Synthetic strategies to a backbone-side chain cyclic SHP-1 N-SH2 ligand containing N-functionalized alkyl phosphotyrosine

Article abstract of DOI:10.2174/092986610791306779

The cyclic peptide EGLNcΨ[CON((CH₂)₃NH)pYNleE(NHCH₂CO)]L-NH₂ (1) was designed and synthesized according to a native interaction partner of tyrosine phosphatase SHP-1. We introduced N-aminopropyl-phosphotyrosine

Full text of DOI:10.2174/092986610791306779

Protein & Peptide Letters, 2010, 17, 809-816
809
Synthetic Strategies to a Backbone-Side Chain Cyclic SHP-1 N-SH2
Ligand Containing N-Functionalized Alkyl Phosphotyrosine†
Kathleen Teichmann1, Tobias Niksch2, Karin Wieligmann3, Martin Zacharias4 and Diana Imhof1,*
1Center for Molecular Biomedicine, Department of Biochemistry, Peptide Chemistry Group, Friedrich-Schiller-
2
University, Hans-Knöll-Str. 2, D-07745 Jena, Germany; Institute of Inorganic and Analytical Chemistry, Friedrich-
Schiller-University Jena, August-Bebel-Str. 2, D-07743 Jena, Germany; 3Max Planck Research Unit for Enzymology of
Protein Folding, Weinbergweg 22, D-06120 Halle, Germany; 4Physik Department T38, Technische Universität
München, James-Franck-Str. 1, D-85748 Garching, Germany
Abstract: The cyclic peptide EGLNcꢀ[CON((CH2)3NH)pYNleE(NHCH2CO)]L-NH2 (1) was designed and synthesized
according to a native interaction partner of tyrosine phosphatase SHP-1. We introduced N-aminopropyl-phosphotyrosine
to enable backbone-side chain cyclization with a glutamic acid derivative as counterpart for cyclization. Different ap-
proaches have been compared to find a strategy for the generation of backbone and backbone-side chain cyclic phos-
phopeptides.
Keywords: N-functionalized alkyl phosphotyrosine; backbone cyclization; SH2 domain, ligand; protein tyrosine phosphatase;
SHP-1.
INTRODUCTION
to their linear counterparts. The fact that restriction of bioac-
tive conformations through cyclization leads to an enhance-
ment of binding affinity was also observed in other studies
where cyclic peptides and, in particular, cyclic phosphopep-
tides were developed as potent inhibitors of a distinct bind-
ing event [9-11].
Protein-protein interactions play a critical role in the
regulation of signal transduction events during various stages
of cell development [1]. Such interactions generally involve
modular adaptor domains that function as docking or scaf-
folding components in proteins [2]. Phosphorylated tyrosine
regions in proteins often exhibit potential binding affinity for
other proteins containing the highly conserved Src homology
2 (SH2) domains. Binding studies for SH2 domains using
phosphotyrosyl peptides with randomized sequences sug-
gested that beside the primary interaction with the phospho-
tyrosyl (pTyr, pY) residue specificity is dictated by the sur-
rounding peptide sequence [3,4]. In our previous work we
have demonstrated that conformationally restricted linear
and cyclic peptides represent high affinity binding partners
of the N-terminal SH2 domain of SHP-1, a cytosolic protein
tyrosine phosphatase (PTP) that is primarily involved in the
regulation of hematopoietic cell signaling [5-7]. Our labora-
tory has synthesized derivatives of the sequence EGL-
NpY2267MVL derived from the epithelial receptor tyrosine
kinase Ros which represents a high-affinity binding partner
of SHP-1 [8]. Some of the synthetic variants showed more
potent binding to SHP-1 N-SH2 than the native lead peptide,
and in addition, cyclic peptides were found to partially in-
hibit SHP-1 phosphatase activity [5]. Within our series of N-
SH2 peptide ligands we also observed that in general the
cyclic peptides retained significant binding affinity relative
Considering the particular situation for ligand binding to
the N-SH2 domain of SHP-1 our previously synthesized pep-
tides were cyclized via side chains or modified side chains
between positions -1 and +2 relative to the pY-residue in
order not to interfere with protein-ligand contacts involving
residues that determine high-affinity binding. Molecular
modeling studies revealed that a shift of the cycle towards
the C-terminus, in particular to the phosphotyrosine residue,
may turn out advantageous with respect to binding affinity
and at the same time may hamper the dissociation process of
the N-SH2-PTP-complex that is responsible for the stimula-
tion of the phosphatase activity. A predicted N-backbone-to-
side-chain cyclic ligand is shown in Fig. (1). Due to the fact
that we are interested in the development of inhibitors of
PTP SHP-1 we first focused our work on the evaluation of
synthetic strategies for the generation of such a cyclic phos-
phopeptide. Compared to the commonly used approach of
introducing suitably protected phosphoamino acid building
blocks into a peptide chain by standard solid phase methods
we were faced with the incorporation of an N-functionalized
alkyl phosphotyrosine residue.
In general, phosphorylation of the hydroxyl amino acids
Ser, Thr and Tyr within a peptide sequence can be achieved
by either postassembly phosphorylation ("global phosphory-
lation") or by insertion of a preformed phosphorylated amino
acid derivative [12]. In our peptide, postassembly phos-
phorylation circumvents the preparation of the phosphoty-
rosine building block, but steric constraints, especially due to
secondary structure formation, may hamper effective phos-
*Address correspondence to this author at the Center for Molecular Bio-
medicine, Department of Biochemistry, Peptide Chemistry Group, Friedrich
Schiller University Jena, Hans-Knöll-Straße 2, D-07745 Jena, Germany;
Tel: +49 3641 949 368; Fax: +49 3641 949 352;
E-mail: diana.imhof@uni-jena.de
†Parts of these studies have been published in: Imhof, D.; et al. In Peptides
2004. Proceedings of the 28th European Peptide Symposium, Kenes Inter-
national: Geneva, 2004, pp. 360-361.
