Light-switched inhibitors of protein tyrosine phosphatase PTP1B based on phosphonocarbonyl phenylalanine as photoactive phosphotyrosine mimetic

Article abstract of DOI:10.1016/j.bmc.2015.03.074

Phosphopeptide mimetics containing the 4-phosphonocarbonyl phenylalanine (pcF) as a photo-active phosphotyrosine isoster are developed as potent, light-switchable inhibitors of the protein tyrosine phosphatase PTP1B. The photo-active inhibitors 6-10 are d

Full text of DOI:10.1016/j.bmc.2015.03.074

Bioorganic & Medicinal Chemistry 23 (2015) 2839–2847
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Light-switched inhibitors of protein tyrosine phosphatase PTP1B
based on phosphonocarbonyl phenylalanine as photoactive
phosphotyrosine mimetic
Stefan Wagner a, Anja Schütz b, Jörg Rademann a,
⇑
a Freie Universität Berlin, Institute of Pharmacy, Medicinal Chemistry, Königin-Luise-Str. 2+4, 14195 Berlin, Germany
b Max-Delbrück-Center for Molecular Medicine (MDC), Robert-Rössle-Str. 10, 13125 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Phosphopeptide mimetics containing the 4-phosphonocarbonyl phenylalanine (pcF) as a photo-active
phosphotyrosine isoster are developed as potent, light-switchable inhibitors of the protein tyrosine phos-
phatase PTP1B. The photo-active inhibitors 6–10 are derived from phosphopeptide substrates and are
prepared from the suitably protected pcF building block 12 by Fmoc-based solid phase peptide synthesis.
Received 19 January 2015
Revised 23 March 2015
Accepted 24 March 2015
Available online 4 April 2015
All pcF-containing peptides are moderate inhibitors of PTP1B with KI values between 10 and 50 lM.
Irradiation of the inhibitors at 365 nm in the presence of the protein PTP1B amplify the inhibitory activity
of pcF-peptides up to 120-fold, switching the KI values of the best inhibitors to the sub-micromolar range.
Photo-activation of the inhibitors results in the formation of triplet intermediates of the benzoylphospho-
nate moiety, which deactivate PTP1B following an oxidative radical mechanism. Deactivation of PTP1B
proceeds without covalent crosslinking of the protein target with the photo-switched inhibitors and
can be reverted by subsequent addition of reducing agent dithiothreitol (DTT).
Keywords:
Protein tyrosine phosphatases
Enzyme inhibitors
Phosphotyrosine mimetics
Protein phosphorylation
PTP1B
Photo-switchable inhibitors
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
as potential drug targets in various diseases including diabetes,
cancer, and inflammation.7–9 The detailed role of different protein
Phosphorylation is one of the most important post-translational
modifications of proteins and plays a central role in the signal trans-
duction cascade transmitting extracellular signals into the cell, for
example, the binding of intercellular mediators, growth factors,
and cytokines, resulting in the activation or deactivation of gene
transcription.1,2 Protein phosphorylation occurs at serine, thre-
onine, histidine, and tyrosine residues and is effected by protein
kinases using O-50-adenosine triphosphate (ATP) as the activated
phosphate donor for the synthesis of protein phosphate
monoesters.3,4 Protein phosphatases are the physiological counter-
parts of protein kinases resulting in the hydrolysis of the phosphate
ester bond and are distinguished according to their substrates as
protein serine/threonine, histidine, and tyrosine phosphatases,
respectively.5 Considering the special significance of tyrosine phos-
phorylation in signal transduction events, for example, in the acti-
vation of receptor tyrosine kinases, protein tyrosine phosphatases
(PTPs) are of primordial importance for the regulation of cell
growth and development.5 Within the human genome ꢀ107 PTP
genes have been identified.6 Several of these have been recognized
tyrosine phosphatases is still under intense research as there is a
lack of specific and cell-permeable chemical tools to inhibit or
modulate these enzymes. Accordingly, despite of the proposed rele-
vance of PTPs as pharmacological targets, there are no clinically
admitted drugs available to date. For these reasons, there is con-
tinuing interest in the development of chemical tools for the mod-
ulation of PTP activities in vitro and in vivo.10,11
Mechanistically, protein tyrosine phosphatases bind the phos-
photyrosine residue in an enzyme pocket containing a protonated
arginine residue at the bottom. The mono- or dianionic phosphate
ester is coordinated by a hydrogen bonding network in which the
protein binding pocket acts as an oligodentate H-bond donor com-
posed of the guanidinium base and several backbone amide bonds
(Fig. 1).12,13 Coordination in the active site of the phosphatase
places the cleaved phosphate monoester next to a nucleophilic cys-
teine residue and together with the protonation of the released tyr-
osine OH the phosphate ester is attacked by the thiolate furnishing
a phosphate thioester intermediate, which is next hydrolyzed
employing water as a nucleophile.
Several non-cleavable phosphotyrosine mimetics have been
employed as inhibitors of PTPs so far, including difluoromethylphos-
phonates,14 aryl sulfonamides,15 isothiazolidinones,16 salicylic
acids,4,17 and pyrazolones18, to name only a few.19
⇑
Corresponding author. Tel.: +49 30 838 53272; fax: +49 30 838 459829.
E-mail address: joerg.rademann@fu-berlin.de (J. Rademann).
URL: http://www.bcp.fu-berlin.de/ag-rademann (J. Rademann).
http://dx.doi.org/10.1016/j.bmc.2015.03.074
0968-0896/Ó 2015 Elsevier Ltd. All rights reserved.

2840
S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
O
O
O
O
HO
H
H2N
O
H
N
H
H2N
H2N
H
H2N
H2N
O
P
O
N
O
P
O
O
O
hydrolysis
HPO4
O
P-O bond break
regenerated PTP1B
S
S
2-
NH
NH
H
H
N
O
O
R
R
HN
HN
O
H
N
H
N
N
OH
R
R
O
O
O
1st Step
2nd Step
O
O
R
R
cysteinyl-phosphate intermediate
O
HO
O
H
H2N
H2N
N
P
O
O
O
no reaction!
S
NH
H
O
R
HN
O
H
N
N
R
O
O
R
Figure 1. Schematic mechanism of the hydrolysis of phosphotyrosine substrates by PTP1B.
Recently, we have discovered benzoylphosphonates of general
structure 1 as photoactive, both chemically and enzymatically
stable phosphotyrosine mimetics (Scheme 1). Irradiation of 1 with
UV light at 365 nm leads to an n–p⁄ transition followed by inter-
system crossing and resulting in the formation of triplet-biradicals
2. The latter were found to undergo photodimerization yielding
diols 3 (R = H or Me), or—if recognized by a phosphotyrosine bind-
ing pocket—led to photocrosslinking of the target protein 4. In
order to enhance the binding affinity of the novel phosphotyrosine
mimetics, they were incorporated into phosphotyrosine peptide
mimetics. For this purpose, the amino acid derivative 4-phos-
phonocarbonyl phenylalanine (pcF) 5 was introduced (Scheme 1).
Indeed incorporation of this non-natural amino acid into the
sequence of a phosphotyrosine peptide recognized by the SH2
domain of the transcription factor STAT5b furnished an enzymati-
cally stable high-affinity ligand, which was suitable for the selec-
tive photocrosslinking, the functional deactivation and the
covalent labelling of the phosphotyrosine recognition domain
STAT5b–SH2.20
In this contribution the significant extension of the previous
work from phosphotyrosine-binding protein domains to the
functional inactivation of protein tyrosine phosphatases is
reported.
