Reversible chemoselective tagging and functionalization of methionine containing peptides

Article abstract of DOI:10.1039/c3cc42214c

Reagents were developed to allow chemoselective tagging of methionine residues in peptides and polypeptides, subsequent bioorthogonal functionalization of the tags, and cleavage of the tags when desired. This methodology can be used for triggered release

Full text of DOI:10.1039/c3cc42214c

ChemComm
View Article Online
View Journal
COMMUNICATION
Reversible chemoselective tagging and
functionalization of methionine containing peptides†
Cite this: DOI: 10.1039/c3cc42214c
Jessica R. Kramer and Timothy J. Deming*
Received 26th March 2013,
Accepted 19th April 2013
DOI: 10.1039/c3cc42214c
www.rsc.org/chemcomm
Reagents were developed to allow chemoselective tagging of methio-
Our lab recently reported the use of Met alkylation as a facile
nine residues in peptides and polypeptides, subsequent bioorthogonal means to irreversibly introduce useful functionality and chemically
functionalization of the tags, and cleavage of the tags when desired. reactive groups onto polypeptides.9 This work was based on the
This methodology can be used for triggered release of therapeutic pioneering studies of Met alkylation in proteins, which were mainly
peptides, or release of tagged protein digests from affinity columns.
focused on use of non-functional alkylating reagents to probe
inhibition of enzyme active sites.10–14 While many of these alkyla-
There is considerable interest in the site specific conjugation of tions were reported to be irreversible,10–12 some studies,13,14 as well
molecules, i.e. ‘‘tags’’, to peptides and proteins.1 These tags may be as subsequent experiments with peptides,15–20 found that Met
used for attachment of probes for imaging, for selective purification alkylation can be reversed under certain conditions. Since each
or detection in complex mixtures, for enhancement of therapeutic study typically employed a different substrate (amino acid, peptide
properties, or as labels to assist in proteomic analysis.2 Such or protein), different alkylating reagents, and different nucleophilic
modifications typically rely on chemoselective reactions with cleavage reagents,10–20 comparison of reactivity and properties of the
natural amino acid functional groups, e.g. cysteine thiols,1 or various alkylated Met sulfonium groups that have been reported is
biosynthetic incorporation of unnatural amino acids that present challenging. This is especially true since Met alkylations with some
functionality for bioorthogonal reactivity, e.g. azide groups.2 While reagents (e.g. benzylic halides)11,20 have been reported to be rever-
many approaches exist for selective tagging of peptides and sible in some cases and irreversible in others.14,18,21 As a further
proteins, few of these, aside from labile disulfides, are reversible complication, reaction of Met sulfoniums with nucleophiles can give
modifications that allow triggered regeneration of unmodified other products, aside from regenerating Met, depending on where
samples.1–5 Tag removal would be advantageous for some applica- nucleophilic attack occurs.15 In light of these uncertainties, we
tions, such as release of therapeutic peptides from a carrier, or sought to evaluate Met sulfonium stability as functions of
recovery of affinity purified, tagged peptide fractions from protein both alkylating reagent and added nucleophile using a model
digests for downstream proteomic analysis.6–8 Here, we report the copolypeptide substrate. Since the chemical modification of remo-
development of reagents for completely reversible, chemoselective vable Met alkylating tags has not been reported, we also sought to
alkylation of natural methionine (Met) residues in peptides and develop tags with this unprecedented feature. We sought to identify
polypeptides (eqn (1)). These reagents have been optimized to give optimized reagents and conditions for stable sulfonium formation,
stable sulfonium products, allow secondary modifications, and introduction of bioorthogonal reactive groups for tag modification,
allow selective tag removal under mild conditions.
and triggered removal of the functional tags when desired.
To study the stability of different Met sulfonium species, we
prepared a statistical copolymer of Met and lysine (KM, 1) and
reacted this model substrate with a variety of alkylating reagents
(Fig. 1). Under the acidic conditions employed, all the Met residues
in KM were chemoselectively alkylated in near quantitative yields,
similar to previous results.9 Note that since peptides and poly-
peptides are routinely handled and manipulated in strongly
acidic media, the acidic alkylation conditions are compatible
with these molecules.4,5,21 The resulting sulfonium containing
copolymers were each reacted with four common sulfur nucleo-
philes as shown in Fig. 1. We chose alkylating agents to cover a
range of properties. The methyl (2a) and carboxymethyl (2b) groups
(1)
Department of Chemistry and Biochemistry, Department of Bioengineering,
HS-SEAS, University of California, 5121 Engineering 5, Los Angeles, CA 90095, USA.