0929-8665/10 $55.00+.00
© 2010 Bentham Science Publishers Ltd.

810 Protein & Peptide Letters, 2010, Vol. 17, No. 7
Teichmann et al.
phorylation [13]. However, the most effective and versatile
phosphorylation procedure for this purpose described is
based on P(III) chemistry converting the hydroxyl amino
acid into phosphono derivative by treatment with either aryl-
or alkyl phosphochloridite or N,N-dialkyl phosphoramidite
[14-16]. Both methods require subsequent in situ oxidation
of the phosphite triester intermediate into the phosphate tri-
ester. Global phosphorylation has been intensively studied
and successfully used for the preparation of short- and me-
dium-sized peptides. Still, the use of properly protected
phosphorylated amino acids in general SPPS is certainly the
preferred method, in particular if the peptide to be synthe-
sized contains amino acids giving rise to side reactions dur-
ing phosphorylation, e.g. Trp and Met in the oxidation step
of the desired phosphorylation procedure [17]. However, the
preparation and use of phosphorylated N-functionalized al-
kyl hydroxyl amino acids for the generation of backbone
cyclic phosphopeptide ligands has not been reported yet.
Herein, we describe the application of different synthetic
approaches to generate the backbone-side chain cyclic pep-
tide EGLNcꢀ[CON((CH2)3NH)pYNleE(NHCH2CO)]L-NH2
(1) containing an N-aminopropyl-phosphotyrosine by using
both a building block as well as the global phosphorylation
approach (Fig. 1). In compound (1), representing a progeni-
tor for further N-SH2 ligands, Nle was used as an isostere for
Met as described earlier, while Met was used in the model
peptide LNYMVL (2) derived from the natural lead peptide
Ros pY2267 (Fig. 1A) [5-7, 18]. The model peptide was
primarily used in order to evaluate different strategies for
global phosphorylation and oxidation due to the occurrence
of a Met residue within the native sequence, whereas the
preparation and phosphorylation of the N-functionalized
alkyl tyrosine and its use was investigated in parallel.
Figure 1. A) The backbone-side chain cyclic peptide
EGLNcꢀ[CON((CH2)3NH)pYNleE(NHCH2CO)]L-NH2 (1) was
derived from consensus sequence class I for SHP-1 N-SH2 ligands
as well as a natural high affinity binding partner, Ros pY2267. B)
Stick illustration of a representative conformation of backbone-side
chain cyclic peptide (1) as ligand for the N-SH2 domain of SHP-1.
C4). MS m/z 545.0 (M+H)+; HPLC tR 30.7 min. Rf 0.89 (n-
butanol/acetic acid/water 4:1:1). Rf 0.64 (chloroform/
methanol 9:1).
MATERIALS AND METHODS
General
Fmoc-N[(CH2)3NHAlloc]Tyr(PO3But2)-OH
The phosphorylation of the Fmoc-N-aminopropyl-
tyrosine was modified from the literature [21]. In brief, to a
solution of Fmoc-N[(CH2)3NHAlloc]Tyr-OH (0.5 g, 0.92
mmol) in dry THF was added DIEA (0.16 ml, 0.92 mmol)
followed by TBDMS-Cl (0.17 g, 1.10 mmol). The solution
was stirred at room temperature for 4 h. To this solution 1H-
tetrazole (0.19 g, 2.76 mmol) and di-tert-butyl-N,N-diethyl-
phosphoramidite (0.28 g, 1.10 mmol) in dry THF was added.
The mixture was stirred at room temperature for 3 h, then t-
BuOOH (0.50 ml, 3 M solution in toluene) in 3 ml di-
chloromethane was added. After 20 min, the solvents were
removed, and the residue was dissolved in EtOAc (150 ml)
and washed with 5% KHSO4 (3-times) and NaCl (3-times),
then dried over Na2SO4. The solvent was removed under
reduced pressure, and the residue was purified by silica gel
column chromatography with chloroform/methanol to afford
Fmoc-N[(CH2)3NHAlloc]Tyr(PO3But2)-OH (0.48 g, 85%) as
a white powder. NMR ꢁ (C13) (CDCl3) 29.67 (PBut CH3),
47.35 (Tyr Cꢂ), 65.57 (Tyr Cꢃ), 76.48/76.98/77.49/78.67
(Fmoc CH2, PBut C), 115.37/117.77/119.91 (Tyr Ar C, Allyl
CH2), 127.20/127.74/130.21 (Fmoc Ar C, Allyl CH),
141.6/143.67 (Fmoc Ar C), 154.58 (Tyr Ar C4). MS m/z
737.02 (M+H)+; HPLC tR 44.37 min. Rf 0.54 (chloro-
form/methanol 9:1). After cleavage of the tert-butyl protec-
tion of the phosphate group using 50% TFA/dichloro-
methane Fmoc-N[(CH2)3NHAlloc]Tyr(PO3H2)-OH was ob-
All amino acid derivatives and coupling reagents were
purchased from Novabiochem (Merck KGaA, Darmstadt,
Germany). Dibenzyl-N,N-diethylphosphoramidite, di-tert-
butyl-N,N-diethylphosphoramidite and tetrakis(triphenyl-
phosphine)palladium(0) were obtained from Sigma-Aldrich
Chemie GmbH (Taufkirchen, Germany). Peptide synthesis
reagents (DIEA, piperidine) and solvents (DMF, DCM) were
of reagent grade from Fluka (Sigma-Aldrich Chemie GmbH,
Taufkirchen, Germany) and Iris Biotech GmbH (Marktred-
witz, Germany), respectively. Solvents for chromatography
were of analytical grade obtained from VWR International
(Dresden, Germany). Purifications of dipeptide building
units by column chromatography were performed on Merck
silica gel 60 (0.040-0.063 mm). NMR experiments were re-
corded at 30°C on a Bruker 400 MHz NMR spectrometer.
