N→O-Acyl shift in Fmoc-based synthesis of phosphopeptides

Article abstract of DOI:10.1039/b718568e

Synthetic phosphopeptides are frequently used as chemical probes to explore protein-protein interactions involved in cellular signal transduction. Most commonly, the solid-phase synthesis of phosphotyrosine-containing peptides is performed by applying the Fmoc-strategy and N-Fmoc-protected tyrosine derivatives bearing acid-labile phospho protecting groups. We observed a side-reaction, the isomerisation at threonine, which furnishes depsipeptides. It is shown that the rate of N→O-acyl migration depends on the sequence context. Depsipeptides were formed most rapidly when the phosphotyrosine was located in the +2 position. Furthermore, different phosphotyrosine building blocks were compared and a suitable method that provides phosphopeptides in enhanced purity and yield is suggested. The Royal Society of Chemistry.

Full text of DOI:10.1039/b718568e

View Article Online / Journal Homepage / Table of Contents for this issue
www.rsc.org/obc
Volume 6 | Number 8 | 21 April 2008 | Pages 1301–1512
ISSN 1477-0520
2WT\XRP[ꢀBRXT]RT
FULL PAPER
Hendrik Eberhard and Oliver Seitz
NO-Acyl shift in Fmoc-based
synthesis of phosphopeptides
In this issue...
1477-0520(2008)6:8;1-9

PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
N→O-Acyl shift in Fmoc-based synthesis of phosphopeptides†
Hendrik Eberhard and Oliver Seitz*
Received 3rd December 2007, Accepted 31st January 2008
First published as an Advance Article on the web 22nd February 2008
DOI: 10.1039/b718568e
Synthetic phosphopeptides are frequently used as chemical probes to explore protein–protein
interactions involved in cellular signal transduction. Most commonly, the solid-phase synthesis of
phosphotyrosine-containing peptides is performed by applying the Fmoc-strategy and
N-Fmoc-protected tyrosine derivatives bearing acid-labile phospho protecting groups. We observed a
side-reaction, the isomerisation at threonine, which furnishes depsipeptides. It is shown that the rate of
N→O-acyl migration depends on the sequence context. Depsipeptides were formed most rapidly when
the phosphotyrosine was located in the +2 position. Furthermore, different phosphotyrosine building
blocks were compared and a suitable method that provides phosphopeptides in enhanced purity and
yield is suggested.
reported to occur in phosphopeptide synthesis; the rearrangement
Introduction
at threonine residues to depsipeptides. We herein report our
investigations into the identification of parameters that might
affect this side-reaction and suggest an improved method that pro-
vides phosphopeptides in enhanced purity and yield. It is shown
that both sequence context and duration of acid-induced global
deprotection critically affect the extent of peptide isomerisation.
While the former can not be changed in a given target sequence
we demonstrate that the latter can be minimised through a careful
choice of phosphate protecting groups.
Phosphorylation of proteins at serine, threonine and tyrosine
hydroxyl groups is a ubiquitous intracellular event used for the
regulation of protein–protein interactions in signal transduction.
Synthetic phosphopeptides allow detailed studies of this key
mechanism of cell regulation. For example, phosphopeptides
have been used in investigations of the binding determinants,1
as protein-diagnostic probes2 and as target-specific inhibitors of
protein–protein interactions.3 Thus, methods that enable a rapid
and efficient synthesis of phosphopeptides are of high interest.
In the synthesis of phosphotyrosine-containing peptides the
reactivity of phosphoric acid phenyl esters has to be consid-
ered. Phosphotyrosine peptides are mainly prepared by two
different approaches: a) global phosphorylation, in which the
phosphate-group is introduced by a phosphorylating agent (e.g.
diarylphosphorochloridate or phosphoramidite) after coupling
of the amino acid, and b) the use of protected, phosphorylated
tyrosine building blocks in the solid phase peptide synthesis
(SPPS) known as the synthon method.4,5 The global approach
allows peptide phosphorylation without alteration of the SPPS-
method, but problems may occur owing to side reactions during
the oxidation step of PIII (e.g. oxidation of cysteine, methionine
or tryptophan residues) or the formation of H-phosphonates.6
The use of preformed phosphorylated amino acid building blocks
avoids these problems. Most commonly, the electrophilic reactivity
of phospho groups in these synthons is masked by means of
protecting groups in order to avoid pyrophosphate formation
and dephosphorylation reactions.7,8 Phosphotyrosine containing
peptides have been synthesised most successfully by applying
the Fmoc-strategy and N-Fmoc-protected tyrosine derivatives
bearing acid-labile phospho protecting groups.9–11 However, we
observed yet another problem that has previously not been
Results and discussion
In a project directed to the regulation of signal transducing kinases
we attempted the solid phase synthesis (Scheme 1) of phosphopep-
tide 1, a part of the immuno-tyrosine-activation-motif (ITAM)
of the SH2-domain of Syk-kinase.12–14 For introduction of the
phosphotyrosine, the commercially available bisdimethylamino-
protected building block was used. However, HPLC analysis of
the crude material obtained after acidolytic cleavage revealed two
products with identical mass in HPLC-MS (Fig. 1).
