Site-specific protein modification through CuI-catalyzed 1,2,3-triazole formation and its implementation in protein microarray fabrication

Article abstract of DOI:10.1002/anie.200600756

(Figure Presented) Out of site: A protein expression system was combined with Cu^I-catalyzed 1,2,3-triazole formation to modify a target protein at its C terminus. The immobilized core protein can be modified by diverse small molecules in a site

Full text of DOI:10.1002/anie.200600756

Communications
Protein Modification
DOI: 10.1002/anie.200600756
Site-Specific Protein Modification through
Cui-Catalyzed 1,2,3-Triazole Formation and
Its Implementation in Protein Microarray
Fabrication**
Po-Chiao Lin, Shau-Hua Ueng, Mei-Chun Tseng, Jia-
Ling Ko, Kuo-Ting Huang, Sheng-Chieh Yu,
Avijit Kumar Adak, Yu-Ju Chen, and Chun-Cheng Lin*
Site-specific modification of proteins is an important yet
challenging task in biochemical and biophysical research. In
particular, during the fabrication of protein microarrays, the
activity of the protein may be significantly influenced by its
orientation on the solid surface.[1] Although many reports
have detailed the noncovalent immobilization of proteins
with orientational control (for example, biotin–streptavidin
and His tag–Ni interactions), there are fewsuch methods for
the site-specific covalent immobilization of proteins.[2] How-
ever, the most commonly used chemistries for protein
modification, such as amide-bond formation and reductive
amination, involve random covalent-bond formation,[3] which
inevitably has a negative impact on protein activity and,
hence, the sensitivity of activity-based assays. Thus, given the
presence of multiple reactive amino acids in proteins under
physiological conditions, there is a considerable need for
highly selective orthogonal reactions.[4] To address this issue,
we have combined the intein fusion protein expression system
with CuI-catalyzed 1,2,3-triazole formation to specifically
modify a target protein at its C terminus. Furthermore, the
use of a diazido linker yielded a covalent link between
monomers of a homodimeric protein. We also applied this
method to the fabrication of a protein microarray by site-
[*] P.-C. Lin, Prof. Y.-J. Chen, Prof. C.-C. Lin
Department of Chemistry
National Tsing Hua University
Hsinchu (Taiwan)
Fax: (+886)3-571-1082
E-mail: cclin66@mx.nthu.edu.tw
and
Chemical Biology and Molecular Biophysics
Taiwan International Graduate Program
Academia Sinica, Taipei (Taiwan)
P.-C. Lin, Dr. S.-H. Ueng, M.-C. Tseng, J.-L. Ko, K.-T. Huang, S.-C. Yu,
Dr. A. K. Adak, Prof. Y.-J. Chen, Prof. C.-C. Lin
Institute of Chemistry and Genomic Research Center
Academia Sinica, Taipei (Taiwan)
J.-L. Ko
Department of Chemistry
National Taiwan Normal University
Taipei (Taiwan)
[**] We acknowledge financial support from the National Tsing Hua
University (Academia Sinica Research Project on Nano Science and
Technology) and the National Science Council, Taiwan.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
4286
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4286 –4290

