
(HepG2) were obtained from ATCC (HB-8065). HepG2 cells were
grown and maintained in Minimum Essential Medium (MEM,
Eagle) with 2 mM L-glutamine, 1 mM sodium pyruvate, and 10%
fetal bovine serum in a 5% CO2 environment.
Immunoblotting. The 60-80% confluent HepG2 cells were
starved for 5 h in MEM without serum. Then, cells were treated
for 1 h with a range of concentrations of prodrug 1b, followed by
stimulation with or without 100 nM insulin for 5 min. The
incubation was terminated by removing the fluid, and the cells were
washed by cold PBS. Cells were scraped and lysed with the lysis
buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% NP-40, 10
mM sodium pyrophosphate, 5 mM iodoacetic acid, 10 mM NaF, 1
mM EDTA, 2 mM orthovanadate, 5 mg/mL leupeptin, 5 mg/mL
aprotinin, 0.1 mg/mL PMSF, and 1 mM benzamidine). After 30
in parts per million from tetramethylsilane unless otherwise
specified. 31P NMR spectra were obtained using broadband 1H
decoupling, and chemical shifts are reported in parts per million
using 1% triphenylphosphine oxide/benzene-d6 as the coaxial
reference (triphenylphosphine oxide/toluene-d8 has a chemical shift
of +24.7 ppm relative to 85% phosphoric acid). Variable temper-
ature 31P NMR kinetics experiments were controlled using the
variable temperature unit of a Bruker 250 MHz spectrometer.
Multiplicities of 13C NMR peaks of 1b were determined from DEPT
experiments employing standard pulse sequences. HPLC analysis
was done using a Beckman System Gold equipped with a 168
detector set at 310 nm, a 126 solvent module, and Econosphere
C18 and phenyl columns from Alltech Associates. A normal-phase
Chiralcel OD column from Daicel Chemical Industries (OD00CE-
FF028) was also employed. Mass spectral analysis was obtained

Protein Tyrosine Phosphatase 1B Inhibitor Prodrug
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 861
min lysing on ice, the cell lysate was centrifuged at 15 000 rpm
for 15 min. The total cellular proteins were separated by SDS-
PAGE and transferred electrophoretically to nitrocellulose mem-
branes, which were then immunoblotted by anti-phospho-IRâ or
anti-phospho-ERK1/2 antibodies. To control for equal sample
loading, the membrane was also probed with anti-actin antibody.
The blots were developed using the Western blot detection kit
(Upstate) according to the manufacturer’s instructions. Western
blotting images were analyzed with Image J (NIH).
In Vitro Inhibition Studies with Prodrug 1b. IC50 values for
prodrug 1b were determined for PTP1B in the following manner.
At fixed substrate p-nitrophenyl phosphate (pNPP) concentration
(2 mM at the Km), the initial rate at various 1b concentrations was
measured by following the production of p-nitrophenol as de-
scribed.36 All experiments were performed at 30 °C in 50 mM 3,3-
dimethylglutarate buffer, pH 7.0, containing 1 mM EDTA with an
ionic strength of 0.15 M adjusted with NaCl.
hexane as eluent to provide 3 (0.18 g, 57%) as a pale yellow oil.
Rf ) 0.33, 20% EtOAc/hexane. H NMR (CDCl3): δ 7.62 (2H,
d); 7.54-7.45 (3H, m); 3.48 (2H, t); 3.21-3.03 (2H, m); 2.80 (3H,
d); 1.71-1.54 (4H, m); 31P NMR (CDCl3) δ +3.40 (t); HRMS
(C12H16Cl2F2NOP) calcd 330.0393, found 330.0403 .
1
O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Difluorometh-
ylphenylphosphonamidate (4a). 5-Nitrofurfuryl alcohol (0.05 g,
0.35 mmol, 1.1 equiv) and 4-(phenylazo)diphenylamine (3 mg) in
THF (2.2 mL) was cooled to -78 °C with stirring under an Ar
atmosphere. A 1.0 M solution of lithium bis(trimethylsilyl)amide
was added dropwise via syringe until the color of the reaction
mixture changed from orange to violet. This required 0.41 mL (0.41
mmol, 1.2 equiv) of base. This mixture was stirred at -78 °C for
15 min while a solution of 3 (0.11 g, 0.34 mmol, 1 equiv) in THF
(2.2 mL) under an Ar atmosphere was prepared and cooled to -78
°C. The alkoxide solution was transferred rapidly by cannula to
the solution of 3, and the resulting mixture was maintained at -78
°C for 10 h. The reaction mixture was quenched with 5 mL of
saturated NH4Cl solution, and the aqueous phase was extracted 3
times with Et2O (10 mL each). The combined organic extracts were
washed with brine, dried over Na2SO4, and concentrated to an amber
oil. The crude material was purified by flash column chromatog-
raphy using gradient elution (25%, 50% EtOAc/hexane) to provide
4a (0.13 g, 88%) as a light amber oil. Rf ) 0.31, 50% EtOAc/
31P NMR Kinetics. A 31P NMR kinetics experiment was carried
out at 37 °C on a 250 MHz Bruker NMR using the Bruker variable
temperature unit to control the probe temperature. Compound 4b
(7.7 mg, 0.021 mmol) in 1 mL of THF was treated with a
suspension of Pd/C, 10% (5 mg), in 1 mL of THF. A balloon filled
with H2 was applied to the flask. After 5 min, the suspension was
filtered through two pieces of filter paper, and the filtrate was
concentrated until the solvent had just evaporated (bath T ) 30
°C). Sodium cacodylate buffer (0.4 M, 0.5 mL) was added to the
resulting oil, and the time of dilution was noted. Triethylamine (0.5
mL) was added to adjust the pH to 7.68, and the sample was
immediately inserted into the spectrometer. The elapsed time from
addition of buffer was noted at the start of acquisition. Spectra were
acquired every 2.5 min for 30 min, then every 5 min for 30 min
and every 10 min for 1 h. A pH of 7.14 was recorded at the end of
the experiment. The relative concentrations of reaction intermediates
were determined by measuring the peak areas, and the relative
concentration for each reactant or product at any given time is
represented as a percent of the total. The phosphonamidate anion
derived from 4b disappeared according to first-order kinetics, and
the half-life (t1/2) was determined to be 44 min at 37 °C by linear
regression of the log(percent remaining) values versus time.
