Imprinted polymers displaying high affinity for sulfated protein fragments

Article abstract of DOI:10.1002/anie.201201314

Leaving a lasting impression: Molecularly imprinted polymers (MIPs) featuring neutral binding sites for phosphotyrosine (P) is also shown to have affinity for sulfotyrosine (S) and a sulfotyrosine-containing peptide. The MIPs were capable of selectively capturing both a phosphorylated and a sulfated peptide in a mixture and could release the peptides separately based on elution conditions. Copyright

Full text of DOI:10.1002/anie.201201314

Angewandte
Chemie
DOI: 10.1002/anie.201201314
Artificial Receptors
Imprinted Polymers Displaying High Affinity for Sulfated Protein
Fragments**
Sudhirkumar Shinde, Anton Bunschoten, John A. W. Kruijtzer, Rob M. J. Liskamp, and
Bçrje Sellergren*
One of the most important challenges in protein science is the
comprehensive mapping of post-translational modifications
(PTMs) in qualitative, quantitative, as well as dynamic terms.
These efforts are driven by the important role of PTMs in the
regulation of cellular processes and the observation that
several PTMs directly correlate with pathogenic conditions.[1]
A PTM of underestimated importance is tyrosine sulfa-
tion, with known targets being membrane proteins, for
example, involved in blood coagulation, cell adhesion, and
hormone regulation.[2,3] The analysis of this modification is
complicated by the lability of sulfotyrosine and isobaric
masses of phospho- and sulfotyrosine.[4,5] Sulfotyrosine read-
ily decomposes at elevated temperature and at low pH values,
which has precluded it from being analyzed by chemical
sequencing experiments. Hence, the scarcity of reports of this
PTM can be attributed to an analytical problem rather than to
a low abundance and importance of this PTM. One approach
to address the problem is to develop selective enrichment
techniques, which could effectively fractionate the various
modified proteins or protein fragments.[5] Sulfotyrosine anti-
bodies or more recently antibody fragments generated by
in vitro display technologies have proven promising in this
regard.[6] However, these are biological receptors, which
suffer from inherent problems related to robustness and cost.
We previously reported that phosphotyrosine molecularly
imprinted polymers (MIPs) can be used to selectively enrich
phosphotyrosine peptides found at trace levels from proteo-
lytic digests.[7] The robustness, reproducibility, and low cost of
these synthetic polymer receptors are particularly attractive
as alternatives to their biologically derived counterparts. As
a continuation of our efforts to develop imprinted polymers
for affinity-based enrichments in proteomics, we demonstrate
herein that MIPs can be used to selectively enrich sulfopep-
tides in a strongly pH-dependent manner.
A common motif in neutral hosts for complexing oxy-
anions are 1,3-disubstituted ureas.[8–10] Acting as a twofold
donor to the acceptor, stable cyclic hydrogen bonds are
formed with the stability decreasing in the order of decreasing
acceptor basicity (i.e. phosphate dianion > carboxylate ꢀ
phosphate monoanion > sulfonate/sulfate).[11] As a first step
in our evaluation of hosts for complexing sulfated biomole-
cules, we decided to investigate host monomer 1 (see
Figure 1) with respect to its complex stability with three
model anions in the form of tetrabutylammonium (TBA) salts
of phenylphosphonic acid (PPA), benzoic acid (BA), and
phenylsulfonic acid (PSA). As monoanion guests, we antici-
pated these to crudely mimic the phenylphosphate or phenyl-
sulfate side chain of the targets. The complex stoichiometries
between 1 and the model anions were determined by Job’s
method of continuous variation and the binding parameters
1
determined by H NMR spectroscopy titration or by isother-
mal titration calorimetry (ITC; see Supporting Information).