Angewandte
Chemie
specific covalent-bond formation, which resulted in the
retention of the protein activity.
the approach, 2 was conjugated with several target azide-
containing molecules, such as fluorescein isothiocyanate
(FITC; 5), biotin (6), N-acetylglucosamine (7), diazide
linker (8), and a glycopeptide (9; Scheme 2).
Recent studies have shown that site-specific modification
of proteins can be achieved either by incorporating non-
natural amino acids containing orthogonally reactive groups[5]
or by the technique of expressed protein ligation (EPL)[6] to
include terminal cysteine-containing molecules at the C ter-
mini. Although EPL has been used to attach a wide variety of
molecules to target proteins[6–11] and to fabricate protein
microarrays,[10–12] it has limited practical application for
diverse modifications because of the instability of thioesters
in proteins. As a result of the rapid expansion of array-based
methods with diverse applications, there is an urgent need to
develop methodologies to establish core template proteins
that may be readily modified in a variety of ways; moreover,
the systems must be amenable to long-term storage.
To develop an efficient and practical ligation method,
Sharpless et al. and Meldal et al. exploited the effective
formation of 1,2,3-triazoles between azides and terminal
acetylenes in the presence of a CuI catalyst.[13] This azide–
alkyne cycloaddition offers good reproducibility and a high
degree of specificity and biocompatibility with water, which
makes it potentially appropriate for a variety of in vitro and in
vivo applications.[14–18] The CuI-catalyzed azide–alkyne cyclo-
addition[13] is highly predictable, very fast, and resistant to side
reactions. Herein, we describe the use of EPL and the CuI-
catalyzed formation of 1,2,3-triazole to achieve C-terminal
protein modification and to fabricate a protein microarray by
site-specific covalent-bond formation.
To demonstrate the concept, maltose binding protein
(MBP) was chosen as the target protein because it has a good
expression yield as an intein fusion protein in Escherichia
coli.[7] MBP was expressed using the commercially available
IMPACT (intein-mediated purification with affinity chitin-
binding tag) vector system. The fusion protein was purified
with chitin beads and treated with cysteine alkyne 1, followed
by a spontaneous S–N acyl shift to give core-alkynated MBP 2
(Scheme 1). The purified MBP 2 had a mass of 43217 Da, as
determined by electrospray mass spectrometry. By a similar
method, we prepared core-modified enhanced green fluores-
cent proteins (EGFPs) 3 and 4, which contain alkyne and
azide functional groups at their respective C termini. Initially,
we used core-alkynated MBP 2 as the core template protein,
with the C-terminal alkyne available for ligation with various
azido molecules. To demonstrate the general applicability of
Scheme 2. Azido molecules for ligation with alkynated MBP 2.
Conjugation of 2 with azido molecules was achieved by
1,2,3-triazole formation catalyzed by CuI, which was produced
from the reaction of CuSO4 (1 mm) with tris(carboxyethyl)-
phosphine (TCEP; 2 mm) in phosphate-buffered saline (PBS;
pH 8.0, 0.1m; Table 1). As unligated and ligated 2 have similar
Table 1: Ligation of various azido compounds with MBP 2.
Azide
Ligation
product
Mobs [Da]
Mcalcd [Da]
5
6
7
8
9
10
11
12
13
14
43704
43539
43596
43619
44838
43704
43544
43594
43606
44836
masses (difference of < 1 kDa) and hydrophobicity,
common protein separation methods, such as SDS-
PAGE and size-exclusion HPLC, could not be used to
distinguish them. Thus, electrospray mass spectrometry
was used to monitor both unligated and ligated 2 during
the reaction and to evaluate the efficiency of cyclo-
addition.
In an attempt to maintain the native protein
structure, the initial ligation between alkynated MBP 2
and 5 was performed overnight at 08C. However, these
conditions resulted in a lowproduct yield. To increase
the reaction rate and yield, tri(triazolyl)amine ligand 15
Scheme 1. Preparation of recombinant alkynated MBP 2, EGFP 3, and EGFP
4.
Angew. Chem. Int. Ed. 2006, 45, 4286 –4290
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4287