Diethyl Difluoromethylbenzylphosphonate (2). The procedure
of Burke and co-workers31 was used. Diethyl benzoylphosphonate
(2.58 g, 10.65 mmol), prepared from benzoyl chloride and triethyl
phosphate as described,30 and (diethylamino)sulfur trifluoride
(DAST) (6.0 mL, 45.41 mmol, 4.3 equiv) were stirred together at
0 °C under an Ar atmosphere. The reaction was warmed to ambient
temperature over 1 h and continued to stir for 20 h. The reaction
mixture was then rechilled to 0 °C and added to a 25% solution of
NaHCO3 in H2O (100 mL) cooled to 0 °C. The resulting suspension
was stirred for 15 min, 150 mL of H2O was added, and the aqueous
phase was extracted with four 50 mL portions of CHCl3. The
combined organic extracts were washed with brine, dried over Na2-
SO4, and concentrated to a brown oil. The crude material was
purified by flash column chromatography using 10% EtOAc/hexane.
The product 2 was isolated (1.18 g, 42%) as an amber oil Rf )
0.23, 20% EtOAc/hexane. 1H NMR (CDCl3): δ 7.62 (2H, d) 7.45
(3H, m) 4.22-4.12 (4H, m) 1.30 (6H, t); 31P NMR (CDCl3) δ
-19.07 (t); HRMS (C11H15F2O3P) calcd 265.0805, found 265.0802.
1
hexane. H NMR (CDCl3): δ 7.55 (2H, d); 7.42 (3H, m); 7.20
(1H, d); 6.53 (1H, d); 5.00 (2H, m); 3.49 (2H, t); 2.95 (2H, m)
2.65 (3H, d); 1.71-1.56 (4H, m); 31P NMR (CDCl3) δ -10.40
(dd); HRMS (C17H20ClF2N2O5P) calcd 459.0664 (M + Na)+, found
459.0664.
O-Benzyl N-Methyl-N-(4-chlorobutyl) Difluoromethylphen-
ylphosphonamidate (4b). Compound 4b was obtained in 70% yield
using the procedure described for the synthesis of 4a. Benzyl alcohol
was substituted for nitrofurfuryl alcohol, and BuLi was used instead
of lithium bis(trimethylsilyl)amide. The crude product was purified
by flash column chromatography (10% EtOAc/hexane) to give 4b
(0.090 g, 70%), Rf ) 0.54, 50% EtOAc/hexane. 1H NMR
(CDCl3): δ 7.59 (2H, d); 7.42 (3H, m); 7.30 (5H, m); 5.03 (2H,
d); 3.47 (2H, t); 2.99 (2H, m); 2.62 (3H, d); 1.67-1.48 (4H, m);
31P NMR (CDCl3): δ -11.32 (dd); HRMS (C19H23ClF2NO2P) calcd
402.1201, found 402.1202.
Methyl 4-Iodophenylacetate (7a). 4-Iodophenylacetic acid (5.56
g, 22.06 mmol) was suspended in MeOH (35 mL) and stirred under
a drying tube at 0 °C. Thionyl chloride (3.20 mL, 43.84 mmol, 2
equiv) was added by addition funnel. The reaction came gradually
to room temperature and was stirred overnight. The solution was
then concentrated to ca. half of its initial volume, diluted with 50
mL of EtOAc, and washed successively with 25 mL portions of
H2O, saturated NaHCO3 solution, and brine. Concentration after
drying over Na2SO4 gave 7a (5.79 g, 95%) as a yellow oil with
spectral data that were consistent with literature values.40 The
product was used without purification. Rf ) 0.13, 15% EtOAc/
hexane.
Methyl N-Fmoc-(L)-4-iodophenylalaninate (7b). Compound 7b
(9.84 g, 96%) was obtained as a white solid using the procedure
described for compound 7a and was used without further purifica-
tion. Spectral data were consistent with published values. Rf ) 0.23,
20% EtOAc/hexane.
