Electrostatic control of peptide side-chain reactivity using amphiphilic homopolymer-based supramolecular assemblies

Article abstract of DOI:10.1021/ja404940s

Supramolecular assemblies formed by amphiphilic homopolymers with negatively charged groups in the hydrophilic segment have been designed to enable high labeling selectivity toward reactive side chain functional groups in peptides. The negatively charged interiors of the supramolecular assemblies are found to block the reactivity of protonated amines that would otherwise be reactive in aqueous solution, while maintaining the reactivity of nonprotonated amines. Simple changes to the pH of the assemblies' interiors allow control over the reactivity of different functional groups in a manner that is dependent on the pK_a of a given peptide functional group. The labeling studies carried out in positively charged supramolecular assemblies and free buffer solution show that, even when the amine is protonated, labeling selectivity exists only when complementary electrostatic interactions are present, thereby demonstrating the electrostatically controlled nature of these reactions.

Full text of DOI:10.1021/ja404940s

Article
pubs.acs.org/JACS
Electrostatic Control of Peptide Side-Chain Reactivity Using
Amphiphilic Homopolymer-Based Supramolecular Assemblies
Feng Wang, Andrea Gomez-Escudero, Rajasekhar R. Ramireddy, Gladys Murage, S. Thayumanavan,*
and Richard W. Vachet*
Department of Chemistry, University of Massachusetts, Amherst, Massachusetts 01003, United States
S
* Supporting Information
ABSTRACT: Supramolecular assemblies formed by amphiphilic
homopolymers with negatively charged groups in the hydrophilic
segment have been designed to enable high labeling selectivity
toward reactive side chain functional groups in peptides. The
negatively charged interiors of the supramolecular assemblies are
found to block the reactivity of protonated amines that would
otherwise be reactive in aqueous solution, while maintaining the
reactivity of nonprotonated amines. Simple changes to the pH of
the assemblies’ interiors allow control over the reactivity of different
functional groups in a manner that is dependent on the pKa of a
given peptide functional group. The labeling studies carried out in
positively charged supramolecular assemblies and free buffer solution show that, even when the amine is protonated, labeling
selectivity exists only when complementary electrostatic interactions are present, thereby demonstrating the electrostatically
controlled nature of these reactions.
these amphiphilic homopolymers can selectively enrich peptides
from complex mixtures based solely on electrostatic inter-
actions.7 In the current work, we investigate whether these
reverse micelle forming polymers can act as a “reaction flask” to
selectively mask specific functional groups, while specifically
labeling certain others in peptides using differences in electro-
static interactions between the functional groups in the peptide
and those in the reverse micelle interiors (Scheme 1). In order to
provide a robust new platform, we stipulated that our
supramolecular approach should concurrently provide the
following characteristics: (i) the reverse micelle-based polymeric
assemblies be capable of selectively extracting peptides from the
aqueous to the organic phase based on electrostatic comple-
mentarity; (ii) the interaction between specific amino acid
residue within the peptide and the polymeric assembly is dictated
by the pKa’s of the polymeric moieties and that of the peptide side
chain functional group; (iii) this electrostatic binding masks the
reactivity of the side chain functional groups, which provides the
opportunity to specifically label functional groups; and (iv) since
the pKa’s dictate the interaction, this interaction and therefore
the reactivity of side chain functional groups can be modulated by
variation of the aqueous phase pH. While the supramolecular
system that exhibits such features could have broader
implications in reaction chemistry, we focus on of improving
our ability to detect peptides in complex mixtures by selectively
enhancing the detection efficiency of desired peptides. Thus,
development of the principles behind this supramolecular
INTRODUCTION
■
Reactions performed inside supramolecular assemblies have
gained a lot of interest recently due to the unique features offered
by their constricted space and organized construction. These
assemblies can be exploited to enable different reactivity or
regioselectivity, compared to those in bulk solution.1 These
molecular confinements have been investigated as a means of
stabilizing intermediates or by preorganizing reagents to obtain
regioselectivity.2 Micelles formed by small surfactants and
amphiphilic polymers have been shown to increase reactivity,
change reaction pathways, or enhance regioselectivity by
increasing local concentrations, stabilizing transition states, or
incorporating a catalytic element.3 Most of these systems have
focused on increasing reactivity or enhancing regioselectivity.4
Similarly, selective labeling in proteins has also been achieved
using methods such as site-directed mutagenesis or chemical
modification and incorporation of artificial amino acids.5
However, development of supramolecular assemblies that can
block specific reaction sites, while leaving others available, is
rarely studied to our knowledge. We have been interested in
utilizing self-assembling systems to selectively mask certain
amino acid side chains, while leaving other reactive amino acids
unmasked, to gain the capability to selectively label amino acids
in peptides.
