The ββα fold of zinc finger proteins as a natural protecting group. Chemoselective synthesis of a DNA-binding zinc finger derivative

Article abstract of DOI:10.1039/c3cc47599a

We report the selective modification of cysteine residues engineered in peptides that have two additional cysteine residues as part of a Cys ₂His₂ zinc finger motif. The chemoselective modification is achieved, thanks to the protecting effect exerted by the zinc cation upon coordination with the native cysteines and histidines of the zinc-finger fold. The strategy allows a straightforward synthesis of DNA binding zinc finger constructs. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3cc47599a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
The bba fold of zinc finger proteins as a ‘‘natural’’
protecting group. Chemoselective synthesis of a
DNA-binding zinc finger derivative†
Cite this: Chem. Commun., 2014,
50, 2258
Received 3rd October 2013,
Accepted 19th December 2013
Jessica Rodrıguez, Jesus Mosquera, Olalla Vazquez, M. Eugenio Vazquez* and
´
´
´
´
´
Jose L. Mascarenas*
´
˜
DOI: 10.1039/c3cc47599a
www.rsc.org/chemcomm
We report the selective modification of cysteine residues engineered in A major problem of this strategy arises when the peptides or
peptides that have two additional cysteine residues as part of a Cys2His2 proteins of interest have more than one cysteine residue, because
zinc finger motif. The chemoselective modification is achieved, thanks to of the competitive formation of polyalkylated products.7
the protecting effect exerted by the zinc cation upon coordination with
Here, we report the selective modification of cysteine residues
the native cysteines and histidines of the zinc-finger fold. The strategy incorporated into a zinc finger peptide that has two additional
allows a straightforward synthesis of DNA binding zinc finger constructs. cysteines as part of a classical Cys2His2 motif. In the presence of
Zn(II) these cysteines are trapped in the bba secondary fold and
The covalent modification of proteins is a powerful strategy to thereby exhibit a much decreased reactivity (Fig. 1). The ligation
modulate their macromolecular function.1 Nature accomplishes such provides a straightforward approach to bisbenzamidine–zinc finger
alterations through a variety of post-translational modifications that conjugates capable of binding to designed DNA target sequences.
modulate the properties of proteins.2 Inspired by nature, scientists Since the zinc finger is a natural motif present in many transcription
have long pursued the development of methodologies that could factors, this work opens a door for the chemoselective access to zinc
emulate these natural post-conjugation reactions and allow the finger derivatives, and for the modification and/or tagging of zinc
introduction of desired groups or labels in specific sites of proteins.1,3 finger proteins.
For a reaction to be of general use, it should selectively target the
residue of interest in the presence of competing side chains of the tion factors, and is responsible for regulating the expression of a
unprotected polypeptide, and in aqueous environments.
myriad of genes controlling fundamental cellular programs.8 The
The zinc finger family is the largest among eukaryotic transcrip-
Many of the strategies developed so far for the site-selective DNA recognition by zinc finger proteins has served as inspiration for
modification of proteins rely on the introduction of designed the engineering of a large variety of highly appealing artificial gene
amino acids equipped with orthogonally reactive groups that can regulators.9 Despite the relevance of zinc finger proteins, work
be coupled to external reactants, for instance, by using the well towards their selective chemical modification has been very scarce.
known click chemistry.4 Although this strategy allows high levels of There are a number of reports on the reactivity of zinc thiolates of
selectivity, it requires the introduction of non-natural amino acids.5 different zinc fingers that suggest that, while the cysteines in
Alternatively, one could pursue the modification of specific natural charged Zn(Cys)4 and Zn(Cys)3His motifs react with electrophiles
amino acids in proteins, but achieving a good combination of or oxidants, neutral Zn(Cys)2(His)2 are less reactive.10 Therefore, we
reactivity and chemoselectivity is extremely challenging. In this reasoned that zinc finger structures might provide a natural protec-
context, most of the successful modifications of unprotected pep- tion for the two cysteines involved in the coordination to the metal
tides or proteins have relied on the nucleophilic reactivity window
provided by the sulfur atom in cysteine side chains,6 which allowed
a number of selective alkylations with a variety of electrophiles.
´
´
´
Centro Singular de Investigacion en Quımica Bioloxica e Materiais Moleculares
´
´
(CIQUS), Departamento de Quımica Organica. Universidade de Santiago de
Compostela, 15782 Santiago de Compostela, Spain.