Identification of the by-product
High-resolution mass spectrometry and fragmentation of both
compounds exposed no difference (Fig. S1†) and racemisation
seemed unlikely because of the relatively large difference in reten-
tion times. For closer examination we carried out a glycine scan
in which each amino acid in phosphopeptide 1 was substituted by
glycine (peptides 2–5, Table 1). The HPLC showed single peaks
for peptides 3 and 5 and two peaks for peptides 1, 2 and 4 (Fig. 2).
Thus, the formation of two products required the usage of both
phosphotyrosine and threonine building blocks.
NMR-Measurements of both isolated products 4a and 4b were
performed to obtain structure information. HMQC and HMBC
measurements (Fig. S2†) enabled a complete assignment of amino
acid protons and carbons in peptides 4a and 4b (Scheme 2).
The chemical shift values of threonine protons and carbons were
most informative. The unusual low field shift of the b-threonine
carbon from 67.2 ppm to 70.2 ppm suggested the presence of
Humboldt-Universita¨t zu Berlin, Brook-Taylor-Str. 2, 12489 Berlin, Ger-
many. E-mail: oliver.seitz@hu-berlin.de; Fax: +49 2093 7266; Tel: +49 2094
7446
† Electronic supplementary information (ESI) available: High-resolution
mass spectra of 1a and 1b. HMQC- and HMBC-spectra of 4a and 4b,
HPLC-traces of 8–11 and 12a. See DOI: 10.1039/b718568e
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1349–1355 | 1349
©

Table 1 Synthesised phosphopeptides
Peptide
Sequence
1a/1b
2a/2b
3
4a/4b
5
pTyr-Glu-Thr-Leu-Gly/depsipeptide
pTyr-Glu-Thr-Gly-Gly/depsipeptide
pTyr-Glu-Gly-Leu-Gly
pTyr-Gly-Thr-Leu-Gly/depsipeptide
Gly-Glu-Thr-Leu-Gly
6a/6b
7a/7b
8
Tyr-Glu-Thr-Leu-Gly/depsipeptide
Gly-Glu-Thr-Leu-Gly/depsipeptide
pTyr-Thr-Ala-Ala-Gly
pTyr-Ala-Thr-Ala-Gly
pTyr-Ala-Ala-Thr-Gly
pTyr-Ala-Ala-Ala-Thr-Gly
Asp-Ile-pTyr-Glu-Thr-Asp-Gly/depsipeptide
9
10
11
12a/12b
Scheme 1 Solid phase peptide synthesis of phosphopeptide 1.
Fig. 2 HPLC traces of crude glycine scan peptides 2–5. All peptides
were prepared using building block Fmoc-TyrPO(NMe2)2. (Conditions:
see Fig. 1.)
Fig. 1 HPLC trace of crude peptides 1 obtained after solid phase
synthesis as described in Scheme 1. (Nucleodur-Gravity C18, 1 ml min−1
3%–50% buffer B [CH3CN, 1% water, 0.1% formic acid].)
,
an electron-withdrawing group at the threonine hydroxy group.
The b-threonine proton experienced a similar low field shift from
4.00 ppm to 4.96 ppm. In contrast the a-threonine carbon and
proton exhibited a high field shift from 58.4 ppm to 55.8 ppm
and from 4.26 ppm to 3.8 ppm, respectively. Furthermore, the
glycine carboxyl group had a higher chemical shift (d = 170.2 ppm)
in peptide 4a than in peptide 4b (d = 165.7 ppm). An O-acyl
structure in peptide 4b formed upon an acid-induced N→O-
acyl shift (Scheme 2) was considered as a plausible cause. In
this case, the chemical shift of the threonine amino group should
provide valuable information. There would be two amine protons
in depsipeptide 4b whereas integration would yield only one
threonine amide proton for peptide 4a. Indeed, the 1H-NMR
resonance of the threonine amino group in 4b integrated as
two protons. Unfortunately, HMBC spectra failed to provide
unambiguous proof as the coupling of the quaternary acyl carbon
Scheme 2 Peptide 4a and depsipeptide 4b. Numbers in italic and plain
style denote 1H- and 13C-NMR resonances, respectively.
from the N-terminal glycine with the b-threonine proton in 4b was
not detected.
To validate the assumed N→O-acyl shift reaction we syn-
thesised the depsipeptide 4b as an authentic reference by cou-
pling Boc-Thr instead of Fmoc-Thr(tBu) according to published
procedures.15 Analysis of authentic 4b by HPLC-MS (Fig. 3a) and
co-injection of the mixture of peptide 4a and 4b (Fig. 3b) obtained
after solid-phase synthesis suggested the identity. The N→O-acyl
shift reaction is reversible and O-acyl peptides have been reported
1350 | Org. Biomol. Chem., 2008, 6, 1349–1355
This journal is
The Royal Society of Chemistry 2008
©

Fig. 4 Time course of depsipeptide formation for peptides 1, 6 and 7 in
TFA.