Given the ubiquity of charged functional groups in peptides
and proteins, we focused on electrostatic interactions as a means
of controlling reactivity. Amphiphilic homopolymers, which are
capable of forming micelles in aqueous solutions and reverse
micelles in nonpolar solvents,6 were used as the supramolecular
hosts. We have previously shown that the reverse micelles of
Received: May 16, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Schematic Illustration of the Labeling of Extracted Peptides Inside Negatively Charged Reverse Micelles and Their
Analysis by MALDI-MS
masking of functional groups in peptides could have implications
in areas such as proteomics and biomarker detection.8
mixture was sonicated for 4 h. Since this polymer is negatively
charged, we used the positively charged peptide malantide
(RTKRSGSVYEPLKI, pI = 10.3) as an initial test substrate. The
lysine residues of malantide should be able to react with sulfo-N-
hydroxysuccinimide acetate (sulfo-NHSA) to acetylate the
amine side chains, as shown in Scheme 3. This labeling event
can be conveniently monitored by MS, because the reaction
should afford an increase in m/z of 42 per labeling event. Sulfo-
NHSA was first added to the toluene solution containing the
reverse micelle, and the resulting mixture was sonicated for ∼15
min. An aqueous solution of the peptide was then immediately
mixed with the toluene solution. After transfer of the peptide into
the organic phase and subsequent analysis of this organic phase,
the mass spectrum clearly shows the presence of malantide. Very
little to no labeling by sulfo-NHSA, however, is observed (Figure
1a). It is possible that this reagent is simply unable to react with
the amines in malantide at pH 7, as illustrated in Scheme 2a.
When malantide is reacted with sulfo-NHSA in the aqueous
phase, though, peaks that correspond to a shift in the m/z of 42
and 84 are indeed observed (Figure 1b), suggesting that the
reaction occurs in at least two amine residues. In fact, there is no
evidence for the presence of unreacted malantide under these
conditions. Analysis of the location of the labeling sites on the
peptide using MS/MS (Figure S2) indicates that the N-terminus
and the two lysine residues can be labeled. Evidently, the peak
with two acetyl groups corresponds to a mixture of three isomers,
where the labels simultaneously react at the N-terminus and
Lys3, the N-terminus and Lys13, and Lys3 and Lys13.
In order to determine that this reaction control was not just
specific to this peptide, we also tested the effect of the reverse
micelles with another positively charged peptide, apelin 13
(QRPRLSHKGPMPA, pI = 12.4). The difference in this
peptide’s reactivity in the aqueous phase and within the reverse
micelles of polymer 1 is indeed similar to that observed for
malantide (see Figure S3). These results are consistent with our
hypothesis that the electrostatic interaction between the polymer
and the peptide could mask the reactivity of certain amino acids.
It is possible that the difference in reactivity observed for the
two peptides above is simply due to the sterics offered by the
reverse micelles or the inability of sulfo-NHSA to diffuse into the
reverse micelle. To test these alternate possibilities, we used the
reverse micelle formed by positively charged polymer 2 to carry
out the binding and labeling. If the reactivity were under steric
control, then the amine groups within this positively charged
Peptides have been chosen as an initial model system, because
they are smaller and simpler than proteins and because they are
the products obtained in protein digestion during protein
identification protocols. We use mass spectrometry (MS) as the
detection method to analyze the versatility of our approach,
because: (i) it is an information-rich detection method, i.e., it
provides direct information regarding the m/z of a molecule
instead of indirect and nonspecific information such as
absorption or fluorescence spectral signatures; (ii) the extent
of the chemical reaction between the peptide substrate and the
labeling reagent can be ascertained using the shift in m/z; (iii) the
specific labeling site in peptides can also be clearly ascertained
using MS/MS; and (iv) MS is a popular method for proteomics
and biomarker identification and therefore demonstrating our
method using mass change would directly test its impact.
RESULTS AND DISCUSSION
■
Amphiphilic homopolymers are capable of forming environ-
ment-dependent amphiphilic assemblies, i.e., these polymers
form micelle-like structures that are capable of sequestering
hydrophobic guest molecules in water and form reverse micelle-
like assemblies in apolar solvents.9 Interestingly, when dispersed
in a biphasic mixture of water and an apolar solvent, the polymers
are kinetically trapped in the solvent in which they are initially
assembled.6 This observation formed the basis for utilizing the
reverse micelles of these polymers to selectively bind and extract
peptides from the aqueous phase into the polymeric interior in
the apolar organic phase.7,10 The driving force for this selectivity
in binding is electrostatics, as it has been consistently observed
that the polymer and the bound peptides were oppositely
charged, i.e., reverse micelles based on a negatively charged
polymer were able to selectively bind peptides with the pI >7,
when the pH of the aqueous phase is ∼7. It is likely that the
electrostatic interaction between the charged peptide side chains
and those of the polymer is responsible for the binding event. If
this were correct, we hypothesized that the reactivity of the
peptide side chain functionalities that are responsible for binding
to the polymer will be significantly modulated (Scheme 2).
To test this hypothesis, we utilized the carboxylic acid based
amphiphilic homopolymer shown as structure 1 (Chart 1). To
form reverse micelles, 2 equiv of water per charged functional
group were added to a solution of polymer 1 in toluene, and the
B
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 2. Schematic Illustration of the Electrostatic Interaction Driven Selectivity and the Detection Event
Chart 1. Structure of the Amphiphilic Homopolymer Based
on Carboxylic Acid Moiety (negatively charged, 1) and
Quaternary Ammonium Moiety (positively charge, 2)
Scheme 3. Reaction of Sulfo-NHSA with Lysine Residue
reverse micelle should remain unreactive. We used the anionic
peptide PSTAIR (EGVPSTAIREISLLKE, pI = 4.6) to bind to
the positively charged reverse micelle (Figure 2a). When treated
with sulfo-NHSA, the amine residues of the peptide are indeed
C
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Mass spectra of malantide reacted with sulfo-NHSA (a) inside the negatively charged polymer reverse micelle at pH 7.0 and (b) in free aqueous
solution at pH 7.0.