Building Block Synthesis
Fmoc-N[(CH2)3NHAlloc]Tyr-OH
The generation of HN[(CH2)3NHAlloc]Tyr-OH and sub-
sequent N-terminal Fmoc-protection was performed accord-
ing to earlier reports [19, 20]. NMR ꢁ (C13) (CDCl3) 47.40
(Tyr Cꢂ), 65.57 (Tyr Cꢃ), 73.3 (Fmoc CH2), 115.4/117.78/
119.9 (Tyr Ar C, Allyl CH2), 127.25/127.78/130.1 (Fmoc Ar
C, Allyl CH), 141.9/143.67 (Fmoc Ar C), 154.58 (Tyr Ar

Synthetic Strategies to a Backbone-Side Chain Cyclic
Protein & Peptide Letters, 2010, Vol. 17, No. 7 811
tained and lyophilized from 80% tert-butanol to yield a white
solid compound (0.39 g, 95%). MS m/z 647.1 (M+Na)+;
HPLC tR 21.9 min; Rf 0.19 (chloroform/methanol 9:1).
in a syringe and incubated and agitated for 30 min at room
temperature. Afterwards the solution was removed and
mCPBA* (39.7 mg, 0.23 mmol) in 1 ml DCM was added.
After an incubation time of 15 min at room temperature, the
resin was washed for three times with DCM (2 min) and
lyophilized. *Instead of mCPBA, also a mixture of t-BuOOH
(53 ꢀl, 3 M solution in toluene) in DCM (320 ꢀl) was added
in selected experiments. The reconversion of oxidized Met
was performed according to Andrews et al. [22].
Peptide Syntheses
All peptides were prepared manually by the solid-phase
method using Fmoc chemistry on Rink amide MBHA resin
(0.54 mmol g-1). In general, coupling reactions were per-
formed using Fmoc amino acids (4 eq.) activated by HBTU
(4 eq.) in the presence of DIEA (8 eq.) for 0.5 – 1 h (double
couplings). Selected couplings were carried out with TFFH
(2 eq.) as the coupling reagent in the presence of Fmoc-
amino acid (2 eq.) and DIEA (4 eq.), preactivation time of
this mixture was 12 min, coupling was performed for 30 min
(double couplings). Fmoc removal was effected by treating
the resin twice with 20% piperidine in DMF (1x 5 min and
1x 15 min). The side chains of trifunctional amino acids
were protected as follows: Asn(Trt) and Glu(OBut). Protect-
ing groups for the side chains to be cyclized were Alloc- or -
OAll as appropriate. All deprotection and coupling steps
were followed by intensive washings using DMF and DCM,
alternately. Capping reactions, if necessary, were performed
using acetic anhydride/N-methylimidazole/DMF (1:2:3) for
30 min. Alloc/OAll-groups were removed with a mixture of
DMF/THF/0.5N HCl/morpholine (2:2:1:0.9) and approxi-
mately 10 mg of Pd(PPh3)4 as the catalyst. Alloc/OAll depro-
tection was followed by washing with DMF/THF (1:1, 3x)
and DMF (3x). Cyclization was achieved using PyBOP (6
eq.) and DIEA (12 eq.) in DMF for at least 3 h. Cleavage of
peptides from the resin with concomitant side-chain depro-
tection was achieved by treating the resins with
TFA/water/TIS (95:2.5:2.5) for 3 h. The crude peptide was
precipitated in diethyl ether, centrifuged and washed three
times with diethyl ether. After lyophilization peptides were
purified by either semipreparative or analytical HPLC.
Peptide Purification and Characterization
Analytical HPLC was carried out on a Shimadzu LC 10A
chromatograph equipped with a UV/VIS detector (ꢀ = 220
nm) and a Vydac 215TP C18 column (5 ꢀm, 4.6 x 250 mm).
The following solvent systems were used: A) 0.1% TFA in
water, B) 0.1% TFA in acetonitrile. Building blocks were
eluted with the gradient 20% to 80% of eluent B in 60 min at
a flow rate of 1 ml/min. Peptides were eluted with the gradi-
ent 10% to 60% of eluent B in 50 min at a flow rate of 1
ml/min. Semipreparative HPLC was carried out on a Shima-
dzu LC 8A instrument equipped with a Knauer Eurospher
100 C18 column (5 ꢀm, 25 x 250 mm) with a gradient from
15% to 65% of eluent B in 120 min at a flow rate of 10
ml/min (ꢀ = 220 nm). Mass spectra were recorded either on a
Laser Tec Research MALDI TOF mass spectrometer
(Perseptive Biosystems) using ꢁ-cyano-4-hydroxycinnamic
acid as matrix or a TSQ Ultra AM mass spectrometer (Finni-
gan). The amino acid composition of all peptides was veri-
fied by amino acid analysis using a LC 3000 system from
Eppendorf-Biotronik (Hamburg, Germany). The peptides
were hydrolyzed using 6N HCl in sealed tubes at 110°C for
24 h. The obtained analyses gave the expected quantitative
results for all peptides.
