A photoswitchable ITAM peptidomimetic: Synthesis and real time surface plasmon resonance (SPR) analysis of the effects of cis-trans isomerization on binding

Article abstract of DOI:10.1016/j.bmc.2007.10.049

The Syk protein plays an important role in immune receptor signaling. The Syk tandem SH2 domain (tSH2)-ITAM interaction is important for recruiting Syk to the receptor complex and for Syk kinase activation. A peptidomimetic ligand for tSH2 was synthesized in which a photoswitchable azobenzene moiety was incorporated. Such a photoswitchable moiety may regulate the distance between the two phosphotyrosine containing ITAM sequences, which bind to tSH2. Different affinities of the cis and trans isomer of the ligand were found by surface plasmon resonance (SPR). By in situ irradiation during SPR measurements the effect of the cis-trans isomerization on binding could be monitored in real time.

Full text of DOI:10.1016/j.bmc.2007.10.049

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 1393–1399
A photoswitchable ITAM peptidomimetic: Synthesis and real
time surface plasmon resonance (SPR) analysis of the effects
of cis–trans isomerization on binding
Joeri Kuil, Loek T. M. van Wandelen, Nico J. de Mol* and Rob M. J. Liskamp
Department of Medicinal Chemistry and Chemical Biology, Utrecht Institute of Pharmaceutical Sciences, Utrecht University,
PO Box 80.082, 3508 TB Utrecht, The Netherlands
Received 9 July 2007; revised 10 October 2007; accepted 17 October 2007
Available online 22 October 2007
Abstract—The Syk protein plays an important role in immune receptor signaling. The Syk tandem SH2 domain (tSH2)–ITAM
interaction is important for recruiting Syk to the receptor complex and for Syk kinase activation. A peptidomimetic ligand for
tSH2 was synthesized in which a photoswitchable azobenzene moiety was incorporated. Such a photoswitchable moiety may reg-
ulate the distance between the two phosphotyrosine containing ITAM sequences, which bind to tSH2. Different affinities of the cis
and trans isomer of the ligand were found by surface plasmon resonance (SPR). By in situ irradiation during SPR measurements the
effect of the cis–trans isomerization on binding could be monitored in real time.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
The Spleen tyrosine kinase (Syk) protein takes part in a
number of receptor signaling cascades.1–6 It is involved
in early events after activation of, for example, the
IgE-, T-cell-, and B-cell receptors. Furthermore, it is in-
volved in IL-15-receptor and integrin signaling. The role
of Syk in the high affinity IgE receptor (FceRI) signaling
in mast cells and basophils is shown in Figure 1. FceRI
consists of an a-, b- and two c-chains. The b- and c-
chains contain a specific intracellular sequence called
the Immunoreceptor Tyrosine based Activation Motif
(ITAM). The ITAM sequence consists of Tyr-Xxx-
Xxx-(Leu/Ile)-(Xxx)n=6–8-Tyr-Xxx- Xxx-(Leu/Ile), in
which Xxx can be any amino acid. The residues in italics
comprise the binding epitopes for the tandem SH2
(tSH2) domain of Syk, when tyrosine is phosphorylated.
We have found that the intervening residues (Xxx)n=6–8
(in FceRI n = 7) were not essential for binding.7,8
Figure 1. Recruitment of Syk to the diphosphorylated c-ITAM of
FceRI results in activation of its kinase domain and ultimately
degranulation.
Upon IgE stimulation of the receptor, the ITAM motif
is diphosphorylated (c-dpITAM). After this, Syk is
recruited to the membrane by binding to c-dpITAM
Keywords: FceRI; Syk; ITAM; SH2; Azobenzene; Conformational
photoswitch; Surface plasmon resonance.
(Fig. 1). This results in a conformational change and
Syk activation.5 Further events involved in activation
of the kinase domain are still subject of investigation.
*
Corresponding author. Tel.: +31 30 2536989; fax: +31 30 2536655;
e-mail: n.j.demol@uu.nl
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.049

1394
J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
They include phosphorylation of tyrosine residues in the
linker region between the tSH2 and kinase domains of
Syk, and in the kinase domain itself.5 Syk activation
eventually leads to cell degranulation and release of
mediators. Overstimulation of this pathway leads to
allergic responses and therefore Syk is an interesting tar-
get for potential anti-allergic therapy.
