Design of a multinuclear Zn(ii) complex as a new molecular probe for fluorescence imaging of His-tag fused proteins

Article abstract of DOI:10.1039/c1cc16263b

A Zn(ii) complex (Zn(ii)-Ida) was designed as the new fluorescent probe for His-tag fused proteins. Thanks to the tight binding ability to histidine-rich sequences and bright fluorescence property of the Cy5-appended Zn(ii)-Ida probes, selective and clear

Full text of DOI:10.1039/c1cc16263b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 594–596
www.rsc.org/chemcomm
COMMUNICATION
Design of a multinuclear Zn(II) complex as a new molecular probe for
fluorescence imaging of His-tag fused proteinsw
Sho-hei Fujishima,a Hiroshi Nonaka,a Sho-hei Uchinomiya,a Yoshiyuki Alex. Kawase,a
Akio Ojidab and Itaru Hamachi*a
Received 8th October 2011, Accepted 11th November 2011
DOI: 10.1039/c1cc16263b
A Zn(II) complex (Zn(II)–Ida) was designed as the new fluorescent
probe for His-tag fused proteins. Thanks to the tight binding ability
to histidine-rich sequences and bright fluorescence property of the
Cy5-appended Zn(II)–Ida probes, selective and clear fluorescent
imaging of the His-tag fused G-protein coupled receptors on live
cell surfaces was carried out.
strong need to develop an alternative probe that can circumvent
these drawbacks to expand the versatility of the tag–probe pair in
biological research.6 To meet these requirements, Hauser and
Tsien recently reported a fluorescent zinc complex HisZiFiT as a
strong binding probe for His-tag.6c In this manuscript, we
describe the design of a Zn(II) complex (Zn(II)–Ida) as the new
fluorescent probe for His-tag fused proteins. The selective binding
ability to histidine-rich sequences and bright fluorescence
properties of Zn(II)–Ida probes tethered with a Cy5 fluorophore
allowed us to clearly visualize the His-tag fused G-protein
coupled receptors (GPCRs) on live cell surfaces.
The complementary recognition pair of an oligo-histidine tag
(His-tag) with a Ni(II)–nitrilotriacetic acid complex (Ni(II)–NTA),
invented by researchers of F. Hoffmann-La Roche,1 is now widely
used as a valuable tool for purification, immobilization, and
detection of proteins2 (Fig. 1a). Owing to the highly selective
binding ability, this pair has recently been applied to protein
bioimaging in living cells. For example, Vogel et al. reported the
first bioimaging study of His-tag fused proteins on cell surfaces
using the fluorophore-appended Ni(II)–NTA probe.3 Piehler et al.
also reported the multinuclear Ni(II)–NTA probes for the fluores-
cence labeling of a His-tag fused protein on live cell surfaces.4
Although these works demonstrated the utility of this pair as a tool
for bioimaging, these labeling methods often suffered from the
dimness of the Ni(II)–NTA probe due to the strong fluorescence
quenching by Ni(II) ions, limiting their application in bioimaging
study. Another problem of Ni(II)–NTA probes is the intrinsic
toxicity of Ni(II) ions as a human carcinogen,5 which may cause
a deleterious effect on living cells and in vivo. Therefore, there is a
We have recently reported that the binuclear Zn(II)–Dpa
complex comprised of two sets of 2,20-dipicolylamine (Dpa)
serves as a molecular probe for selective binding to proteins
tethered with a tetra-aspartate tag (D4-tag; DDDD).7 While
the Zn(II)–Dpa complex showed a strong binding affinity to
D4-tag peptide (Boc-DDDD-NH2) (Kd = B1.5 mM) under
the neutral aqueous conditions (50 mM HEPES, pH 7.2), the
binding affinity to a His6-tag peptide (HHHHHH) was negligible
(Kd > 1 mM). Our design strategy of the molecular probe for
His-tag is based on exchange of the coordination ligand (Fig. 1b).
We here assumed that the replacement of the nitrogen-containing
pyridine rings of Dpa in the Zn(II)–Dpa complex with carboxyl-
ates may give rise to a zinc complex exhibiting a high binding
affinity to the His-tag which instead possesses the multiple
nitrogen-containing imidazole rings in lieu of the carboxylates
of D4-tag. We thus designed the new binuclear Zn(II) complex
(Zn(II)–Ida) possessing two sets of iminodiacetic acid (Ida) and
synthesized its derivatives (Fig. 2).
