A phosphoarginine containing peptide as an artificial SH2 ligand

Article abstract of DOI:10.1039/c1cc13341a

We have developed a synthesis of phosphoarginine containing peptides using a bis(2,2,2-trichloroethyl) protected phosphoarginine derivative as building block. Binding studies and computer modelling demonstrate the ability of the SH2 domain from Src kinase to recognize a phosphoarginine-containing peptide in a phosphoryl group-dependent manner.

Full text of DOI:10.1039/c1cc13341a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 10335–10337
www.rsc.org/chemcomm
COMMUNICATION
A phosphoarginine containing peptide as an artificial SH2 ligandw
Frank T. Hofmann,a Claudia Lindemann,a Helen Salia,a Philipp Adamitzki,a
John Karanicolasb and Florian P. Seebeck*ac
Received 5th June 2011, Accepted 3rd August 2011
DOI: 10.1039/c1cc13341a
We have developed a synthesis of phosphoarginine containing
peptides using a bis(2,2,2-trichloroethyl) protected phosphoarginine
derivative as building block. Binding studies and computer
modelling demonstrate the ability of the SH2 domain from Src
kinase to recognize a phosphoarginine-containing peptide in a
phosphoryl group-dependent manner.
peptide synthesis (SPPS). For this building block approach
the choice of protecting groups for the phosphoguanidinium
moiety is crucial. Most widely used phosphate protecting
groups require acidic conditions for their removal which
would also cleave acid labile N–P bonds. Alternatively,
trichlorethyl (Tc) protecting groups may be cleaved by
hydrogenolysis at slightly alkaline pH after global deprotection
of the peptide by acidic cleavage cocktails (Scheme 1).16–19
If so, Tc protecting groups may prove suitable for the
preparation of N-phosphorylated peptides.
Reversible phosphorylation of protein side chains is a central
mechanism in signal transduction and is therefore of key
interest to further the understanding of cell signalling and
the development of therapeutic interventions. Consequently,
much attention has been focused on the complex network of
kinases and phosphatases that control the phosphorylation
state of thousands of serine, threonine and tyrosine side
chains. In addition, phosphorylation of histidines, arginines
and possibly lysines1–3 may also contribute to phosphoproteomes,
but their role in signal transduction is far less well understood.
Recent accounts on phosphohistidine (pHis) containing proteins
in mammalian cells,2,4 the development of stable pHis analogs
and pHis-specific antibodies5,6 promise exciting insight into
the physiological role of this posttranslational modification
(PTM). Meanwhile, the discovery of a bacterial arginine
kinase which phosphorylates a heat-shock regulator provided
the first example of a signalling mechanism based on
phosphoarginine (pArg).7 This well characterized system
complements previous reports on protein arginine kinase
activity in several mammalian tissues and a number of pArg
containing proteins.8–12 However, more systematic
prospecting of cellular pArg sites will be necessary to gauge
their physiological role.
To test this idea, we prepared bis(2,2,2-trichloroethyl) and
Fmoc protected pArg (Scheme 1, 1) in a four-step synthesis from
Z-Arg-OH with an overall yield of 37% (see ESIw) and prepared
three phosphopeptides using Fmoc-SPPS.20 Peptides 2 and 3
(Scheme 2) represent the acetylated N- and C-termini of histone 3
which have been identified as substrates for arginine kinase
activity in mammalian cells.8 Peptide 4a was designed as a
potential ligand for the phosphotyrosine (pTyr) binding SH2
domain from Src kinase as detailed below. After peptide
assembly, cleavage from the solid support with trifluoroacetic
acid (TFA) yielded essentially pure bis(2,2,2-trichloroethyl)
protected phosphoarginyl peptides, confirming that 1 is well suited
for routine SPPS (Fig. S1, S4 and S7, ESIw).20 Hydrogenolytic
Synthetic access to specifically modified peptides has proven
to be an important tool for the study of PTMs but current
methods for arginine phosphorylation are either unspecific13
or designed to prepare pArg as the free amino acid.14,15 To
overcome this limitation we here present the synthesis of
peptides containing specific pArg residues by solid phase
a Department of Physical Biochemistry, Max Planck Institute of
Molecular Physiology, Otto-Hahn Strasse 11, Dortmund, Germany
b Center for Bioinformatics and Department of Molecular Biosciences,
The University of Kansas, 2030 Becker Drive, Lawrence, USA
c Department of Chemistry, University of Basel, St. Johanns-Ring 19,
Basel, Switzerland. E-mail: florian.seebeck@unibas.ch
w Electronic supplementary information (ESI) available: Fig. S1–S14,
Schemes S1–S4, detailed experimental procedures and characterization of
1 and synthetic intermediates thereof. See DOI: 10.1039/c1cc13341a
Scheme 1 (i) BnBr, NMP, 5 h; (ii) bis(2,2,2-trichloroethyl)phosphor-
yloxychloride, NEt3, CH3CN, 7 h; (iii) H2, Pd/C, AcOH/TFA; (iv)
Fmoc-OSu, NEt3, CH3CN, 1 h; (v) Fmoc-SPPS, 92.5% TFA; sca-
vengers; (vi) H2, Pd/C, pH 9.2.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10335–10337 10335

View Article Online
Table 1 Binding affinity (Kd) of SH2 complexes with compounds 4- 7
determined by a fluorescence polarization based displacement assaya
or by direct fluorescence polarizationb
Compound
Kd/mM
4aa
4ba
4ca
6ab
6bb
7a
550 Æ 150
0.13 Æ 0.094
>10 000
0.052 Æ 0.012
2.9 Æ 0.31
270 Æ 35
a
b
Fluorescence polarization based displacement assay. Fluorescence
polarization.