Figure 2. Mass spectra of PSTAIR reacted with sulfo-NHSA (a) inside the positively charged polymer reverse micelle at pH 7.0 and (b) in free aqueous
solution at pH 7.0.
Figure 3. Mass spectra of β-Amyloid 10−20 reacted with sulfo-NHSA (a) inside the negatively charged polymer reverse micelle at pH 7.0 and (b) inside
the positively charged polymer reverse micelle at pH 7.0.
D
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
quite extensively labeled once the peptide is inside the reverse
micelles, as illustrated in Scheme 2b. MS/MS data confirm that
both the lysine and N-terminus are acetylated by sulfo-NHSA
(Figure S4). In fact, the MS results indicate that PSTAIR is more
reactive with sulfo-NHSA inside the positively charged reverse
micelle than in free solution (Figures 2b and S5). This could be
attributed to the high local peptide concentration inside the
reverse micelles or the favorable positioning of sulfo-NHSA in
the positively charged interiors.
achieved with the negatively charged reverse micelles, three
different types of peptides were examined: (i) a peptide with
lysine and a free N-terminus; (ii) a peptide with histidine and a
free N-terminus; and (iii) a peptide with histidine, lysine, and a
free N-terminus.
Malantide (RTKRSGSVYEPLKI, pI = 10.3) contains two
lysine residues and obviously an N-terminus but does not have a
histidine residue. When this peptide is reacted with DEPC in free
solution and inside the reverse micelle at pH 7.0 (Figure 4), both
the lysine residues and the N-terminus are labeled in free solution
(aqueous phase), as evidenced by the three DEPC additions to
the peptide (Figure 4a). Note that the lysines are indeed
protonated at pH 7.0 and therefore might be less available for
reaction in the aqueous phase. It seems, though, that the
equilibrium between the protonated and unprotonated forms is
sufficient for this reaction to occur in the aqueous phase. In
contrast, when the peptide is extracted and reacted inside the
negatively charged reverse micelle at pH 7.0, the lysines no
longer react with DEPC (Figure 4b). These results are consistent
with our hypothesis that lysine is protected from the reaction
through electrostatic interactions within the reverse micelle
interior, as illustrated in Scheme 2c. Most of the N-termini
should be protonated (90% in theory); however, a percentage of
these functionalities are presumably available for reaction.
Indeed, only the N-terminus is labeled at pH 7.0 inside the
reverse micelles (see Figure S9 in for MS/MS data). These
results with DEPC rule out electrostatic repulsion between a
negatively charged sulfo-NHSA and the negatively charged
reverse micelle interiors as the reason for the observed reactivity
differences in Figure 1. It should be noted, however, that sulfo-
NHSA did not react with the N-terminus of malantide in the
reverse micelles at this pH (Figure 1). This is most likely due to
the fact that DPEC is much more reactive than sulfo-NHSA.
We then tested the reactivity of malantide at pH 11.2, a
condition under which the lysine residues and the N-terminus
should not be protonated and thus should not be available for
electrostatic interactions with the polymer, as illustrated in
Scheme 2d. In this case, we find that all these residues react with
DEPC both in free solution and inside the reverse micelles
(Figure 4c,d). In addition to the lysine residues and the N-
terminus, MS/MS data also show that Tyr9 is also labeled at this
pH in both media (Figures S11 and S12). This is understandable,
because tyrosine’s pKa is ∼10; as such, it should be deprotonated
at pH 11.2 and becomes a good nucleophile to react with DEPC.
Our prior work has shown that the nature of the interaction
between the polymer-based reverse micelles and the extracted
molecules is based on electrostatics, not only using mass
spectrometry as the analytical technique but also using
absorption and fluorescence spectroscopy.6,10a To test whether
the interaction between the peptide and polymer assembly is
really due to electrostatics and more importantly to test whether
the reactivity variations are due to the electrostatic interactions,
we carried out DEPC modification of malantide in the presence
of the carboxylate polymer 1 with varying salt concentrations
(100, 250, and 500 mM and 1 M) at pH 7.0. The degree of lysine
reactivity masking by the polymer gradually decreases with
increasing ionic strength (Figure S13). In addition, when the
DEPC labeling of malantide was carried out in the presence of a
charge-neutral oligoethylene glycol-based reverse micelle (Chart
S1), the peptide was not extracted from the aqueous solution,
and there was no selectivity in peptide labeling (Figure S14). All
these results clearly demonstrate the electrostatic nature of the
reactivity control.