After having confirmed the 1:1 stoichiometry, a 1:1 binding
model was used to determine the respective association
constants (K). In agreement with our predictions, the oxourea
monomer formed the most stable complex with the benzoate
anion (K = 8820mÀ1), followed by the phosphomonoanion
(K = 1926mÀ1), whereas the complex with the sulfonate was
too weak to be detected (Supporting Information, Table S1
and Figure S1).
Thereafter, the imprinted polymers were prepared (P1
and P2, Figure 1) and characterized (see Supporting Infor-
mation and Table S2–S3) using the urea host monomer 1 in
a 2:1 (pTyr) and 1:1 (sTyr) stoichiometric ratio to the template
Fmoc-pTyr-OEt in its dianionic form and Fmoc-sTyr-OEt in
its monoanionic form; two different cross-linking monomers
were used, EGDMA (hydrophobic) and PETA (hydrophilic).
Nonimprinted polymers (PN1 and PN2) were prepared
identically to the imprinted polymers but omitting the
template. Methacrylamide (MAAM) or acrylamide (AA)
were added as supplementary monomers to provide addi-
tional hydrogen-bond stabilization. Imprinting effects were
investigated by chromatography using the crushed polymer
monoliths as the stationary phases. Thus, Fmoc-protected
amino acids were injected onto the columns in an acetonitrile-
rich mobile phase buffered with either triethylamine (TEA)
or trifluoroacetic acid (TFA; Figure 2). Basic conditions will
promote deprotonation of the template, potentially allowing
it to bind to the MIP through stable quadruple hydrogen
bonds.[12] P1 and P2 exhibited strong affinity for Fmoc-pTyr-
OH in this mobile phase (Supporting Information, Figure S2)
[*] MSc S. Shinde, Priv.-Doz. Dr. B. Sellergren
INFU, Technische Universitꢀt Dortmund
Otto Hahn Strasse 6, 44221 Dortmund (Germany)
E-mail: b.sellergren@infu.uni-dortmund.de
Dr. A. Bunschoten, Dr. J. A. W. Kruijtzer, Prof. Dr. R. M. J. Liskamp
Medicinal Chemistry & Chemical Biology, Utrecht Institute for
Pharmaceutical Sciences, Utrecht University
P.O. Box 80082, Utrecht, 3508 TB (The Netherlands)
[**] This project was in part financed by the Deutsche Forschungsge-
meinschaft (DFG) (Se 777/9-1). S.S. gratefully acknowledges
a Matching Funds fellowship from the Deutsche Akademische
Austauschdienst (DAAD). We thank Dr. Marc Lamshçft for
assistance with LC-MS measurements.
Supporting information for this article (experimental details) is
available on the WWW under http://dx.doi.org/10.1002/anie.
201201314.
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
polymer P1S in spite of it being programmed to
bind this analyte (Figure 2B).
Considerably more interesting was the reten-
tion behavior of the sTyr analyte in mobile
phases containing TFA as acidic modifier
(Figure 2 and Figure S3). In agreement with
our previous report, the acidic conditions led to
a strong enhancement in the retention of pTyr
over Tyr. We previously explained this behavior
in terms of the relative strengths of the acids
involved (Supporting Information, Table S4).
The pH of a 0.1% TFA solution is not suffi-
ciently low to fully protonate the phosphate
groups, which still carry one oxygen-localized
partial negative charge, a potent hydrogen bond
À
acceptor, for interacting with the donor N Hs of
the binding site.[12] At the same time, the analyte
is less strongly hydrated in this state (i.e. it
should cost less energy for it to transfer from the
bulk solvent to the imprinted site). The type of
cross-linking monomer strongly influenced this
Figure 1. Possible prepolymerization complex of monomer 1 and Fmoc-pTyr-OEt and
recognition of an sTyr containing peptide by the resulting MIP. Fmoc=fluorenylmeth-
yloxycarbonyl group; HPMP=1,2,2,6,6-pentamethylpiperidinium.
behavior. Hence MIPs prepared using the hydrophilic cross-
linking monomer PETA (P2) displayed much enhanced
retentivity and selectivity than those prepared using the
hydrophobic monomer EGDMA (P1). Apart from the more
hydrophilic character of the PETA cross-linker, it also leads
to higher imprinting factors, notably for the pTyr template.