The binding affinity of Zn(II)–Ida 1 to His6-tag peptide
(Ac-WAHHHHHH-NH2) was evaluated by isothermal titration
calorimetry (ITC) under the neutral aqueous conditions (50 mM
HEPES, 100 mM NaCl, pH 7.2). Fig. 3a indicates that their
interaction was an enthalpy-driven exothermic process (0 >
DH). The resultant binding isotherm was fitted with a curve
derived from the 1 : 1 binding algorithm to yield a binding
constant of (3.2 ꢀ 0.2) ꢁ 103 Mꢂ1 (n = 1.4).8 This value is
almost 20-fold lower than that of Ni(II)–NTA (Ka = 7.1 ꢁ
104 Mꢂ1) reported in the literature,9 though the binding isotherm
was not apparently 1 : 1 (n = 0.35). On the other hand, our ITC
experiment revealed that the interaction between the Zn(II)
complex of nitrilotriacetic acid (Zn(II)–NTA) 2 and His6 peptide
was apparently weaker than that of Zn(II)–Ida (103 Mꢂ1 > Ka,
Fig. 1 (a) Conventional tag–probe pair of His-tag/Ni(II)–NTA. (b)
Molecular design of the new probe with a binding affinity to His-tag.
a Department of Synthetic Chemistry and Biological Chemistry,
Graduate School of Engineering, Kyoto University, Katsura Campus,
Kyoto, 615-8510, Japan. E-mail: ihamachi@sbchem.kyoto-u.ac.jp
b Graduate School of Pharmaceutical Sciences, Kyushu University,
3-1-1, Maidashi, Higashi-ku, Fukuoka, 812-8582, Japan
w Electronic supplementary information (ESI) available: Supplementary
figures and methods. See DOI: 10.1039/c1cc16263b
c
594 Chem. Commun., 2012, 48, 594–596
This journal is The Royal Society of Chemistry 2012

View Article Online
Table 1 Summary of the ITC titration experiments between various
Zn(II) complexes and oligo-His peptidesa,b
His6 peptide
His10 peptide
1
2
3
n
1.40 ꢀ 0.07
(3.23 ꢀ 0.18) ꢁ 103
ꢂ9.93 ꢀ 0.61
ꢂ5.13
0.91 ꢀ 0.01
Kapp
DH
TDS
o103
(7.74 ꢀ 1.33) ꢁ 106
ꢂ18.37 ꢀ 0.32
ꢂ8.97
c
—
a
n = stoichiometry, Kapp = binding constant (Mꢂ1), DH = enthalpy
(kcal molꢂ1), TDS = entropy (kcal molꢂ1). Measurement conditions:
50 mM HEPES (pH 7.2), 100 mM NaCl, 25 1C. Heat formation was
scarcely detected (data not shown).
b
c
The cyanine dye is a suitable platform for the design of a multi-
functionalized fluorescent probe. Thus, the probes 4 and 6 were
newly designed, which have two and four sets of Zn(II)–Ida units
as the binding sites at both ends of the Cy5 fluorophore,
respectively.10 The fluorescent titration experiment revealed that
the binding affinity of 4 to a His10-tag fused EGFP (enhanced
green fluorescent protein) was 1.3 ꢁ 106 Mꢂ1, which was almost
comparable to the case of 3 to His10 peptide (Fig. S1, ESIw). 6
showed a remarkably strong binding affinity (8.1 ꢁ 108 Mꢂ1),
which is more than 600-fold larger than that of 4 (Fig. S2w).
These results encouraged us to examine the fluorescence bioimaging
under live cell conditions. We initially employed bradykinin
receptor type2 (B2R) as a target protein expressing on cell
surfaces, which possesses a His10-tag and an EGFP domain at
the exoplasmic N-terminus (His10-EGFP-B2R). When HEK
293 cells transiently expressing His10-EGFP-B2R were incubated
with a low concentration of 4 (0.5 mM), the fluorescence image
due to Cy5 was observed on cell surfaces by confocal laser
scanning microscopy (CLSM) (Fig. 4a). This fluorescence image
was well overlapped with that of EGFP fused to the B2R,
indicative of co-localization of the probe with the B2R receptor.