Fig. 3 HPLC analysis of a) authentic depsipeptide 4b; b) co-injection of
authentic 4b and the mixture 4a–4b obtained after solid phase synthesis
of 4a and c) peptide 4a obtained after treatment of depsipeptide 4b with
base.
was varied while keeping the length of the peptide constant. The
analysis of the time course of depsipeptide formation (Fig. 5,
Fig. S3†) clearly showed that there is an influence of the position of
the phosphotyrosine residue on the rate of N→O-acyl migration.
The shift reaction was slowest, furnishing only 8% depsipeptide
after 6 hours of TFA treatment, when phosphotyrosine was
positioned in the direct N-terminal neighbourhood (see 8). A
remarkably fast N→O-acyl migration occurred when pY was
located in the Thr+2-position. In this case, almost 50% of
depsipeptide was obtained after 6 hours exposure to TFA. We
speculate that the protonated tyrosine phosphate acts as an
internal acid that activates acyl groups for migration. Following
this notion, it would be difficult to transfer a proton via a cyclic
intermediate from tyrosine phosphate to the tyrosine carboxyl
group which may explain the low migration rate found for 8.
to be amenable to an O→N-acyl shift when exposed to pH > 7.4.
Thus, base-induced conversion of O-acyl peptide 4b to the desired
N-acyl peptide 4a can serve as an additional means of validation.
Indeed, both the authentic depsipeptide 4b and the peptide in the
mixture could be converted to the desired peptide 4a by adding
base (Fig. 3c).
Sequence dependence of the N→O-acyl shift
The N→O-acyl shift in peptides, which leads to depsipeptides,
is a possible side reaction that has been reported to occur in
Boc-based solid-phase peptide synthesis.16 This reaction was first
reported by Bergmann et al. in c-benzamido-b-hydroxy-butanoic
acid.17 Usually, strong acids such as H2SO4, HF or HCl are
required to promote the N→O-acyl transfer.18–20 Mild acids such
as trifluoroacetic acid (TFA) used in the final deprotection in
Fmoc-based solid-phase peptide synthesis have not been regarded
as troublesome as far as N→O-acyl shift reactions at serine and/or
threonine are concerned. However, a recent report by Carpino and
co-workers demonstrated that the incorporation of D-amino acids
can increase the susceptibility to depsipeptide formation.21 An
influence of the primary structure on the N→O-acyl migration
was proposed.
We considered the feasibility of the phosphotyrosine residue
affecting the rate of the N→O-acyl shift. Addressing this issue, the
phosphotyrosine in peptide 1 was substituted by tyrosine (peptide
6a) and glycine (peptide 7a). Thereupon, peptides 1, 6a and 7a
were exposed to TFA and the conversion into depsipeptides 1b, 6b
and 7b, respectively, was monitored by HPLC-analysis. It became
apparent that the phosphotyrosine containing peptide 1 was more
susceptible to N→O-acyl migration than the glycine-containing
analogue 6a (Fig. 4). For example, only 7.9% depsipeptide was
formed when 6a was exposed to a 6 hours TFA treatment while 1
furnished 14.5% depsipeptide. This kinetic analysis revealed that
both the amino acid residue and the phospho group affect the
N→O-acyl migration rates.
Fig. 5 Time course of depsipeptide formation for peptides 8–11 in TFA.
Minimising N→O-acyl migration in solid phase synthesis
In most cases, phosphopeptide synthesis is performed in order to
rapidly and efficiently provide homogeneous material for subse-
quent biological studies. We sought for a method that minimises
the problem of N→O-acyl migration, thereby avoiding the need
to reverse potential N→O-acyl shift reactions by a precautionary
exposure to basic conditions. We supposed that refinement of
the TFA treatment required to remove the phospho protecting
groups may allow the desired improvements of product purity.
The initial synthesis of phosphopeptide 1a was performed by cou-
pling the bisdimethylamino-masked phosphotyrosine 13 (Fig. 6).
For hydrolysis of the phosphoramidate prolonged treatments in
Next, the distance dependence of the N→O-acyl shift was
examined. Four model peptides (8–11) comprising alanine, threo-
nine and phosphotyrosine were synthesised. In these peptides the
number of alanine units between phosphotyrosine and threonine
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1349–1355 | 1351
©

cresol).9,25 The HPLC analysis of the crude product revealed an
unacceptably high extent (40%) of depsipeptide formation (entry 3,
Table 2). Again, shortening of the deprotection time resulted in the
expected decrease of the depsipeptide content (compare Fig. 7b),
however, at the cost of incomplete protecting group removal
which lead to a rather low content (77%) of phosphopeptide 1
(entry 4, Table 2). It was concluded that the N→O-acyl migration
proceeded equally fast regardless of the water content of the
cleavage solution. Thus, an augmentation of the acid lability of
the phospho protecting groups was considered necessary. The
commercially available monobenzyl-protected phosphotyrosine
15 allows the application of milder cleavage conditions.26–28 Indeed,
a 90 minute treatment with a mixture of TFA–TIS–EDT–H2O (90 :
2.5 : 2.5 : 5, EDT is 1,2-ethanedithiol) was sufficient to accomplish
quantitative deprotection and to deliver phosphopeptide 1a in
high 98% purity of the crude material (Fig. 7c, entry 5 Table 2).