The relative orientation of the SH2 domains in tSH2 dis-
plays remarkable conformational flexibility. Recently,
we have found that binding of Syk tSH2 to an ITAM
peptide causes a significant change in the dynamics of
tSH2.1,9 This change could imply that a fixation in the
inter SH2 distance is necessary for Syk activation by
ITAM binding. To gain more insight into the function-
ing of Syk tSH2, we decided to take advantage of a
photoswitchable building block and incorporate this
into the ITAM sequence.
Azobenzene as part of (4-aminomethyl)phenylazoben-
zoic acid (AMPB) is the most widely used photoswitch
because of its fast and fully reversible photoisomeriza-
tion.10–12 Moreover, the synthesis of AMPB is well de-
scribed and the cis isomer is relatively stable in
solution, when not exposed to light. AMPB is compati-
ble with solid phase peptide synthesis and can easily be
incorporated into a peptide sequence.
Scheme 1. Synthesis of the photoswitchable building block 9 (Fmoc-
AMPB-OH).
Next, hydroxylamine 6 was treated with FeCl3 to pro-
duce the corresponding nitroso compound 7. Due to
the instability of this compound, it was immediately
used in the condensation reaction with Fmoc-protected
amine 3. The allyl group was easily removed afterwards
with Pd(PPh3)4 and phenylsilane as scavenger, which
gave no undesired reduction of the azo group.
Binding of the cis and trans isomers of the photoswitch
containing ITAM mimic was determined using surface
plasmon resonance (SPR). In a convenient approach,
the cis–trans transition and its effect on binding could
be monitored in real time using in situ irradiation in
the SPR measuring cell.
NMR spectra of 100% trans azobenzene moiety 9 and
mixtures with cis and trans conformers revealed that a
maximum cis content of 78% could be obtained for 9
upon irradiation with 366 nm light.
2. Results and discussion
The photoswitchable peptidomimetics 1a–b (Fig. 2) were
synthesized using standard Fmoc/tBu chemistry includ-
ing 9 as a building block. A small amount of TIS could
be used in the cleavage cocktail without reducing the
azobenzene. This was also recently observed by Hilvert
et al.19
The length of the seven intervening amino acids in c-
ITAM, which are not essential for binding, is 14.1–
13
˚
16.4 A, according to the X-ray structure. The photo-
˚
˚
switch AMPB is only 6.2 A in the cis- and 12.0 A in
the trans-conformation.14,15 Therefore, an extra glycine
residue was added to each side of AMPB and in this
˚
way a linker of 18.6 A in the trans conformation was ob-
tained.16 The linker length, including the two glycine
After preparative RP-HPLC, a small amount of the pep-
tide was dissolved in water to record UV-VIS spectra.
First, the sample was irradiated with visible light to ob-
tain trans isomer 1a. NMR spectra of compounds 8 and
9 showed that the amount of trans isomer after irradia-
tion with visible light was nearly 100%. This 100% trans
isomer cannot be obtained by irradiation alone, due to
partial overlap of the n-p* and p–p* transitions.11 To
reach 100% trans also thermal isomerization has to oc-
cur. After the spectrum of 1a was measured, the sample
was irradiated with 366 nm light for several time inter-
vals and spectra were recorded (Fig. 3). This was re-
peated until no further change in the spectrum was
observed after 120 s. The spectra showed three isosbestic
points at 240, 286, and 403 nm.
˚
residues, is only 7.2 A in the cis conformation.