In the control experiment, the negligible fluorescence of the probe
was observed on the cells expressing B2R lacking the His10-tag
(Fig. 4b). These results indicate that 4 selectively binds to the
His10-tag site so as to enable fluorescence visualization of the B2R
on cell surfaces. The labeling with 4 is also applicable to the
visualization of another GPCR, muscarinic acetylcholine receptor
tethered with a His10 tag and EGFP at the exoplasmic N-terminus
(His10-EGFP-m1AchR). As shown in Fig. 4c, clear fluorescence
due to 4 was observed on cell surfaces, which co-localized well with
that of EGFP. In contrast, the Cy5 fluorescence was scarcely
observed in the case of m1AchR lacking His10-tag (Fig. 4d).
To demonstrate the advantage of the Zn(II)–Ida complex 4,
we carefully compared the fluorescence bioimaging data of
B2R using other metal complexes. The fluorescence of the
Ni(II)–NTA probe 5–2Ni(II) was almost co-localized with that
of the EGFP on the cell surfaces as shown in Fig. 4e. However,
the fluorescence intensity was apparently weaker than that of
4. The quantitative analysis revealed that the fluorescence
intensity of 4 relative to the EGFP fluorescence on the same
cell surface was 0.87 ꢀ 0.16 (N > 10), which is 3-fold greater
than the case of 5–2Ni(II) (0.31 ꢀ 0.07 (N > 10)). The dim
fluorescence image of 5–2Ni(II) is reasonably ascribed to its
lower fluorescence quantum yield (F = 0.048) compared to
that of 4 (F = 0.24). In the case of the Zn(II)–NTA type probe
Fig. 2 (a) Molecular structures of the metal complexes and the
fluorescent derivatives used in the present study. (b) Selective labeling
of a His-tag fused GPCR protein on live cell surfaces with the
Cy5-appended fluorescent Zn(II)–Ida probes.
Fig. 3 ITC titration curve (upper) and processed data (lower) in the
titration of (a) His6 peptide (200 mM) with 1 (6 mM, 10 mL ꢁ 24 injections),
or (b) His10 peptide (10 mM) with 3 (0.2 mM, 10 mL ꢁ 24 injections).
Measurement conditions: 50 mM HEPES (pH 7.2), 100 mM NaCl, 25 1C.
data not shown). Fortunately, the rather low binding affinity of 1
to His-tag was dramatically improved by increasing the number of
the Zn(II)–Ida units in the probe. As shown in Fig. 3b and Table 1,
the binding affinity of the bis-Zn(II)–Ida complex 3 to His10
peptide (Ac-WAHHHHHHHHHH-NH2) was evaluated to be
(7.7 ꢀ 1.1) ꢁ 106 Mꢂ1 (n = 0.91) by ITC measurement, the value
of which is ca. 2000-fold larger than that of 1 to His6 peptide.
We next evaluated the affinity of the multinuclear Zn(II)–Ida
probe to a tag-fused protein in a test tube experiment.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 594–596 595

View Article Online
One limitation of Zn(II)–Ida probe 4 was that the fluorescence
imaging is needed to be conducted in the presence of excess of 4
(0.5 mM). Due to the rather low affinity of 4 toward His10-tag,
the fluorescence image of 4 bound to the GPCRs easily dis-
appeared when the cells were washed with HBS buffer (Fig. 4g).
Fortunately, this was overcome by probe 6. We found that 6
provided a clear and lasting fluorescence image of His10-EGFP-
B2R on live cell surfaces even after removal of the excess of 6 by
the single wash with HBS buffer (Fig. 4h). This result clearly
demonstrates the validity of our design strategy of the His-tag
binding probe based on multivalent coordination chemistry.
In conclusion, we successfully developed Zn(II)–Ida com-
pounds as a new molecular probe for His-tag based on the two
strategies involving exchange of the ligand coordination and
multiple coordination chemistry. We further demonstrated
that Zn(II)–Ida is a versatile alternative to the conventional
Ni(II)–NTA in the fluorescent bioimaging of the His-tag fused
proteins, especially by harnessing the less fluorescence quenching
property of the Zn(II)–Ida than that of Ni(II)–NTA. We believe
that, due to its simple structure and ease of functionalization,
Zn(II)–Ida would become a molecular scaffold widely applicable
in purification, handling, and detection of His-tag proteins in
many biological research. To this end, we are going to improve
their binding affinity through optimization of the His-tag sequence.
Our research is now on going along this way.