EGLNcꢀ [CON((CH2)3NH)pYNleE(NHCH2CO)]L-NH2
(1). MALDI-MS 1147.5 (M+Na)+; HPLC tR 33.05 min; Rf
0.54 (n-butanol/acetic acid/water 48:18:24). LNYMVL-NH2
(2). MALDI-MS 722.9 (M+H)+; HPLC tR 20.58 min; Rf 0.63
(n-butanol/acetic acid/water 48:18:24). LNpYMVL-NH2 (3).
MALDI-MS 801.8 (M+H)+; HPLC tR 16.38 min; Rf 0.45 (n-
butanol/acetic acid/water 48:18:24).
Peptide Phosphorylations
The phosphorylation of the peptides was performed on
the solid phase as well as in solution. In case of the phos-
phorylation in solution the linear peptide (46 mg, 0.06 mmol)
and the phosphorylation reagent di-tert-butyl-N,N-
diethylphosphoramidite (20 ꢀl, 0.07 mmol) were dissolved
in 1 ml THF and 1H-tetrazole (15.1 mg, 0.216 mmol) was
added. After 15 min (20°C) the solution was cooled down to
-40°C. A mixture of mCPBA* (18.1 mg, 0.11 mmol) in
DCM (150 ꢀl) was added. The reaction batch was agitated
for 15 min at -20 °C. Afterwards 10% Na2S2O5 (602 ꢀl) and
diethyl ether (3.0 ml) were added. The solution was taken up
in diethyl ether and washed twice with 10% NaHCO3 and
saturated aqueous NaCl solution, dried over Na2SO4 and
evaporated to dryness. The product, a yellow oil was taken
up in 80% tert-butanol and lyophilized. *In case t-BuOOH
was used instead of mCPBA, then a mixture of t-BuOOH (25
ꢀl, 3 M solution in toluene) in DCM (150 ꢀl) was added.
The reconversion of oxidized Met was performed according
to Andrews et al. [22].
Molecular Modeling
Structural models for the complex of SHP-1 N-SH2 with
phosphopeptide ligand were generated using the molecular
modeling package InsightII (Accelrys Inc., San Diego, CA)
based on homology to the SHP-2 N-SH2 domain in complex
with a peptide (SVLpYTAVQP) for which a structure is
available (pdb1aya) [23]. The molecular modeling method
used has been presented in more detail earlier [5]. In order to
test if the peptide cyclization is compatible with the known
linear peptide binding geometry the necessary amino acid
substitutions were introduced in silico assuming a peptide
backbone structure identical to the template structure
(pdb1aya). After introducing additional chemical bonds be-
tween side chains (cyclization) the resulting structural model
was energy minimized in complex with the SH2-domain.
The process was repeated for several different starting geo-
metries of the cyclic structure and the lowest energy model
was retained.
The phosphorylation on the solid phase was performed
using 50 mg resin-bound peptide, 1H-tetrazole (32.5 mg,
0.46 mmol) and di-tert-butyl-N,N-diethylphosphoramidite
(43 ꢀl, 0.15 mmol) in 2 ml DCM. The mixture was taken up

812 Protein & Peptide Letters, 2010, Vol. 17, No. 7
Teichmann et al.
RESULTS AND DISCUSSION
ing high binding affinity, yet reduced capacity to stimulate
SHP-1. One such ligand (1) containing N-aminopropyl
phosphotyrosine at position 0 is shown in Fig. (1). However,
considering accessibility of this backbone-side chain cyclic
peptide a variety of synthetic approaches were conceivable
keeping in mind that a suitably protected N-functionalized
alkyl phosphotyrosine building block was not available at the
time we started with our investigations. In Scheme 1, differ-
ent strategies that can be applied in order to generate peptide
(1) are summarized. Starting from the evaluation of several
variants for the introduction of the phosphate group in the
model peptide (2) (Scheme 1A) and concomitant attempts to
introduce the N-aminopropyl tyrosine (nonphosphory-
lated/phosphorylated) in the suggested backbone-side chain
cyclic peptide (Scheme 1B), the approaches finally used to
synthesize peptide (1) are represented in Scheme 1C.
A major consideration in our work on N-SH2 ligands for
protein tyrosine phosphatase SHP-1 was the availability of
different conformationally constrained peptides that allow
for a more extensive biochemical examination of SH2 do-
main ligand interactions. In previous studies we employed
constrained amino acids as well as side chain to side chain
cyclization to study the topographical and conformational
determinants for peptide ligands belonging to the different
binding modes known for SHP-1 N-SH2 association [5-7].
These binding modes dictate whether a peptide is able to
efficiently stimulate SHP-1 activity or inhibit the phospha-
tase due to an imperfect fit at position pY+3 [7]. Prior to the
synthesis, a molecular modeling approach was used to sug-
gest modifications with increased binding affinities com-
pared to a natural lead peptide (Ros pY2267) by docking of
the structures into the N-SH2 domain of SHP-1. In the
course of these investigations it was also suggested that cy-
clization involving the phosphotyrosine residue might be
advantageous with respect to induce a conformation possess-
The chemical phosphorylation itself can be either per-
formed on the level of a suitably protected peptide having
the side chain of phosphorylation still unprotected or by us-
ing properly protected phosphorylated amino acids in the
Scheme 1. Different synthetic pathways to backbone-side chain cyclic peptide (1). A) model peptide LNYMVL-NH2 was used in order to
test different global phosphorylation strategies (R = tert-butyl), B) introduction of N-functionalized alkyl tyrosine in the phosphorylated and
non-phosphorylated form (R = tert-butyl, X = H, tert-butyl), and C) selected strategies for generation of the backbone-side chain cyclic pep-
tide (1) resulting from investigations performed according to A and B (R = tert-butyl, X = H, tert-butyl).