For synthesis of the Fmoc protected azobenzene con-
taining building block 9 (Scheme 1) a literature proce-
dure was adapted.12,17,18 First, (4-amino)benzylamine
was selectively protected using Fmoc-OSu to yield the
Fmoc protected amine 3. Initially, it was attempted to
synthesize AMPB from 4-nitrobenzoic acid tert-butyl
protected as is described in the literature.12,18 After the
tert-butyl protection, the nitro group was converted into
a hydroxylamine using activated zinc dust. However, in
our hands under these conditions also the tert-butyl es-
ter was removed, possibly due to traces of HCl left in the
zinc after activation, resulting in the undesired 4-
hydroxylaminebenzoic acid. Therefore, 4-nitrobenzoic
acid (4) was instead allyl protected and now the prepa-
ration of hydroxylamine allylester 6 went smoothly.
Integration of the signals from NMR spectra of trans
isomer 1a (8.61 and 8.95 ppm) and cis isomer 1b (8.49

J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
1395
Figure 2. The native c-ITAM peptide (derived from FceRI) and the photoswitchable ITAM mimics 1a and 1b. The indicated distances are between
the SH2 binding epitopes.14,15
tide immobilized on the SPR chip as previously de-
scribed.1,7,8 Prior to addition, the competing ligand 1
was irradiated with UV light (366 nm) or visible light
to give 60% cis isomer or 100% isomer, respectively.
The tSH2 and the irradiated cis or trans isomer in differ-
ent concentrations were injected into the SPR cuvette
and the binding of the tSH2 to the sensor chip was mea-
sured. From these measurements the dissociation con-
stants (KD) could be calculated as described previously
Figure 3. UV-absorption spectra of trans isomer 1a (t = 0) upon
increasing time of irradiation with 366 nm light. From top (335 nm)
spectrum: t = 0 s, then t = 10, 20, 30, 60, 90, and 120 s.
and 8.79 ppm) revealed that a maximum cis isomer per-
centage of 60% could be reached upon irradiation with
366 nm light. The lower maximal cis percentage for the
peptide as compared to its building block 9 has also
been observed in other peptides and might be due to a
higher degree of steric hindrance.20
The interaction between the two isomers and the Syk
tSH2 protein was measured with SPR. Competition
experiments were performed with the native ITAM pep-
Figure 4. SPR competition experiments with Syk tSH2 (50 nM) in the
presence of various concentraties of peptide 1 in cis (ꢀ) or trans (•)
conformation. The data are fitted and KD derived as described.21

1396
J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
(Fig. 4).21 The KD value of the trans isomer was
65 8 nM and the observed KD of the mixture contain-
ing 60% cis isomer was 146 11 nM. Since the mixture
still contained 40% trans isomer 1a, the actual KD,cis can
be calculated with Eq. 1 and amounts to 860 nM.
displacing more Syk tSH2 from the sensor surface in the
competition experiment. Within 500 s the signal reached
the level of binding of only the trans isomer (bottom
curve), showing that a 100% trans state could be ob-
tained in reasonably short time. The cis–trans isomeriza-
tion could be perfectly fitted by a mono-exponential
function (not shown), showing pseudo first order kinet-
ics, as was expected for a photochemical process with
constant irradiation intensity.
Although the difference in affinity is more than 10-fold,
these numbers show that the cis isomer, in which the
SH2 domains are forced to be much closer to each other
than in the complex with native ITAM, still has an affin-
ity that is significantly higher than for a monovalent
interaction, which has a KD of 27 lM.22 Syk tSH2 has
a very flexible coiled-coil linker which connects the
two SH2 domains.1 Therefore, it accommodates fairly
easy ITAM sequences containing different linker
lengths. Furthermore, the glycine residues also have a
moderate degree of flexibility, which can result in a
somewhat larger linker length in the cis isomer 1b, when
bound to tSH2 compared to the free linker, which was
used for initial modeling.
3. Conclusions
In conclusion, we have shown that we can design and
synthesize an ITAM peptidomimetic with significant dif-
ference in binding affinity between the cis and trans iso-
mer for Syk tSH2. Using SPR the change in binding
affinity due to photoisomerization could be studied in
real time. To our knowledge, this is the first report on
the effect of isomerization of a photoswitchable ligand
on binding to a protein that is monitored during an
SPR binding assay. In the near future we aim to design
a photoswitchable Syk tSH2 ligand, which can enter
cells, having a larger difference in affinity between cis
and trans isomers. Such a ligand might be an interesting
tool to study the possibility of instantaneously switching
intracellular processes on and off by irradiation.