This reflects an enhanced fidelity of the imprinted site,
possibly involving an active participation of the matrix OH
groups in interacting with the guest. Also striking was the
“turning on” of retention of the sTyr analytes in presence of
TFA. This led to k-values of Fmoc-sTyr-OH exceeding that of
Fmoc-pTyr-OH (notably on P2) and contrasted with the weak
retention in the presence of base (TEA; Supporting Infor-
mation, Figure S2). This interesting retention behavior is also
linked to the protonation state of the functional groups
(Supporting Information, Table S4). As in the case of pTyr,
the TFA-buffered mobile phase should render sTyr less
charged and hence less strongly hydrated.
Comparing a water-poor (Figure 2A) with a water-rich
(Figure 2B) mobile phase revealed a strong influence of the
hydrophilicity of the polymer on the retention and selectivity
of the MIPs for the anions. Whereas the EGDMA based
polymers displayed a decrease in retention when changing to
a water-rich mobile phase, the hydrophilic polymers (P2 and
PN2) on the contrary retained the analytes more strongly in
this mobile phase. This suggests that the PETA scaffold is
better adapted to the analysis of aqueous biological samples.
For practical applications, the sTyr selective MIPs should
also display affinity for sulfated peptides and proteins. The
sulfopeptide affinity must prevail when the MIP is challenged
with complex biological samples containing an excess of not
only unmodified but also phosphorylated protein fragments,
the latter offering a strong competition for binding to the MIP
host.
Figure 2. Retention factors (k) for the indicated solutes on pTyr- (P1
and P2), sTyr-imprinted (P1S) polymers, and nonimprinted control
polymers (PN1 and PN2). The mobile phase was A) MeCN:water 95:5
(0.1% TFA) and B) MeCN:water 50:50 (0.1% TFA). The imprinting
factors (kMIP/kNIP) rounded to a single digit are shown above the MIP
data bars.
with nearly 70% of the injected analyte remaining on the
column. Meanwhile, the other amino acids were not retained
and PN1 and PN2 exhibited no affinity for any of the analytes.
These results can be explained as follows: the phospho-
monoester exists under these conditions in the dianionic form
(Supporting Information, Table S4), which corresponds to the
ionization state of the templated species. In addition, this
anion is the most basic of those tested, thus exhibiting the
highest affinity to the urea-group donors in the MIP host.[12]
On the other hand, sulfate is a strongly hydrated anion[15]
(Supporting Information, Table S4) and is therefore not
retained by P1 and P2 in basic media. Furthermore, the
sTyr analytes were only weakly retained by the sTyr imprinted
To test this possibility, the synthetic sulfated peptide
[13]
C5aR10-18S2 (Figure 1) was chosen. C5aR is a G protein-
coupled receptor involved in the inflammatory response in
the presence the proinflammatory agent C5a. The two sTyr
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
residues (sTyr-11 and sTyr-14) are located near the N-
terminus and are essential for high affinity binding of C5a
by C5aR. This sequence is hence a target for inflammation
inhibitors.[14]
cannot be used to distinguish the sulfated peptide from its
nonsulfated counterpart (Supporting Information, Table S5).
In this context, the non-imprinted polymer failed to retain any
of the peptides tested.