Synthetic Strategies to a Backbone-Side Chain Cyclic
Protein & Peptide Letters, 2010, Vol. 17, No. 7 813
actual synthesis [17]. It is generally accepted that the use of
the preformed phosphorylated residues in solid phase synthe-
sis is preferred towards the global phosphorylation approach.
However, in which way a phosphorylated N-functionalized
alkyl amino acid can be introduced in a peptide sequence has
not been described yet, though recently backbone cyclic pep-
tides containing phosphoserine residues were reported to
inhibit NF-ꢀB [24]. Nevertheless, these peptides did not con-
tain an N-modified phosphoserine, and backbone cyclization
was achieved via N-carboxyalkyl and N-aminoalkyl amino
acids, e.g. derived from Asp and Gly, respectively. Using the
model peptide LNYMVL derived from the lead compound
Ros pY2267, three different strategies of post-assembly
phosphorylation were evaluated that were extracted from
literature reports in this field. While in strategy AI phos-
phorylation was performed in solution after completion of
peptide synthesis according to Nomizu et al. [25], in AII and
AIII phosphorylation was achieved on the solid phase, yet at
different stages of the synthesis protocol [26-30]. Certainly,
optimal phosphorylation can be achieved by using differen-
tially protected phosphoramidites as efficient phosphoryla-
tion reagents [12,13,26-30]. For the subsequent oxidation of
the tyrosinyl phosphotriester derivative (PIII to PV) several
reagents were successfully used in the past, and m-
chloroperoxybenzoic acid (mCPBA) turned out the most
effective in this respect. However, our lead peptide contains
a Met residue at position pY+1 and future backbone cyclic
peptides will be more related to the natural interaction part-
ner Ros pY2267 rather than to the consensus sequence class
I containing bulky amino acids at pY+1 (Phe, Nle) due to
sterical hindrance introduced by the building blocks required
for cyclization. Andrews et al. have demonstrated that the
use of tert-butylhydroperoxide (t-BuOOH) for oxidation and
subsequent treatment with N-methyl mercaptoacetamide is
suitable for regeneration of methionine residues, while this
amino acid is irreversibly oxidized if using m-
chloroperoxybenzoic acid [22]. Both, the resin-bound pep-
tide (2) as well as the purified peptide underwent phosphory-
lation according to Perich et al. by using di-tert-butyl-N,N-
the desired peptide (1), we performed several experiments
according to the strategies shown in Scheme 1B. Thus, it was
necessary to seek methods for the preparation of N-
aminopropyl phosphotyrosine which would be applicable to
the Fmoc-chemistry in solid phase synthesis, while the gen-
eration of nonphosphorylated N-functionalized alkyl amino
acids has been reported earlier by our lab and others
[19,20,24,31]. Already in 1989 Perich and Johns described a
versatile one-pot synthesis of a Boc-protected dialkyl phos-
photyrosine derivative using dialkyl-N,N-diethylphos-
phoramidites which was initiated by a temporary tert-
butyldimethylsilyl protection [32]. This method has later
been applied to Fmoc-protected tyrosines, too [29,33]. The
effective use of DIEA as the base during phosphorylation
instead of N-methyl morpholine was later demonstrated by
Mathe et al. for the introduction of S-acyl-2-thioethyl-
enzyme-labile phosphate protection [21]. This synthetic ap-
proach was used herein and has allowed us to prepare in a
suitable and easy way the N-functionalized alkyl phosphoty-
rosine derivative as demonstrated in Fig. (2), while the omis-
sion of the temporary tert-butyldimethylsilyl protection did
not yield the desired product.
diethylphosphoramidite
and
dibenzyl-N,N-diethylphos-
phoramidite, respectively. In all cases, independent on solu-
tion or solid phase global phosphorylation, the yields
achieved in the literature [26-30], 96% for dibenzyl-N,N-
diethylphosphoramidite and 95% for di-tert-butyl-N,N-
diethylphosphoramidite, could be confirmed. In further ex-
periments we only used the tert-butyl protection due to pro-
longed cleavage times required for benzyl-protected phos-
phopeptides as reported earlier [5-7]. The same was found
for the oxidation step of the phosphotriester intermediate that
could successfully be performed with m-chloroperoxy-
benzoic acid as well as with t-BuOOH, while the latter is
more suitable with respect to the required reconversion of
Met. However, because of the reduced number of purifica-
tion steps for the peptide prepared by solid phase synthesis
and the higher overall yield we finally decided to only incor-
porate the solid phase post-assembly phosphorylation strat-
egy for the generation of the backbone-side chain cyclic pep-
tide (Scheme 1C, strategies CI and CII).
Figure 2. A) Synthetic route to Fmoc-N[(CH2)3NHAlloc]Tyr
(PO3But2)-OH, reagents: i) TBDMS-Cl (1.2 eq.)/DIEA/THF [19],
ii) 1H-tetrazole (3 eq.)/(Et)2NP[OBut]2 (1.2 eq.) [26-30], iii) t-
BuOOH (70%) [22]; and B) HPLC-analysis of the conversion of
the starting compound to the phosphorylated analogue with (a) and
without (b) TBDMS-Cl; R = tert-butyl. Gradient: 20–80% eluent B
in 60 min with A: 0.1% TFA in water and B: 0.1% TFA in acetoni-
trile; flow rate at 1 ml/min, 220 nm.
In order to demonstrate that the application of a suitably
protected N-functionalized alkyl phosphotyrosine derivative
is equally well or even better suitable for the preparation of

814 Protein & Peptide Letters, 2010, Vol. 17, No. 7
Teichmann et al.