Now that different affinities for cis isomer 1b and trans
isomer 1a have been established, it is interesting to mon-
itor the conversion from cis to trans, and the effect on
binding in real time with SPR. For this, an experiment
was performed, making use of the convenient cuvette
design of the IBISII SPR instrument, which makes it
possible to irradiate the sample during the SPR binding
assay. First, the sensograms of only tSH2 and tSH2 with
1a were recorded (Fig. 5, top - and bottom curve). Then
the mixture containing 60% 1b with tSH2 was injected
into the cuvette and after 800 s a steady state situation
was almost reached (Fig. 5, middle curve). Then the
sample in the SPR cuvette was irradiated with visible
light with the SPR instrument in the measuring mode.
This resulted in a decrease of SPR signal, as the weaker
binding 1b was transformed to the stronger binding 1a,
4. Experimental
All chemicals were obtained from commercial sources
and used without further purification. Solvents, which
were used for the solid phase peptide synthesis, were
˚
stored at 4 A molsieves, except for MeOH, which was
˚
stored at 3 A molsieves. Prior to use, zinc was activated
by stirring in 2% HCl for 30 min. The zinc was filtered,
washed with water, ethanol, and diethyl ether, and dried
under reduced pressure. The reactions were performed
at room temperature unless stated otherwise. The reac-
tions were monitored and the Rf values were determined
by thin layer chromatography (TLC). The TLC plates
were obtained from Merck and were coated with silica
gel 60 F-254 (0.25). The spots were visualized by UV
light and ninhydrin staining. Solvents were removed un-
der reduced pressure at a temperature of 40 ꢁC. Column
chromatography was performed on silica gel. Photoiso-
merization was performed using a visible light emitting
lamp (Schott/Paes KL-1500) in combination with a glass
plate to filter the UV light or a 6 W 366 nm handheld
TLC lamp (Konrad Benda).14
1H NMR spectra were measured on a Varian Mercury
plus 300 MHz spectrometer and the chemical shifts are
given in ppm (d) relative to TMS. 13C NMR spectra
were measured on a Varian Mercury plus 75 MHz spec-
trometer and the chemical shifts are given in ppm (d) rel-
ative to CDCl3 or DMSO-d6. The 13C NMR spectra
were measured using the attached proton test (APT).
NMR spectra of peptide 1 were recorded on a Varian
Inova 500 MHz spectrometer in H2O/D2O 9:1 and were
calibrated on dioxane. UV-VIS spectra were recorded
with a UV1 spectrophotometer (Thermo Electron Cor-
Figure 5. Effect of conversion of cis isomer 1b to trans isomer 1a in an
SPR competition experiment within situ irradiation with visible light.
Top curve: Real time SPR signal for binding of 50 nM Syk tSH2 to
immobilized native diphosphorylated c-ITAM peptide on the SPR
sensor surface. Bottom curve: SPR signal of 50 nM Syk tSH2 in the
presence of 375 nM 1a. Middle curve: 50 nM Syk tSH2 in the presence
of 375 nM 60% 1b, after 800 s the sample in the SPR cuvette was
irradiated with visible light as indicated by the arrow.

J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
1397
poration). Analytical HPLC was measured on a Shima-
dzu HPLC system with a UV detector operating at 220
˚
and 254 nm using an Alltech Adsorbosphere XL C8 90A
4.5. Allyl N-Fmoc-(4-aminomethyl)phenylazobenzoate (8)
Compound 8 was synthesized according to a procedure
described in literature,18 except for the following. The
reaction time was 16 h. The solvent was removed under
reduced pressure and the resulting orange solid was dis-
solved in EtOAc and water. The organic layer was re-
moved and the water layer was extracted with EtOAc
(2·). During the extraction a slightly orange precipitate
occurred. This precipitate was removed by filtration and
the organic layers were combined, washed twice with
brine, dried over Na2SO4 and filtered. The solvent was re-
moved under reduced pressure and the resulting orange
solid was purified by silica gel chromatography with
CH2Cl2/MeOH 198:1 yielding 220 mg (0.42 mmol, 65%)
of diazocompound 8 as an orange solid. Rf = 0.83
(CH2Cl2/MeOH 98:2). 1H NMR (CDCl3, 300 MHz)
d = 4.20–4.22 (t, 1H, CH Fmoc), 4.43–4.51 (2d, 4H,
CH2NH + CH2 Fmoc), 4.85–4.87 (d, 2H, COOCH2
CHCH2), 5.25 (bs, 1H, NH), 5.29–5.46 (2dd, 2H,
COOCH2CHCH2), 6.00–6.12 (m, 1H, COOCH2
CHCH2), 7.24–7.60 (m, 6H, 2 ar + 4 ar Fmoc), 7.74–
7.77 (d, 2H, ar Fmoc), 7.88–7.90 (d, 2H, ar Fmoc),
7.92–7.95 (m, 4H, ar), 8.19–8.22 (d, 2H, ar).13C NMR
(CDCl3, 75 MHz) d = 45.0 (CH2NH), 47.6 (CH Fmoc),
66.1 (COOCH2CHCH2), 66.9 (CH2 Fmoc), 118.8
(COOCH2CHCH2), 120.2, 125.2, 127.3, 127.9, 141.6
and 144.1 (ar Fmoc), 122.9, 128.4, 142.5 and 152.2
(C6H4CH2NH), 123.7, 132.1, 132.3 and 155.4 (C6H4
COO), 130.9 (COOCH2CHCH2), 156.4 (NHCOO),
165.9 (COOCH2CHCH2).
5 lm (250 · 4.6 mm) column. For the analytical HPLC
a gradient from 100% buffer A (15 mM TEA in H2O
titrated at pH 6 with 85% H3PO4) to 100% buffer B (buf-
fer A/acetonitrile 1:9) in 20 min was used. For the pre-
parative HPLC a Gilson system with a UV detector
operating at 220 and 254 nm equipped with an Alltech
˚
Adsorbosphere XL C8 90 A 10 lm (250 · 22 mm) col-
umn was used. A gradient of 100% buffer A (0.1%
TFA in H2O/acetonitrile 95:5) to 100% buffer B (0.1%
TFA in H2O/acetonitrile 5:95) was used. For the use
in HPLC buffers TEA was distilled over ninhydrin and
KOH.
SPR measurements were performed on a double channel
IBISII SPR instrument (Eco Chemie, Utrecht, The
Netherlands) equipped with a CM 5 sensor chip (BIA-
core AB, Uppsala, Sweden).
4.1. N-Fmoc(4-amino)benzylamine (3)
Compound 3 was synthesized according to a procedure
described in the literature.17 Yield: 10.47 g (30.42 mmol,
61%) of a white solid.
4.2. Allyl 4-nitrobenzoate (5)
Allylbromide (2.50 mL, 30 mmol) and K2CO3 (4.15 g,
30 mmol) were added to a solution of 4-nitrobenzoic
acid (3.36 g, 20 mmol) in DMF (100 mL). After 5 min
of stirring the solvent was removed under reduced pres-
sure. The resulting solid was dissolved in CH2Cl2 and
water. The organic layer was washed with 5% NaHCO3
(2·), 1 M KHSO4, and brine, dried over Na2SO4, fil-
tered, and concentrated yielding 3.30 g (15.9 mmol,
79%) of 5 as a brown solid. Rf = 0.61 (hexane/EtOAc
5:1) 1H NMR (CDCl3, 300 MHz) d = 4.87–4.89 (d,
2H, OCH2), 5.32–5.48 (2dd, 2H, CH2), 6.00–6.12 (m,
1H, CH), 8.22-8.30 (m, 4H, ar.) 13C NMR (CDCl3,
75 MHz) d = 66.6 (OCH2), 119.3 (CHCH2), 123.7,
130.9, 135.7 and 150.7 (C6H4), 131.70 (CHCH2), 164.5
(COO).
4.6. N-Fmoc-(4-aminomethyl)phenylazobenzoic acid
(Fmoc-AMPB-OH, 9)
Phenylsilane (56 lL, 0.46 mmol) was added to a solution
of 8 (220 mg, 0.42 mmol) in THF (20 mL) and stirred
under a nitrogen atmosphere. Pd(PPh3)4 (49.8 mg,
0.008 mmol) was added under an argon atmosphere
and the reaction mixture was stirred for 2 h under a
nitrogen atmosphere. The solvent was removed under
reduced pressure and the resulting solid was dissolved
in DMF and purified by silica gel chromatography with
CH2Cl2/MeOH/CH3COOH 97:2:1 yielding 180 mg
(0.37 mmol, 90%) of Fmoc-AMPB-OH 9 as an orange
solid. Rf = 0.65 (CH2Cl2/MeOH 95:5) 1H NMR
(DMSO, 300 MHz) d = 4.25 (d, 2H, CH2NH), 4.28–
4.30 (d, 1H, CH Fmoc), 4.38–4.40 (d, 2H, CH2 Fmoc),
7.31–7.36 (t, 2H, ar), 7.40–7.46 (t, 4H, ar Fmoc), 7.70–
7.72 (d, 2H, ar Fmoc), 7.89–7.97 (m, 6H, 4 ar + 2 ar
Fmoc), 8.12–8.15 (d, 2H, ar). 13C NMR (DMSO,
75 MHz) d = 43.5 (CH2NH), 46.8 (CH Fmoc), 65.3
(CH2 Fmoc), 120.1, 125.1, 127.0, 127.6, 140.8 and
144.3 (ar Fmoc), 122.5, 127.4, 143.8 and 155.9
(C6H4CH2NH), 122.9, 130.6, 132.7 and 154.3
(C6H4COOH), 156.4 (NHCOO), 166.7 (COOH).
4.3. Allyl 4-hydroxylaminebenzoate (6)
Compound 6 was synthesized according to a procedure
described in the literature.23 Yield: 524 mg (2.71 mmol,
68%) of 6 as an orange solid. Rf = 0.35 (CH2Cl2/MeOH
98:2) 1H NMR (CDCl3, 300 MHz) d = 4.74–4.76 (d, 2H,
OCH2), 5.22–5.39 (2dd, 2H, CH2), 5.92–6.05 (m, 1H,
CH), 6.92 + 7.20 (2 bs, 2H, NH + OH) 6.87–6.89 (m,
2H, ar) 7.90–7.92 (d, H, ar.) 13C NMR (CDCl3,
75 MHz) d = 65.9 (OCH2), 113.4, 123.1, 131.5 and
154.9 (C6H4), 118.6 (CHCH2), 132.8 (CHCH2), 167.3
(COO).
4.7. Solid phase peptide synthesis
4.4. Allyl 4-nitrosobenzoate (7)
Peptide 1 was assembled on Tentagel[ꢂ]-Rink-NH-Fmoc
resin (0.65 g, 0.15 mmol, loading 0.23 mmol/g) using
standard Fmoc/tBu chemistry. The Fmoc protecting
group was removed using 20% piperidine in NMP
Compound 7 was synthesized according to a procedure
described in the literature.24 The product was used
immediately without further purification.

1398
J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
(3 · 8 min) followed by washing steps with NMP
(3 · 2 min), CH2Cl2 (3 · 2 min) and NMP (3 · 2 min).
The amino acid coupling mixtures were prepared by dis-
solving 4 equivalents of amino acid (0.60 mmol), 4 equiv-
alents of HOBt and HBTU, and 8 equivalents of DiPEA
in NMP (10 mL) and coupled during a coupling time of
60 min. The resin was washed with NMP (3 · 1 min)
and CH2Cl2 (3 · 1 min) after every coupling step, fol-
lowed by Fmoc deprotection. The coupling steps and
deprotection steps were monitored using the Kaisertest.25
When the Fmoc deprotection was not complete, the
deprotection steps were repeated. The amino acid build-
ing blocks Fmoc-Leu-OH, Fmoc-Thr(tBu)-OH, Fmoc-Glu
(OtBu)-OH, Fmoc-Tyr(OP(OBn)OH)-OH, Fmoc-Gly-
OH, Fmoc-AMPB-OH (9), Fmoc-Gly-OH, Fmoc-Leu-OH,
Fmoc-Gly-OH, Fmoc-Thr(tBu)-OH, and Fmoc-Tyr(O-
P(OBn)OH)-OH were subsequently coupled. For the cou-
pling of Fmoc-Tyr(OP(OBn)OH)-OH, 2 equivalents of
amino acid, 2 equivalents of the coupling reagents HATU
and HOAt, and 5 equivalents of DiPEA were used. For
the coupling of Fmoc-AMPB-OH also 2 equivalents of
amino acid and 2 equivalents of the coupling reagents
HATU and HOAt were used and 4 equivalents of DiPEA.