N-Methyl-N-(4-chlorobutyl)
Difluoromethylphenylphos-
Methyl 4-[(Diethyl Phosphono(difluoromethyl)]phenylacetate
(8a). The procedure of Qabar32 with a slight modification was
employed as follows. Cadmium powder (7.5 g, 66.7 mmol, 3.2
equiv) in 74 mL of DMF was treated with TMSCl (0.27 mL, 2.12
mmol, 0.1 equiv). After 5 min, diethyl (bromodifluoromethyl)-
phosphonate (11.5 mL, 64.7 mmol, 3.1 equiv) was added dropwise
from an addition funnel. The reaction mixture became warm and
cloudy. If no exothermicity was observed after adding approxi-
mately one-third of the phosphonate, an additional portion of
chlorotrimethylsilane was added before continuation of the addition
of phosphonate. Formation of BrCdCF2P(O)(OEt)2 was deemed
complete by the absence of starting phosphonate (TLC, 15% EtOAc/
hexane). Aryl iodide 7a (5.79 g, 21.0 mmol, 1 equiv) and CuCl
(8.3 g, 83.8 mmol, 4 equiv) were suspended in 17 mL of DMF. To
phonamidochloridate (3). Compound 2 (0.25 g, 0.95 mmol) was
dissolved in thionyl chloride (5 mL), three drops of pyridine were
added, and the mixture was refluxed under an atmosphere of Ar
for 5 days at which time TLC (10% EtOAc/hexane) showed no
starting material remaining. Thionyl chloride was removed under
reduced pressure to give the corresponding dichloride (31P NMR
(CDCl3) δ +5.93 (t)), which was taken up in dry CH2Cl2 (12 mL)
and was cooled to -78 °C with stirring under an Ar atmosphere.
N-Methyl-N-(4-chlorobutyl)amine hydrochloride (0.14 g, 0.89
mmol, 0.94 equiv) was added followed by triethylamine (0.27 mL,
1.94 mmol, 2.0 equiv). The reaction was maintained at -78 °C for
20 h. Solvent was removed under reduced pressure, and the residue
was purified by flash column chromatography using 10% EtOAc/

862 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Boutselis et al.
1
this suspension, the organocadmium reagent was added by cannula.
The resulting mixture was stirred under Ar at room temperature
for 4 days. The reaction was diluted with 75 mL of Et2O and filtered
through celite. The filtrate was washed with brine, dried over Na2-
SO4, and concentrated to a gold oil. The oil was subjected to
Kugelrohr distillation at reduced pressure (1.5 Torr, 80 °C) to
remove volatile byproducts. Compound 8a (6.40 g, 91%) was
obtained from the distillation bulb as a yellow oil and was used
without further purification. Rf ) 0.37, 5% Et2O/CH2Cl2. 1H NMR
(CDCl3): δ 7.56 (2H, d); 7.37 (2H, d); 4.18 (4H,m); 3.70 (3H, s);
3.66 (2H, s); 1.33 (6H, t); 31P NMR (CDCl3): δ -19.20 (t); HRMS
(C14H19F2O5P) calcd 336.0938, found 336.0938.
Methyl 4-[(Diethyl Phosphono(difluoromethyl)]-N-Fmoc-(L)-
phenylalaninate (8b). The procedure utilized for the synthesis of
8a was employed to prepare 8b (4.92 g, 80%). Purification was
effected by Kugelrohr distillation (80 °C, 1.55 Torr) to remove
volatile side products. The remaining oil was further purified by
flash column chromatography (gradient elution: 0, 20, 50% EtOAc/
hexane). Rf ) 0.31, 10% Et2O/CH2Cl2. Spectral data were consistent
with published values.32
Rf ) 0.50, 70% EtOAc/hexane. H NMR (CDCl3): δ 7.72 (2H,
d); 7.53-7.17 (11H, m); 6.50 (1H, bs); 5.02-4.87 (2H, m); 4.63
(1H, m); 4.35 (2H, m); 4.15 (1H, t); 3.68 (3H, s); 3.48 (2H, t);
3.16-2.98 (4H, m); 2.62 (3H, d); 1.68-1.53 (4H, m); 31P NMR
(CDCl3): δ -10.50 (dd); HRMS (C36H37ClF2N3O9P) calcd 782.1822
(M + Na)+, found 782.1816.
General Procedure A: Methyl Ester Hydrolysis using LiOH.
The methyl ester (0.35 mmol) in 5.5 mL of THF/H2O (5:1) was
cooled to 0 °C, and 0.3 M LiOH/H2O (2 equiv) was added dropwise
in the following increments: 1.1, 0.4, 0.4, and 0.4 mL. The reaction
was monitored by TLC (2% MeOH, 1% AcOH/CH2Cl2) after each
addition of base. The reaction was quenched with 10 mL of 10%
citric acid 10 min after the final addition of base. The reaction
mixture was extracted with EtOAc (4 × 10 mL), and the combined
extracts were washed with brine, dried over Na2SO4, and concen-
trated to an oil that was purified by silica gel chromatography.
General Procedure B: Methyl Ester Hydrolysis using
Me3SnOH. The hydrolysis procedure of Nicolaou and co-workers35
was employed. The methyl ester (0.05 mmol) was dissolved in 1,2-
dichloroethane (1 mL), and Me3SnOH (0.22 mmol, 4.4 equiv) was
added. The reaction mixture was heated to reflux for 20 h at which
time the disappearance of the starting material was observed by
TLC (50% EtOAc/hexane). Solvent was removed under reduced
pressure, and the residue was taken up in 5 mL of EtOAc. The
solution was washed with brine, dried over Na2SO4, and concen-
trated to an oil that was purified by silica gel chromatography.
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)] Phenylacetic Acid (11). General procedure B
with 10a (0.040 g, 0.08 mmol), reaction time ) 3 h. The product
was purified by silica gel chromatography (20%EtOAc, 1%
HCO2H/hexane; 50% EtOAc, 1% HCO2H/hexane; 70% EtOAc, 1%
HCO2H/hexane) and isolated as a pale yellow oil, 0.027 g (69%).