First we incubated polymers P1 and PN1 with C5aR10-18S2
in MeCN:water 80:20 (0.1% TFA) under equilibrium con-
ditions and measured the specific binding to each of the
polymers by C18 HPLC quantification of the nonbound
peptide. A strong imprinting effect was seen, with more
than 80% of the peptide bound to P1 (ca. 0.8 mmolgÀ1),
whereas PN1 showed only minimal affinity for the peptide
(Supporting Information, Figure S4). We then employed
a simple on/off solid-phase extraction (SPE) protocol com-
prised of one loading step and one elution step, followed by
reverse-phase HPLC analysis of the eluted fractions. Figur-
es S5–S6 show the recovery of the model peptide after SPE on
P1 and P2 using an optimized extraction protocol. Also, in
this test P2 performed better, offering a near quantitative
recovery of the peptide at lower sample loads—significantly
exceeding the recoveries obtained by the other polymers.
By increasing sample complexity, P2 and PN2 were then
subjected to two stringent tests. In the first, an artificial
peptide mixture was prepared containing C5aR10-18S2, its
desulfated parent peptide, and methylated peptides, resulting
from hydrolytic treatment of C5aR in acidified methanol (2%
TFA in MeOH; Supporting Information, Table S5). The
mixture was loaded onto the P2 or PN2 columns and SPE
performed as described in Figure 3A. From the chromato-
grams shown in this Figure, we conclude that the sTyr peptide
was selectively retained by P2 with minimal retention of the
non-sulfated peptides and that it could be cleanly eluted from
the column (recovery approximately 40%). It is noteworthy
that mass spectrometry, in contrast to the MIP sorbents,
In a final test, we loaded an equimolar mixture of the
monophosphorylated peptide pYY-ZAP70 (see Supporting
Information for the sequence) and C5aR10-18S2 on P2 and PN2
and performed SPE according to two different methods. In
the first method, the SPE consisted of only a loading and an
elution step (SPE1), whereas the second method had, in
addition, an intermediate aqueous wash step (SPE2). Fig-
ure 3B shows the chromatograms of the elution fractions and
Figure 3C the corresponding MALDI-TOF mass spectra.
First we observed that the non-optimized load/elute SPE1
(red traces) resulted in a approximately 40% total peptide
recovery with both peptides present in roughly equal amounts
(see Supporting Information). This suggests that the peptides
do not compete for the binding site under these conditions. In
the load–wash–elute SPE2 (blue traces), on the other hand,
the elution fraction contained pure sulfopeptide whereas the
phosphorylated peptide had been removed in the wash step.
The recoveries based on peak areas were approximately 20%
for C5aR10-18S2 and less than 1% for pYY-ZAP70.[16]
Hence, MIPs combined with different SPE methods may
be used to fractionate peptides according to the nature and/or
number of phospho- or sulfotyrosines (Figure 4). We attribute
Figure 4. Principle of selective fractionation of sulfopeptides and
phosphopeptides by stepwise elution from an MIP affinity column.
Figure 3. Reversed phase HPLC chromatogram of a mixture of
A) C5aR10-18S2 and its hydrolysis and esterification products (green; see
Table S5: 1=Ac-DsYGHsYDDKD (C5aR10-18S2); 2=Ac-DYGHYDDKD; 3
and 4=Ac-DYGHYDDKD(Me); 5=Ac-DYGHYDDKD(Me)2) and the
elution fraction after SPE on P2 (red) and PN2 (blue). SPE conditions:
loading: MeCN (0.1% TFA); washing: MeCN:water 95:5 (0.1% TFA);
elution: MeOH (0.1% TFA). B) An equimolar mixture of C5aR10-18S2
and pYY-ZAP70 (GADDSpYYTAR; green) and the elution fraction after
SPE on P2 (SPE1=red; SPE2=blue) and PN2 (black dashes). SPE1
conditions: loading: MeCN; elution: MeOH (0.1% TFA). SPE2: load-
ing: MeCN; washing: MeCN:water 90:10 (0.1% TFA); elution: MeOH
(0.1% TFA). C) MALDI-TOF MS spectra of the P2 elution fractions
after SPE1 (red) and SPE2 (blue). The signal at m/z 1199 is assigned
to pYY-ZAP70 and m/z 1169 is desulfated C5aR10-18S2.
this unique behavior to the presence of neutral binding sites in
the MIP, presenting urea based hydrogen-bond donors in
a complementary arrangement to the sulfo- or phosphotyr-
osine anion acceptors.