E-mail: demingt@seas.ucla.edu; Fax: +1 310-794-5956; Tel: +1 310-267-4450
† Electronic supplementary information (ESI) available: Experimental procedures
and spectral data for all new compounds, ESI mass spectra, GPC-LS data, and
additional dealkylation and stability studies. See DOI: 10.1039/c3cc42214c
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.

View Article Online
Communication
ChemComm
sulfonium dealkylation, and also shows low reactivity with disulfides
(see ESI,† Fig. S2). While excess nucleophile was used in these
studies, stoichiometric PyS was also found to effect quantitative
sulfonium dealkylation with longer reaction times (see ESI†).
To identify optimal alkylating reagents, we focused on the alkyne
containing polymers 3c, 3d, and 3g, which differ in linkage structure.
While all of these copolypeptides were quantitatively dealkylated by
PyS back to KM as the sole product, it was found that 3c was less
desirable since it reacted much slower compared to 3d and 3g
(Fig. 2, see ESI,† Fig. S1). The propargyl sulfonium 3d also had
drawbacks since it was found to be unstable in basic aqueous media
and upon prolonged storage as a solid (see ESI†). Consequently, the
benzylic sulfonium derivatives 3f and 3g, were chosen since they
provide an excellent combination of facile formation, stability
Fig. 1 Structures of KM substrate, alkylating reagents (R-X), and nucleophiles
(Nuc) used in this study. GSH = glutathione. Reagent 2e was reacted with a Met against hydrolysis (pH 10), and rapid, facile dealkylation back to
homopolymer instead of KM.
KM when treated with PyS. It is also worth noting that 3g was found
to be completely stable in PBS buffer at 20 1C for 2 weeks, and that
no peptide chain cleavage was detected after alkylation and deal-
kylation reactions (see ESI,† Fig. S8).
To showcase the potential of this optimized system, we performed
proof of concept tag, modify, and release studies using the copoly-
peptide 3g (Fig. 3). A sample of 3g was prepared from KM as described
above, and its alkyne tags were then modified via copper catalyzed
cycloadditions using a variety of functionalized azides.2 Polyethylene
glycol (PEG) chains and glucose units, which may be useful for
improving biological lifetimes of peptide based therapeutics,23 were
quantitatively attached to all alkylated Met residues in 3g (Fig. 3). As a
model probe, 5-azidoacetamido-fluorescein was also attached to 3g
Fig. 2 Dealkylation of polymers 3b, 3c, and 3g over time using different Nuc
(0.1 M in PBS, 37 1C). (A) PyS. (B) GSH.
(ca. 1 per polypeptide chain), to give the fluorescent derivative 5c
(see ESI,† Fig. S3). Treatment of each of these derivatives 5a, 5b, or 5c
were chosen as controls with non-reactive side-chains, and their with PyS resulted in their quantitative conversion back to parent KM,
sulfoniums 3a and 3b were found to be stable to all four nucleophiles, confirming the facile release of the modified tags. We envision that a
as well as strong base (pH 10) and heat (80 1C) in water (Fig. 2, wide variety of azide, alkyne, or cycloalkyne containing molecules or
see ESI†). Although some reports state that these sulfoniums can substrates could be used to modify methionine containing peptide
dealkylate,14,18 our results are consistent with many studies that show samples that have been tagged with one of the alkylating reagents 2f
these groups to be inert under similar conditions.10–12 The reagents or 2g, making this chemistry attractive for applications when eventual
2c, 2d, and 2g were chosen to introduce desirable alkyne functionality release of the modified tag is desired.
that is useful for subsequent modification of the tagged copoly-
For broad utility in tagging of peptides, Met alkylation needs to be
peptides under bioorthogonal conditions.2 An azide containing a chemoselective process that is compatible and doesn’t interfere with
analog (2f) was also used to showcase the ability to incorporate other peptide functional groups. In peptides and proteins, there are
different reactive groups, and finally a galactose containing reagent many nucleophilic functional groups that can react with alkylating
(2e) was used to introduce a model biofunctional side-chain.