(4c) in place of pTyr (4b). These measurements were per-
formed using a fluorescence polarization based displacement
assay24–26 in which a complex between the SH2 domain and a
fluorescein-labelled sensor peptide (6a) is titrated with the
unlabeled analytes 4a–c. We demonstrated that this assay is
suitable for detecting low affinity interactions by measuring a
Kd of 270 Æ 35 mM (Fig. S11, ESIw) for the SH2 complex with
phenyl phosphate (7) which is consistent with previous
measurements using scintillation proximity binding.27 We then
determined the dissociation constants for the SH2 complexes
with peptides 4b (Kd = 130.0 Æ 9.4 nM) and 4a (Kd = 550 Æ
150 mM) but we were unable to detect any interaction with 4c
below 10 mM (Table 1, Fig. S11–S12, ESIw). These data
clearly establish phosphoryl-group dependent recognition of
4a by the SH2 domain. However, exchanging pTyr with pArg
Scheme 2 pArg containing peptides an artificial SH2 ligands. 2 and 3
are phosphorylated derivatives of the N- and C- terminus of histone 3.
4 - 7 are potential SH2 ligands.
cleavage of the Tc groups in ammonia buffered aqueous ethanol
at pH 9.2 released 69–75% 2, 3, and 4a while causing less than
2% dephosphorylation (Fig. S2, S5 and S8, ESIw). Minor side
products were removed by preparative HPLC using an
acetonitrile gradient in alkaline aqueous buffer as a mobile phase
(Fig. S3, S6 and S9, ESIw). We also enquired whether phospho-
peptides generated by this methodology are stereochemically
homogeneous. As an example, peptide 4a was dephosphorylated
in 50 mM HCl and compared with the two authentic
diastereomers 4c and 5 (Scheme 2). HPLC analysis of the three
peptides showed that more than 99% of dephosphorylated 4a
co-eluted with 4c suggesting that racemization is insignificant
throughout the synthesis of 1 and that this building block is not
prone to epimerization during Fmoc-SPPS (Fig. S10, ESIw).
As a first application of synthetic pArg containing peptides
we studied a model protein–protein interaction mediated by
this PTM. Modulation of protein complex stabilities is a major
mechanism by which phosphorylated residues control protein
activity as exemplified by SH2 domains, which recognize pTyr
within short peptide motifs.21 The human genome codes for
more than one hundred SH2 domains which direct the cellular
localization of a vast array of proteins by phosphate-
dependent interactions. Because analogous pArg receptors are
still unknown, we were interested in whether an SH2 domain
could also recognize pArg in place of pTyr, and could ultimately
serve as a blue print for designing phosphoarginine receptors.
The SH2 domain from Src kinase binds a phosphorylated
peptide motif (pTyr-Glu-Glu-Ile) with a Kd of 180 nM while
discriminating against the unphosphorylated version by
104-fold.22 Peptides with phosphoserine (pSer)22 or pHis23 in place
of pTyr are also poor ligands, demonstrating that the SH2
domain is highly sensitive to the length and geometry of the
phosphate-bearing side chain. The side chain of pArg is longer
than that of pSer and more flexible than that of pHis, and
might therefore be a better match for the pTyr binding pocket.
To test this possibility we compared the affinities of the SH2
domain for peptides containing either pArg (4a) or arginine
results in a
4
Â
103-fold lower SH2 affinity for the
phosphopeptide.