2. Results and discussion
PTP1B was selected as the model protein tyrosine phosphatase
as this enzyme is the PTP studied most extensively for several
decades.21 In this work we used the catalytic domain of PTP1B
(amino acids 1-321) that was produced from Escherichia coli
Rosetta (DE3) as previously described.20 The calculated molecular
weight of 37,456 Da could be verified with MALDI-mass spec-
trometry with a detected average mass of 37,461 Da.
2.1. Design of phosphotyrosine peptide mimetics
The hexapeptides Ac-DADEY⁄L-NH2 and Ac-NY⁄ILDL-NH2
(Y⁄ = phosphotyrosine) have been reported as highly active
OH
OH
PO3
2-
2-O3P
isopropanol
(R = H,or Me)
OH
P
OH
P
HO
HO
O
O
irradiation at
365 nm
R
R
3
O
O
OH
2-O3P
R
R
buffer
1
2
OH
P
R
HO
O
4
O
Non natural
amino acid
OH
H2N
O
4-phosphonocarbonyl phenylalanine
5
Scheme 1. Photoactivation of benzoylphosphonates 1: Irradiation of 1 generates a triplet biradical species 2, which undergoes photodimerization to photoproduct 3 in
isopropanol and can crosslink with protein binding sites.

S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
2841
phosphotyrosine-containing substrates of the enzyme PTP1, which
is the rat homologue of the human PTP1B, with about 97% identical
amino acids between the positions 1-322.22 Thus, it was our plan
to construct the non-cleavable, photoactive mimetics Ac-
DADEpcFL-NH2 (6) and Ac-NpcFILDL-NH2 (7) of these reported
phosphotyrosine peptides by incorporation of the novel phos-
photyrosine-mimicking amino acid, 4-phosphonocarbonyl pheny-
lalanine (pcF) in place of the native phosphotyrosine residues
(Y⁄) (Fig. 1). In order to check for covalent protein binding N-fluo-
rescein-modified peptide derivatives 8 and 9 were designed and
for control experiments, peptide 10 was envisaged as a pcF-con-
taining peptide with no reported affinity for the active site of
PTP1B. This peptide contains a sidechain biotinylated lysine for
the potential use in affinity chromatography applications.
sodium-loaded AmberliteÒ IR-120 ion exchange resin furnishing
the Fmoc-protected pcF building block 12, which could be directly
used in peptide synthesis.
For the synthesis of the 4-phosphonomethyl-peptide 11, the
preparation of a suitably protected 4phosphonomethyl phenylala-
nine building block 15 was required. N-Fmoc-4-carboxy phenylala-
nine methyl ester 14, an intermediate of the synthesis of building
block 12, was converted to the di-O-ethyl-4-phosphonomethyl
phenylalanine building block 15. First, the aromatic carboxylic acid
14 was reduced by the reaction with BH3⁄dimethylsulfide complex
to the benzylic alcohol 16. The alcohol was brominated with triph-
enylphosphine and N-bromosuccinimide furnishing benzyl bro-
mide 17. Alkylation of tri-O-ethyl phosphite with 17 yielded the
phosphonic ester intermediate 18, which was saponified with
potassium hydroxide in the presence of CaCl2 to preserve the
base-labile Fmoc-group providing the carboxylic acid 15, which
was a suitable starting point for the synthesis of 4-phospho-
nomethyl phenylalanine (pmF)-containing control peptides.
In order to compare the novel photoactive phosphotyrosine
mimetics containing 4-phosphonocarbonyl phenylalanine with
known phosphotyrosine mimetics, the peptide CF-NpmFILDL-NH2
11 was designed. It contained the 4-phosphonomethyl phenylala-
nine building block as a well-studied phosphotyrosine mimetic.
2.3. Synthesis of phosphotyrosine-mimetic peptides
2.2. Synthesis of phosphotyrosine-mimetic building blocks
The photoactive, pcF-containing building block which was
incorporated into the peptides using Fmoc-based solid phase pep-
tide synthesis on Rink amide resin. Coupling of the pcF building
block to a free amino terminus required only minor deviations from
the commonly used solid phase peptide synthesis protocol.
Activation of the pcF-building block succeeded by using N,N,N0,N0-
tetramethyl-O-(benzotriazol-1-yl)uronium tetrafluoroborate (TBTU)
as condensation reagent with superior coupling efficiency com-
pared to the commonly used N,N’-diisopropylcarbodiimide/1-
hydroxybenzotriazol (DIC/HOBt) system. Fmoc-deprotections after
incorporation of the photoreactive phosphotyrosine mimetic were
conducted with 2% of the non-nucleophilic base 1,8-diazabicyclo
[5.4.0]undec-7-en (DBU) instead of piperidine in order to avoid
the nucleophilic cleavage of the phosphonate residues furnishing
piperidylamides as products. In the final coupling step prior to pep-
tide cleavage, the N-terminal amino group of the hexapeptides was
acetylated employing acetic anhydride yielding peptides 6, 7 after
cleavage from the resin. Alternatively, 5,6-carboxyfluorescein was
coupled with DIC/HOBt (8, 9, 10, 11). Since activated carboxyfluo-
rescein can form oligomeric phenolic ester, which were cleaved
by treatment with 2% DBU.25 For the synthesis of the pcF-contain-
ing, fluorescein and biotin-labeled peptide, Fmoc-protected
N6-(biotin)-lysine was prepared according to literature starting
from Fmoc-N6-(Boc)-Lys-OH. After the cleavage of the tert-
butyloxycarbonyl group and adjusting the pH to 8, the sidechain
For the purpose of synthesizing the planned pcF-containing
peptides, the N-Fmoc- and O-mono-benzyl phosphonate protected
derivative of pcF, amino acid building block 12 was prepared from
L-N-Boc-O-triflyl-tyrosine methyl ester 13 in seven steps employ-
ing palladium-catalyzed hydro-carbonylation as the key step as
reported recently (Scheme 2).23 First, the triflate was reacted with
gaseous carbon monoxide in the presence of base, the catalyst
Pd(Ac)2 and the ligand 1,3-bis(diphenylphosphino)propane (dppp)
to furnish a carboxylic acid. After exchanging the N-protection
group from Boc to Fmoc compound 14 was obtained. The aromatic
carboxylate was activated as the corresponding acid chloride and
reacted with tribenzylphosphite in a Michaelis Arbuzow reaction.