E-mail: joseluis.mascarenas@usc.es; Fax: +34 981 595 012;
Tel: +34 981 576 541-14405
Fig. 1 Schematic illustration of the approach for the chemoselective
synthesis of zinc finger derivatives. (a) Unfolded peptide with three
nucleophilic sites; (b) zinc-promoted folding and cysteine coordination
leaves only a single nucleophilic site for modification.
† Electronic supplementary information (ESI) available: Peptide synthesis, full
experimental procedures and analytical data of the peptides and products
obtained. See DOI: 10.1039/c3cc47599a
2258 | Chem. Commun., 2014, 50, 2258--2260
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
(Fig. 3, left, and ESI†). Interestingly, when the reaction was carried out
in the presence of ZnSO4 we observed a different outcome, with the
exclusive formation of a monoalkylated derivative (retention time
18.42 min). The other peaks correspond to the starting peptide 1
(16.73 min) and benzyl bromoacetate (Fig. 3, right).
For the full identification of the monoalkylated product of the
Zn containing reaction we did a MS/MS study of the fragments
obtained after digestion of the modified peptide with trypsin. After
2 h of digestion, HPLC of the mixture showed two peaks with
double-charged signals at 851.36347 and 787.31604 (see the ESI†),
which correspond to fragments cleaved at Lys171 (y14 fragment)
and Arg170. The deconvoluted MS/MS spectra of the first peak show
signals corresponding to the y ion series of the partial sequence
GC159AGPFAC165DIC168GRK, where Cys159 is the cysteine alkylated
with benzyl bromoacetate, and Cys165 and Cys168 are carbamido-
methyl cysteines,14 which are formed during the treatment of the
sample before digestion (see the ESI†). The mass difference of
251.04275 between the signals at 1602.68829 (y13) and 1351.62554
(y12) can only be accounted for if the cysteine alkylated with benzyl
bromoacetate is next to the acetylated glycine (derivative 3, Fig. 4).
Interestingly, the monoalkylated derivatives obtained in the
presence or absence of Zn(II) are different, which suggests that
the intrinsic reactivity of the terminal cysteine is lower than that of
the internal ones (Fig. 3, and ESI†).
Peptide 2 is analogous to one we previously used to construct
DNA binding hybrids containing a zinc finger fragment and a
minor groove binder;13 therefore it seemed specially appropriate
to apply the chemoselective strategy for the preparation of DNA
binders. Previous synthesis of this and related conjugates required
the incorporation of the minor groove-binding agent while the
peptide is still protected and attached to the resin, which represents
an important drawback in terms of efficiency and versatility.15
Therefore we first analyzed the reactivity of peptide 2 with benzyl
bromoacetate, and found that, in the absence of Zn(II), it is fully
converted into the trialkylated adduct after only 5 min at 5 1C
(25.09 min, Fig. 5, left). However, in the presence of ZnSO4 the
reaction resulted mainly in the formation of a monoalkylated
product (22.17 min, for details see the ESI†).
Fig. 2 Structure of model peptides derived from Zif268 (1) and GAGA (2).
Arrows point to the thiols to be selectively modified.
ion, and thereby favor the selective modification of other cysteines
present in the peptide. This would provide a convenient and versatile
way of making zinc finger derivatives.
To test the approach we synthesized model peptides 1 and 2,
which exhibit the sequence of two different types of Cys2His2
fragments, together with an additional cysteine engineered either
at the N-terminus (1) or in the side chain of an internal lysine (2)
(Fig. 2). The zinc finger sequence of peptide 1 is based on a fragment
of the C-terminal Cys2His2 unit of the trimeric Zif268 transcription
factor (residues Pro162 to Arg187 in the crystal structure).11 Peptide 2
contains the Cys2His2 unit of the DNA-binding domain of the GAGA
transcription factor of Drosophila melanogaster (from residues Ser28
to Phe58 in the reference structure).12 Peptide 2 includes a mutation
of Arg44 to Lys to allow the selective introduction of a cysteine group
in its side chain.13 Both peptides were assembled using standard
Fmoc protocols and HBTU as a coupling agent. The Lys residue of
peptide 2 was introduced as alloc derivative, for its selective manip-
ulation in the solid phase (see the ESI†).
We tested the selectivity in the alkylation of the sulfur side chains
in the peptides with benzyl bromoacetate, a very good electrophile
that is commercially available. The reactions were carried out in
parallel, in the absence and presence of 1.5 equiv. of ZnSO4, and
in a deoxygenated phosphate buffer (pH 7.5). The reactions were
monitored by RP-HPLC and the identity of the products was con-
firmed by mass spectrometry. For peptide 1 the alkylation was carried
out at rt in the presence of 4 equiv. of benzyl bromoacetate. After
10 min, LC-MS analysis of an aliquot of the reaction carried out in
the absence of Zn(II) showed a mixture of several products, the
major being the trialkylated adduct (retention time 22.36 min). The
HPLC peak at 19.45 min corresponds to a monoalkylated derivative
Circular dichroism spectral changes displayed by this monoalky-
lated product in the presence of Zn(II) are similar to those observed for
the precursor peptide 2 (increase in negative ellipticity at 208/222 nm).