After the first Fmoc-Tyr(OP(OBn)OH)-OH coupling an
extra washing step (2 · 10 min) with a mixture of 1 M
TFA/1.1 M DiPEA in NMP was performed after each
Fmoc deprotection step. This was done to replace the
piperidine counterion of Tyr(OP(OBn)OH) for DiPEA.
When all the coupling steps were completed the peptide
was acetylated using a capping solution of Ac2O
(4.72 mL, 42.7 mmol), DiPEA (2.18 mL, 22.8 mmol),
and HOBt (0.23 g, 1.7 mmol) in 100 mL NMP for
2 · 20 min. The peptide was cleaved from the resin and
the side chains were deprotected with a solution of
TFA/H2O/TIS (95/2.5/2.5) for 3 h. The resin was removed
from the solution by filtration. The peptide was precipi-
tated with MTBE/hexane 1:1 v/v at ꢁ20 ꢁC and lyophi-
lized from CH3CN/H2O 1:1 v/v yielding 141 mg
(0.093 mmol, 62%) of the crude peptide. The peptide
was purified by preparative HPLC. About half
(62.4 mg) of the crude peptide was purified yielding
9.7 mg of 1. If all of the crude peptide was purified with
the same yield, the overall yield would have been
21.3 mg (16%).
The other half of the sample was irradiated with UV light
of 366 nm to obtain the maximum percentage of cis con-
former 1b and subjected to the SPR assay under exclusion
of light. KD values have been derived from the competi-
tion experiments as was earlier described.21 The intrinsic
affinity of 1b (KA,cis = 1/KD,cis) is derived by applying
Eq. 1, in which Kobs is the observed association constant,
f is the fraction of 1 in the cis isomer, and KA,trans is the
affinity assayed for the pure trans conformer.
Kobs ¼ f :KA;cis þ ð1 ꢁ f ÞKA;trans
ð1Þ
To monitor the effect of conversion of 1a to 1b on bind-
ing in real time, experiments were performed with in situ
irradiation in the SPR instrument. The isomer mixture
1ab was first irradiated with 366 nm light to obtain a
maximum percentage of cis-isomer. Then the sample
was injected into the SPR cuvette under exclusion of
light. On reaching binding equilibrium the sample in
the SPR cuvette was irradiated with visible light for con-
version to trans isomer 1a, and the change in SPR signal
was recorded.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmc.
2007.10.049.
References and notes
1. De Mol, N. J.; Catalina, M. I.; Dekker, F. J.; Fischer, M. J. E.;
Heck, A. J.; Liskamp, R. M. J. ChemBioChem 2005, 6, 2261.
2. Kumaran, S.; Grucza, R. A.; Waksman, G. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 14828.
3. Sada, K.; Yamamura, H. Curr. Mol. Med. 2003, 3, 85.
4. Simon, M.; Vanes, L.; Geahlen, R. L.; Tybulewicz, V. L. J.
J. Biol. Chem. 2005, 280, 4510.
5. Siraganian, R. P.; Zhang, J.; Suzuki, K.; Sada, K. Mol.
Immunol. 2001, 38, 1229.
6. Colgan, J.; Rothman, P. Curr. Opin. Allergy Clin. Immu-
nol. 2007, 7, 51.
7. Dekker, F. J.; de Mol, N. J.; van Ameijde, J.; Fischer, M.
J. E.; Ruijtenbeek, R.; Redegeld, F. A. M.; Liskamp, R.
M. J. ChemBioChem 2002, 3, 238.
8. Dekker, F. J.; de Mol, N. J.; Fischer, M. J. E.; Liskamp,
R. M. J. Bioorg. Med. Chem. Lett. 2003, 13, 1241.
9. Catalina, M. I.; Fischer, M. J. E.; Dekker, F. J.; Liskamp,
R. M. J.; Heck, A. J. R. J. Am. Soc. Mass Spectrom. 2005,
16, 1039.
HPLC retention times: 1b 9.8 min, 1a 10.4 min. HRMS
(ESI): [M + Na]+ calculated 1533.547, found 1533.664;
[M + 2Na]2+ calculated 778.268, found 778.366.
4.8. Surface plasmon resonance
Peptide 1 (1.02 mg, 0.675 lmol) was dissolved in HBS
buffer (675 lL) to obtain a stock solution with a concen-
tration of 1 mM. A series of samples was prepared in
HBS buffer containing 50 nM Syk tSH2 and 1 with con-
centrations ranging from 0 to 3750 nM. The sensor chip
was immobilized with the native ITAM peptide as de-
scribed.7 Competition experiments were performed with
the trans isomer 1a and with the isomer mixture contain-
ing the maximum percentage of cis isomer 1b. For this
240 lL samples were irradiated with visible light for
complete conversion to the trans isomer 1a. Then, half
of the sample was injected into the SPR apparatus.
10. Ulysse, L.; Cubillos, J.; Chmielewski, J. J. Am. Chem. Soc.
1995, 117, 8466.
11. Renner, C.; Moroder, L. ChemBioChem 2006, 7, 868.
12. Renner, C.; Kusebauch, U.; Loeweneck, M.; Milbradt, A.
G.; Moroder, L. J. Peptide Res. 2005, 65, 4.
13. Futterer, K.; Wong, J.; Grucza, R. A.; Chan, A. C.;
Waksman, G. J. Mol. Biol. 1998, 281, 523.
14. Park, S. B.; Standaert, R. F. Bioorg. Med. Chem. 2001, 9,
3215.
15. Standaert, R. F.; Park, S. B. J. Org. Chem. 2006, 71, 7952.
16. For all calculations YASARA (http://www.yasara.org) was
˚
used. The simulation cell was extended 5.0 A along each
axis, the Amber99 forcefield was used, the temperature

J. Kuil et al. / Bioorg. Med. Chem. 16 (2008) 1393–1399
1399
control was step-10 annealing and the simulation time was
10000 fs each run. All molecules were modeled as their N-
terminal acetamide and C-terminal amide derivatives. All
distances are measured form the N-terminal nitrogen atom
to the carbon atom of the C-terminal amide.
21. De Mol, N. J.; Gillies, M. B.; Fischer, M. J. E. Bioorg.
Med. Chem. 2002, 10, 1477.
22. Ruijtenbeek, R.; Kruijtzer, J. A. W.; van de Wiel, W.;
Fischer, M. J. E.; Fluck, M.; Redegeld, F. A. M.;
Liskamp, R. M. J.; Nijkamp, F. P. ChemBioChem 2001,
2, 171.
23. Nakayama, K.; Endo, M.; Majima, T. Bioconjugate Chem.
2005, 16, 1360.
24. Park, S. B.; Standaert, R. F. Tetrahedron Lett. 1999, 40,
6557.
17. Behrendt, R.; Schenk, M.; Musiol, H.-J.; Moroder, L.
J. Peptide Sci. 1999, 5, 519.
18. Ulysse, L.; Chmielewski, J. Bioorg. Med. Chem. Lett.
1994, 4, 2145.
19. Aemissegger, A.; Hilvert, D. Nat. Protocols 2007, 2, 161.
20. Rovero, P.; Pegoraro, S.; Vigano, S.; Amato, C.; Vaccari,
L.; Balestreri, E.; Felicioli, R. Lett. Pept. Sci. 1995, 2, 27.
25. Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I.
Anal. Biochem. 1970, 34, 595.