As the dibenzylphosphonate reacts rapidly with nucleophiles like
piperidine24 which is used in Fmoc-deprotection, nucleophilic
monodebenzylation of the dibenzylphosphonate intermediate
was accomplished with lithium bromide yielding the correspond-
ing monobenzyl phosphonate. Next, the methyl ester was
saponified. Even though the transformation into a monobenzyl
protected phosphonate furnished sufficient stability towards base,
the conditions used for aqueous saponification of the methyl ester
led to cleavage of the ketophosphonate moiety. To avoid strong
basic media the ester was deprotected enzymatically with subtil-
isin Carlsberg (‘alcalase’) in NH4HCO3 at pH 8. After lyophilization
the carboxylic acid was converted into its sodium salt using
O
O
O
TfO
P
HO
solid phase
peptide synthesis
BnO
O
d, e, f
a, b, c
photoactive peptidomimetics
6, 7 8, 9 10
,
,
2Na
OMe
HO
O
BocHN
OMe
FmocHN
12
FmocHN
O
O
13
14
O
a, b, c
O
O
Br
P
P
solid phase
EtO
EtO
OEt
FmocHN
18
OEt
FmocHN
15
peptide synthesis
g
h
i
j
control pmF peptide
14
11
OMe
OMe
OMe
OH
FmocHN
FmocHN
O
O
O
O
16
17
Scheme 2. Reagents and conditions: (a) dppp, Pd(OAc)2, CO(g), DIPEA, DMF/H2O 3:1, 2 h, 70 °C, 89%; (b) TFA, DCM, 1 h, rt; (c) FmocOSu, Na2CO3, acetone/H2O, 3 h, 0 °C to rt,
87%; (d) (COCl)2, DMF, DCM, 3.5 h, rt; P(OBn)3, toluene, 1 h, 50 °C, 75%; (e) LiBr, ACN, 2 h, rt, 80%; (f) subtilisin Carlsberg, aq NH4HCO3, pH 8, 2 h, 60 °C; AmberliteÒ IR-120, Na+-
form, H2O, 1 h, rt, 92%;20 (g) BH3⁄DMS, THF, 16 h, rt, 99%; (h) NBS, P(Ph)3, THF, 4.5 h, rt, 77%, (i) P(OEt)3, 4 h, 120 °C, microwave, 100%; (j) KOH, 0.8 M CaCl2, isopropanol/H2O,
rt, 82%.

2842
S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
O
HN
H
NH
H
O
S
O
O
OH
O
P
O
P
O
2Na
HN
OH
HO
O
O
O
O
O
O
O
O
O
4Na
H
N
H
N
H
N
NH
NH2
OH
R
O
O
O
N
H
N
H
N
H
NH2
O
O
H
N
O
H
H
N
O
N
O
O
O
N
N
H
N
H
N
N
N
H
H
O
O
O
O
O
O
O
O
6, 8: R=Ac, 5,6-carboxyfluoresceinyl(CF)
10
OH
O
O
O
NH2
O
O
O
O
O
O
H
H
H
NH2
O
O
O
O
N
N
N
O
H
H
N
H
N
N
N
N
NH2
O
R
N
H
H
H
O
N
H
N
H
N
H
O
NH2
O
O
O
O
O
O
O
P
O
O
O
O
O
O
O
O
4Na
P
HO
3Na
HO
7, 9: R=Ac, 5,6-carboxyfluoresceinyl(CF)
11
Figure 2. Synthesized peptides with pcF and pmF as non-hydrolyzable Y⁄ mimetics
deprotected amino acid was reacted with formerly prepared biotin-
NHS furnishing the derivatized lysine. This building block was also
coupled with TBTU as a condensation agent. After these final cou-
pling steps the peptides were deprotected and cleaved from the
Rink amide resin with 100% TFA. Acidic cleavage cocktails usually
contain additives like thiols and silanes to reduce side reactions
by scavenging reactive species, however, the investigated ben-
zoylphosphonates showed a strong reactivity in TFA towards thio-
phenol by forming the supposed thioacetal. The addition of
alkylated silanes like triisopropylsilane lead to unidentified side
products. Since the utilization of water as a scavenging agent did
not enhance the product formation, we omitted any additives for
the cleavage reaction by using neat TFA. The ketophosphonates
were the main peptide products (purity >70%) according to analytic
HPLC–MS. The crude peptides 6–10 were purified by preparative
reversed phase HPLC using an NH4HCO3 buffer at pH 8.0. The
change towards the basic buffered HPLC conditions led to a slight
loss of separation performance, however, HPLC separations in acidic
media, were not applicable, since the isolated peptides rapidly
degraded during lyophilization. Stable isolated peptides 6–10
(Fig. 2) were obtained in purities higher than 95% and the products
were analyzed by high resolution mass spectrometry to confirm the
identity of the photoactive substrate mimetics.
inhibition of PTP1B was assayed by monitoring the hydrolysis of
the chromogenic substrate para-nitrophenyl phosphate (pNPP) in
a tris(hydroxymethyl) aminomethane (TRIS) buffered medium at
405 nm at pH 7.0. The used buffer contained common additives
for the stabilization of proteins including EDTA, glycerol, and the
detergent Brij 35. In order to avoid partial quenching of biradicals
no additional reducing agents like dithiothreitol (DTT) were added
to the assay buffer. Activities of PTP1B were normalized against
negative and positive controls and the IC50 values were determined
with non-linear regression analysis. Determined IC50 values were
subsequently converted into the inhibition constant KI by applying
the Cheng Prusoff equation.26 Prior to activation with light all pcF-
peptides (6–10) were moderately active inhibitors of PTP1B with KI
values in the low micromolar range. Irradiation of protein samples
with or without photoactive ligands was conducted in sealed ves-
sels in a cold room (4 °C) to prevent the mixture from overheating.
In order to determine the effects of irradiation on the protein, sam-
ples of PTP1B were irradiated without photoactive ligands present.
Depending on the applied light source the irradiated protein sam-
ples can show a moderate reduction of enzyme activity. This effect
is probably caused by warming of the samples during irradiation
and did not interfere with the light induced inhibition of the
enzyme and the IC50-values of inhibitors. Next inhibition of
PTP1B was investigated following to the photoactivation of the
PcF-containing phosphatase inhibitors 6–10. After irradiation for
all cases the inhibition constants decreased significantly
(Table 1). Photoactivation of the tested inhibitors was quantified
by calculating the photodeactivation factor KI/KI(UV) for each com-
pound. Significant photodeactivation was observed for all tested
compounds and ranged between 9 and 120. The largest deactiva-
Synthesis of the pmF-containing control peptide 11 was con-
ducted also on Rink amide resin using the standard DIC/HOBt pro-
tocol for activation of all building blocks including the
phosphotyrosine mimetic 15. After completion of the sequence
the phosphonic ester was saponified with TMSBr followed by the
complete cleavage of the peptide with trifluoroacetic acid.
Peptide 11 was purified under standard HPLC conditions with
TFA as an additive.
tion factor KI/KI(UV) of 120 was determined for peptide
7
(Table 1). Irradiation of peptides 6 and 8 in the presence of 1 mM
DTT showed almost no enhanced activity suggesting a radical
mechanism for the photodeactivation of PTP1B. Noticeable, the
photodeactivation factors of all fluorescein-labelled peptides 8, 9,
and 10 were significantly lower than those observed for the N-
acetylated peptides 6 and 7. This observation might be explicable
by the partial absorption of the irradiation light by the fluorescein
2.4. Inhibition of PTP1B with photoactive phosphopeptide
mimetics
Next, the obtained phosphotyrosine mimetic peptides were
tested as inhibitors of phosphotyrosine phosphatase PTP1B with-
out and with irradiation for 45 min at 365 nm (Fig. 3). The

S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
2843
100
50
0
100
100
50
0
irradiation
no irradiation
irradiation
no irradiation
irradiation
no irradiation
50
0
-9
-8
-7
-6
-5
-4
-3
-9
-8
-7
-6
-5
-4
-3
-8
-7
-6
-5
-4
-3
-2
log [6] (M)
log [7] (M)
log [11] (M)
100
50
0
100
50
0
100
50
0
irradiation
no irradiation
irradiation
no irradiation
irradiation
no irradiation
-9
-8
-7
-6
-5
-4
-3
-8
-7
-6
-5
-4
-3
-8
-7
-6
-5
-4
-3
-2
log [9] (M)
log [10] (M)
log [8] (M)
Figure 3. Representative normalized deactivation curves.