Received: February 16, 2012
Revised: April 26, 2012
Published online: && &&, &&&&
Keywords: molecularly imprinted polymers ·
.
post-translational modifications · receptors · sulfated peptides ·
synthetic receptors
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[10] A. J. Hall, P. Manesiotis, M. Emgenbroich, M. Quaglia, E.
De Lorenzi, B. Sellergren, J. Org. Chem. 2005, 70, 1732.
[11] T. R. Kelly, M. H. Kim, J. Am. Chem. Soc. 1994, 116, 7072.
[12] M. Emgenbroich, C. Borrelli, S. Shinde, I. Lazraq, F. Vilela, A. J.
Hall, J. Oxelbark, E. De Lorenzi, J. Courtois, A. Simanova, J.
Verhage, K. Irgum, K. Karim, B. Sellergren, Chem. Eur. J. 2008,
14, 9516.
[1] A. Alonso, J. Sasin, N. Bottini, I. Friedberg, A. Osterman, A.
Godzik, T. Hunter, J. Dixon, T. Mustelin, Cell 2004, 117, 699.
[2] J. W. Kehoe, C. R. Bertozzi, Chem. Biol. 2000, 7, R57.
[3] C. Seibert, T. P. Sakmar, Biopolymers 2008, 90, 459.
[4] D. Balsved, J. R. Bundgaard, J. W. Sen, Anal. Biochem. 2007,
363, 70.
[5] Y. Yu, A. J. Hoffhines, K. L. Moore, J. A. Leary, Nat. Methods
2007, 4, 583.
[6] A. R. M. Bradbury, S. Sidhu, S. Dꢀbel, J. McCafferty, Nat.
Biotechnol. 2011, 29, 245.
[7] S. Helling, S. Shinde, F. Brosseron, A. Schnabel, T. Mꢀller, H. E.
Meyer, K. Marcus, B. Sellergren, Anal. Chem. 2011, 83, 1862.
[8] E. Fan, S. A. van Arman, S. Kincaid, A. D. Hamilton, J. Am.
Chem. Soc. 1993, 115, 369.
[13] A. Bunschoten, J. A. W. Kruijtzer, J. H. Ippel, C. J. C. de Haas,
J. A. G. van Strijp, J. Kemmink, R. M. J. Liskamp, Chem.
Commun. 2009, 2999.
[14] J. H. Ippel, C. J. C. de Haas, A. Bunschoten, J. A. G. van Strijp,
J. A. W. Kruijtzer, R. M. J. Liskamp, J. Kemmink, J. Biol. Chem.
2009, 284, 12363.
[15] J. J. Heymann, K. D. Weaver, T. A. Mietzner, A. L. Crumbliss, J.
Am. Chem. Soc. 2007, 129, 9704.
[9] D. E. Gomez, L. Fabbrizzi, M. Licchelli, E. Monzani, Org.
Biomol. Chem. 2005, 3, 1495.
[16] As indicated in Figure S5, higher overall recoveries can be
expected at lower sample loads.
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Artificial Receptors
Leaving a lasting impression: Molecularly
imprinted polymers (MIPs) featuring
neutral binding sites for phosphotyrosine
(P) is also shown to have affinity for
sulfotyrosine (S) and a sulfotyrosine-
containing peptide. The MIPs were capa-
ble of selectively capturing both a phos-
phorylated and a sulfated peptide in
a mixture and could release the peptides
separately based on elution conditions.
S. Shinde, A. Bunschoten,
J. A. W. Kruijtzer, R. M. J. Liskamp,
B. Sellergren*
&&&& — &&&&
Imprinted Polymers Displaying High
Affinity for Sulfated Protein Fragments
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!