The principles of the strategies selected according to the
results obtained from the experiments in AI-BIII or in combi-
nation with these experiments are demonstrated for cyclic
octapeptide (1) in Scheme 1C. We decided to use the follow-
ing approaches: CI) solid phase assembly of peptide - cycli-
zation - phosphorylation, CII) solid phase assembly of pep-
tide - phosphorylation - cyclization, CIII) solid phase assem-
bly up to N-aminopropyl-tyrosine (position 0) - phosphoryla-
tion - prolongation - cyclization, and CIV) solid phase assem-
bly using N-aminopropyl-phosphotyrosine - cyclization. As
reported earlier, introduction of N-functionalized alkyl
amino acids for N-backbone cyclization is difficult in solid
phase synthesis due to sterical hindrance for the subsequent
acylation step. Therefore, the preparation of dipeptide build-
ing blocks has been suggested in several publications in the
past [19,20,31]. However, we also found herein that if using
TFFH (tetramethylfluoroformamidinium hexafluorophos-
phate) for the in situ generation of amino acid fluorides cou-
pling onto the building block can be achieved in acceptable
yields (40-50%) and time-consuming additional deprotection
and purification steps for the dipeptide building block can be
omitted. Fig. (3) demonstrates the coupling yields for each
step according to the synthetic pathways CI-CIV. Though
yields in strategy CI and CII are reduced after coupling of
Fmoc-Asn(Trt) to N-aminopropyl-tyrosine, the complete
sequence could be prepared. In contrast, introduction of the
di-tert-butyl-protected N-aminopropyl-phosphotyrosine de-
rivative led to sequence termination at the stage of Asn(Trt)
coupling. Following consideration of the reaction, it was
concluded that a possible reason for the not feasible coupling
onto the building unit could be the sterical hindrance of the
additional protection at the phosphate group of tyrosine. We
therefore tried to introduce the deprotected N-aminopropyl-
phosphotyrosine instead (CIV). As demonstrated in Fig. (3),
we were indeed able to prolongate the sequence by Asn at
position pY-1 relative to phosphotyrosine and beyond. The
coupling yield (43%) was only slightly reduced in compari-
son to the same reaction in strategies CI and CII (48%). We
can therefore conclude that both, post-assembly as well as a
building block approach led to the desired linear precursor
peptide. Cyclization of the linear peptide has been performed
as described earlier [31]. In addition, according to the results
obtained for peptide (1) it is recommended to perform cycli-
zation after phosphorylation of the complete linear peptide
(strategies CI and CIV), as was shown to be successful for
cyclic N-SH2-ligands containing a phosphotyrosine residue
[5-7]. However, in contrast to previous preparations of back-
bone cyclic peptides it was necessary to repeat cyclization at
least two times in order to increase the yields. Nevertheless,
side-products, e.g. linear peptide, linear Alloc-/OAll-
protected peptide and partially protected linear peptide were
still identified in the crude material (Fig. 4A). Fig. (4B) rep-
resents the purified peptide (1) obtained from the crude ma-
terial. Thus, continuing work is focused on the optimization
of the Alloc-/OAll-deprotection and the cyclization reaction.
Figure 3. Monitoring of coupling yields (%) obtained for the as-
sembly of the linear sequence of peptide 1 according to different
synthetic strategies.
Figure 4. HPLC analysis of cyclic peptide (1) in the crude mixture
obtained from strategy CIV (A), and after purification (B). Peak (a)
represents the linear peptide, (b) the monoallyl-protected peptide,
and (c) the diallyl-protected precursors. Peptides were eluted with a
gradient of 10-60% in 50 min, where B was 0.1% TFA in acetoni-
trile.
ing the post-assembly phosphorylation as well as a building
block approach. The latter required the availability of Fmoc-
N-aminopropyl-phosphotyrosine that was successfully ob-
tained by phosphorylation of the nonphosphorylated precur-
sor amino acid using dialkyl-N,N-diethylphosphoramidites
following temporary tert-butyldimethylsilyl protection of the
starting compound. Coupling of this building unit to the
CONCLUSIONS
In this study, the backbone-side chain cyclic peptide
EGLNcꢀ[CON((CH2)3NH)pYNleE(NHCH2CO)]L-NH2 has
been prepared according to different synthetic ways includ-

Synthetic Strategies to a Backbone-Side Chain Cyclic
Protein & Peptide Letters, 2010, Vol. 17, No. 7 815
Ros receptor tyrosine kinase signaling. An epithelial function of the
SH2 domain protein tyrosine phosphatase SHP-1. J. Cell Biol.,
2001, 152(2), 325-334.
resin-bound peptide and coupling of the subsequent amino
acid (Asn) to N-aminopropyl-phosphotyrosine was demon-
strated to proceed most efficiently if using TFFH as the cou-
pling reagent. In contrast, the acylation of the phosphate-
diprotected N-aminopropyl-phosphotyrosine was not at all
possible for obvious sterical hindrance through the additional
protecting groups. However, further work dealing with the
optimization of the successful strategies, e.g. optimization of
the cleavage of allyl-type protecting groups and the cycliza-
tion reaction, are currently in progress.
[9]
Barchi, J.J.; Nomizu, M.; Otaka, A.; Roller, P.P.; Burke, T.R. Con-
formational analysis of cyclic hexapeptides designed as constrained
ligands for the SH2 domain of the p85 subunit of phosphatidyli-
nositol-3-OH kinase. Biopolymers, 1996, 38(2), 191-208.
Burke, T.R.; Nomizu, M.; Otaka, A.; Smyth, M.S.; Roller, P.P.;
Case, R.D.; Wolf, G.; Shoelson, S.E. Cyclic Peptide Inhibitors of
Phosphatidylinositol 3-Kinase P85 Sh2 Domain Binding. Biochem.