Table 1
Inhibition data
Cpd-#
Peptide
Yielda (%)
20%
KI, without UVb
KI, with UVb
Factor KI/KI(UV)
6
Ac-DADEpcFL-NH2
Ac-DADEpcFL-NH2
Ac-NpcFIDLD-NH2
CF-DADEpcFL-NH2
CF-DADEpcFL-NH2
CF-NpcFIDLD-NH2
CF-K(biotin)GpcFLSLPPW-NH2
CF-NpmFIDLD-NH2
PHPS1
30.0 ( 4.7)
17.3 ( 4.5)
48.8 ( 7.9)
19.1 ( 1.5)
10.6 ( 1.1)
10.8 ( 1.4)
21.4 ( 1.8)
58.1 ( 6.5)
lM
lM
lM
lM
lM
lM
lM
lM
0.39 ( 0.03)
l
M
76
1.4
120
12
0.7
9
11
11
2.5
6c
7
12.9( 1.8)
lM
30%
16%
0.41 ( 0.05)
lM
8
1.62 ( 0.2)
14.8 ( 1.4)
1.16 ( 0.09)
1.87 ( 0.06)
5.14 ( 0.23)
2.57 ( 0.69)
l
M
8c
9
lM
20%
12%
17%
l
l
l
l
M
10
11
19
M
M
M
6.2 ( 1.2) lM
a
Yields of isolated products.
b
c
Determined with Cheng–Prusoff equation.26
Inhibition data in the presence of 1 mM DTT.
fluorophore that could reduce the excitation energy available for
the formation of the biradical species from the alpha keto group.
Partial absorption of the UV light (365 nm) by the fluorescein pos-
sessing a maximal absorption at 490 nm, is indicated by the visible
green fluorescence of the reaction mixtures.
Peptide 11, containing the phosphonomethyl phenylalanine
(pmF) as a non-hydrolysable phosphotyrosine isoster, was synthe-
sized as a negative control for the peptides bearing the photoactive
pcF isoster. To our surprise, however, a significant photoactivating
effect was also observed for peptide 11, resulting in the pho-
todeactivation of PTP1B by a factor of 11. There have been occa-
sional reports on the photo-oxidative cleavage of (benzyl)
phosphonates though further studies will be necessary to explain
the observations reported here.27,28 For another control, the small
molecule inhibitor of PTP1B, PHPS1 (19), which was synthesized
using a published protocol,18,30 was tested with and without
irradiation displaying a deactivation factor of 2.5.
modification, for example, on an electrophoresis gel. To examine
if the protein tyrosine phosphate PTP1B could be modified cova-
lently, higher concentrations (ꢀ600 nM) of the protein were
exposed to the peptides 8 and 9. The samples were irradiated
under the same conditions as in the inhibition experiments and
were subsequently separated by sodium dodecyl sulfate polyacry-
lamide gel electrophoresis, SDS–PAGE. The polyacrylamide gels
were analyzed under UV-light and subsequently stained with the
protein dye coomassie brilliant blue to verify the presence of suffi-
cient amounts of PTP1B. In all experiments, no covalent modifica-
tion of the protein could be detected, even at concentrations of up
to 500 lM of the peptide 8 or 9. These findings led to the conclu-
sion that the reaction between the photoreactive peptides and
the phosphatase furnished no stable covalent photoreaction
product.
Additional experiments were conducted to scrutinize the
mechanism of the observed protein deactivation. It had already
been demonstrated that the inhibitory effects of the irradiated
ketophosphonates, pcF peptides 6 and 8 could be almost abolished
by addition of 1 mM dithiothreitol (DTT) (Table 1). Thus, suppres-
sion of photo-switching of PTP1B was investigated at various con-
centrations of DTT between 0 and 1 mM and at fixed peptide
2.5. Crosslinking experiments
Peptides containing ketophosphonates in form of the 4-phos-
phonocarbonyl phenyl alanine building block have been applied
recently for the selective and specific covalent modification the
(phosphotyrosine-binding) SH2-domain of the transcription factor
STAT5b.20 Therefore, we also investigated photocrosslinking of the
light-switched inhibitors with the target protein, phosphatase
PTP1B. Peptides 8, 9, and 10 carried suitable labels like a fluores-
cent dye or biotin to enable the sensitive detection of protein
concentrations of 6 and 8 (35 lM) (Fig. 4, left). 100 lM of DTT
was sufficient to block most of the photo-deactivation of PTP1B,
while lower concentrations of DTT inhibited the photo-switching
only partially. In order to elucidate the effect of DTT on the pho-
toreactions, a 10 mM solution of 4-methyl benzoylphosphonate 1
(Scheme 1, R = Me) was irradiated in isopropanol/water in the

2844
S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
100
80
60
40
20
0
0.8
no irradiation,
followed by DTT addition
peptide
peptide
6
8
0.6
0.4
0.2
irradiation,
followed by DTT addition
irradiation, not followed by
DTT addition
0
200
400
c [DTT] µM
600
800
1000
0
1000
2000
t in s
3000
Figure 4. Inhibition after irradiation depending on DTT concentrations, with fixed peptide concentration of 35
with/without irradiation and following addition of 1 mM DTT.
lM for peptide 6 and 8, PTP1B activity at ꢀ10 lM peptide 6
presence of an equilmolar amount of either DTT or the 2,2,6,
6-tetramethyl-piperidinyloxyl (TEMPO) radical. In both cases the
reported photodimerization was completely suppressed.
Therefrom one can conclude that DTT blocks the photo-
deactivation of PTP1B by interaction with the biradical
intermediates formed from the irradiated ketophosphonates.
Finally, we investigated whether the photo-deactivation of
PTP1B could be reverted by the addition of a reducing agent such
as DTT after irradiation and completed photo-deactivation of the
enzyme. PTP1B was irradiated in the presence of peptide 7
insight into the structure of the finally formed oxidized enzyme,
although, structures of oxidized PTP1B have been reported.31
3. Conclusions
Benzoyl phosphonates are enzymatically stable phos-
photyrosine mimetics, which can be activated by irradiation with
UV light to furnish reactive triplet biradicals. Here these phos-
photyrosine isosters have been investigated as photo-induced
deactivation agents of the protein tyrosine phosphatase PTP1B. A
benzoyl phosphonate containing amino acid, 4-phosphonocar-
bonyl phenylalanine 5, was incorporated as a suitably protected
building block into peptides known as substrates of PTP1B replac-
ing the native phosphotyrosine residue. The obtained phosphopep-
tide mimetic peptides were investigated as inhibitors of PTP1B
with and without irradiation at 365 nm. While the non-irradiated
pcF-containing peptides were moderate inhibitors with KI-values
in the micromolar range, irradiation induced the strong enhance-
ment of the inhibitory effects. The strongest photoactivation was
(10 lM) leading to complete deactivation of the enzyme as indi-
cated in the subsequent pNPP assay (Fig. 4, right). Addition of
DTT (1 mM) after the irradiation led to the time-dependent reacti-
vation of the enzyme reaching the activity of the non-irradiated
enzyme after 10–20 min as indicated by the rate of pNPP
hydrolysis.
From these observations it can be concluded that the phos-
phatase PTP1B is deactivated by irradiated ketophosphonates in a
reversible radical mechanism leading to the oxidative deactivation
of the enzyme without formation of a covalent cross-linking pro-
duct (Fig. 5). The mechanism is clearly distinct from that observed
for the photo-crosslinking of STAT5–SH2 domain earlier. We sug-
gest that the presence of the reported redox-reactivity of PTP1B
is responsible for the reversible deactivation of the enzyme.
Irradiation of the photoactive peptides 6–10 at 365 nm forms the
triplet intermediate, which can be quenched by reaction with
DTT. When bound to the active site of PTP1B this triplet inter-
mediate will abstract a hydrogen radical from the thiol of Cys215
thereby oxidizing the thiol residue to a thiol-radical. This reaction
can be expected to proceed, as a thiol radical is significantly more
stable than radicals formed by H-abstraction from carbon, oxygen
or nitrogen atoms being present in the active site of the enzyme.
The thio-radical intermediate (see Fig. 5) results subsequently in
the formation of the observed inactive PTP1B without the covalent
attachment of photoactive peptide. We do not have a further
observed for the phosphopeptide mimetic Ac-NpcFIDLD-NH2
7
with an activation factor KI/KI(UV) of 120 resulting in a sub-
micromolar inhibitor; the second strongest photoactivation was
observed for Ac-DADEpcFL-NH2 6 displaying a deactivation factor
of 76. The deactivation of PTP1B was found to proceed via an
oxidative radical mechanism and could be reverted by addition
of DTT as reducing agent. The deactivation of PTP1B was con-
siderably stronger for peptides with N-terminal acetylation than
with N-terminal carboxy fluoresceinyl residue. Interestingly, the
photodeactivation of the enzyme PTP1B was detected not only
for the phosphatase inhibitors containing the photoactive pcF
building block, but also for the phosphopeptide mimetic contain-
ing the pmF residue. Though this finding is surprising at first
glance, it is in accordance with earlier reports on the oxidative pho-
tolysis of phosphonates. No covalent photocrosslinking was
observed for the fluorescein-labeled pcF peptides
8 and 9
O
HO
O
P
O
O
P
O
O
P
O
H
N
HN
O
O
O
H2N
O
O
HO
H
H
365 nm
inactive, oxidized PTP1B
without covalent crosslink;
activated by DTT
S
S
S
NH
NH
NH
H
N
H
N
H
N
O
O
O
R
R
R
HN
HN
HN
H
H
H
N
N
N
R
R
R
O
O
O
O
O
O
O
O
O
R
R
R
triplet activation
expected thio-radical intermediate
Figure 5. Proposed mechanism of for the photo-deactivation of the phosphatase PTP1B: Irradiation of the 4-phosphonocarbonyl-phenylalanine (pcF) containing peptide
ligands 6–10 at 365 nm generates a triplet biradical intermediate, which oxidizes the enzyme at Cys215, resulting in the formation of inactivated, oxidized PTP1B.

S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
2845
presumably due to a photo-induced, reversible oxidation reaction
with the active site cysteine residue. The observed strong de-ac-
tivation together with the absence of photocrosslinking products
in the gel electrophoresis (SDS–PAGE) indicates clearly distinct
effects of pcF-containing peptides on protein tyrosine phos-
phatases compared to the earlier studied effects on non-enzymatic
phosphotyrosine-binding SH2-domains. The results suggest that
the proteome-wide covalent photocrosslinking of pcF-containing
peptides in cell lysates might be relatively specific for the covalent
modification of phosphotyrosine binding peptides and might not
result in the pull-down of phosphotyrosine phosphatases. In addi-
tion, acetylated pcF-peptides should be further investigated in the
photodeactivation of PTPs in more complex systems, possibly
including cell lysates or even living cells and might possibly be
useful tools in the time- and spatially resolved switching of PTP
activities.
(d, J = 8.2 Hz, 1H, NH), 4.66 (m, 4H, CH2, CH), 4.40 (m, 2H, CH2-
Fmoc), 4.20 (t, J = 7.0 Hz, 1H, CH-Fmoc), 3.74 (3H, Me), 3.10 (m,
2H, CH2-Phe); 13C NMR (75 MHz, CDCl3): d 172.0 (C@O, Phe),
155.7, (C@O, Fmoc), 144.0, 143.9, 141.5, 139.9, 135.3, 129.6,
127.9, 127.4, 125.2, 120.1, 67.0 (CH2-Fmoc), 65.2 (CH2), 54.9
(CH), 52.5 (CH3), 47.3 (CH-Fmoc), 38.1 (CH2-Phe); MS (ESI, [m/z]):
calculated: [M+Na+]+: 454.16, [Mꢂ18+H+]+: 414.17; found
[M+H+]+: 494.0, [M-18+H+]+: 414.0.
4.1.2. N-Fluorenyl-9-methoxycarbonyl-4-bromomethyl-pheny-
lalanine methyl ester 17
N-Fluorenyl-9-methoxycarbonyl-4-hydroxymethyl-phenylala-
nine methyl ester 16 (240 mg, 0.557 mmol) and triphenylphos-
phine (425 mg, 1.62 mmol, 2.9 equiv) were dissolved in dry THF
(10 ml). N-Bromosuccinimide (287 mg, 1.62 mmol, 2.9 equiv)
was dissolved in dry THF and was added dropwise to the solution
through a syringe. The solution was stirred at room temperature
for 4.5 h. Silica 60 was added and the solvent was removed under
reduced pressure. The crude product was purified by column
chromatography (SiO2, hexane +1% AcOH/ethyl acetate +1%
AcOH, 100–40% hexane) to afford 17 as a pale yellow solid
(213 mg, 77%).
4. Experimental section
4.1. Materials and methods
1H NMR (300 MHz, CDCl3): d 7.77 (d, J = 7.6 Hz, 2H, H-Fmoc),
7.57 (m, 2H, H-Fmoc), 7.41 (t, J = 7.3 Hz, 2H, H-Fmoc), 7.32 (m,
4H, H-Fmoc, Hortho-Phe), 7.07 (d, J = 7.8 Hz, 2H, Hmeta-Phe), 5.25
(d, J = 8.2 Hz, 1H, NH), 4.67 (m, 1H, CH), 4.45 (s, 2H, CH2), 4.41
(m, 2H, CH2-Fmoc), 4.20 (t, J = 7.0 Hz, 1H, CH-Fmoc), 3.70 (3H,
Me), 3.12 (m, 2H, CH2-Phe); 13C NMR (75 MHz, CDCl3): d 171.9
(C@O, Phe), 155.6, (C@O, Fmoc), 143.9, 141.5, 136.8, 129.9, 129.4,
127.9, 127.2, 125.2, 120.2, 120.1, 67.1 (CH2-Fmoc), 54.8 (CH),
52.6 (CH3), 47.3 (CH-Fmoc), 38.1 (CH2-Phe), 33.3 (CH2-Br); MS
(ESI, [m/z]): calculated: [M+H+]+: 494.10; found [M+H+]+: 494.0.
All reactions under dry conditions were conducted under nitro-
gen using Schlenk methodology. For these purposes commercially
available anhydrous solvents stored over molecular sieves were
used. Chemicals were purchased from common suppliers and were
used without any purification. Non high resolution mass spec-
trometry was conducted with an LC of the Agilent 1100 series
equipped with a diode array detector, coupled with a single quad-
rupole mass spectrometer with an ESI source from Agilent
Technologies. For preparative separations, a preparative HPLC of
the 1100 series from Agilent Technologies was used equipped with
a multi-wavelength detector. As eluents, mixtures of water and
acetonitrile were used with addition of 0.1 % trifluoroacetic acid.