In the examples above, the polymer and the peptide were
simultaneously varied to test whether electrostatics is the main
reason for the observed difference in reactivity. The support for
our hypothesis would be stronger, if we were to use the same
peptide with two different polymers. This could be tested,
because we have previously shown that peptides can be extracted
by both negatively and positively charged polymers if the pH of
the solution is close to the peptide’s pI.8b,10a Accordingly, β-
amyloid 10−20 (YEVHHQKLVFF, pI = 7.9) was reacted with
sulfo-NHSA in both negatively and positively charged reverse
micelles at pH 7.0. As shown in Figure 3a, there is no reactivity of
the peptide toward sulfo-NHSA inside the negatively charged
reverse micelle (akin to Scheme 2a). Inside the positively charged
reverse micelle, however, the peptide is very reactive as shown in
Figure 3b (akin to Scheme 2b). MS/MS confirms that both
lysine and the N-terminus are labeled (Figure S6). Interestingly,
a third labeling site is observed in the presence of the positively
charged polymers, which could be due to the labeling of the
glutamine or tyrosine side chains. MS/MS experiments are not
able to unambiguously confirm the location of the third label. It is
important to note that the reaction of the peptide in free aqueous
solution at pH 7.0 shows a labeling pattern that is similar to that
of the positively charged reverse micelle (Figure S7 and S8), but
the extent of labeling in the positively charged reverse micelle is
significantly greater.
An alternate explanation for all the observed reactions is that
sulfo-NHSA is less reactive in the presence of negatively charged
polymers, because of the electrostatic repulsion between the
reverse micellar interior and the inherent charge of the reagent.
To test this possibility, we studied a charge neutral reagent,
diethylpyrocarbonate (DEPC). In addition to testing the above-
mentioned possibility, DEPC’s reactivity toward functional
groups beyond amines provides the opportunity to further test
the versatility of this approach. DEPC can also react with
histidines to provide an increase in m/z of 72 per label (Scheme
4). We were interested in testing whether selective labeling of
different peptide sites can be achieved based on pKa differences of
the side chain functionalities. DEPC can react with the N-
terminus, histidine, and lysine residues; the reactive moieties in
these residues exhibit pKa values of about 8, 6, and 10,
respectively. To illustrate the reaction selectivity that can be
Scheme 4. DEPC Labeling of Histidine Residue Increases m/z
by 72 Units
E
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Mass spectra of malantide reacted with DEPC (a) in free aqueous solution at pH 7.0, (b) inside the negatively charged reverse micelles at pH
7.0, (c) in free solution at pH 11.2, and (d) inside the negatively charged reverse micelles at pH 11.2.
Next, we investigated a peptide containing a reactive N-
terminus and a histidine residue. Kinetensin (IARRHPYFL, pI =
11.0) was reacted with DEPC at pH 5.5 and 7.0 to investigate the
pH selectivity of this reaction (Figure 5). At a pH of 5.5, the side
chain of histidine and the N-terminus should be protonated.
Nonetheless, these functional groups still react with DEPC in
free solution as evidenced by the two DEPC additions to the
peptide (Figures 5a and S15). When the peptide is extracted and
reacted inside the negatively charged reverse micelle at pH 5.5,
however, both of these sites should be blocked for reactivity at
this pH. Indeed, we find that only the unreacted peptide is found
in the presence of the negatively charged reverse micelles (Figure
5b). If this failure to react is indeed due to electrostatic blocking,
then the peptide should be more reactive at pH 7.0 because the
histidine residue should not be protonated. When the reaction in
free solution is conducted at pH 7.0, we again find the addition of
two DEPC molecules in free solution (Figures 5c and S16) with
the MS/MS data confirming that both His5 and the N-terminus
are labeled. In contrast, we find the addition of only one DEPC
molecule when the peptide is reacted inside the reverse micelle at
pH 7.0 (Figure 5d). MS/MS data indicate that the peptide ions
with one DEPC added are a mixture of two isomers, a
predominant one in which His5 is labeled and another in
which the N-terminus is labeled (Figure S17). His5 is the
predominantly labeled site because at this pH most of the
histidine residues in the peptides are not protonated. With a
typical pKa around 8, a small fraction (∼10% in theory) of the N-
termini in the peptides should also not be protonated and could
therefore be reactive, but not to the same extent as the His
residues. Together these data suggest that simple changes to pH
can enable selective labeling of amino acid side chains inside the
reverse micelles of these amphiphilic homopolymers. As with
malantide, we also carried out DEPC labeling of kinetensin inside
negatively charged reverse micelle at different solution ionic
strengths at pH 5.5 (Figure S18). While the interaction between
malantide’s lysine and the polymer was fully screened at much
higher salt concentrations, the anticipated weaker interaction
between the histidine units of kinetensin and the carboxylate
polymer at pH 5.5 was screened even at 100 mM salt
concentration. As with malantide, kinetensin also was not
extracted into the organic phase, and no selectivity in labeling was
observed with the charge-neutral oligoethylene glycol-based
reverse micelle (Figure S19).
We then studied a peptide containing lysine, histidine, and N-
terminal functional groups. Apelin 13 (QRPRLSHKGPMPA, pI
= 12.4) was reacted with DEPC at pH 5.5, 7.0, and 10.0 (Figure
6). At pH 5.5, the peptide reacts in free solution (Figures 6a and
S20), but no reactivity is observed inside the reverse micelles
F
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. Mass spectra of kinetensin reacted with DEPC (a) in free aqueous solution at pH 5.5, (b) inside the negatively charged reverse micelles at pH
5.5, (c) in free solution at pH 7.0, and (d) inside the negatively charged reverse micelles at pH 7.0.