Rf ) 0.48, 70% EtOAc, 1% HCO2H/hexane. 1H NMR (CDCl3): δ
7.52 (2H, d); 7.34 (2H, d); 7.20 (1H, d); 6.49 (1H, d); 5.02 (2H,
dd); 3.68 (3H, s); 3.52 (2H, t); 3.05 (2H, m); 2.72 (3H, d); 1.70
(4H, m); 31P NMR (CDCl3): δ -10.78 (dd); HRMS (C19H22-
ClF2N2O7P) calcd 517.0713 (M + Na)+, found 517.0720.
Methyl 4-[Chloro-N-methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)]phenylacetate (9a). The procedure of Landry and
co-workers13 was used to activate diethylphosphonate 8a. Thus,
8a (3.02 g, 8.98 mmol) in 70 mL of dry CH2Cl2 was treated with
bromotrimethylsilane (3.50 mL, 26.5 mmol, 3 equiv), and the
mixture was stirred under a drying tube for 18 h. Solvent was
removed under reduced pressure, and the residue was coevaporated
with CH2Cl2 (2 × 30 mL) to remove excess bromotrimethylsilane.
The resulting oil was taken up in CH2Cl2 (70 mL), and oxalyl
chloride (2.40 mL, 27.4 mmol, 3 equiv) was added followed by 2
drops of dry DMF. The reaction mixture was stirred under an Ar
atmosphere for 1.5 h, solvent was removed under reduced pressure,
and the residue was coevaporated with CH2Cl2 (2 × 30 mL) to
remove excess oxalyl chloride. The resulting oil was taken up in
CH2Cl2 (70 mL) and was cooled to -20 °C with stirring under an
Ar atmosphere. N-Methyl-N-(4-chlorobutyl)amine hydrochloride
(1.56 g, 9.87 mmol, 1.1 equiv) was added followed by dropwise
addition of diisopropylethylamine (3.40 mL, 19.5 mmol. 2.2 equiv).
The reaction mixture was maintained at -20 °C for 20 h. Solvent
was removed under reduced pressure, and the residue was purified
by flash column chromatography using 3% Et2O/CH2Cl2 as eluent
providing 9a (2.54 g, 70%) as a yellow oil. Rf ) 0.56, 5% Et2O/
CH2Cl2. 1H NMR (CDCl3): δ 7.58 (2H, d); 7.40 (2H, d); 3.71 (3H,
s); 3.68 (2H, s); 3.49 (2H, t); 3.25-3.02 (2H, m); 2.81 (3H, d);
1.72-1.52 (4H, m); 31P NMR (CDCl3): δ +3.99 (t); HRMS
(C15H20Cl2F2NO3P) calcd (M + Na)+ 424.0418, found 424.0437.
Methyl 4-[Chloro N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)]-N-Fmoc-(L)-phenylalaninate (9b). The same
procedure used for the synthesis of 9a was employed to provide
9b (2.89 g, 85%) after purification by flash column chromatography
(2% Et2O/CH2Cl2). Rf ) 0.35, 5% Et2O/CH2Cl2. 1H NMR
(CDCl3): δ 7.75 (2H, d); 7.53 (4H, m); 7.40 (2H, t); 7.31 (2H, d);
7.22 (2H, d); 5.55 (1H, t); 4.66 (1H, m); 4.43-4.35 (2H, m); 4.19
(1H, t); 3.70 (3H, s); 3.46 (2H, t); 3.18-3.05 (4H, m); 2.75 (3H,
dd); 1.62-1.44 (4H, m); 31P NMR (CDCl3): δ +3.95 (t); HRMS
(C31H33Cl2F2N2O5P) calcd (M + Na)+ 675.1370, found 675.1375.
Methyl 4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phos-
phono(difluoromethyl)] Phenylacetate (10a). Compound 10a
(4.11 g, 60%) was obtained using the same method employed in
the synthesis of 4a. The compound was purified by flash column
chromatography using gradient elution (50%, 60% EtOAc/hexane).
Methyl 4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phos-
phono(difluoromethyl)]-N-Fmoc-(L)-phenylalanine (12a). Com-
pound 10b (1.60 g, 2.10 mmol, 1 equiv) in THF/H2O (30 mL/6
mL) reacted with LiOH/H2O (0.3 M, 14 mL) as described in
General Procedure A to provide the crude product as a foam.
Purification by silica gel chromatography (25% EtOAc/hexane; 50%
EtOAc/hexane; then 50% EtOAc, 1% HCO2H/hexane) provided
the product as a foam, 1.30 g (83%). Rf ) 0.52, 70% EtOAc, 1%
1
HCO2H/hexane. H NMR (CDCl3): δ) 7.78 (2H, d); 7.58-7.21
(11H, m); 6.49 (1H, dd); 5.52 (1H, m); 4.97 (2H, m); 4.66 (1H,
m); 4.44-4.35 (2H, m); 4.18 (1H, t); 3.50 (2H, m); 3.19 (2H, m);
3.00 (2H, m); 2.68 (3H, d); 1.60 (4H, m); 31P NMR (CDCl3): δ
-10.58 (t); HRMS (C35H35ClF2N3O9P) calcd 790.1479 (M +
2Na)+, found 790.1469.