Communications
(2 mm)[15] was added and the reaction temperature was
increased to 258C. As depicted in Figure 1a, the electrospray
mass spectrum revealed the time-dependent reactant/product
ratio during the reaction. The m/z value of 43704 clearly
(10 mm) followed by capping with ethanolamine (Figure 2).
The alkynated EGFP 3 (2.7 mgmLÀ1) and azidated EGFP 4
(2.4 mgmLÀ1) were spotted on the azide- and alkyne-func-
tionalized glass slides, respectively, and unmodified EGFP
Figure 1. a) Ligation of alkynated MBP 2 and FITC (5) with reaction
times of 0, 1, and 6 h. b) SDS-PAGE of homoligation of alkynated MBP
2 and 13.
showed that ligation was complete in 6 hours, thus forming
FITC-modified MBP 10 and indicating the efficiency of C-
terminal modification. To demonstrate the application and
versatility of the site-specific modification method, we tested
some common reagents in bioevaluation, such as fluorescent
probes and affinity tags. The observed masses of the modified
protein products under different ligation reactions are shown
in Table 1. The consistency between the calculated and
observed masses of ligated proteins unambiguously shows
that the correct ligation products were obtained. The
observed yields clearly demonstrate that alkyne–azide cyclo-
addition presents a very efficient method for protein con-
jugation. In general, all ligations were complete within 12 h.
In addition to ligation with small molecules, the linker-
conjugated MBP 13 was designed to demonstrate the applic-
ability of the method to protein–protein conjugation. Alky-
nated MBP 2 was ligated with excess diazido linker 8 to give
MBP 13, which was again conjugated with MBP 2 to yield a
covalently linked MBP homodimer (see Figure 1b). SDS-
PAGE of the ligation product showed a new band migrating at
approximately 86 kDa, which is in good agreement with the
expected mass of a covalent homodimer of MBP 2 (monomer
43 kDa). The ligation reaction was inefficient, however,
probably because of the short length of the polyethylene
glycol linker, which resulted in steric hindrance during close
contact with the second MBP molecule. This putative steric
effect was more significant with shorter linkers (data not
shown). Further optimization of linker length, thereby
minimizing steric hindrance, could improve the effective
yield of the protein modification.
Figure 2. Fabrication of an EGFP microarray by 1,2,3-triazole formation
for a) alkynated EGFP 3 and b) native EGFP with azido slide; c) azido-
EGFP 4 and d) native EGFP with alkynated slide. Reaction conditions:
i) immersed in a solution of glutaric acid bis-N-hydroxysuccinimide
ester (10 mm) in DMF at room temperature for 24 h; ii) a solution of
3-azidopropanylamine (10 mm) in DMF, and iii) a solution of mono-
propargylamine (10 mm) in DMF incubated with activated slide in the
presence of N,N-diisopropylethylamine (100 mm) at room temperature
for 24 h.
was used as control for both functionalized glass slides. The
ligation reaction was incubated for 12 h at room temperature
in the presence of CuSO4 (2 mm) and TCEP (4 mm) in PBS
buffer (pH 8.0), and the protein-modified slide was then
extensively washed with PBST (PBS + 0.04% Tween 20).
Positive EGFP fluorescence, as measured at 530 nm, was
evident for both ligations on the glass slides, whereas the
negative control gave no signal (Figure 2). These results
indicate that chemoselective attachment was achieved by
1,2,3-triazole formation and that the protein tertiary structure
was maintained.
However, the azide-modified slide presented a much
better surface for efficient protein ligation relative to the
alkyne-modified surface (Figure 2a versus 2c). This effect
may be because CuI coordinates with alkynes in solution more
rapidly and with higher affinity than with the azide, thereby
enhancing the rate of 1,2,3-triazole formation with the surface
azido group.[19] A similar preference for the surface functional
group has also been reported when labeling azido-modified
cowpea mosaic virus with an alkynated dye molecule through
triazole formation.[15b] Significantly, in the current study a
protein concentration of only 1 mm was needed to achieve
well-detectable levels of immobilization, which demonstrates
that CuI-catalyzed 1,2,3-triazole formation is a very efficient
reaction in surface chemistry.[12]
To demonstrate the utility of 1,2,3-triazole formation
during the fabrication of a protein microarray, the alkyne-
modified EGFP 3 and azide-modified EGFP 4 were used as
fluorescent proteins that were covalently and site-specifically
attached to glass slides by triazole formation, respectively. To
prepare the alkynated and azidated glass slides, a slide coated
with amino groups was incubated with glutaric acid bis-N-
hydroxysuccinimide ester (10 mm) and then treated with 3-
azidopropanylamine (10 mm) or monopropargylamine
To address the importance of orientation in protein
immobilization, alkynated MBP 2 was covalently attached
to a glass slide in a site-specific manner by triazole formation,
as described in the fabrication of the EGFP microarray.
4288 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4286 –4290