Biophys. Res. Commun., 1994, 201(3), 1148-1153.
Li, P.; Peach, M.L.; Zhang, M.C.; Liu, H.P.; Yang, D.J.; Nicklaus,
M.; Roller, P.P. Structure-based design of thioether-bridged cyclic
phosphopeptides binding to Grb2-SH2 domain. Bioorg. Med.
Chem. Lett., 2003, 13(5), 895-899.
[10]
[11]
ACKNOWLEDGMENT
[12]
[13]
McMurray, J.S.; Coleman, D.R.; Wang, W.; Campbell, M.L. The
synthesis of phosphopeptides. Biopolymers, 2001, 60(1), 3-31.
Valerio, R.M.; Bray, A.M.; Maeji, N.J.; Morgan, P.O.; Perich, J.W.
Preparation of O-Phosphotyrosine-Containing Peptides by Fmoc
Solid-Phase Synthesis - Evaluation of Several Fmoc-Tyr(Po(3)
R(2))-Oh Derivatives. Lett. Pept. Sci., 1995, 2(1), 33-40.
Mcbride, L.J.; Caruthers, M.H. Nucleotide Chemistry .10. An In-
vestigation of Several Deoxynucleoside Phosphoramidites Useful
for Synthesizing Deoxyoligonucleotides. Tetra. Lett., 1983, 24(3),
245-248.
Kitas, E.A.; Knorr, R.; Trzeciak, A.; Bannwarth, W. Alternative
Strategies for the Fmoc Solid-Phase Synthesis of O-4-Phospho-L-
Tyrosine-Containing Peptides. Helv. Chim. Acta, 1991, 74(6),
1314-1328.
Lacombe, J.M.; Andriamanampisoa, F.; Pavia, A.A. Solid-Phase
Synthesis of Peptides Containing Phosphoserine Using Phosphate
Tert-Butyl Protecting Group. Int. J. Pept. Protein Res., 1990, 36(3),
275-280.
We thank Prof. Siegmund Reissmann for useful scientific
discussions. This work was supported in part by the
Deutsche Forschungsgemeinschaft (SFB 604), the University
of Jena and the Fonds der Chemischen Industrie.
[14]
[15]
[16]
ABBREVIATIONS
Standard abbreviations for amino acids, peptides, protect-
ing groups and peptide synthesis reagents were used as rec-
ommended in the guide published in J. Peptide Sci., 2006;
12, 1-12.
OTHER ABBREVIATIONS USED
t-BuOOH
mCPBA
Pd(PPh3)4
N-SH2
=
=
=
=
=
tert-butylhydroperoxide
[17]
[18]
Bannwarth, W.; Trzeciak, A.A Simple and Effective Chemical
Phosphorylation Procedure for Biomolecules. Helv. Chim. Acta,
1987, 70(1), 175-186.
Meta-chloroperoxybenzoic acid
Tetrakis(triphenylphosphine)palladium(0)
N-terminal Src homology 2 domain
SH2 domain protein tyrosine phosphatase-1
Sweeney, M.C.; Wavreille, A.S.; Park, J.; Butchar, J.P.; Tridanda-
pani, S.; Pei, D. Decoding protein-protein interactions through
combinatorial chemistry: Sequence specificity of SHP-1, SHP-2,
and SHIPSH2 domains. Biochemistry, 2005, 44(45), 14932-14947.
Muller, B.; Besser, D.; Kleinwachter, P.; Arad, O.; Reissmann, S.
Synthesis of N-carboxyalkyl and N-aminoalkyl functionalized
dipeptide building units for the assembly of backbone cyclic pep-
tides. J. Pept. Res., 1999, 54(5), 383-393.
Schumann, C.; Seyfarth, L.; Greiner, G.; Reissmann, S. Synthesis
of different types of dipeptide building units containing N- or C-
terminal arginine for the assembly of backbone cyclic peptides. J.
Pept. Res., 2000, 55(6), 428-435.
Mathe, C.; Perigaud, C.; Gosselin, G.; Imbach, J.L. Phosphopeptide
prodrug bearing an S-acyl-2-thioethyl enzyme-labile phosphate
protection. J. Org. Chem., 1998, 63(23), 8547-8550.
Andrews, D.M.; Kitchin, J.; Seale, P.W. Solid-Phase Synthesis of a
Range of O-Phosphorylated Peptides by Post-Assembly Phosphity-
lation and Oxidation. Int. J. Pept. Protein Res., 1991, 38(5), 469-
475.
Klingmuller, U.; Lorenz, U.; Cantley, L.C.; Neel, B.G.; Lodish,
H.F. Specific recruitment of SH-PTP1 to the erythropoietin recep-
tor causes inactivation of JAK2 and termination of proliferative
signals. Cell, 1995, 80(5), 729-38.
Qvit, N.; Hatzubai, A.; Shalev, D.E.; Friedler, A.; Ben-Neriah, Y.;
Gilon, C. Design and Synthesis of Backbone Cyclic Phosphory-
lated Peptides: The IkB Model. Biopolymers, 2009, 91(2), 157-168.
Nomizu, M.; Otaka, A.; Burke, T.R.; Roller, P.P. Synthesis of
Phosphonomethyl-Phenylalanine and Phosphotyrosine-Containing
Cyclic-Peptides as Inhibitors of Protein-Tyrosine Kinase/Sh2 Inter-
actions. Tetrahedron, 1994, 50(9), 2691-2702.