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)]-N-Fmoc-(L)-phenylalanine Amide (12b). Com-
pound 12a (0.41 g, 0.55 mmol, 1 equiv) in THF (5.7 mL) was
treated with N-hydroxysuccinimide (0.06 g, 0.55 mmol, 1 equiv)
and was cooled to 0 °C. Diisopropylcarbodiimide (0.13 mL, 0.82
mmol, 1.5 equiv) was added by syringe. The reaction mixture was
stirred at 0 °C for 1.5 h, and a 1 M solution of NH4HCO3 (2.7 mL,
2.7 mmol, 5 equiv) was added. The reaction mixture was stirred
for 20 h as the temperature gradually came to ambience. Solvent
was removed under reduced pressure, CH2Cl2 (10 mL) was added,
and the mixture was washed successively with 10 mL portions of
1 M HCl, saturated NaHCO3, and brine. The organic phase was
dried over Na2SO4 and concentrated. Purification by silica gel
chromatography (50% EtOAc/hexane; 75% EtOAc/hexane) pro-
vided the product as a yellow oil, 0.21 g (52%). Rf ) 0.26, 100%
1
Rf ) 0.29, 60% EtOAc/hexane. H NMR (CDCl3): δ 7.54 (2H,
d); 7.36 (2H, d); 7.25 (1H, d); 6.54 (1H, d); 5.03 (2H, m); 3.72
(3H, s); 3.68 (2H, s); 3.54 (2H, t); 3.07 (2H, m); 2.72 (3H, d);
1.78-1.66 (4H, m); 31P NMR (CDCl3): δ -10.54 (t); HRMS
(C20H24ClF2N2O7P) calcd (M + Na)+ 531.0875, found 531.0872.
Methyl 4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phos-
phono(difluoromethyl)]-N-Fmoc-(L)-phenylalaninate (10b). The
procedure utilized for the synthesis of 4a was applied to obtain
10b (1.92 g, 41%) The compound was purified by flash column
chromatography using gradient elution (30, 60, 80% Et2O/hexane).
1
EtOAc. H NMR (CDCl3) δ 7.72 (2Η, δ); 7.54-7.25 (10H, m);
7.18 (1Η, d); 6.51 (1H, d); 6.25 (1H, bs); 5.93 (1H, bs); 5.74 (1H,
m); 4.94 (2H, m); 4.47-4.35 (3H, m); 4.15 (1H, t); 3.50 (2H, t);
3.11-2.99 (4H, m); 2.66 (3H, d); 1.67-1.57 (4H, m); 31P NMR

Protein Tyrosine Phosphatase 1B Inhibitor Prodrug
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4 863
(CDCl3): δ -10.33 (dd); HRMS (C35H36ClF2N4O8P) calcd 767.1820
(M + Na)+, found 767.1821.
Fmoc-(L)-phenylalanine Amide (1b). Compound 15 (0.067 g,
0.058 mmol) in CH2Cl2 (1.1 mL) was treated with TFA (0.57 mL),
and the reaction mixture was stirred at room temperature for 1.5 h.
The reaction was quenched with 5 mL of H2O and extracted with
EtOAc (4 × 10 mL). The combined organic extracts were washed
successively with 5 mL portions of water and brine, then were dried
over Na2SO4 and concentrated to a yellow oil. Purification by silica
gel chromatography (100% EtOAc; 2% MeOH, 1% AcOH/CH2-
Cl2; 4% MeOH, 1% AcOH/CH2Cl2) provided the product 1b as a
pale yellow oil, 0.049 g (77%). Rf ) 0.21, 4% MeOH, 1% HCO2H/
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)] Phenylacetyl-(L)-aspartic Acid â-t-Butyl Ester
R-Methyl Ester (14a). Compound 13 (0.25 g, 0.59 mmol) was
dissolved in CH2Cl2 (3 mL), and the solution was cooled to 0 °C.
A 0.17 M solution of DBU in CH2Cl2 (3.8 mL, 0.65 mmol, 1.1
equiv) was added. While this mixture stirred, a solution of 11 (0.29
g, 0.59 mmol, 1.0 equiv) and PyBOP (0.31 g, 0.59 mmol, 1.0 equiv)
in CH2Cl2 (3 mL) was prepared and cooled to 0 °C. TLC (2%
MeOH, 1% AcOH/CH2Cl2) of the first reaction mixture after 20
min at 0 °C showed no starting material remaining. Deprotected
13 was transferred to the solution of 11, and DIEA (0.22 mL, 1.26
mmol, 2.1 equiv) was added. The solution was stirred at 0 °C for
10 min and at room temperature for 50 min. TLC (75% EtOAc/
hexane) showed no 11 remaining and the appearance of a new less
polar spot. The reaction mixture was quenched with saturated NH4-
Cl (10 mL). The phases were separated, and the aqueous phase
was extracted with CH2Cl2 (2 × 10 mL). The combined organic
phases were washed with brine, dried over Na2SO4, and concen-
trated to a dark, viscous oil. Purification by silica gel chromatog-
raphy (50% EtOAc/hexane; 75% EtOAc/hexane) provided the
product 14a as yellow oil, 0.32 g (79%) Rf ) 0.35, 75% EtOAc/
hexane. 1H NMR (CDCl3): δ 7.55 (2Η, d); 7.35 (2H, d); 7.24 (1H,
d); 6.55 (1H, d); 6.46 (1H, d); 5.00 (2H, m); 4.77 (1H, m); 3.71
(3H, s); 3.63 (2H, s); 3.53 (2H, t); 3.07 (2H, m); 2.91 (1H, dd);
2.69 (4H, m); 1.71-1.57 (4H, m); 1.36 (9H, s); 31P NMR
(CDCl3): δ -10.37 (dd); HRMS (C28H37ClF2N3O10P) calcd
702.1771 (M + Na)+, found 702.1764.
1
CH2Cl2. H NMR (CD3OD): δ 8.10 (4H, bs); 7.47-7.35 (10H,
m); 6.75 (2H, m); 5.09 (4H, m); 4.65 (1H, t); 4.57 (1H, m); 3.61-
3.54 (6H, m); 3.10-2.95 (6H, m); 2.68 (6H, d); 2.63 (2H, m); 1.68-
1.60 (8H, m); 31P NMR (CD3OD): δ -9.85 (t); 13C NMR
(CD3OD): δ 173.67 (s); 173.32 (s); 173.04 (s); 172.82 (s); 172.66
(s); 153.79 (s); 153.32 (s); 142.03 (s); 139.94 (s); 132.45 (s); 130.60
(d); 127.34 (d); 115.19 (d); 113.23 (d); 59.92 (t); 55.44 (d); 51.72
(d); 51.41 (d); 49.32 (t); 45.39 (t); 43.02 (t); 38.28 (t); 36.61 (t);
33.78 (q); 30.55 (t); 26.10 (t); HPLC: flow rate ) 1 mL/min,
retention time ) 14.5 min, C18 column, 50% CH3CN/H2O; 5.4
min, phenyl column, 30% CH3CN/H2O.
Acknowledgment. We appreciate the helpful discussions
with Christine O’Day throughout the course of this project.
Financial Support from the NCI (R01 CA34619 to R.F.B.) and
NIDDK (RO1 DK68447 to Z.-Y.Z.) is gratefully acknowledged.
Support from the Purdue Cancer Center Support Grant P30
CA23168 for services provided by the NMR and Mass
Spectrometry Shared Resources is appreciated.
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)] Phenylacetyl-(L)-aspartic Acid â-t-Butyl Ester
(14b). Compound 14a (0.24 g, 0.35 mmol, 1 equiv) in THF/H2O
(4.6 mL/0.91 mL) was reacted with 0.3 M LiOH/H2O (2.30 mL,
0.70 mmol, 2 equiv) according to General Procedure A. The crude
product was isolated as a yellow oil. Purification by silica gel
chromatography (75%EtOAc/hexane, 100% EtOAc) provided the
product 14b as a pale yellow oil, 0.22 g (94%). Rf ) 0.19, 4%
References
(1) Hunter, T. Signalings2000 and beyond. Cell 2000, 100, 113-27.
(2) Zhang, Z. Y. Protein tyrosine phosphatases: prospects for therapeu-
tics. Curr. Opin. Chem. Biol. 2001, 5, 416-23.
(3) Lund, I. K.; Andersen, H. S.; Iversen, L. F.; Olsen, O. H.; Moller,
K. B.; Pedersen, A. K.; Ge, Y.; Holsworth, D. D.; Newman, M. J.;
Axe, F. U.; Moller, N. P. Structure-based design of selective and
potent inhibitors of protein-tyrosine phosphatase beta. J. Biol. Chem.
2004, 279, 24226-35.
(4) Blume-Jensen, P.; Hunter, T. Oncogenic kinase signalling. Nature
2001, 411, 355-65.
(5) Ventura, J. J.; Nebreda, A. R. Protein kinases and phosphatases as
therapeutic targets in cancer. Clin. Trans. Oncol. 2006, 8, 153-60.
(6) Goldman, J. M.; Melo, J. V. Targeting the BCR-ABL tyrosine kinase
in chronic myeloid leukemia. N. Engl. J. Med. 2001, 344, 1084-6.
(7) Berman, J.; O’Leary, T. J. Gastrointestinal stromal tumor workshop.
Hum. Pathol. 2001, 32, 578-82.
(8) Li, L.; Dixon, J. E. Form function and regulation of protein tyrosine
phosphatases and their involvement in human diseases. Semin.
Immunol. 2000, 12, 75-84.
(9) Bourdeau, A.; Dube, N.; Tremblay, M. L. Cytoplasmic protein
tyrosine phosphatases regulation and function: the roles of PTP1B
and TC-PTP. Curr. Opin. Cell Biol. 2005, 17, 203-9.
(10) Zhang, Z. Protein tyrosine phosphatases: structure and function,
substrate specificity, and inhibitor development. Ann. ReV. Pharma-
col. Toxicol. 2002, 42, 209-34.
(11) Burke, T. R., Jr; Zhang, Z.-Y. Protein-tyrosine phosphatases:
structure mechanism and inhibitor discovery. Biopolymers 1998, 47,
225-41.
(12) Zhang, Z.-Y. Inhibitors of protein tyrosine phosphatases. In Handbook
of Cell Signaling, 1st ed.; Bradshaw, R., Dennis, E., Eds; Academic
Press: New York, 2003; pp 677-684.
(13) Burke, T. R., Jr; Kole, H. K.; Roller, P. P. Potent inhibition of insulin
receptor dephosphorylation by a hexamer peptide containing the
phosphotyrosyl mimetic F2Pmp. Biochem. Biophys. Res. Commun.
1994, 204, 129-34.
(14) Chen, L.; Wu, L.; Otaka, A.; Smyth, M. S.; Roller, P. P.; Burke, T.
R., Jr; den Hertog, J.; Zhang, Z.-Y. Why is phosphonodifluoromethyl
phenylalanine a more potent inhibitory moiety than phosphonomethyl
phenylalanine toward protein-tyrosine phosphatases? Biochem.
Biophys. Res. Commun. 1995, 216, 976-84.
(15) Shen, K.; Keng, Y. F.; Wu, L.; Guo, X. L.; Lawrence, D. S.; Zhang,
Z.-Y. Acquisition of a specific and potent PTP1B inhibitor from a
novel combinatorial library and screening procedure. J. Biol. Chem.
2001, 276, 47311-9.
1
MeOH, 1% HCO2H/CH2Cl2. H NMR (CDCl3): δ 7.47 (2H, d);
7.33 (2H, d); 7.21 (1H, d); 6.82 (1H, t); 6.54 (1H, d); 4.99 (2H,
m); 4.74 (1H, m); 3.62 (2H, s); 3.50 (2H, t); 3.02 (2H, m); 2.88
(1H, dd); 2.69 (4H, m); 1.67-1.55 (4H, m); 1.34 (9H, s); 31P NMR
(CDCl3): δ -10.57 (dd); HRMS (C27H35ClF2N3O10P) calcd
710.1428 (M + 2Na)+, found 710.1427.
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)]phenylacetyl-(L)-aspartyl-â-t-butyl Ester 4-[O-
Nitrofurfuryl-N-methyl-N-(4-chlorobutyl) Phosphono(difluo-
romethyl)]-N-Fmoc-(L)-phenylalanine Amide (15). Compound
12b (0.16 g, 0.21 mmol) was dissolved in CH2Cl2 (3 mL), and the
solution was cooled to 0 °C. A 0.17 M solution of DBU in CH2Cl2
(1.73 mL, 0.29 mmol, 1.4 equiv) was added. While this mixture
stirred, a solution of 14b (0.18 g, 0.27 mmol, 1.3 equiv) and PyBOP
(0.14 g, 0.27 mmol, 1.3 equiv) in CH2Cl2/THF (3 mL/1 mL) was
prepared and cooled to 0 °C. TLC (2% MeOH, 1% AcOH/CH2-
Cl2) of the first reaction mixture after 30 min at 0 °C showed no
Fmoc-protected starting material remaining. Deprotected 12b was
transferred to the solution of 14b, and DIEA (0.10 mL, 0.56 mmol,
2.7 equiv) was added. The reaction mixture was stirred at 0 °C for
10 min and at room temperature for 30 min. The reaction was
quenched with saturated NH4Cl (10 mL). The phases were
separated, and the aqueous phase was extracted with CH2Cl2 (3 ×
10 mL). The combined organic phases were washed with brine,
dried over Na2SO4, and concentrated to a dark oil. Purification by
silica gel chromatography (2% MeOH, 1% AcOH/CH2Cl2) provided
the product 15 as a pale yellow oil, 0.090 g (37%). Rf ) 0.38, 2%
MeOH, 1% HCO2H/CH2Cl2. 1H NMR (CD3COCD3): δ 7.62-7.37
(10H, m); 6.81 (2H, d); 5.16 (4H, m); 4.77 (2H, m); 3.67 (2H, s);
3.58 (4H, t); 3.06-2.95 (6H, m); 2.75 (2H, m); 2.68 (6H, d); 1.72-
1.55 (8H, m); 1.38 (9H, s); 31P NMR (acetone-d6): δ -9.90 (t).
4-[O-Nitrofurfuryl N-Methyl-N-(4-chlorobutyl) Phosphono-
(difluoromethyl)]phenylacetyl-(L)-aspartyl-4-[O-nitrofurfuryl
N-Methyl-N-(4-chlorobutyl) Phosphono(difluoromethyl)]-N-

864 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 4
Boutselis et al.
(16) Dufresne, C.; Roy, P.; Wang, Z.; Asante-Appiah, E.; Cromlish, W.;
Boie, Y.; Forghani, F.; Desmarais, S.; Wang, Q.; Skorey, K.;
Waddleton, D.; Ramachandran, C.; Kennedy, B. P.; Xu, L.; Gordon,
R.; Chan, C. C.; Leblanc, Y. The development of potent non-peptidic
PTP 1B inhibitors. Bioorg. Med. Chem. Lett. 2004, 14, 1039-42.
(17) Lau, C. K.; Bayly, C. I.; Gauthier, J. Y.; Li, C. S.; Therien, M.;
Asante-Appiah, E.; Cromlish, W.; Boie, Y.; Forghani, F.; Desmarais,
S.; Wang, Q.; Skorey, K.; Waddleton, D.; Payette, P.; Ramachandran,
C.; Kennedy, B. P.; Scapin, G. Structure-based design of a series of
potent and selective non-peptidic PTP 1B inhibitors. Bioorg. Med.
Chem. Lett. 2004, 14, 1043-8.
(18) 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. Increased insulin sensitivity and obesity resistance in mice
lacking the protein tyrosine phosphatase 1B gene. Science 1999, 283,
1544-8.
(19) Klaman, L. D.; Boss, O.; Peroni, O. D.; Kim, J. K.; Martino, J. L.;
Zabolotny, J. M.; Moghal, N.; Lubkin, M.; Kim, Y. B.; Sharpe, A.
H.; Stricker-Krongrad, A.; Shulman, G. I.; Neel, B. G.; Kahn, B. B.
Increased energy expenditure decreased adiposity and tissue-specific
insulin sensitivity in protein--tyrosine phosphatase 1B deficient mice.
Mol. Cell Biol. 2000, 20, 5479-89.
(20) Zhang, Z.-Y.; Lee, S. Y. PTP1B inhibitors as potential therapeutics
in the treatment of type 2 diabetes and obesity. Expert Opin. InVestig.
Drugs 2003, 12, 223-33.
(21) Xie, L.; Lee, S. Y.; Andersen, J. N.; Waters, S.; Shen, K.; Guo, X.
L.; Moller, N. P.; Olefsky, J. M.; Lawrence, D. S.; Zhang, Z.-Y.
Cellular effects of small molecule PTP1B inhibitors on insulin
signaling. Biochemistry 2003, 42, 12792-804.