(Figure 6b). This observation is consistent with the expectation
that N-terminus, histidine, and lysine groups are all protonated
and protected from reaction at this pH (akin to Scheme 2a). At
pH 7.0 however, the peptide reacts with DEPC inside the reverse
micelles (Figure 6c), but only histidine and the N-terminus are
labeled (akin to Scheme 2c). The lysine moiety is blocked,
because it is still protonated at this pH (see Figure S21 for MS/
MS data). When the pH is increased to 10.0, the peptide is even
more reactive (Figure 6d and Scheme 2d), and a small degree of
labeling of the lysine residue can indeed be found (see Figure S22
for MS/MS data). Again, these results are consistent with our
hypothesis that electrostatic interactions within the reverse
micelles control reactivity and also that this reactivity can be
tuned by varying aqueous solution pH. As a general control for
the DEPC labeling experiments, we also investigated the
reactivity of peptides in positively charged polymers and found
again that reactive functional groups were not protected from
reaction inside the positively charged reverse micelles (Figures
S23−S25).
is not extracted, the side chains in that peptide might not be
protected even though the pKa of the side chain is suitable for
electrostatic masking. To test this possibility, we studied the
sulfo-NHSA labeling of malantide (pI 10.3) and PSTAIR (pI 4.6)
inside negatively charged supramolecular assemblies at pH 7.0.
Considering the negative charge of the reverse micelle assembly,
the positively charged malantide should be selectively extracted
inside, while PSTAIR should be left behind in the aqueous phase.
Indeed, as seen in Figure 7, the lysine residue in malantide is
blocked from reacting with sulfo-NHSA, while the lysine in
PSTAIR was labeled by the reagent (Figures S26 and S27). This
demonstrates that both the pI of the peptide and the pKa of the
functional groups in the peptide can be used to control labeling
selectivity in the presence of the appropriately charged
amphiphilic homopolymer-based supramolecular assemblies.
The versatility of this approach was further confirmed by testing
the phenomenon with DEPC as the labeling agent and by using
different peptide mixtures of PSTAIR and apelin 13 (Figure
S28−S33).
We were interested in finding whether we can concurrently
perform selective binding between the two phases and execute
the selective labeling of the peptides based on the electrostatic
masking. In this case, we anticipated that we would be able to
achieve selectivity based on both the pI of the peptide and the
pKa of the specific side chain. If the pI of the peptide is such that it
CONCLUSIONS
■
We have developed an amphiphilic homopolymer-based supra-
molecular assembly that allows selective labeling of side chain
functionalities in peptides. We have shown that: (i) peptide side
chain functional groups are masked for reaction when the peptide
G
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 6. Mass spectra of apelin 13 reacted with DEPC (a) in free aqueous solution at pH 5.5, inside negatively charged reverse micelles at (b) pH 5.5,
(c) pH 7.0, and (d) pH 10.0. Note: The unlabeled peaks in each spectrum correspond to oxidative modifications to the side chain of methionine. These
oxidative modifications are also observed in control experiments, indicating that the peptide is oxidized in the pure sample.
Figure 7. Mass spectra of malantide and PSTAIR reacted with sulfo-NHSA in negatively charged assemblies at pH 7.0 (a) malantide and (b) PSTAIR.
is electrostatically bound to the polymer; (ii) even if a peptide is
extracted into the reverse micelle, reactivity masking of a
functional group occurs only when they are electrostatically
engaged in a binding interaction with the complementary
polymer functional groups; (iii) by varying the pH of the
solution, site selective labeling of amino acid residues can be
achieved by taking advantage of the inherent differences in their
pKa; and (iv) the combination of pI of the peptide and the pKa of
the functional group can be used to achieve labeling selectivity
with respect to both the functional group and the peptide. We
H
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Summary of the Selective Labeling of Peptides in the Presence of Polymeric Reverse Micelles
a
peptide
malantide
labeling reagent
sulfo-NHSA
charge of reverse micelle
pH
labeling result
negative
7.0
7.0
7.0
7.0
11.2
11.2
7.0
7.0
5.5
7.0
7.0
7.0
5.5
5.5
7.0
7.0
7.0
7.0
5.5
5.5
7.0
7.0
10.0
10.0
RTKRSGSVYEPLKI
RTKRSGSVYEPLKI
RTKRSGSVYEPLKI
RTKRSGSVYEPLKI
RTKRSGSVYEPLKI
RTKRSGSVYEPLKI
EGVPSTAIREISLLKE
EGVPSTAIREISLLKE
EGVPSTAIREISLLKE
EGVPSTAIREISLLKE
YEVHHQKLVFF
aqueous phase
negative
DEPC
aqueous phase
negative
aqueous phase
positive
PSTAIR
sulfo-NHSA
DEPC
aqueous phase
positive
positive
β-amyloid 10−20
sulfo-NHSA
DEPC
negative
positive
YEVHHQKLVFF
kinetensin
negative
IARRHPYFL
aqueous phase
negative
IARRHPYFL
IARRHPYFL
aqueous phase
negative
IARRHPYFL
apelin 13
sulfo-NHSA
DEPC
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
QRPRLSHKGPMPA
aqueous phase
negative
aqueous phase
negative
aqueous phase
negative
aqueous phase
a
The labeled sites, identified through MS/MS, are in bold and underlined.
the low m/z range were found to be the most diagnostic ions for
determining the amino acids that were labeled.
have summarized the experimental findings that led to these
conclusions in Table 1. Selective labeling of specific function-
alities among a multitude of available sites using our system could
open a potentially new and convenient approach for structural
elucidation studies and functional investigations of biomolecules.