SHP-1
[19]
[20]
TBDMS-Cl = tert-butyldimethylsilyl chloride
TFFH
=
Tetramethylfluoroformamidinium hex-
afluorophosphate
TIS
=
Triisopropyl silane
REFERENCES
[21]
[22]
[1]
[2]
[3]
[4]
Yaffe, M.B. Phosphotyrosine-binding domains in signal transduc-
tion. Nat. Rev. Mol. Cell Biol., 2002, 3(3), 177-186.
Cesarini, G.; Gimona, M.; Sudol, M.; Yaffe, M. Modular Protein
Domains; Wiley-VCH Verlag: Weinheim, 2005.
Pawson, T. Dynamic control of signaling by modular adaptor pro-
teins. Curr. Opin. Cell Biol., 2007, 19(2), 112-116.
Songyang, Z.; Shoelson, S.E.; Chaudhuri, M.; Gish, G.; Pawson,
T.; Haser, W.G.; King, F.; Roberts, T.; Ratnofsky, S.; Lechleider,
R.J.; et al. SH2 domains recognize specific phosphopeptide se-
quences. Cell, 1993, 72(5), 767-778.
[23]
[24]
[25]
[5]
Imhof, D.; Wieligmann, K.; Hampel, K.; Nothmann, D.; Zoda,
M.S.; Schmidt-Arras, D.; Zacharias, M.; Bohmer, F.D.; Reissmann,
S. Design and biological evaluation of linear and cyclic phos-
phopeptide ligands of the N-terminal SH2 domain of protein tyro-
sine phosphatase SHP-1. J. Med. Chem., 2005, 48(5), 1528-1539.
Imhof, D.; Nothmann, D.; Zoda, M.S.; Hampel, K.; Wegert, J.;
Bohmer, F.D.; Reissmann, S. Synthesis of linear and cyclic phos-
phopeptides as ligands for the N-terminal SH2-domain of protein
tyrosine phosphatase SHP-1. J. Pept. Sci., 2005, 11(7), 390-400.
Hampel, K.; Kaufhold, I.; Zacharias, M.; Bohmer, F.D.; Imhof, D.
Phosphopeptide ligands of the SHP-1 N-SH2 domain: effects on
binding and stimulation of phosphatase activity. Chem. Med.
Chem., 2006, 1(8), 869-877.
[6]
[7]
[8]
[26]
[27]
[28]
Perich, J.W.; Nguyen, D.L.; Reynolds, E.C. The Facile Synthesis of
Ala-Glu-Tyr(P)-Ser-Ala
by
Global
Di-Tert-Butyl
N,N-
Diethylphosphoramidite Phosphite-Triester Phosphorylation of a
Resin-Bound Peptide. Tetrahedron Lett., 1991, 32(32), 4033-4034.
Perich, J.W. Efficient Solid-Phase Synthesis of Mixed Thr(P)-
Containing, Ser(P)Containing and Tyr(P)-Containing Phosphopep-
tides by Global Phosphite-Triester Phosphorylation. Int. J. Pept.
Protein Res., 1992, 40(2), 134-140.
Keilhack, H.; Muller, M.; Bohmer, S.A.; Frank, C.; Weidner, K.M.;
Birchmeier, W.; Ligensa, T.; Berndt, A.; Kosmehl, H.; Gunther, B.;
Muller, T.; Birchmeier, C.; Bohmer, F.D. Negative regulation of
Perich, J.W.; Meggio, F.; Pinna, L.A. Solid phase synthesis of
pp60(src)-related phosphopeptides via 'global' phosphorylation and

816 Protein & Peptide Letters, 2010, Vol. 17, No. 7
Teichmann et al.
their use as substrates for enzymatic phosphorylation by casein
kinase-2. Bioorg. Med. Chem., 1996, 4(2), 143-150.
zation of octapeptide somatostatin analogues with backbone cycli-
zation: Comparison of different strategies, biological activities and
enzymatic stabilities. J. Prakt. Chem., 2000, 342(6), 537-545.
Perich, J.W.; Johns, R.B. An Improved One-Pot Synthesis of N-
Alpha-(Tert-Butoxycarbonyl)-O-(O',O''-Dialkylphosphoro)-L-Tyro-
sines Using Dialkyl N,N-Diethylphosphoramidites. Synthesis,
1989, 9, 701-703.
Perich, J.W.; Reynolds, E.C. The Facile One-Pot Synthesis of N-
Alpha-(9-Fluorenylmethoxycarbonyl)-O-(O',O'-Dialkylphosphoro)-
L-Tyrosines Using Dialkyl N,N-Diethylphosphoramidites. Synlett,
1991, 8, 577-578.
[29]
[30]
Perich, J.W.; Ruzzene, M.; Pinna, L.A.; Reynolds, E.C. Efficient
Fmoc Solid-Phase Peptide-Synthesis of O-Phosphotyrosyl-
Containing Peptides and Their Use as Phosphatase Substrates. Int.
J. Pept. Protein Res., 1994, 43(1), 39-46.
Hoffmann, R.; Wachs, W.O.; Berger, R.G.; Kalbitzer, H.R.;
Waidelich, D.; Bayer, E.; Wagnerredeker, W.; Zeppezauer, M.
Chemical Phosphorylation of the Peptides Ggxa (X=S, T, Y) - an
Evaluation of Different Chemical Approaches. Int. J. Pept. Protein
Res., 1995, 45(1), 26-34.
[32]
[33]
[31]
Besser, D.; Muller, B.; Kleinwachter, P.; Greiner, G.; Seyfarth, L.;
Steinmetzer, T.; Arad, O.; Reissmann, S. Synthesis and characteri-
Received: May 20, 2009
Revised: July 23, 2009
Accepted: August 26, 2009