(22) Lee, S. Y.; Liang, F.; Guo, X. L.; Xie, L.; Cahill, S. M.; Blumenstein,
M.; Yang, H.; Lawrence, D. S.; Zhang, Z.-Y. Design, construction,
and intracellular activation of an intramolecularly self-silenced signal
transduction inhibitor. Angew. Chem., Int. Ed. 2005, 44, 4242-4.
(23) Liang, F.; Lee, S. Y.; Liang, J.; Lawrence, D. S.; Zhang, Z.-Y. The
role of protein-tyrosine phosphatase 1B in integrin signaling. J. Biol.
Chem. 2005, 280, 24857-63.
(24) Lee, H.; Xie, L.; Luo, Y.; Lee, S. Y.; Lawrence, D. S.; Wang, X. B.;
Sotgia, F.; Lisanti, M. P.; Zhang, Z.-Y. Identification of phospho-
caveolin-1 as a novel protein-tyrosine phosphatase 1B substrate.
Biochemistry 2006, 45, 234-40.
(25) Dang, Q. Organophosphonic acids as drug candidates. Expert Opin.
Ther. Patents 2006, 16, 343-8.
(27) Freel Meyers, C. L.; Hong, L.; Joswig, C.; Borch, R. F. Synthesis
and biological activity of novel 5-fluoro-2′-deoxyuridine phosphor-
amidate prodrugs. J. Med. Chem. 2000, 43, 4313-8.
(28) Tobias, S. C.; Borch, R. F. Synthesis and biological studies of novel
nucleoside phosphoramidate prodrugs. J. Med. Chem. 2001, 44,
4475-80.
(29) Garrido-Hernandez, H.; Moon, K. D.; Geahlen, R. L.; Borch, R. F.
Design and synthesis of phosphotyrosine peptidomimetic prodrugs.
J. Med. Chem. 2006, 49, 3368-76.
(30) Terauchi, K.; Sakurai, H. Photochemical studies of esters of
aroylphosphonic acids. Bull. Chem. Soc. Jpn. 1970, 43, 883-90.
(31) Burke, T. R.; Smyth, M. S.; Otaka, A.; Roller, P. P. Synthesis of
4-phosphono(difluoromethyl)-D,L-phenylalanine and N-boc and N-
fmoc derivatives suitably protected for solid phase synthesis of
nonhydrolyzable phosphotyrosyl peptide analogues. Tetrahedron Lett.
1993, 34, 4125-8.
(32) Qabar, M. N.; Urban, J.; Kahn, M. A facile solution and solid phase
synthesis of phosphotyrosine mimetic L-4-[diethylphosphono(difluo-
romethyl)]-phenylalanine (F2Pmp(EtO)2 derivatives. Tetrahedron
1997, 53, 11171-8.
(33) Yang, G.; Chun, J.; Arakawa-Uramoto, H.; Wang, X.; Gawinowicz,
M. A.; Zhao, K.; Landry, D. W. Anti-cocaine catalytic antibodies:
a synthetic approach to improved antibody diversity. J. Am. Chem.
Soc. 1996, 118, 5881-90.
(34) Namikoshi, M.; Kundu, B.; Rinehart, K. L. Use of tetrabutylammo-
nium fluoride as a facile deprotecting reagent for 4-nitrobenzyl, 2,2,2-
trichloroethyl, and phenacyl esters of amino acids. J. Org. Chem.
1991, 56, 5464-6.
(35) Nicolaou, K. C.; Estrada, A. A.; Zak, M.; Lee, S. H.; Safina, B. S.
A mild and selective method for the hydrolysis of esters with
trimethyltin hydroxide. Angew. Chem., Int. Ed. 2005, 44, 1378-82.
(36) Guo, X. L.; Shen, K.; Wang, F.; Lawrence, D. S.; Zhang, Z.-Y.
Probing the molecular basis for potent and selective protein-tyrosine
phosphatase 1B inhibition. J. Biol. Chem. 2002, 277, 41014-22.
(37) Sun, J. P.; Fedorov, A. A.; Lee, S. Y.; Guo, X. L.; Shen, K.;
Lawrence, D. S.; Almo, S. C.; Zhang, Z.-Y. Crystal structure of
PTP1B complexed with a potent and selective bidentate inhibitor. J.
Biol. Chem. 2003, 278, 12406-14.
(38) White, M. F. The IRS-signaling system: a network of docking
proteins that mediate insulin action. Mol. Cell Biochem. 1998, 182,
3-11.
(39) Saltiel, A. R.; Kahn, C. R. Insulin signaling and the regulation of
glucose and lipid metabolism. Nature 2001, 414, 799-806.
(40) Clampit, J. E.; Meuth, J. L.; Smith, H. T.; Reilly, R. M.; Jirousek,
M. R.; Trevillyan, J. M.; Rondinone, C. M. Reduction of protein-
tyrosine phosphatase 1B increases insulin signaling in FAO hepatoma
cells. Biochem. Biophys. Res. Commun. 2003, 300, 261-7.
(26) Stankovic, C. J.; Surendran, N.; Lunney, E. A.; Plummer, M. S.;
Para, K. S.; Shahripour, A.; Fergus, J. H.; Marks, J. S.; Herrera, R.;
Hubbell, S. E.; Humblet, C.; Saltiel, A. R.; Stewart, B. H.; Sawyer,
T. K. The role of 4-phosphonodifluoromethyl- and 4-phosphono-
phenylalanine in the selectivity and cellular uptake of SH2 domain
ligands. Bioorg. Med. Chem. Lett. 1997, 7, 1904-9.
JM061146X