The use of the monobenzyl-protected phosphotyrosine building
block 15 may give rise to reasons for concern. It is conceivable
that coupling reactions that are performed after incorporation
of the monoprotected phosphotyrosine building block may be
impeded by concomitant reactions at the remaining, poten-
tially nucleophilic phospho oxygen. The solid phase synthesis
of phosphopeptide 12a, which is part of the human insulin
receptor,29,30 addressed this issue. The HPLC-analysis of crude
materials confirmed the rather low content of desired 12a (<65%)
in crude materials that was obtained when bisdimethylamino-
and dimethyl-protected phosphotyrosine derivatives 13 and 14,
respectively, were incorporated (Fig. 8a–c, entries 6–8 in Table 2).
The usage of Fmoc-Tyr(PO(OBzl)OH) 15 furnished, again, the
highest 98% purity of crude products (Fig. 8d, entry 9 in
Table 2). HPLC-MS analysis showed minor peaks corresponding
to truncation products (Fig. S4†). However, the application of
double couplings (2 × 6 eq.) completely solved this problem and
provided the fully deprotected phosphopeptide 12a in high purity
and an isolated overall yield of 52%.
Fig. 6 Protected phosphotyrosine building blocks for Fmoc-based solid
phase peptide synthesis.
aqueous TFA solution are required. It has been recommended to
firstly detach the phosphotyrosine containing peptide from the
resin and to secondly add 10% of H2O to the TFA–TIS–m-cresol–
H2O (85 : 5 : 5 : 5, TIS is triisopropylsilane) filtrate and extend
the reaction time to 18 hours.22–24 We first explored whether the
reaction time can be shortened. However, the formation of the
depsipeptide 1b was detected already after 90 minutes (Fig. 7a).
After this short reaction time deprotection of the phospho group
was not complete and, as a result, the crude product contained
only 64% of the desired phosphopeptide 1a (entry 2, Table 2).
As an alternative, the use of the dimethyl-protected building
block 14 was examined. The removal of the methyl protecting
groups can be accomplished via an SN2-like mechanism under non-
aqueous conditions. The reported procedure for deprotection of
the dimethylphosphotyrosine moiety involved a 5 hour treatment
with 1 M trimethylsilylbromide–thioanisole in TFA (0.05% m-
Conclusions
In summary, we have shown that N→O-acyl migration can occur
at threonine residues during the Fmoc-based solid-phase synthesis
of phosphotyrosine-containing peptides. The phosphotyrosine
residues were introduced by means of N-Fmoc-protected tyro-
sine derivatives bearing acid-labile phospho protecting groups
typically used in phosphopeptide synthesis. HPLC-MS-Data,
Fig. 7 HPLC-Analysis of crude product 1 obtained after 90 min
deprotection by using a) building block 13 and method A; b) building
block 14 and method B and c) building block 15 and method C. The peaks
marked with an asterisk denote protected products.
Table 2 Purity and yield of synthesised phosphopeptides
Building block
Method
Crude product
Isolated yield
1
2
3
4
5
6
7
8
9
13
13
14
14
15
13
13
14
15
A, 18 h
A, 1.5 h
B, 5 h
B, 1.5 h
C, 1.5 h
A, 18 h
A, 1.5 h
B, 1.5 h
C, 1.5 h
69% 1a, 31% 1b
64% 1a, 3% 1b, 33% protected
60% 1a, 40% 1b
77% 1a, 10% 1b, 13% protected
98% 1a, 2% 1b
59% 12a, 30% 12b
65% 12a, 3% 12b, 33% protected
65% 12a, 24% 12b, 11% protected
98% 12a, 2% 12b
n.d.
n.d.
10%
n.d.
29%
12%
n.d.
n.d.
52%
A: i) TFA–TIS–m-cresol–H2O (85 : 5 : 5 : 5), 60 min; ii) addition of 10% H2O; B: 1 M trimethylsilylbromide–thioanisole in TFA (0.05% m-cresol); C:
TFA–TIS–EDT–H2O (90 : 2.5 : 2.5 : 5). n.d., not determined.
1352 | Org. Biomol. Chem., 2008, 6, 1349–1355
This journal is
The Royal Society of Chemistry 2008
©

product was isolated by ether precipitation. Analytical HPLC-MS
was performed on an Agilent 1100 HPLC-MS system equipped
with a UV–Vis-detector and a VL-quadrupole mass spectrometer
using a thermostated (55 ◦C) analytical CC 125/4 Nucleodur-C18
gravity, 3 l column (Macherey-Nagel) and detection wavelength
k = 210 nm. Eluents A (H2O : MeCN : HCOOH = 98.9 : 1 : 0.1
[v/v/v]) and B (MeCN : H2O : HCOOH = 98.9 : 1 : 0.1 [v/v/v])
were used in a linear gradient (gradient 1: 0–20 min, 3–50% B
in A or gradient 2: 0–20 min, 3–20% B in A) at a flow rate of
1 mL min−1. 1H- and 13C-NMR spectra were recorded on a Bruker
spectrometer at 300 MHz and 75 MHz, respectively.
General procedure for the synthesis of the phosphopeptides
Preloaded resin was suspended in DMF for 1 hour and treated
with 25% piperidine–DMF (1 ml) for 2 min and washed with
25% piperidine–DMF, DMF, dichloromethane and DMF again.
The automated synthesis was commenced with a deprotection
step, which included treatment with 1 ml DMF–piperidine (4 : 1
[v/v]) for 2 min and subsequent washing with DMF. For coupling
6 eq. amino acid (0.2 M) in DMF were preactivated using 12 eq.
N-methylmorpholine and 6 eq. 1-[bis(dimethylamino)methylene]-
5-chloro-1H-benzotriazolium-3-oxide tetrafluoroborate (TCTU).
This solution was added to the resin. Double couplings were
performed after the incorporation of the fourth amino acid. The
resin was capped with 1 ml of a solution of DMF–2,6-lutidine–
acetic anhydride (90 : 5 : 5 [v/v/v]). The terminal capping was
performed twice. Prior to final cleavage the resin was washed
ten times with dichloromethane.
Cleavage method A: in the case of building block 13 the resin
was shaken for 90 min with 1 ml of a solution that contained
850 ll TFA, 50 ll triisopropylsilane, 50 ll m-cresol, 50 ll water
and 5 mg cysteine methyl ester. The resin was washed with 200 ll
TFA. Then, 120 ll water were added to the combined filtrates.
After 18 h shaking the solution was concentrated to 1/5 of its
volume. The crude product precipitated upon addition of cold
diethylether. The pellet obtained after centrifugation and disposal
of the supernatant was washed with cold diethylether and collected
by centrifigation.
Fig. 8 HPLC-Analysis of crude peptide 12 obtained by using a) building
block 13 and method A, 18 h; b) building block 13 and method A, 90 min;
c) building block 14 and method B and d) building block 15 and method
C. The peaks marked with an asterisk and with a plus denote protected
products and aspartamide by-products, respectively.
NMR-spectroscopy and the reversibility of the reaction at basic
conditions have provided ample evidence for this acid-induced
side reaction which results in the formation of depsipeptides. This
side reaction occurred during the acidolytic cleavage required
to remove the dimethylamino- or methyl protecting groups of
the tyrosine phosphate group. Furthermore, the analysis of
initial kinetics suggested an influence of the phosphotyrosine
residue. The N→O-acyl shift proceeded most rapidly when the
phosphotyrosine was located in the +2 position. In contrast, rather
slow N→O-acyl migration was observed for peptides in which
the phosphotyrosine was the N-terminal neighbour of threonine.
The duration of the acid-induced global deprotection was found
to most affect the extent of peptide isomerisation. It is demon-
strated that the use of the more acid-labile monobenzyl-protected
phosphotyrosine building block Fmoc-Tyr(PO(OBzl)OH) allows
improvements of both purity and yield of phosphopeptides. We
wish to note that the formation of depsipeptides also occurs
at serine residues (data not shown). The findings are expected
to be of interest for those who are involved in the synthesis of
depsipeptides (e.g. as precursors to switch peptides)31–33 as well as
for those who seek methods that enable rapid and efficient access to
homogeneous phosphopeptides for subsequent biological studies.
Cleavage method B: when building block 14 was used the
resin was shaken for 5 h with 1 ml of a 1 M solution of
trimethylsilylbromide–thioanisol and 50 ll m-cresol in TFA at
5 ◦C. The crude product was concentrated, precipitated with cold
ether and collected by centrifugation.
Cleavage method C: in the case of building block 15 the resin
was shaken for 90 min with a mixture of 950 ll TFA, 25 ll water,
12.5 ll 1,2-ethanedithiol and 12.5 ll triisopropylsilane. Further
work-up was performed as described for method A.
Experimental
General
All organic starting materials were purchased in analytically pure
grade and used without further purification. The amino acids
were purchased from SennChemicals. Building block 13 was
synthesised using methods described in the literature.20 Building
blocks 14 and 15 were purchased from Bachem and Novabiochem,
respectively. HPLC-Grade acetonitrile was purchased from Acros,
DMF from Biosolve. The solid phase peptide synthesis was
performed using a Respep Synthesizer from Intavis Bioanalytical
Instruments AG. Cleavage of the peptide resins was performed
in 2 ml PET-syringes from MultsynTech/Witten, which were
equipped with Teflon filters (pore size 50 lm). The combined cleav-
age filtrates were concentrated by rotary evaporation and the crude
Synthesis of Ac-pTyr-Glu-Thr-Leu-Gly-OH (1)
Method A: Fmoc-Gly-Wang-resin (12.5 mg, 10 lmol) and Fmoc-
Tyr(PO(NMe2)2) 13 were used. Analytical HPLC-MS (gradient 1)
exposed two major products at rt = 5.1 min (30.8% of integral area
at k = 210 nm) and rt = 8.2 min (69.2%). [M + H]+: m/z 704.2 for
peptide 1b and 1a (C28H42N5O14P1: 703.25 g mol−1).
Method B: Fmoc-Gly-Wang-resin (2 lmol) and building block
14 were used. Analytical HPLC-MS (gradient 1) exposed two
major products at rt = 5.1 min (40.3%) and rt = 8.2 min (59.7%)
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1349–1355 | 1353
©

([M + H]+: m/z 704.2). Purification by prep. HPLC resulted in
10.4% overall yield of peptide 1a determined with an extinction
coefficient for phosphotyrosine at k = 260 nm of 652 M−1 cm−1.34
Method C: Fmoc-Gly-Wang-resin (2 lmol) and building block
15 were used. Analytical HPLC-MS (gradient 1) revealed 97.9%
peptide 1a and 2.1% 1b. Purification by prep. HPLC furnished
peptide 1a in 29.1% overall yield.
13C-NMR (DMSO-d6), 75 MHz, d 16.8 (1 C, CH3-Thr); 22.0,
22.9 (2 C, 2 × CH3-Leu); 23.6 (1 C, CH-Leu); 24.5 (1 C, CH3-
acetyl); 37.0 (1 C, CH2-pTyr); 40.1 (1 C, CH2-Gly); 51.3 (1 C,
CH-Leu); 54.8 (1 C, CH-pTyr); 55.8 (1 C, CH-Thr); 70.2 (1 C,
CH-O-Thr); 120.1 (2 C, CH-pTyr-arom.); 130.5 (2 C, CH-pTyr-
arom.); 134.1 (1 C, Cq-pTyr); 150.4 (1 C, Cq-O-pTyr); 165.7 (1 C,
=
=
=
Gly-C O); 170.0 (1 C, Thr-C O); 170.2 (1 C, acetyl-C O); 171.5
=
=
=
(1 C, Leu-C O); 172.3 (1 C, pTyr-C O); 172.8 (1 C, Glyterm.-C O)
ppm.
Synthesis of peptides 2–7
Synthesis of depsipeptide Ac-pTyr-Gly-Thr-Leu-Gly-NH2 (4b)
Fmoc-Gly-TGR-resin (42 mg,
5
lmol) and Fmoc-
Tyr(PO(NMe2)2) 13 were used. HPLC-MS analysis (gradient 1)
exposed two peaks for peptide 2 (1.9 min and 3.0 min, [M +
H]+: m/z 647.2), one peak for peptide 3 (6.8 min, [M + H]+:
m/z 659.2), two peaks for peptide 4 (4.4 min and 7.2 min, [M +
H]+: m/z 631.2), one peak for peptide 5 (5.9 min, [M + H]+: m/z
517.3), two peaks for peptide 6 (gradient 2, 5.6 min and 8.1 min,
[M + H]+: m/z 623.3) and two peaks for peptide 7 (2.9 min and
5.9 min, [M + H]+: m/z 517.3).
Fmoc-Gly-TGR-resin (17 mg, 2 lmol) and Fmoc-Tyr(PO3H2)
were used. The first amino acids were coupled according to
the general protocol. After coupling of Boc-threonine the resin
was washed three times with DMF and dichloromethane. Subse-
quently, the resin was submitted to a double coupling with 2 mg
Fmoc-glycine (297.3 g mol−1, 3 eq.) in 50 ll dichloromethane and
DMF with 0.3 eq. DMAP and 1.1 ll diisopropylcarbodiimide
(3 eq.). The synthesis was continued as described in the general
method. HPLC-MS analysis (gradient 1) showed one major
product at rt = 4.7 min ([M + H]+: m/z 631.2) for depsipeptide 4b
(C25H39N6O11P1: 630.24 g mol−1).
Synthesis of Ac-pTyr-Gly-Thr-Leu-Gly-OH (4a/4b)
Fmoc-Gly-Wang-resin (230 mg, 150 lmol) and Fmoc-
Tyr(PO(NMe2)2) 13 were used. HPLC-MS analysis (gradient 2)
exposed two major peaks at rt = 7.5 min and rt = 13.4 min ([M +
H]+: m/z 704.2) for peptide 4a and 4b (C28H42N5O14P1: 703.25 g
mol−1). The crude product was purified by prep. HPLC to yield
16.1 mg peptide 4a (15.3%) and 35 mg depsipeptide 4b (33.2%)
after lyophylisation.
Synthesis of peptides 8–11
Fmoc-Gly-Wang-resin (3 mg, 2 lmol) was used. HPLC-MS
analysis (gradient 1) showed two major peaks for peptide 8 (rt =
3.8 min and 5.6 min, [M + H]+: m/z 604.2), for peptide 9 (rt =
7.4 min and 8.1 min, [M + H]+: m/z 604.2), for peptide 10 (rt =
3.1 min and 5.8 min, [M + H]+: m/z 604.2) and for peptide 11 (rt =
6.4 min and 8.8 min, [M + H]+: m/z 675.2).
1H-NMR for 4a (DMSO-d6), 300 MHz, d 0.83–0.89 (dd, J =
6.5 Hz, J = 12.4 Hz, 6 H, 2 × CH3-Leu); 1.03 (d, J = 6.3 Hz,
3 H, CH3-Thr); 1.50 (t, J = 7.2 Hz, 3 H, CH2-Leu); 1.63 (m, 1
H, CH-Leu); 1.78 (s, 3 H, Ac-CH3); 2.85 (ddd, J = 7.2 Hz, J =
14.0 Hz, J = 24.0 Hz, 2 H, CH2-pTyr); 3.65–3.86 (m, 4 H, 2 ×
CH2-Gly); 3.97–4.04 (m, 1 H, CH-OH-Thr); 4.26 (dd, J = 4.1 Hz,
J = 8.3 Hz, 1 H, CH-Thr); 4.31–4.37 (m, 1 H, CH-Leu); 4.39–
4.48 (m, 1 H, CH-pTyr); 7.13 (dd, J = 8.1 Hz, J = 50.3 Hz, 4 H,
CHarom.-pTyr); 7.71 (d, J = 8.3 Hz, 1 H, NH-Thr); 7.89 (d, J =
8.4 Hz, 1 H, NH-Leu); 8.17–8.23 (m, 2 H, NH-pTyr, NH-Gly);
8.39 (t, J = 5.6 Hz, 1 H, NH-Gly) ppm.
Synthesis of peptide Asp-Ile-pTyr-Glu-Thr-Asp-Gly 12
Fmoc-Gly-Wang-resin (3 mg, 2 lmol) was used.
Method A: analytical HPLC-MS (gradient 1) showed two major
products at rt = 5.9 min (30.0%) and rt = 8.2 min (58.5%) ([M +
H]+: m/z 934.3) for peptides 12b and 12a (C36H52N7O20P1: 933.30 g
mol−1), respectively, and minor peaks for aspartamide by-products
(11.5% at rt = 6.4 min, 8.1 min and 8.9 min, [M + H]+: m/z
916.2). After 90 min deprotection 64.8% 12a, 2.6% 12b and 32.6%
protected peptide (9.8 min and 12.0 min, [M + H]+: m/z 988.3)
were found. Purification by prep. HPLC furnished 11.9% overall
yield.
Method B: HPLC-MS analysis of crudes obtained after 90 min
deprotection time revealed products (gradient 1) at rt = 5.9 min
(23.8%, 12b), rt = 7.2 min (65.2%, 12a) ([M + H]+: m/z 934.3) and
11.0% methyl-protected peptide (rt = 7.3 min and 8.4 min [M +
H]+: m/z 948.3).
13C-NMR: (DMSO-d6), 75 MHz, d 19.9 (1 C, CH3-Thr); 22.0,
22.9 (2 C, 2 × CH3-Leu); 23.6 (1 C, CH-Leu); 24.5 (1 C, CH3-
acetyl); 37.0 (1 C, CH2-pTyr); 42.7 (1 C, CH2-Gly); 51.3 (1 C,
CH-Leu); 54.8 (1 C, CH-pTyr); 58.4 (1 C, CH-Thr); 67.2 (1 C,
CH-OH-Thr); 120.1 (2 C, CH-pTyr-arom.); 130.5 (2 C, CH-pTyr-
arom.); 134.1 (1 C, Cq-pTyr); 150.4 (1 C, Cq-O-pTyr); 169.4 (1 C,
=
=
=
Gly-C O); 170.0 (1 C, Thr-C O); 170.2 (1 C, acetyl-C O); 171.5
=
=
=
(1 C, Leu-C O); 172.3 (1 C, pY-C O); 172.8 (1 C, Glyterm.-C O)
ppm.
Method C: HPLC-MS analysis of crudes obtained after 90 min
deprotection time showed one major peak for 12a (98%) and one
minor peak for 12b (2.0%). Purification by prep. HPLC furnished
peptide 12a in 52.2% overall yield.
1H-NMR for 4b (DMSO-d6), 300 MHz, d 0.83–0.92 (m, 6 H, 2 ×
CH3-Leu); 1.25 (d, J = 6.4 Hz, 3 H, CH3-Thr); 1.50 (t, J = 7.2 Hz,
3 H, CH2-Leu); 1.63 (m, 1 H, CH-Leu); 1.78 (s, 3 H, Ac-CH3);
2.85 (ddd, J = 7.2 Hz, J = 14.0 Hz, J = 24.0 Hz, 2 H, CH2-pTyr);
3.68–3.81 (m, 2 H, CH2-Gly); 3.82–4.28 (m, 3 H, CH2-Gly, CH-
Thr); 4.34–4.63 (m, 2 H, CH-Leu, CH-pTyr); 4.96 (p, J = 6.3 Hz,
References
1 H, CH-O-Thr); 7.13 (dd, J = 8.1 Hz, J = 50.3 Hz, 4 H, CHarom.
-
1 (a) U. Reimer, U. Reineke and J. Schneider-Mergener, Curr. Opin.
Biotechnol., 2002, 13, 315–320; (b) R. Frank, Tetrahedron, 1992, 48,
9217–9232.
pTyr); 8.19–8.23 (m, 2 H, NH-Gly, NH-Leu); 8.37–8.49 (m, 3 H,
NH-pTyr, NH2-Thr); 8.82 (d, J = 8.1 Hz, 1 H, NH-Gly) ppm.
1354 | Org. Biomol. Chem., 2008, 6, 1349–1355
This journal is
The Royal Society of Chemistry 2008
©

2 D. M. Rothmann, M. D. Shults and B. Imperiali, Trends Cell Biol.,
2005, 15, 502–510.
3 P. L. Toogood, J. Med. Chem., 2002, 45, 1453–1558.
4 G. K. To´th, Z. Kele and G. Va´radi, Curr. Org. Chem., 2007, 11, 409–426.
5 J. S. McMurray, D. R. Coleman, W. Wang and M. L. Campbell,
Biopolymers, 2001, 3, 3–31.
6 E. A. Ottinger, L. L. Shekels, D. A. Bernlohr and G. Barany,
Biochemistry, 1993, 32, 4354–4361.
7 C. Garcia-Echeverria, Lett. Pept. Sci., 1995, 2, 93–98.
8 B. W. Gibson, A. M. Falick, A. L. Burlingame, L. Nadasdi, A. C.
Nguyen and G. L. Kenyon, J. Am. Chem. Soc., 1987, 109, 5343–5348.
9 E. A. Kitas, J. W. Perich, J. D. Wade, R. B. Johns and G. W. Tregear,
Tetrahedron Lett., 1989, 30, 6229–6232.
18 D. F. Elliott, Biochem. J., 1952, 50, 542–550.
19 J. Lenard and G. P. Hess, J. Biol. Chem., 1964, 239, 3275–3281.
20 D. Levy and F. H. Carpenter, Biochemistry, 1970, 9, 3215–3222.
21 L. A. Carpino, E. Krause, C. D. Sferdean, M. Bienert and M.
Beyermann, Tetrahedron Lett., 2005, 46, 1361–1364.
22 H.-G. Chao, B. Leitling, P. D. Reiss, A. L. Burkhardt, C. E. Klimas,
J. B. Bolen and G. R. Matsueda, J. Org. Chem., 1995, 60, 7710–
7711.
23 M. Ueki, J. Tachibana, Y. Ishii, J. Okumura and M. Goto, Tetrahedron
Lett., 1996, 37, 4953–4956.
24 D. R. Coleman, Z. Ren, P. K. Mandal, A. G. Cameron, G. A. Dyer, S.
Muranjan, M. Campbell, X. Chen and J. S. McMurray, J. Med. Chem.,
2005, 48, 6661–6670.
10 Combined use of Bpoc-protection: T. J. Attard, E. C. Reynolds and
J. W. Perich, Org. Biomol. Chem., 2007, 5, 664–670.
11 Use in the synthesis of PNA–phosphopeptide chimera: L. Ro¨glin,
M. R. Ahmadian and O. Seitz, Angew. Chem., Int. Ed., 2007, 46, 2704–
2707.
25 H. Fretz, Lett. Pept. Sci., 1996, 3, 343–348.
26 E. A. Kitas, J. D. Wade, R. B. Johns, J. W. Perich and G. W. Tregear,
J. Chem. Soc., Chem. Commun., 1991, 338–339.
27 J. W. Perich, N. J. Ede, S. Eagle and A. M. Bray, Lett. Pept. Sci., 1999,
6, 91–97.
12 R. Ruijtenbeek, J. A. W. Kruijtzer, W. van de Wiel, M. J. E. Fischer,
M. Flu¨ck, F. A. M. Redegeld, R. M. J. Liskamp and F. P. Nijkamp,
ChemBioChem, 2001, 2, 171–179.
13 F. J. Dekker, N. J. de Mol, J. van Ameijde, M. J. E. Fischer, R.
Ruijtenbeek, F. A. M. Redegeld and R. M. J. Liskamp, ChemBioChem,
2002, 2, 238–242.
14 F. J. Dekker, N. J. de Mol, M. J. E. Fischer and R. M. J. Liskamp,
Bioorg. Med. Chem. Lett., 2003, 13, 1241–1244.
15 Y. Sohma, Y. Hayashi, M. Kimura, Y. Chiyomori, A. Taniguchi, M.
Sasaki, T. Kimura and Y. Kiso, J. Pept. Sci., 2005, 11, 441–451.
16 M. Bodanszky and J. Martinez, Synthesis, 1981, 5, 333–356.
17 M. Bergmann, E. Brand and F. Weinmann, Hoppe-Seyler’s Z. Physiol.
Chem., 1923, 131, 1–17.
28 F. J. Dekker, N. J. de Mol, P. Bultnick, J. Kemmink, H. W. Hilbers and
R. M. J. Liskamp, Bioorg. Med. Chem., 2003, 11, 941–949.
29 J. King and G. J. Sale, FEBS Lett., 1988, 237, 137–140.
30 W. Bannwarth and E. A. Kitas, Helv. Chim. Acta, 1992, 75, 707–714.
31 L. A. Carpino, E. Krause, C. D. Sferdean, M. Schu¨mann, H. Fabian,
M. Bienert and M. Beyermann, Tetrahedron Lett., 2004, 45, 7519–7523.
32 S. Dos Santos, A. Chandravarkar, B. Mandal, R. Mimna, K. Murat, L.
Sauce`de, P. Tella, G. Tuchscherer and M. Mutter, J. Am. Chem. Soc.,
2005, 127, 11888–11889.
33 T. Yoshiya, Y. Sohma, T. Kimura, Y. Hayashia and Y. Kiso, Tetrahedron
Lett., 2006, 47, 7905–7909.
34 R. A. Grucza, K. Fu¨tterer, A. C. Chan and G. Waksman, Biochemistry,
1999, 38, 5024–5033.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1349–1355 | 1355
©