High resolution mass spectrometry were conducted using an
Agilent 6210 ESI-TOF spectrometer coupled with an HPLC of the
1100 series from Agilent Technologies. NMR spectra were recorded
on a Bruker AV300 or Varian Mercury 300 MHz or Varian Mercury
400 MHz. 1H and 13C NMR spectra were referenced to the solvent
residual peak or tetramethylsilane (TMS). For referencing 31P
NMR spectra, external calibration with phosphoric acid
(d = 0 ppm) was used. 13C NMR and 31P NMR spectra were recorded
with 1H decoupling unless noted. Broad signals were marked either
as ‘br’ (broad) or ‘vb’ (very broad). Irradiation experiments were
conducted on a UVP transilluminator (4 ꢁ 25 W) at the wavelength
of 365 nm or a Biostep UXDT-20ML-15 K.
4.1.3. N-Fluorenyl-9-methoxycarbonyl-4-(diethoxyphosphoryl)-
methyl-phenylalanine methyl ester 18
N-Fluorenyl-9-methoxycarbonyl-4-bromomethyl-phenylalanine
methyl ester 17 (93 mg, 0.118 mmol) and triethylphosphite (64 ll,
0.377 mmol, 2 equiv) were mixed in a microwave vessel and irradi-
ated at 120 °C for 4 h. The crude product was purified by column
chromatography (SiO2, DCM/MeOH, 100–90% DCM) to afford 18
as a colorless oil containing minor impurities of triethylphosphate
and diethylphosphite (113 mg, 100%).
1H NMR (300 MHz, CDCl3): d 7.77 (d, J= 7.6 Hz, 2H, H-Fmoc),
7.57 (dd, J = 7.7 Hz, 3.3 Hz, 2H, H-Fmoc), 7.40 (t, J = 7.6 Hz, 2H, H-
Fmoc), 7.31 (m, 2H, H-Fmoc,), 7.22 (dd, J = 7.8 Hz, 2.4 Hz, 2H,
Hortho-Phe), 7.04 (d, J = 7.7 Hz, 2H, Hmeta-Phe), 5.27 (d, J = 8.2 Hz,
1H, NH), 4.65 (q, J = 6.0 Hz, 1H, CH), 4.39 (m, 2H, CH2-Fmoc), 4.0
(m, 4H, CH2-Et), 3.72 (3H, Me), 3.11 (d and m, J = 15 Hz, 2H,
–CH2–P, and CH2-Phe), 1.22 (t, J = 7.1 Hz, 6H, CH3–CH2–O–P); 13C
NMR (75 MHz, CDCl3): d 172.0 (C@O, Phe), 155.7, (C@O, Fmoc),
143.9, 141.4, 130.2 (d, J = 6.8 Hz), 129.6(J = 3.4 Hz), 127.9, 127.2,
125.2 (J = 4.7 Hz), 120.1 (J = 1.7 Hz), 120.1, 67.1 (CH2-Fmoc), 62.4
(J = 8.8 Hz, CH3–CH2–O–P), 54.9 (CH), 52.5 (CH3), 47.3 (CH-Fmoc),
38.1 (CH2-Phe), 33.5 (J = 138.5 Hz, Ph–CH2–P–O), 16.5 (J = 6.4 Hz,
CH3–CH2–O–P); 31P NMR (162 MHz, CDCl3): d 27.5 (br m, 1P, P);
MS (ESI, [m/z]): calculated: [M+H+]+: 552.21; found [M+H+]+: 551.8.
4.1.1. N-Fluorenyl-9-methoxycarbonyl-4-hydroxymethyl-pheny-
lalanine methyl ester 16
N-Fluorenyl-9-methoxycarbonyl-4-carboxy-phenylalanine
methyl ester 14 (253 mg, 0.569 mmol) was dissolved in dry THF
(3 ml) at 0 °C. A solution of BH3⁄DMS in THF (600
ll, 0.6 mmol,
1.1 equiv) was added dropwise. The solution was stirred at room
temperature for 16 h and cooled down to 0 °C. A solution of
NaHCO3 was added until no more hydrogen was formed. The
organic solvent was removed under reduced pressure. Ethyl acet-
ate was added and the aqueous phase extracted twice with ethyl
acetate. The organic phase was washed with 1 M HCl and brine.
The combined organic phases were combined and dried over
Na2SO4. The solvents were removed under reduced pressure and
the residue was purified by column chromatography (SiO2,
hexane/ethyl acetate, 100–20% hexane) to afford 16 as a clear oil
(240 mg, 99%).
4.1.4. N-Fluorenyl-9-methoxycarbonyl-4-(diethoxyphosphoryl)-
methyl-phenylalanine 15
N-Fluorenyl-9-methoxycarbonyl-4-(diethoxyphosphoryl)methyl-
phenylalanine methyl ester 18 (104 mg, 0.118 mmol) was
dissolved in 5 ml of a 0.8 M CaCl2 solution (isopropanol/water
70:30). A 1 M solution of KOH in ethanol was added and the con-
version was monitored via HPLC. After the addition of a final vol-
1H NMR (300 MHz, CDCl3): d 7.77 (d, J = 7.6 Hz, 2H, H-Fmoc),
7.57 (m, 2H, H-Fmoc), 7.41 (m, 2H, H-Fmoc), 7.31 (m, 3H,
ume of 440
ll (0.44 mmol, 3.7 equiv) the ester was completely
H-Fmoc,
Hortho-Phe), 7.09 (d, J= 7.9 Hz, 2H, Hmeta-Phe), 5.26
saponified. Ethyl acetate and sat. NaHCO3 were added and the

2846
S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
layers separated. The aqueous phase was washed with ethyl acet-
ate and acidified with 1 M HCl. The mixture was extracted with
ethyl acetate and the organic phases were combined. Drying over
Na2SO4 and evaporation of the organic solvent furnished 15 as a
colorless solid (83 mg, 82%).
4.2.5. Fmoc-cleavage
The Fmoc-group was cleaved by using a piperidine/DMF (20:80
v/v) solution. The resin was suspended in the basic cocktail and
shaken either for 45 min and washed, or shaken once for 1 min,
twice for 10 min, with washing after each step.