reagents.24 Of these, all except Met exist in protonated forms at low
Upon treatment with sulfur nucleophiles, the copolypeptides 3c, pH, which greatly decreases their reactivity.25 While alkylations of
3d, 3f, and 3g all showed some dealkylation back to parent KM as proteinaceous functional groups, such as thiols, are common practice
the sole product, while glycopolymer 3e was found to be stable under at high pH,26 Met is the only functional group in proteins able to react
these conditions (Fig. 2, see ESI,† Fig. S1). The stability of 3e, like 3a, with alkylating reagents at low pH.11,13,16,27
is likely due to the lack of an electron withdrawing substituent on the
alkylating carbon, resulting in the sulfonium being less electrophilic.
The alkylating carbons of samples 3c, 3d, 3f, and 3g all have an
activating substituent (carbonyl, alkyne or phenyl), which greatly
increases the electrophilicity of these sulfoniums. The least reactive
nucleophile was found to be glutathione (GSH), but it did give high
yields of dealkylated KM over time (Fig. 2), which is relevant for
applications in vivo. 2-Mercaptoethanol, thiourea,22 and 2-mercapto-
pyridine (PyS) were all effective for quantitative dealkylation of
sulfonium groups to regenerate KM (Fig. 2, see ESI,† Fig. S1),
and PyS was chosen as the optimal reagent since it provides rapid Fig. 3 Schematic showing tag, modify, and release studies on KM copolypeptide.
c
Chem. Commun.
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
reagents is modular and straightforward, we envision that related
compounds can be readily prepared to introduce other desirable
features, such as isotopic labels to assist MS analysis.7 Once
installed, the tags can be further modified using bioorthogonal
reactions to introduce additional functionalities, such as affinity
tags, or for attachment to substrates for purification.6–8 Finally,
facile removal of these modified tags provides a unique advantage
of this tag and modify method, where tag release may be useful for
peptide concentration and purification from solid supports, as well
as for release of unmodified peptide therapeutics from carriers.
This work was supported by the IUPAC Transnational Call in
Polymer Chemistry and the NSF under award No. MSN 1057970.
Notes and references
1 J. M. Chalker, G. J. L. Bernardes and B. G. Davis, Acc. Chem. Res.,
2011, 44, 730–741.
2 J. A. Prescher and C. R. Bertozzi, Nat. Chem. Biol., 2005, 1, 13–21.
3 J. Olejnik, S. Sonar, E. Krzymanska-Olejnik and K. J. Rothschild,
Proc. Natl. Acad. Sci. U. S. A., 1995, 92, 7590–7594.
4 A. Foettinger, A. Leitner and W. Lindner, J. Chromatogr., A, 2005,
1079, 187–196.
5 A. Foettinger, A. Leitner and W. Lindner, J. Proteome Res., 2007, 6,
3827–3834.
6 W.-C. Lin and T. H. Morton, J. Org. Chem., 1991, 56, 6850–6856.
7 H. Zhou, J. A. Ranish, J. D. Watts and R. Aebersold, Nat. Biotechnol.,
2002, 19, 512–515.
8 K. D. Park, R. Liu and H. Kohn, Chem. Biol., 2009, 16, 763–772;
M. A. Nessen, G. Kramer, J. W. Back, J. M. Baskin, L. E. J. Smeenk,
L. J. de Koning, J. H. van Maarseveen, L. de Jong, C. R. Bertozzi,
H. Hiemstra and C. G. de Koster, J. Proteome Res., 2009, 8, 3702–3711;
J. Szychowski, A. Mahdavi, J. J. L. Hodas, J. D. Bagert, J. T. Ngo,
P. Landgraf, D. C. Dieterich, E. M. Schuman and D. A. Tirrell, J. Am.
Chem. Soc., 2010, 132, 18351–18360.
Fig. 4 Reversible alkylation of PHCRKM with 2g. (A) Reaction scheme. MALDI-
MS spectra of (B) PHCRKM (M+), (C) PHCKRM alkylated with 2g to give 6 (MR+),
and (D) 6 after treatment with PyS to regenerate PHCRKM. M(O)R+ is some 6
that had oxidized during MS ionization.