To characterize the structural basis for this behaviour we built
a model of the SH2 domain in a complex with a set of pArg
rotamers based on commonly occurring Arg rotamers in the
protein data bank (PDB).28 We used the Rosetta software suite29
to screen each of these rotamers by replacing the pTyr side chain
in the crystal structure of the SH2 domain bound to pYEEI
(PDB access code 1SHD).30 Consistent with the observed
recognition of peptide 4a by the SH2 domain we found several
pArg rotamers to project the phosphoryl group to a similar
position and orientation as that of pTyr, allowing it to form
equivalent interactions with ArgbB5 and ArgaA2 of the SH2
domain (Fig. 1, numbering according to ref. 22). The two salt
bridges between these arginine side chains and the dianionic
phosphate group on pTyr are crucial for high-affinity recognition
by the src SH2 domain.22 It is possible that the interaction with
the monoanionic phosphoguanidinium side chain of pArg
(Scheme 2) is weak predominantly because of a missing salt
bridge. To quantify the effect of this missing charge we compared
the affinities of the SH2 domain for peptide 6a and the
corresponding sulfotyrosine containing peptide 6b (Scheme 2).
The two compounds also differ by one negative charge but are
otherwise strictly isosteric.31 In a fluorescence polarization assay
the sulfopeptide proved a 55-fold weaker SH2 ligand (Table 1,
Fig. S13, ESIw) than 6a, suggesting that charge difference alone is
not the dominating factor in the 4 Â 103-fold discrimination
between pArg and pTyr.
Reinspection of our model points to additional challenges in
pArg recognition. The SH2 domain recognizes pTyr by two
cation-p interactions involving the side chains of ArgaA2 and
c
10336 Chem. Commun., 2011, 47, 10335–10337
This journal is The Royal Society of Chemistry 2011

View Article Online
3 P. G. Besant, P. V. Attwood and M. J. Piggott, Curr. Protein Pept.
Sci., 2009, 10, 536–550.
4 P. G. Besant and P. V. Attwood, Mol. Cell. Biochem., 2009, 329,
93–106.
5 T. E. McAllister, M. G. Nix and M. E. Webb, Chem. Commun.,
2011, 47, 1297–1299.
6 J. M. Kee, B. Villani, L. R. Carpenter and T. W. Muir, J. Am.
Chem. Soc., 2010, 132, 14327–14329.
7 J. Fuhrmann, A. Schmidt, S. Spiess, A. Lehner, K. Turgay,
K. Mechtler, E. Charpentier and T. Clausen, Science, 2009, 324,
1323–1327.
8 B. T. Wakim and G. D. Aswad, J. Biol. Chem., 1994, 269,
2722–2727.
9 B. T. Wakim, M. M. Picken and R. J. Delange, Biochem. Biophys.
Res. Commun., 1990, 171, 84–90.
10 M. E. Wilson and R. A. Consigli, Virology, 1985, 143, 516–525.
11 M. Sikorska and J. F. Whitfield, Biochim. Biophys. Acta, Protein
Struct. Mol. Enzymol., 1982, 703, 171–179.
12 F. Levyfavatier, M. Delpech and J. Kruh, Eur. J. Biochem., 1987,
166, 617–621.
13 A. Kumon, H. Kodama, M. Kondo, F. Yokoi and H. Hiraishi,
J. Biochem., 1996, 119, 719–724.
14 J. Bolte and G. M. Whitesides, Bioorg. Chem., 1984, 12,
170–175.
15 F. Cramer, A. Vollmar and E. Scheiffele, Chem. Ber., 1962, 95,
1670–1682.
16 Y. Liu, I. F. F. Lien, S. Ruttgaizer, P. Dove and S. D. Taylor, Org.
Lett., 2004, 6, 209–212.
Fig. 1 Overlay of a pArg rotamer onto the Src SH2 domain in a
complex with a pTyr containing peptide. ArgbB5 and ArgaA2 shown
in space filling representation.
17 A. Paquet, Int. J. Pept. Protein Res., 1992, 39, 82–86.
18 A. Paquet, B. Blackwell, M. Johns and J. Nikiforuk, J. Pept. Sci.,
1997, 50, 262–268.
19 A. Paquet and M. Johns, Int. J. Pept. Protein Res., 1990, 36,
97–103.
20 M. Ueki, K. Hashimoto and Y. Oka, Peptide Sci., 2004, 169–172.
21 R. B. Jones, A. Gordus, J. A. Krall and G. MacBeath, Nature,
2006, 439, 168–174.