Notes and references
1 E. Hochuli, H. Dobeli and A. Schacher, J. Chromatogr., A, 1987,
¨
411, 177–184.
2 (a) K. Tomizaki, K. Usui and H. Mihara, ChemBioChem, 2005, 6,
782–789; (b) J. Arnau, C. Laurizen, G. E. Petersen and J. Pedersen,
Protein Expression Purif., 2006, 48, 1–13.
3 E. G. Guignet, R. Hovius and H. Vogel, Nat. Biotechnol., 2004, 22,
440–444.
4 S. Lata, M. Gavutis, R. Tamp and J. Piehler, J. Am. Chem. Soc.,
2006, 128, 2365–2372.
5 M. Valko, H. Morris and M. T. D. Cronin, Curr. Med. Chem.,
2005, 12, 1161–1208.
6 (a) N. Soh, D. Seto, K. Nakano and T. Imato, Mol. BioSyst., 2006,
2, 128–131; (b) C. R. Goldsmith, J. Jaworski, M. Sheng and
S. J. Lippard, J. Am. Chem. Soc., 2006, 128, 418–419;
(c) C. T. Hauser and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A.,
2007, 104, 3693–3697; (d) M. Kamoto, N. Umezawa, N. Kato and
T. Higuchi, Chem.–Eur. J., 2008, 14, 8004–8012; (e) S. Uchinomiya,
H. Nonaka, S. Fujishima, S. Tsukiji, A. Ojida and I. Hamachi, Chem.
Commun., 2009, 5880–5882.
7 (a) A. Ojida, K. Honda, D. Shinmi, S. Kiyonaka, Y. Mori and
I. Hamachi, J. Am. Chem. Soc., 2006, 128, 10452–10459;
(b) K. Honda, E. Nakata, A. Ojida and I. Hamachi, Chem.
Commun., 2006, 4024–4026; (c) K. Honda, S. Fujishima,
A. Ojida and I. Hamachi, ChemBioChem, 2007, 8, 1370–1372;
(d) H. Nonaka, S. Tsukiji, A. Ojida and I. Hamachi, J. Am. Chem.
Soc., 2007, 129, 15777–15779; (e) H. Nonaka, S. Fujishima,
S. Uchinomiya, A. Ojida and I. Hamachi, J. Am. Chem. Soc.,
2010, 132, 9301–9309.
Fig. 4 Fluorescence visualization of the His-tag fused GPCR
proteins on live cell surfaces. HEK 293 cells expressing the GPCRs
(B2R or m1AchR) were cultivated on a poly-Lys coated glass base
dish. The cells were treated with (a, b, c, d, g) 0.5 mM of 4, (e) 0.5 mM
of 5–2Ni(II), (f) 0.5 mM of 5–2Zn(II) or (h) 0.1 mM of 6 in HBS buffer
for 10 min at rt. The cells were observed by CLSM without washing
(a–f) or after single wash with HBS buffer (g & h). In each labeling
experiment, the fluorescence images were obtained using two different
channels corresponding to (i) EGFP as a protein expression marker
and (ii) Cy5. The transmission image is shown in (iii), and the overlay
image of (i) and (ii) is shown in (iv). Scale bars, 20 mm.
8 The binding affinity of 1 to His6-tag (Kd = 120 mM) is much
weaker than that of the binuclear HisZiFiT (Kd = 40 nM) reported
by Hauser and Tsien (ref. 6c), which might be ascribed to the lower
Lewis acidity of the zinc ions in the negatively charged Zn(II)–Ida
complex compared to the positively charged HisZiFiT.
5–2Zn(II), on the other hand, the fluorescence due to Cy5 was
scarcely observed on cell surfaces (Fig. 4f), despite the fact that
5–2Zn(II) has a comparable fluorescence quantum yield (F = 0.26)
to that of 4. This result was probably because of the lower binding
affinity of Zn(II)–NTA 2 to His-tag than that of 1 (Table 1).
´
9 S. Lata, A. Reichel, R. Brock, R. Tampe and J. Piehler, J. Am.
Chem. Soc., 2005, 127, 10205–10215.
10 A. N. Kapanidis, Y. W. Ebright and R. H. Ebright, J. Am. Chem.
Soc., 2001, 123, 12123–12125.
c
596 Chem. Commun., 2012, 48, 594–596
This journal is The Royal Society of Chemistry 2012