To demonstrate this selectivity, we attempted to alkylate only the
Met residues in the antioxidant peptide PHCKRM, which also
contains highly nucleophilic histidine, cysteine and lysine residues
(Fig. 4A). Treatment of PHCKRM with alkylating agent 2g in 0.2 M
aqueous formic acid (pH 2.4) gave a single product (6) in 92% isolated
yield, where only the Met residue was alkylated. The composition of 6
9 J. R. Kramer and T. J. Deming, Biomacromolecules, 2012, 13, 1719–1723.
10 H. G. Gundlach, W. H. Stein and S. Moore, J. Biol. Chem., 1959, 234,
1754–1760; P. J. Vithayathil and F. M. Richards, J. Biol. Chem., 1960,
235, 2343–2351.
was confirmed using MALDI MS (Fig. 4B and C), as well as 1H NMR 11 H. J. Schramm and W. B. Lawson, Z. Physiol. Chem., 1963, 332, 97–100.
12 H. J. Goren and E. A. Barnard, Biochemistry, 1970, 9, 959–973;
B. H. Landis and L. J. Berliner, J. Am. Chem. Soc., 1980, 102, 5350–5354.
13 F. Naider, Z. Bohak and J. Yariv, Biochemistry, 1972, 11, 3202–3208.
and ESI-MS analysis (see ESI,† Fig. S4–S6), where the addition of a
single 186 Da 2g tag in MS, and shift of the Met methyl resonance
in NMR, were observed. Control experiments where Na-Z-histidine, 14 C. Kleanthous and J. R. Coggins, J. Biol. Chem., 1990, 265, 10935–10939.
15 H. G. Gundlach, S. Moore and W. H. Stein, J. Biol. Chem., 1959, 234,
Na-Z-lysine, and Na-Z-cysteine were each reacted with benzyl bromide
at pH 2.4 also showed no alkylation under these conditions (see ESI†).
1761–1764.
16 F. Naider and Z. Bohak, Biochemistry, 1972, 11, 3208–3211.
For contrast, reaction of PHCKRM with 2g (2 eq.) in carbonate buffer 17 R. L. Noble, D. Yamashiro and C. H. Li, J. Am. Chem. Soc., 1976, 98,
2324–2328.
(pH 8.3) showed the formation of a complex mixture of multiply
alkylated peptides (see ESI,† Fig. S7). These results demonstrate that
18 M. Bodanszky and M. A. Bednarek, Int. J. Pept. Protein Res., 1982, 20,
408–413.
peptides containing a variety of nucleophilic natural amino acid side- 19 J. T. Doi and G. W. Luehr, Tetrahedron Lett., 1985, 26, 6143–6146.
20 M. Taichi, T. Kimura and Y. Nishiuchi, Int. J. Pept. Res. Ther., 2009,
chains can be chemoselectively, and near quantitatively, modified at
Met residues at low pH.
15, 247–253.
21 G. E. Perlman and E. Katchalski, J. Am. Chem. Soc., 1962, 84,
The alkylated peptide 6 was also readily dealkylated by addition of
PyS to give unmodified PHCKRM as the sole product along with the
alkylated PyS byproduct (Fig. 4D). This tag removal reaction is also
selective, as we have found that Met sulfoniums can be dealkylated
using concentrations of PyS that do not react with the disulfide bond
in cystine under identical conditions (see ESI†), which is an advan-
tage in using PyS instead of 2-mercaptoethanol.
Overall, we have developed functionalized alkylating reagents
that were optimized for high yield, chemoselective tagging of Met
residues in peptides and polypeptides. Since the synthesis of these
452–457.
22 H. M. R. Hoffmann and E. D. Hughes, J. Chem. Soc., 1964,
1252–1258.
23 G. Pasut and F. M. Veronese, Adv. Drug Delivery Rev., 2009, 61,
1177–1188.
24 J. B. Jones and D. W. Hysert, Can. J. Chem., 1971, 49, 3012–3019.
25 K. Lindorff-Larson and J. R. Winther, Anal. Biochem., 2000, 286,
308–310.
26 J. M. Chalker, G. J. L. Bernardes, Y. A. Lin and B. G. Davis,
Chem.–Asian J., 2009, 4, 630–640.
27 L. H. Cohen, Annu. Rev. Biochem., 1968, 37, 695–726; B. L. Vallee and
J. F. Riordan, Annu. Rev. Biochem., 1969, 38, 733–794; A. N. Glazer,
Annu. Rev. Biochem., 1970, 39, 101–130.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.