These changes, consistent with the expected Zn-promoted
bba-folding, confirm the exclusive modification of the cysteine
in the side chain of Lys44, since alkylation of other cysteines
would prevent such folding (see Fig. S15 of the ESI†).16
Fig. 3 HPLC traces of the reactions of peptide 1 (200 mM) with benzyl
bromoacetate (4 equiv.), in the absence (left) and in the presence (right) of
1.5 equiv. of ZnSO4, after 10 minutes of reaction at rt, in a deoxygenated
phosphate buffer (pH 7.5). Peaks with asterisks: benzyl bromoacetate.
Fig. 4 (left) Schematic representation of the fragmentations of the y ion.
(right) Structure of the monoalkylated peptide 3.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 2258--2260 | 2259

View Article Online
ChemComm
Communication
Xunta de Galicia GRC2010/12, GRC2013-041 and INCITE09 209
´
´
084PR. J. R. thanks the Fundacion Gil Davila and the Xunta de
Galicia and J. M. the Spanish MCINN for their PhD fellowships.
We specially thank Dr Manuel Marcos for his invaluable help in
the MS identification of product 3.
Notes and references
Fig. 5 HPLC chromatograms after 5 minutes of the reaction of peptide 2
(200 mM) with benzyl bromoacetate (4 equiv.), in the absence (left) and in
the presence (right) of 1.5 equiv. of ZnSO4 (deoxygenated phosphate
buffer, pH 7.5, 5 1C).
1 (a) C. T. Walsh, Posttranslational Modification of Proteins: Expanding
Nature’s Inventory, Roberts and Co., Englewood, Colorado, 2005;
(b) C. T. Walsh, S. Garneau-Tsodikova and G. J. Gatto Jr., Angew.
Chem., Int. Ed., 2005, 44, 7342; (c) T. L. Foley and M. D. Burkart,
Curr. Opin. Chem. Biol., 2007, 11, 12; (d) D. Qi, C. M. Tann, D. Haring
and M. D. Distefano, Chem. Rev., 2001, 101, 3081; (e) I. S. Carrico,
Chem. Soc. Rev., 2008, 37, 1423.
2 D. P. Gamblin, S. I. van Kasteren, J. M. Chalker and B. G. Davis, FEBS
J., 2008, 275, 1949.
´
3 (a) E. Basle, N. Joubert and M. Pucheault, Chem. Biol., 2010, 17, 213;
(b) M. Rashidian, J. K. Dozier and M. D. Distefano, Bioconjugate
Chem., 2013, 24, 1277.
4 (a) H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004; (b) S. S. Van Berkel, M. B. van Eldijk and J. C. M. van
Hest, Angew. Chem., Int. Ed., 2011, 50, 8806.
5 (a) V. Muralidharan and T. W. Muir, Nat. Methods, 2006, 3, 429;
(b) A. J. de Graaf, M. Kooijman, W. E. Hennink and
E. Mastrobattista, Bioconjugate Chem., 2009, 20, 1281.
6 (a) G. T. Hermanson, Bioconjugate Techniques, Academic Press,
London, 2008; (b) J. M. Chalker, G. J. L. Bernardes, Y. A. Lin and
B. G. Davis, Chem.–Asian J., 2009, 4, 630.
7 (a) M. W. Crankshaw and G. A. Grant, Modification at Cysteine,
Wiley, Unit 15.1, 1996; (b) J. J. Smith, D. W. Conrad, M. J. Cuneo and
H. W. Hellinga, Protein Sci., 2005, 14, 64; (c) J. M. Kuiper, R. Pluta,
W. H. C. Huibers, F. Fusetti, E. R. Geertsma and B. Poolman, Protein
Sci., 2009, 18, 1033.
8 C. O. Pabo, E. Peisach and R. A. Grant, Annu. Rev. Biochem., 2001,
70, 313.