Our polymeric supramolecular assembly could also have the
potential to improve targeted proteomics or biomarker studies
by selectively labeling peptides of interest in complex mixtures
with functional groups that enhance their detection efficiency by
MS.
Peptide Extraction. Aqueous solutions of the peptides were
prepared using a peptide concentration of 5.0 × 10−7 M in 25 mM
MOPS, and NH4OH or NaOH was added to achieve the desired
aqueous phase pH. The carboxylate polymer at a concentration of 1.0 ×
10−4 M and the quaternary ammonium polymer at a concentration of 3.3
× 10−5 M were prepared separately in toluene. The critical micelle
concentrations (cmc) of the carboxylate and quaternary ammonium
polymer are 0.34 mg/mL (1.5 × 10−5 M) and 0.78 mg/mL (7.3 × 10−6
M), respectively (see Figure S1), and so the concentrations used in the
experiments exceed the cmc’s for both polymers. The concentrations of
the peptide in the aqueous phase and polymer in the toluene phase were
also chosen to ensure complete transfer of the peptide into the reverse
micelle. These concentrations were chosen based on previous reverse
micelle capacity measurements, as described previously.10a A two-phase
extraction was performed by first mixing 1 mL of the peptide-containing
aqueous solution with 200 μL of the toluene solution containing the
reverse micelles. The mixture was vortexed for 2 h before centrifuging
the sample to separate the two phases. The toluene phase was dried with
flowing N2 gas, redissolved in 10 μL THF, mixed with 20 μL of an α-
CHCA matrix solution (22.5 mg in 350 μL THF/150 μL H2O/6 μL
TFA), spotted on a MALDI target, and then readied for analysis by
MALDI-MS. For the control experiments in free aqueous phase, a
solution containing the peptide at a concentration of 5 × 10−7 M in 25
mM MOPS at the desired pH was reacted with the labeling reagent for
1−2 h. The amount of labeling reagent added was the same molar ratio
added to the reverse micelle solutions. After reaction, these control
solutions were immediately analyzed by MALDI-MS after 10 μL this
solution was mixed with 10 μL matrix solution and then spotted it on a
MALDI target.
EXPERIMENTAL METHODS
■
Reagents. Kinetensin, malantide, apelin 13, PSTAIR, and β-amyloid
10−20 were obtained from the American Peptide Company. Sulfo-N-
hydroxy succinimideacetate (sulfo-NHSA) was purchased from Thermo
Scientific. 3-(N-morpholino)propanesulfonic acid (MOPS), diethylpyr-
ocarbonate (DEPC), α-cyano-hydroxycinnamic acid (α-CHCA), and
trifluoroacetic acid (TFA) were purchased from Sigma-Aldrich. Water,
used in these experiments, was purified first using a Milli-Q water
purification system (Millipore, Bedford, MA). Tetrahydrofuran (THF),
toluene, ammonium hydroxide (NH4OH), and sodium hydroxide
(NaOH) were purchased from Fisher Scientific. THF was distilled over
Na/Ph2CO before use. All other chemicals were used as obtained from
commercial sources.
Instrumentation. MALDI-MS analyses were performed on a
Bruker Autoflex III time-of-flight mass spectrometer. All mass spectra
were acquired in the reflectron mode, and an average of 200 laser shots
at an optimized power (20−60%) was used. Collision-induced
dissociation (CID) via the fragmentation analysis and structural TOF
(FAST) mode was used for the tandem MS (MS/MS) experiments. In
the FAST mode, a specific precursor ion mass of interest was chosen,
and the settings were modified so that the acquisition of at least 10
segments of product ions was possible. Argon gas was used for CID, and
the pressure was adjusted to obtain optimum results. BioTools was used
to facilitate the analysis of the tandem mass spectra. Immonium ions in
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthesis, other experimental details, and tandem mass spectra.
This material is available free of charge via the Internet at http://
pubs.acs.org.
I
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(i) Chen, I.; Howarth, M.; Lin, W. Y.; Ting, A. Y. Nat. Methods 2005, 2,
99−104. (j) Hong, V.; Kislukhin, A. A.; Finn, M. G. J. Am. Chem. Soc.
2009, 131, 9986−9994. (k) Nonaka, H.; Fujishima, S. H.; Uchinomiya,
S. H.; Ojida, A.; Hamachi, I. J. Am. Chem. Soc. 2010, 132, 9301−9309.
(l) Antos, J. M.; Chew, G. L.; Guimaraes, C. P.; Yoder, N. C.;
Grotenbreg, G. M.; Popp, M. W. L.; Ploegh, H. L. J. Am. Chem. Soc.
2009, 131, 10800−10181. (m) George, N.; Pick, H.; Vogel, H.;
Johnsson, N.; Johnsson, K. J. Am. Chem. Soc. 2004, 126, 8896−8897.
(n) Deuss, P. J.; Popa, G.; Botting, C. H.; Laan, W.; Kamer, P. C. J.
Angew. Chem., Int. Ed. 2010, 49, 5315−5317. (o) Ratner, V.; Kahana, E.;
Eichler, M.; Haas, E. Bioconjugate Chem. 2002, 13, 1163−1170.