Angewandte
Chemie
Biotinylated maltose 16 was then used as the probe to interact
with 2 on the glass surface, followed by fluorescence visual-
ization with streptavidin-Cy3. As shown in Figure 3a, the
relative intensity of the signal measured at 570 nm indicated
that chemoselective attachment was achieved by 1,2,3-tri-
azole formation.
protein attachment methods, thereby significantly improving
assay sensitivity.
In conclusion, by combining the intein fusion protein
expression system with CuI-catalyzed 1,2,3-triazole forma-
tion, the target protein was efficiently modified at its
C terminus. This method allows the immobilized core tem-
plate protein to be modified by diverse small molecules in a
site-specific manner. In addition, site-specific protein–protein
conjugation was achieved by using a diazido linker, a method
that should be applicable to the preparation of protein
heterodimers. Moreover, we demonstrated the importance of
site-specific immobilization of a protein in an array-based
assay; when protein molecules are oriented in an optimal
manner, the native conformation is more likely to be
preserved, thereby retaining greater activity. Given the
flexibility and ease of this site-specific modification/immobi-
lization technique, the procedure reported herein constitutes
an efficient and reproducible method for the fabrication of
protein microarrays or other solid-phase assays.
Received: February 27, 2006
Published online: May 31, 2006
Keywords: copper · dimerization · ligation ·
.
nitrogen heterocycles · protein modifications
[1] T. Cha, A. Guo, X.-Y. Zhu, Proteomics 2005, 5, 416 – 419.
[2] a) K.-Y. Tomizaki, K. Usui, H. Mihara, ChemBioChem 2005, 6,
782 – 799; b) A. D. De Araffljo, J. M. Palomo, J. Cramer, M.
Köhn, H. Schröder, R. Wacker, C. Niemeyer, K. Alexandrov, H.
Waldmann, Angew. Chem. 2006, 118, 302 – 307; Angew. Chem.
Int. Ed. 2006, 45, 296 – 301.
[3] G. T. Hermanson, Bioconjugate Techniques, Academic Press,
San Diego, 1996.
[4] For a review, see: G. A. Lemieux, C. R. Bertozzi, Trends
Biotechnol. 1998, 16, 506 – 513.
[5] L. Wang, P. G. Schultz, Angew. Chem. 2004, 116, 34 – 66; Angew.
Chem. Int. Ed. 2005, 44, 34 – 66.
[6] T. W. Muir, Annu. Rev. Biochem. 2003, 72, 249 – 289.
[7] T. J. Tolbert, C.-H. Wong, J. Am. Chem. Soc. 2000, 122, 5421 –
5428.
[8] J. A. Pezza, K. N. Allen, D. R. Tolan, Chem. Commun. 2004,
2412 – 2413.
Figure 3. Fabrication of a protein microarray by a) site-specific cova-
lent-bond (1,2,3-triazole) formation or b) random amide-bond forma-
tion. c) X-ray structure of MBP, with blue moieties indicating Lys and
Arg residues.
[9] T. Durek, K. Alexandrov, R. S. Goody, A. Hildebrand, I.
Heinemann, H. Waldmann, J. Am. Chem. Soc. 2004, 126,
16368 – 16378.
[10] M. Lovrinovic, R. Seidel, R. Wacker, H. Schroeder, O. Seitz, M.
Engelhard, R. S. Goody, C. M. Niemeyer, Chem. Commun. 2003,
822 – 823.
[11] M.-L. Lesaicherre, R. Y. P. Lue, G. Y. J. Chen, Q. Zhu, S. Q. Yao,
J. Am. Chem. Soc. 2002, 124, 8768 – 8769.
[12] J. A. Camarero, Y. Kwon, M. A. Coleman, J. Am. Chem. Soc.
2004, 126, 14730 – 14731.
[13] a) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708 – 2721; Angew. Chem. Int. Ed.
2002, 41,2596 – 2599; b) C. W. Tornøe, C. Christensen, M.
Meldal, J. Org. Chem. 2002, 67, 3057 – 3064; c) for a review,
see: V. D. Bock, H. Hiemstra, J. H. Van Maarseveen, Eur. J. Org.
Chem. 2006, 51 – 68.
[14] a) H. C. Kolb, K. B. Sharpless, Drug Discovery Today 2003, 8,
1128 – 1137; b) A. Krasinski, Z. Radic, R. Manetsch, J. Raushel,
P. Taylor, K. B. Sharpless, H. C. Kolb, J. Am. Chem. Soc. 2005,
127, 6686 – 6692.
To compare the effect of site-specific or random conju-
gation on maltose binding activity, MBP 2 was immobilized
on N-hydroxysuccinimide (OSu)-activated slides by random
amide-bond formation (Figure 3b). The site-specific immobi-
lization technique yielded higher protein binding activity
(Figure 3a versus 3b). The X-ray crystal structure of MBP[20]
reveals that seven of 41 lysine and arginine residues on the
protein surface are located near the maltose binding site
(Figure 3c). These amino-containing residues may react with
reagents involved in protein immobilization, which could,
upon covalent attachment to a solid surface, cause a partial or
complete blockade of the maltose binding site. By contrast,
the C terminus of MBP is far from the binding site, and thus
immobilization at this site would have minimal impact on the
native conformation of the maltose binding site. Our data
demonstrate that the site-specific fabrication of a protein
array preserves greater protein activity relative to random
Angew. Chem. Int. Ed. 2006, 45, 4286 –4290
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4289

Communications
[15] a) T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett.
2004, 6, 2853 – 2855; b) Q. Wang, T. R. Chan, R. Hilgraf, V. V.
Fokin, K. B. Sharpless, M. G. Finn, J. Am. Chem. Soc. 2003, 125,
3192 – 3193.
[16] A. E. Speers, B. F. Cravatt, Chem. Biol. 2004, 11, 535 – 546.
[17] a) J. P. Collman, N. K. Devaraj, C. E. D. Chidsey, Langmuir
2004, 20, 1051 – 1053; b) N. K. Devaraj, G. P. Miller, W. Ebina, B.
Kakaradov, J. P. Collman, E. T. Kool, C. E. D. Chidsey, J. Am.
Chem. Soc. 2005, 127, 8600 – 8601; c) J. K. Lee, Y. S. Chi, I. S.
Choi, Langmuir 2004, 20, 3844 – 3847.
[18] M. C. Bryan, F. Fazio, H.-K. Lee, C.-Y. Huang, A. Chang, M. D.
Best, D. A. Calarese, O. Blixt, J. C. Paulson, D. Burton, I. A.
Wilson, C.-H. Wong, J. Am. Chem. Soc. 2004, 126, 8640 – 8641.
[19] S. Punna, J. Kuzelka, Q. Wang, M. G. Finn, Angew. Chem. 2005,
117, 2255 – 2260; Angew. Chem. Int. Ed. 2005, 44, 2215 – 2220.
[20] N. V. Adikesavan, S. S. Mahmood, S. Stanley, Z. Xu, N. Wu, M.
Thibonnier, M. Shoham, Acta Crystallogr. F 2005, 61, 341.
4290 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 4286 –4290