1H NMR (300 MHz, CDCl3): d 7.77 (d, J = 7.5 Hz, 2H, H-Fmoc),
7.59 (t, J = 6.8 Hz, 2H, H-Fmoc), 7.40 (t, J = 7.5 Hz, 2H, H-Fmoc),
7.31 (m, 2H, H-Fmoc), 7.22 (dd, J = 8.0 Hz, 2.6 Hz, 2H, Hortho-Phe),
7.14 (d, J = 7.7 Hz, 2H, Hmeta-Phe), 5.40 (d, J = 7.6 Hz, 1H, NH),
4.71 (m, 1H, CH), 4.42 (m, CH2-Fmoc), 4.23 (t, J = 7.0 Hz, 1H,
CH-Fmoc), 4.00 (m, 4H, CH3–CH2–O–P), 3.11 (m, 2H, –CH2–P, and
CH2-Phe), 1.22 (2 t, J = 7.1 Hz, 6H, CH3–CH2–O–P); 13C NMR
(75 MHz, CDCl3): d 173.0 (C@O, Phe), 155.7, (C@O, Fmoc), 144.0,
144.0, 141.5, 135.0, 130.1 (d, J = 6.5 Hz), 129.7 (d, J = 9.2 Hz),
127.9, 127.2, 125.3 (J = 8.1 Hz), 120.1 (J = 1.9 Hz), 67.0
(CH2-Fmoc), 63 (2d, J = 6.9 Hz, CH3–CH2–O–P), 54.5 (CH), 47.4
(CH-Fmoc), 37.3 (CH2-Phe), 33.1 (J = 137.1 Hz, Ph–CH2–P–O), 16.4
(2d, J = 6.9 Hz, CH3–CH2–O–P); 31P NMR (162 MHz, CDCl3): d 28.8
(m, 1P, P); MS (ESI, [m/z]): calculated: [M+H+]+: 538.20,
[MꢂH+]ꢂ: 536.18; found [M+H+]+: 538.3, [MꢂHꢂ]ꢂ: 536.0.
Synthesized peptides including pcF that underwent Fmoc-
cleavage with piperidine/DMF showed significant substitution of
the phosphonate group by piperidine. Side product formation with
piperidine could be avoided by using DBU/DMF. Therefore, all
Fmoc-cleavages following pcF were conducted with DBU/DMF
(2:98 v/v), while the same reaction time as for piperidine/DMF
was used.
4.2.6. Acetylation of the peptides
For N-acetylation
a solution of acetic anhydride in DCM
(50:50 v/v) was used with a reaction time of 1 h.
4.2.7. Coupling of 5,6-carboxyfluorescein
5,6-Carboxyfluorescein (10 equiv) was preactivated with DIC/
HOBt (10 equiv/10 equiv) for 2 min and pipetted to the resin.
After 1 h, the resin was washed and DBU/DMF (2:98 v/v) was
added to cleave formed CF-oligomers. After 15 min the resin
was washed and again mixed with DBU/DMF. This procedure
was repeated until the solution stayed clear.
4.1.5. N-Fluorenyl-9-methoxycarbonyl-N6-(biotin)-lysine
N-Fluorenyl-9-methoxycarbonyl-N6-(biotin)-lysine was pre-
pared according to literature.29
4.2. Protocols for peptide synthesis
4.2.8. Acidic peptide cleavage
The phosphonocarbonyl-phenylalanine peptides 6, 7, 8, 9, 10
were cleaved with 100% TFA and shaking for 2 h.
The phosphonomethylphenylalanine-containing peptide, was
Peptide synthesis was conducted on Rink amide resin using
Fmoc-strategy. PP/PE syringes equipped with a PE-frit were used
as reaction vessels. Fmoc-amino acids and reagents were pur-
chased from common suppliers. The coupling reactions were moni-
tored using the Kaiser test and quantified via UV-photometric
determination of the dibenzofulvene product following to Fmoc
cleavage.
deprotected with a 1:1 mixture of TFA and TMSBr (each 50 ll) in
a microwave vial for 1 h at 60 °C and cleaved subsequently with
100% TFA. The cleaved peptides were filtered and the remaining
resin washed with TFA. The TFA was removed with a stream of
dry nitrogen and the residue dried in high vacuum.
4.2.1. Coupling of Fmoc-amino acids
4.2.9. HPLC purification of peptides 6, 7, 8, 9, 10, 11
In general, amino acids were coupled with N,N0-diisopropylcar-
bodiimide and N-hydroxybenzotriazole in a minimal volume of
DMF. Three equivalents (with respect to the loading of the resin)
of Fmoc-amino acid were preactivated with DIC/HOBt for 5 min
and added to the resin. Fmoc-Pro-OH was coupled with HATU
(5 equiv) and 2,4,6-collidine (10 equiv).
The phosphomethylphenylalanine peptide 11 was purified with
the standard HPLC buffer (0.1% TFA in water/acetonitrile; 90% to
30% in 20 min). The crude peptides 6, 7, 8, 9, 10 were dissolved
in NH4HCO3 (50 mM, pH 8) and purified using a basic buffered
HPLC system (50 mM NH4HCO3/acetonitrile; 95–30% in 20 min).
The product fractions were directly lyophilized, dissolved in water
and stirred over Amberlite resin (IR120 sodium form) for 20 min.
The resin was filtered off and the filtrate was lyophilized to yield
the sodium salts of the corresponding peptides.
4.2.2. Coupling of Fmoc-phosphonocarbonyl building block 12
(pcF)
The unnatural amino acid 12 was coupled with TBTU and
DIPEA. First, 4 equivalents (in respect to loading of the resin) of
12 were dissolved in a minimal amount of DMF. Then, DIPEA
(2 equiv) was added and the solution mixed for 2 min. TBTU
(4 equiv) was added, the solution mixed for 3 min and pipetted
to the resin. After 4 h the resin was washed and the conversion
analyzed with the Kaiser test.
4.3. Synthesized peptides
4.3.1. Ac-DADEpcFL-NH2
6
50 mg preloaded resin (0.72 mmol/g) were used for cleavage.
6: 6.2 mg = 20%; HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
856.2772; found [MꢂH+]ꢂ: 856.2911.
4.2.3. Coupling of the amino acid following the introduction of
pcF
4.3.2. Ac-NpcFIDLD-NH2 7
30 mg preloaded resin (0.54 mmol/g) were used for cleavage.
7: 4.3 mg = 30%; HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
883.3244; found [MꢂH+]ꢂ: 883.3357.
The first amino acid after incorporation of the pcF building
block was coupled with full conversion using the combination of
Fmoc-AA-OH/TBTU/DIPEA (4 equiv/4 equiv/4 equiv) for activation.
Subsequent coupling steps were again conducted with DIC/HOBt
activation.
4.3.3. CF-DADEpcFL-NH2
8
15 mg preloaded resin (0.72 mmol/g) were used for cleavage.
8: 2 mg = 16%; HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
1172.3143; found [MꢂH+]ꢂ: 1172.3238.
4.2.4. Washing of the resin
The resin was washed after every coupling and cleavage proce-
dure. First, the resin was washed with 5 syringe volumes of DMF,
then with 5 volumes of THF, DCM and Et2O.
4.3.4. CF-NpcFIDLD-NH2
9
25 mg preloaded resin (0.54 mmol/g) were used for cleavage.

S. Wagner et al. / Bioorg. Med. Chem. 23 (2015) 2839–2847
2847
9: 3.3 mg = 20%; HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
1199.3616; found [MꢂH+]ꢂ: 1199.3753.
This work was supported by the DFG (SFB 765). The Protein
Sample Production Facility at the Max Delbrück Center is funded
by the Helmholtz Association of German Research Centres.
4.3.5. CF-K(biotin)GpcFLSLPPW-NH2 10
30 mg preloaded resin (0.48 mmol/g) were used for cleavage.
10: 3.1 mg (12%); HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
1733.6757; found [MꢂH+]ꢂ: 1733.6907.
References and notes
1. Hunter, T. Cell 2000, 100, 113.
2. Soulsby, M.; Bennett, A. M. Physiol. (Bethesda) 2009, 24, 281.
3. Sun, H.; Tonks, N. K. Trends Biochem. Sci. 1994, 19, 480.
4. Burke, T. R. Jr.; Lee, K Acc. Chem. Res. 2003, 36, 426.
5. Barford, D.; Das, A. K.; Egloff, M.-P. Annu. Rev. Biophys. Biomol 1998, 27, 133.
6. Alonso, A.; Sasin, J.; Bottini, N.; Friedberg, I.; Friedberg, I.; Osterman, A.; Godzik,
A.; Hunter, T.; Dixon, J.; Mustelin, T. Cell 2004, 117, 699.
4.3.6. CF-NpmFIDLD-NH2 11
25 mg preloaded resin (0.54 mmol/g) were used for cleavage.
11: 2.7 mg = 17%; HRMS (ESI, [m/z]): calculated: [MꢂH+]ꢂ:
1185.3823; found [MꢂH+]ꢂ: 1185.3929.
7. Tonks, N. K. Nat. Rev. Mol. Cell Biol. 2006, 7, 833.
8. Lessard, L.; Stuible, M.; Tremblay, M. L. Biochim. Biophys. Acta 2010, 1804, 613.
9. De Munter, S.; Köhn, M.; Bollen, M. ACS Chem. Biol. 2013, 8, 36.
10. Barr, A. J. Future Med. Chem. 2010, 2, 1563.
4.4. Biochemical procedures
11. Zhang, S.; Zhang, Z.-Y. Drug Discovery Today 2007, 12, 373.
12. Pedersen, A. K.; Guo, X.-L.; Møller, K. B.; Peters, G. H.; Andersen, H. S.; Kastrup,
J. S.; Mortensen, S. B.; Iversen, L. F.; Zhang, Z.-Y.; Møller, N. P. H. Biochem. J.
2004, 378, 421.
Activity and affinity measurements were conducted using a
Safire2 wellplate reader and 384-well microtiter plates. For the
analysis of the data the non linear regressing ‘log (agonist) versus
normalized response—Variable slope’ from GraphPad Prism 5 was
used. Determinated IC50 values were converted into the
corresponding KI values applying the Cheng Prusoff equation
KI ¼ IC50=ð1 þ ½Sꢃ=KMÞ, with KI: inhibition constant, [S]: used
substrate concentration, KM: Michaelis constant. The KM values of
irradiated and non-irradiated PTP1B were determined as 2.5 mM
and 2.3 mM. For activity measurements a substrate concentration
of 2.36 mM was used.
13. Zhang, Z.-Y. Crit. Rev. Biochem. Mol. Biol. 1998, 33, 1.
14. 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.
Bioorg. Med. Chem. Lett. 2004, 14, 1039.
15. Patel, D.; Jain, M.; Shah, S. R.; Bahekar, R.; Jadav, P.; Joharapurkar, A.; Dhanesha,
N.; Shaikh, M.; Sairam, K. V. V. M.; Kapadnis, P. Bioorg. Med. Chem. Lett. 2012,
22, 1111.
16. Zhang, X.; He, Y.; Liu, S.; Yu, Z.; Jiang, Z. X.; Yang, Z.; Dong, Y.; Nabinger, S. C.;
Wu, L.; Gunawan, A. M.; Wang, L.; Chan, R. J.; Zhang, Z.-Y. J. Med. Chem. 2010,
53, 2482.
17. Combs, A. P.; Zhu, W.; Crawley, M. L.; Glass, B.; Polam, P.; Sparks, R. B.; Modi,
D.; Takvorian, A.; McLaughlin, E.; Yue, E. W.; Wasserman, Z.; Bower, M.; Wei,
M.; Rupar, M.; Ala, P. J.; Reid, B. M.; Ellis, D.; Gonneville, L.; Emm, T.; Taylor, N.;
Yeleswaram, S.; Li, Y.; Wynn, R.; Burn, T. C.; Hollis, G.; Liu, P. C. C.; Metcalf, B. J.
Med. Chem. 2006, 49, 3774.
Irradiation experiments were performed in a sealed black 384
well plate with UV transparent bottom on a transilluminator at
4 °C.
18. Hellmuth, K.; Grosskopf, S.; Lum, C. T.; Würtele, M.; Röder, N.; von Kries, J. P.;
Rosario, M.; Rademann, J.; Birchmeier, W. Proc. Natl. Acad. Sci. U.S.A. 2008, 105,
7275.
4.5. Activity measurements of PTP1B using a pNPP assay
19. Schmidt, M. F.; Groves, M. R.; Rademann, J. ChemBioChem 2011, 12, 2640.
20. Horatscheck, A.; Wagner, S.; Ortwein, J.; Kim, B. G.; Lisurek, M.; Beligny, S.;
Schütz, A.; Rademann, J. Angew. Chem., Int. Ed. 2012, 51, 9441.
21. Yip, S.-C.; Saha, S.; Chernoff, J. Trends Biochem. Sci. 2010, 35, 442.
22. Zhang, Z.-Y.; Maclean, D.; McNamara, D. J.; Sawyer, T. K.; Dixon, J. E.
Biochemistry 1994, 33, 2285.
The stock buffer used for the determination of inhibition con-
stants was composed as follows: TRIS⁄HCl (pH 7.0): 50 mM,
NaCl: 100 mM, glycerol: 20%, EDTA: 4 mM, Brij 35: 0.02%. The pH
was adjusted by the addition of 1 M NaOH. The buffer was pre-
pared and frozen at ꢂ20 °C and was diluted with an equal volume
of deionized water prior to use. Stock solutions (2 mM) of the
tested substrate mimetics were prepared in H2O/DMSO (50:50
v/v) and were diluted serially on the wellplate (Corning 3544 or
23. Reaction conditions modified from Grimm, J. B.; Wilson, K. J.; Witter, D. J.
Tetrahedron Lett. 2007, 48, 4509.
24. Sekine, M. S.; Satoh, M.; Yamagata, H.; Hata, T. J. Org. Chem. 1980, 45, 4162.
25. Fischer, R.; Mader, O.; Jung, G.; Brock, R. Bioconjugate Chem. 2003, 14, 653.
26. Cheng, C. Y.-C.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
27. Okamoto, Y.; Iwamoto, N.; Takamuku, J. J. Chem. Soc., Chem. Commun. 1986,
1516.
28. Krosley, K. W.; Collard, D. M.; Adamson, J.; Fox, M. A. J. Photochem. Photobiol. A:
Chem. 1993, 69, 357.
29. Sun, H.; Panicker, R. C.; Shao, Q. Y. Biopolymers (Pept. Sci.) 2007, 88, 141.
30. Grosskopf, S.; Eckert, C.; Arkona, C.; Radetzki, S.; Böhm, K.; Heinemann, U.;
Wolber, G.; von Kries, J.-P.; Birchmeier, W.; Rademann, J. ChemMedChem. 2015,
10, in press. http://dx.doi.org/10.1002/cmdc.201500015.
3711) with buffer solution, and 4
in an assay volume of 18 l. The mixture was incubated for 30 min.
After the irradiation with 365 nm for 45 min in a 4 °C cooling room,
l of pNPP (13 mM in H2O) were added and the mixture was sha-
ll of PTP1B (10 lg/ml) resulting
l
4
l
ken for 3 min. The final concentration of PTP1B was ꢀ50 nM.
Acknowledgements
31. Salmeen, A.; Andersen, J. N.; Myers, M. P.; Meng, T.-C.; Hinks, J. A.; Tonks, N. K.;
Barford, D. Nature 2003, 423, 769.
We thank Regina Reppich-Sacher from the Universität Leipzig
for providing the MALDI MS data.