(p) Zaccolo, M.; De Giorgi, F.; Cho, C. Y.; Feng, L.; Knapp, T.;
Negulescu, P. A.; Taylor, S. S.; Tsien, R. Y.; Pozzan, T. Nat. Cell Biol.
2000, 2, 25−29.
AUTHOR INFORMATION
■
Corresponding Authors
rwvachet@chem.umass.edu
thai@chem.umass.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported through the National Cancer Institute
of the National Institutes of Health (CA169140). Partial support
from NSF through the Center for Hierarchical Manufacturing
(NSF CMMI-1025020) is also acknowledged.
(6) Basu, S.; Vutukuri, D. R.; Thayumanavan, S. J. Am. Chem. Soc. 2005,
127, 16794−16795.
REFERENCES
■
(7) (a) Combariza, Y.; Savariar, E. N.; Vutukuri, D. R; Thayumanavan,
S.; Vachet, R. W. Anal. Chem. 2007, 79, 7124−7130. (b) Rodthongkum,
N.; Chen, Y.; Thayumanavan, S.; Vachet, R. W. Anal. Chem. 2010, 82,
8686−8691. (c) Rodthongkum, N.; Chen, Y.; Thayumanavan, S.;
Vachet, R. W. Anal. Chem. 2010, 82, 3686−3691. (d) Azagarsamy, M.
A.; Gomez-Escudero, A.; Yesilyurt, V.; Vachet, R. W.; Thayumanavan, S.
Analyst 2009, 134, 635−649.
(8) (a) Khidekel, N.; Ficarro, S. B.; Clark, P. M.; Bryan, M. C.; Swaney,
D. L.; Rexach, J. E.; Sun, Y. E.; Coon, J. J.; Peters, E. C.; Hsieh-Wilson, L.
C. Nat. Chem. Biol. 2007, 3, 339−348. (b) Beynon, R. J.; Pratt, J. M. Mol.
Cell. Proteomics 2005, 4, 857−872. (c) Pan, S.; Zhang, H.; Rush, J.; Eng,
J.; Zhang, N.; Patterson, D.; Comb, M. J.; Aebersold, R. Mol. Cell.
(1) (a) Rebek, J. Acc. Chem. Res. 2009, 42, 1660−1668. (b) Pluth, M.
D.; Bergman, R. G.; Raymond, K. N. Acc. Chem. Res. 2009, 42, 1650−
1659. (c) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int.
̀
Ed. 2009, 48, 3418−3438. (d) Vriezema, D. M.; Aragones, M. C.;
Elemans, J. A. A. W.; Cornelissen, J. J. L. M.; Rowan, A. E.; Nolte, R. J. M.
Chem. Rev. 2005, 105, 1445−1489. (e) Palivan, C. G.; Onaca, O. F.;
Delcea, M.; Itel, F.; Meier, W. Chem. Soc. Rev. 2012, 41, 2800−2823.
(2) (a) Liu, S.; Gan, H. Y.; Hermann, A. T.; Rick, S. W.; Gibb, B. C. Nat.
Chem. 2010, 2, 847−852. (b) Restorp, P.; Rebek, J. J. Am. Chem. Soc.
2008, 130, 11850−11851. (c) Poul, N. L.; Campion, M.; Douziech, B.;
Rondelez, Y.; Clainche, L. L.; Reinaud, O.; Mest, Y. L. J. Am. Chem. Soc.
2007, 129, 8801−8810. (d) Brumaghim, J. L.; Michels, M.; Raymond, K.
N. Eur. J. Org. Chem. 2004, 4552−4559. (e) Wang, Z. J.; Brown, C. J.;
Bergman, R. G.; Raymond, K. A.; Toste, F. D. J. Am. Chem. Soc. 2011,
133, 7358−7360.
Proteomics 2005, 4, 182−190. (d) Charles, R. L.; Schroder, E.; May, G.;
̈
Free, P.; Gaffney, P. R. J.; Wait, R.; Begum, S.; Heads, R. J.; Eaton, P. Mol.
Cell. Proteomics 2007, 6, 1473−1484. (e) Bose, R.; Molina, H.;
Patterson, A. S.; Bitok, J. K.; Periaswamy, B.; Bader, J. S.; Pandey, A.;
Cole, P. A. Proc. Natl. Acad. Sci. 2006, 103, 9773−9778. (f) Hu, L.; Zhou,
H.; Li, Y.; Sun, S.; Guo, L.; Ye, M.; Tian, X.; Gu, J.; Yang, S.; Zou, H.
Anal. Chem. 2009, 81, 94−104. (g) Johnson, A. E.; Jeppsson, F.; Sandell,
(3) (a) Tascioglu, S. Tetrahedron 1996, 52, 11113−11152.
(b) Romsted, L. S.; Bunton, C. A.; Yao, J. Curr. Opin. Coll. Interfaces
1997, 2, 622−628. (c) Holmberg, K. Eur. J. Org. Chem. 2007, 731−742.
(d) Oehme, G. In Aqueous-Phase Organometallic Catalysis-Concept and
Applications; Cornils, B., Herrmann, C. W., Eds.; Wiley-VCH:
Weinheim, 1998. (e) Mandal, H. K.; Majumdar, T.; Mahapatra, A. Int.
J. Chem. Kinet. 2011, 43, 579−589. (f) Kaanumalle, L. S.;
Nithyanandhan, J.; Pattabiraman, M.; Jayaraman, N.; Ramamurthy, V.