22 J. M. Bradshaw, V. Mitaxov and G. Waksman, J. Mol. Biol., 1999,
293, 971–985.
23 L. Senderowicz, J. X. Wang, L. Y. Wang, S. Yoshizawa,
W. M. Kavanaugh and C. W. Turck, Biochemistry, 1997, 36,
10538–10544.
24 M. Hafner, E. Vianini, B. Albertoni, L. Marchetti, I. Grune,
C. Gloeckner and M. Famulok, Nat. Protoc., 2008, 3, 579–587.
25 J. A. Cruz-Aguado and G. Penner, Anal. Chem., 2008, 80,
8853–8855.
LysbD6 on either side of the aromatic ring (Fig. S14, ESIw).
While these cationic sidechains form stabilizing interactions
that contribute additional affinity for pTyr, we expect their
contribution on the basic portion of the pArg side chain to
instead be destabilizing due to direct electrostatic repulsion.
The lack of a favourable electrostatic environment for pArg is
further compounded by the greater loss of conformation
entropy required for localization of the phosphoryl group in
pArg relative to pTyr. Structural analysis of other SH2
domains may reveal those most suitable for pArg recognition,
on the basis of the chemical character of the side chains
forming the binding pocket.
In conclusion, we have presented a general method for the
synthesis of pArg-containing peptides using 1 as a building
block for SPPS coupled with selective deprotection of the
phosphoryl group. Affinity measurements based on fluorescence
polarization demonstrated phosphoryl group-dependent
recognition of an arginine-phosphorylated peptide by the
SH2 domain from Src kinase. This result in conjunction with
computational modelling suggests that pTyr-binding proteins
may serve as starting points for development of high-affinity
reagents for the detection of proteins with phosphorylated
arginine side chains.
26 T. J. Burke, K. R. Loniello, J. A. Beebe and K. M. Ervin, Comb.
Chem. High Throughput Screening, 2003, 6, 183–194.
27 D. Lesuisse, G. Lange, P. Deprez, D. Benard, B. Schoot,
G. Delettre, J. P. Marquette, P. Broto, V. Jean-Baptiste,
P. Bichet, E. Sarubbi and E. Mandine, J. Med. Chem., 2002, 45,
2379–2387.
28 R. L. Dunbrack and F. E. Cohen, Protein Sci., 1997, 6, 1661–1681.
29 A. Leaver-Fay, M. Tyka, S. M. Lewis, O. F. Lange, J. Thompson,
R. Jacak, K. Kaufman, P. D. Renfrew, C. A. Smith, W. Sheffler,
I. W. Davis, S. Cooper, A. Treuille, D. J. Mandell, F. Richter,
Y. E. Ban, S. J. Fleishman, J. E. Corn, D. E. Kim, S. Lyskov,
M. Berrondo, S. Mentzer, Z. Popovic, J. J. Havranek,
´
J. Karanicolas, R. Das, J. Meiler, T. Kortemme, J. J. Gray,
B. Kuhlman, D. Baker and P. Bradley, Methods Enzymol., 2011,
487, 545–574.
We would like to thank R.S. Goody for financial support,
H.D. Arndt and M.M. Muller for critical comments and
¨
30 T. Gilmer, M. Rodriquez, S. Jordan, R. Crosby, K. Alligood,
M. Green, M. Kimery, C. Wagner, D. Kinder, P. Charifson,
A. M. Hassell, D. Willard, M. Luther, D. Rusnak,
D. D. Sternbach, M. Mehrotra, M. Peel, L. Shampine, R. Davis,
J. Robbins, I. R. Patel, D. Kassel, W. Burkhart, M. Moyer,
T. Bradshaw and J. Berman, J. Biol. Chem., 1994, 269,
31711–31719.
S. Gentz for technical assistance. F.T. Hofmann was supported
by the International Max Planck Research School in Chemical
Biology. J.K. was supported by the Alfred P. Sloan Fellowship.
Notes and references
1 H. R. Matthews, Pharmacol. Ther., 1995, 67, 323–350.
2 P. V. Attwood, M. J. Piggott, X. L. Zu and P. G. Besant,
Amino Acids, 2007, 32, 145–156.
31 I. Catrina, P. J. O’Brien, J. Purcell, I. Nikolic-Hughes,
J. G. Zalatan, A. C. Hengge and D. Herschlag, J. Am. Chem.
Soc., 2007, 129, 5760–5765.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10335–10337 10337