9 (a) A. Klug, Annu. Rev. Biochem., 2010, 79, 213; (b) F. D. Urnov,
E. J. Rebar, M. C. Holmes, H. S. Zhang and P. D. Gregory, Nat. Rev.
Genet., 2010, 11, 636; (c) N. Holt, J. Wang, K. Kim, G. Friedman,
X. Wang, V. Taupin, G. M. Crooks, D. B. Kohn, P. D. Gregory,
M. C. Holmes and P. M. Cannon, Nat Biotechnol., 2010, 28, 839.
Fig. 6 (top) Synthesis of conjugate 4. (bottom) EMSA results; lanes 1–10: [4] = 0,
100, 200, 300, 400, 500, 600, 800, 900, 1000 nM with 50 nM of ATꢀGAGA
dsDNA (SYBR gold staining). Oligonucleotide sequence (binding site underlined,
only one strand shown): ATꢀGAGA: 50-GACGGAATTTGAGAGCGTCG -30.
´ `
10 M. Isaac, J. M. Latour and O. Seneque, Chem. Sci., 2012, 3, 3409.
11 M. Elrod-Erickson, M. A. Rould, L. Nekludova and C. O. Pabo,
Structure, 1996, 4, 1171 (PDB ID: 1AAY).
12 J. G. Omichinski, P. V. Pedone, G. Felsenfeld, A. M. Gronenborn and
G. M. Clore, Nat. Struct. Biol., 1997, 4, 122 (PDB ID: 1YUI).
With this information at hand we assayed the viability of
coupling a more challenging electrophile containing a bisben-
zamidine DNA binder (bb-Br).17 The reaction took place effi-
ciently at rt, in the presence of 1.5 equiv. of ZnSO4 and 1 equiv.
of the bromide (phosphate buffer, pH 7.5, Fig. 6). Importantly,
the resulting conjugate is an efficient and sequence selective
DNA binder, as demonstrated by EMSA (Fig. 6).18 Therefore, the
chemoselective ligation provides a very simple approach to
functional zinc finger peptides, and opens the possibility of
running related reactions with recombinant zinc fingers.
The Zn(II) cation can be exploited both as a folding agent
and as a protecting group. This double role allows the selective
modification of cysteines present in zinc finger peptides that
are not involved in the formation of the Cys2His2–Zn fold. The
strategy provides a direct approach to functional DNA binding
zinc finger conjugates, and promises to be useful for the post-
transcriptional modification of zinc finger proteins.
´
´
13 (a) O. Vazquez, M. E. Vazquez, J. B. Blanco, L. Castedo and
˜
J. L. Mascarenas, Angew. Chem., Int. Ed., 2007, 46, 6886;
´
´
´
˜
(b) M. I. Sanchez, O. Vazquez, M. E. Vazquez and J. L. Mascarenas,
Chem.–Eur. J., 2013, 19, 9923.
14 Positions correspond to the numbering scheme in the Zif268
structure 1AAY in ref. 11.
´
15 (a) J. B. Blanco, V. I. Dodero, M. E. Vazquez, M. Mosquera,
˜
L. Castedo and J. L. Mascarenas, Angew. Chem., Int. Ed., 2006,
´
45, 8210; (b) J. B. Blanco, M. E. Vazquez, L. Castedo and
˜
J. L. Mascarenas, ChemBioChem, 2005, 6, 2173; (c) J. B. Blanco,
˜
O. Vazquez, J. Martinez-Costas, L. Castedo and J. L. Mascarenas,
´
Chem.–Eur. J., 2005, 11, 4171; (d) J. B. Blanco, M. E. Vazquez,
˜
J. Martinez-Costas, L. Castedo and J. L. Mascarenas, Chem. Biol.,
´
˜
´
2003, 10, 713; (e) M. E. Vazquez, A. M. Caamano, J. Martınez-Costas,
˜
L. Castedo and J. L. Mascarenas, Angew. Chem., Int. Ed., 2001,
40, 4723.
16 A. D. Frankel, J. M. Berg and C. O. Pabo, Proc. Natl. Acad. Sci. U. S. A.,
1987, 84, 4841.
´
17 (a) M. I. Sanchez, O. Vazquez, J. Martinez-Costas, M. E. Vazquez and
˜
´
J. L. Mascarenas, Chem. Sci., 2012, 3, 2383; (b) O. Vazquez, M. I.
´
´
´
˜
Sanchez, J. Martınez-Costas, M. E. Vazquez and J. L. Mascarenas,
Org. Lett., 2010, 12, 216.
We acknowledge the support from the Spanish grants
SAF2010-20822-C02, CTQ2009-14431/BQU, CSD2007-00006, the
18 As shown in the ESI† (Fig. S12), hybrid 4 does not bind to DNA
sequences mutated in the peptide binding site.
2260 | Chem. Commun., 2014, 50, 2258--2260
This journal is ©The Royal Society of Chemistry 2014