J. Am. Chem. Soc. 2004, 126, 8999−9006. (g) Zayas, H. A.; Lu, A.;
Valade, D.; Amir, F.; Jia, Z.; O’Reilly, R. K.; Monteiro, M. J. ACS Macro
Lett. 2013, 2, 327−331. (h) Lu, A.; O’Reilly, R. Curr. Opin. Biotech. 2013,
24, 639−645. (i) Otto, S.; Engberts, J. B. F. N.; Kwak, J. C. T. J. Am.
Chem. Soc. 1998, 120, 9517−9525. (j) Klijn, J. E.; Engberts, J. B. F. N. J.
Am. Chem. Soc. 2003, 125, 1825−1833. (k) Wang, H.; Wang, X. S.;
Winnik, M. A.; Manners, I. J. Am. Chem. Soc. 2008, 130, 12921−12930.
(l) Arumugam, S.; Vutukuri, D. R.; Thayumanavan, S.; Ramamurthy, V.
J. Am. Chem. Soc. 2005, 127, 13200−13206.
́
J.; Wensbo, D.; Neelissen, J. A. M.; Jureus, A.; Strom, P.; Norman, H.;
̈
Farde, L.; Svennson, S. P. S. J. Neurochem. 2009, 108, 1177−1186.
(h) Beaudette, P.; Rossi, N. A.; Huesgen, P. F.; Yu, X.; Shenoi, R. A.;
Doucet, A.; Overall, C. M.; Kizhakkedathu, J. N. Anal. Chem. 2011, 83,
6500−10. (i) Kleifeld, O.; Doucet, A.; Prudova, A.; auf dem Keller, U.;
Gioia, M.; Kizhakkedathu, J. N.; Overall, C. M. Nat. Protoc. 2011, 6,
1578−1611.
(9) (a) Basu, S.; Vutukuri, D.; Shyamroy, S.; Sandanaraj, B. S.;
Thayumanavan, S. J. Am. Chem. Soc. 2004, 126, 9890−9891.
(b) Sandanaraj, S. B.; Bayraktar, H.; Krishnamoorthy, K.; Knapp, M.;
Thayumanavan, S. Langmuir 2007, 23, 3891−3897. (c) Savariar, E. N.;
Aathimanikandan, S. V.; Thayumanavan, S. J. Am. Chem. Soc. 2006, 128,
16224−16230. (d) Kale, T.; Klaikherd, A.; Popere, B.; Thayumanavan,
S. Langmuir 2009, 25, 9660−9670.
(4) (a) Tanner, P.; Onaca, O.; Balasubramanian, V.; Meier, W.;
Palivan, C. G. Chem.−Eur. J. 2011, 17, 4552−4560. (b) Yfrach, G. S.;
Liddell, P. A.; Hung, S. C.; Moore, A. L.; Gust, D.; Moore, T. A. Nature
1997, 385, 239−241. (c) Stano, P.; Carrara, P.; Kuruma, Y.; Souza, T. P.;
Luisi, P. L. J. Mater. Chem. 2011, 21, 18887−18902. (d) Smuleac, V.;
Butterfield, D. A.; Bhattacharyya, D. Langmuir 2006, 22, 10118−10124.
(e) Bi, H.; Qiao, L.; Busnel, J. M.; Liu, B.; Girault, H. H. J. Proteome Res.
2009, 8, 4685−4692. (f) Chopineau, J.; Lesieur, S.; Taravella, B. C.;
Ollivon, M. Biochimie 1998, 80, 421−435. (g) Conrado, R. J.; Varner, J.
D.; DeLisa, M. P. Curr. Opin. Biotechnol. 2008, 19, 492−499. (h) Nardin,
C.; Thoeni, S.; Widmer, J.; Winterhalter, M.; Meier, W. Chem. Commun.
2000, 1433−1434.
(10) (a) Rodthongkum, N.; Savariar, E. N.; Thayumanavan, S.; Vachet,
R. W. Anal. Chem. 2009, 81, 5046−5053. (b) Rodthongkum, N.;
Ramireddy, R.; Thayumanavan, S.; Vachet, R. W. Analyst 2012, 137,
1024−1030.
(5) (a) Liu, C. C.; Schultz, P. G. Annu. Rev. Biochem. 2010, 79, 413−
444. (b) Young, T. S.; Schultz, P. G. J. Biol. Chem. 2010, 285, 11039−
11044. (c) Ngo, J. T.; Tirrell, D. A. Acc. Chem. Res. 2011, 44, 677−685.
(d) Johnson, J. A.; Lu, Y. Y.; van Deventer, J. A.; Tirrell, D. A. Curr. Opin.
Chem. Biol. 2010, 14, 774−780. (e) Stephanopoulos, N.; Francis, M. B.
Nat. Chem. Biol. 2011, 7, 876−884. (f) Seim, K. L.; Obermeyer, A. C.;
Francis, M. B. J. Am. Chem. Soc. 2011, 133, 16970−16976. (g) Carrico, I.
S. Chem. Soc. Rev. 2008, 37, 1423−1431. (h) Qi, D. F.; Tann, C. M.;
Haring, D.; Distefano, M. D. Chem. Rev. 2001, 101, 3081−3111.
J
dx.doi.org/10.1021/ja404940s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX