The Scaffold Design of Trivalent Chelator Heads Dictates Affinity and Stability for Labeling His-tagged Proteins in vitro and in Cells

Article abstract of DOI:10.1002/anie.201802746

Small chemical/biological interaction pairs are at the forefront in tracing protein function and interaction at high signal-to-background ratios in cellular pathways. However, the optimal design of scaffold, linker, and chelator head still deserve systematic investigation to achieve the highest affinity and kinetic stability for in vitro and especially cellular applications. We report on a library of N-nitrilotriacetic acid (NTA)-based multivalent chelator heads (MCHs) built on linear, cyclic, and dendritic scaffolds and compare these with regard to their binding affinity and stability for the labeling of cellular His-tagged proteins. Furthermore, we describe a new approach for tracing cellular target proteins at picomolar probe concentrations in cells. Finally, we outline fundamental differences between the MCH scaffolds and define a cyclic trisNTA chelator that displays the highest affinity and kinetic stability of all reported reversible, low-molecular-weight interaction pairs.

Full text of DOI:10.1002/anie.201802746

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Scaffold design of trivalent chelator heads dictates high-affinity
and stable His-tagged protein labeling in vitro and in cellulo
Authors: Karl Gatterdam, Eike F. Joest, Volker Gatterdam, and Robert
Tampé
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201802746
Angew. Chem. 10.1002/ange.201802746
Link to VoR: http://dx.doi.org/10.1002/anie.201802746
http://dx.doi.org/10.1002/ange.201802746

10.1002/anie.201802746
Angewandte Chemie International Edition
COMMUNICATION
Scaffold design of trivalent chelator heads dictates high-affinity
and stable His-tagged protein labeling in vitro and in cellulo
Karl Gatterdam#, Eike F. Joest#, Volker Gatterdam, and Robert Tampé*
#These authors contributed equally to this work.
Abstract: Small chemical/biological interaction pairs are at the
forefront in tracing proteins’ function and interaction at high signal-to-
background ratio in cellular pathways. Pharma ventures have eager
plans to develop trisNTA probes for in vitro and in vivo screening of
His-tagged protein targets. However, the optimal design of scaffold,
linker, and chelator head yet deserves systematic investigations to
achieve highest affinity and kinetic stability for in vitro and especially
cell applications. In this study, we report on a library of N-nitrilotriacetic
histidine tag for site-specific labeling of targets even in a crowded
cellular environment with low toxicity[6-7]
Therefore, the
combination of high-affinity binding properties and an ultra-small,
non-disturbing affinity tag of smaller than 1 kDa enable (i) protein
.
labeling with organic dyes[8], quantum dots[9], and streptavidin[10]
,
(ii) protein tethering to membranes[11]
, , glass
liposomes[12]
surfaces[13], gold interfaces[7a,14], and nanoparticles[15], (iii) guided
high-affinity protein trans-splicing and 'traceless' labeling[16], as
acid (NTA) based multivalent chelator heads (MCHs) built up on linear, well as (iv) stochastic sensing of single molecules in nanopores[17]
,
cyclic, and dendritic scaffolds and contrast these with regard to their
binding affinity and stability for labeling of cellular His-tagged proteins.
Furthermore, we assign a new approach for tracing cellular target
proteins at picomolar probe concentrations in cells. Finally, we
describe fundamental differences between the MCH scaffold and
define a cyclic trisNTA chelator, which displays the highest affinity and
kinetic stability of all reversible, low-molecular weight interaction pairs.
single-molecule tracking[18], or super-resolution microscopy even
in living cells[7b,19]. However, the overwhelming contingences
connecting the NTA groups led to an inscrutable offer of MCH
probes, e.g. even sharing the same nomination, despite a
fundamentally different chemical structure.
Here, we disclose substantial differences between cyclic, linear,
and dendritic MCHs, underlining that the structural organization of
the NTA elements is important for the high affinity and kinetic
stability. Furthermore, we describe the fine-tuned optimization of
trivalent chelator heads towards the highest affinity (i.e. lower
probe concentration) and especially kinetic stability as well as
specificity (lower background). As a final milestone, we determine
which of all currently available MCH probes are optimal for life
science applications.
Decoding of protein tracks and interactions is essential towards a
system-based understanding of biological processes as well as
their dysregulation in diseases[1]. Generic tools are eagerly
awaited to trace proteins by small chemical probes at high
spatiotemporal resolution[2]. Protein labeling can be realized by
either covalent bond formation, enzyme or affinity mediated
attachment of certain precursors[3]. These methods offer versatile
ways to analyze the activity, conformational states, the cellular
localization, and interactions of the target protein.
We first focused on the flexibility of the NTA-linker arm and the
rigidity of the cyclic scaffold. For that purpose, new cyclic cyclam-
Asp-trisNTA (2) and cyclen-Glu-trisNTA (3) were compared with
cyclam-Glu-trisNTA (1), linear trisNTA (4) and (5), dendritic NTA-
Dab-trisNTA (6), and NTA-Orn-trisNTA (7) (Figure 1a). The
synthesis of cyclam-Asp-trisNTA (2) and cyclen-Glu-trisNTA (3) is
related to the one of cyclam-Glu-trisNTA (1)[6] (Figure S1). NTA-
Orn-trisNTA (7) was synthesized as previously described[20], while
NTA-Dab-trisNTA (6) was obtained via a new synthesis route
(Figure S2). Linear trisNTA (4) and (5) were synthesized as
recently described[21] (Figure S3). To compare their kinetic and
thermodynamic parameters (ka, kd, KD), all trivalent chelator
heads were conjugated to biotin via a flexible spacer to allow a
highly controlled immobilization for interaction analysis with His-
tagged proteins using surface plasmon resonance (SPR). The
MCHs were immobilized at the lowest possible surface density to
exclude any bias by multivalency effect at the sensor interface.
This important aspect is often neglected when multivalent
compounds are analyzed at interfaces. Based on the maximum
binding capacity of the neutravidin-functionalized SPR chip (300
resonance units (RU) for cyclam-Glu-trisNTA (1)), we ensured a
very low immobilization density of 10 RU for all biotin-PEG4-
trisNTA compounds. Utilizing the His6-tagged maltose-binding
protein (MBP-H6), the kinetic and thermodynamic binding
properties of all MCHs were compared by SPR spectroscopy
(Figure 1b-d), resulting in binding affinities in the range of 10-
60 nM. All data were corrected for unspecific binding and were
The metal chelating N-nitrilotriacetic acid (NTA) and the
oligohistidine-tag (His-tag) constitute
a
well-established
interaction pair, which has first been described in immobilized
metal ion affinity chromatography (IMAC)[4]. Specific binding is
based on the formation of an octahedral complex of bivalent metal
ions, such as Ni(II) or Co(II), with histidines (Figure 1e). The
interaction is reversible and can be released by competition with
histidine or imidazole. So far, the His-tag is the smallest and most
commonly used affinity tag in life sciences, making it an attractive
target for site-specific protein labeling and in vitro or in vivo
screening approaches. Initial attempts of His-tag labeling utilized
Ni-NTA modified dyes[5]. However, the low affinity (KD ≈ 10 µM)
and the fast off-rate (kd ≈ 1/s) resulted in a very transient labeling
by the His-tag/Ni-NTA pair[5a,6]. To overcome this limitation, we
introduced the concept of multivalent chelator heads (MCHs),
exploiting a molecular avidity effect. The MCHs contain two to six
NTA units leading to (sub-)nanomolar binding affinity towards a
K. Gatterdam, E. F. Joest, Dr. V. Gatterdam, Prof. Dr. R. Tampé
Institute of Biochemistry, Biocenter, Goethe-University Frankfurt
Max-von-Laue-Str. 9, 60438 Frankfurt/M (Germany)
E-mail: tampe@em.uni-frankfurt.de
Homepage: http://www.biochemie.uni-frankfurt.de
fitted by a Langmuir 1:1 isotherm[22]
.
The German Research Foundation (GRK 1986, SFB 902, SFB 807)
and the Volkswagen Foundation (Az. 91 067) supported this work.
This article is protected by copyright. All rights reserved.

10.1002/anie.201802746
Angewandte Chemie International Edition
COMMUNICATION
Figure 1. a) Chemical structures of trivalent chelator heads. The flexibility of the linker arms and the rigidity of the cyclic scaffold were fine-tuned, leading to cyclam-
Asp-trisNTA (2) and cyclen-Glu-trisNTA (3), which were compared to the trivalent MCHs cyclam-Glu-trisNTA (1), linear trisNTA (4) and (5), NTA-Dab-trisNTA (6),
and NTA-Orn-trisNTA (7). For surface immobilization, a biotin-PEG4 linker of ~20 Å was introduced. b-d) Association and dissociation kinetics of MCHs with MBP-
H6. Sensorgrams at different concentrations of MBP-H6 binding to Ni(II)-loaded b) cyclam-Glu-trisNTA (1), cyclam-Asp-trisNTA (2), and cyclen-Glu-trisNTA (3), c)
linear trisNTAs (4) and (5), d) NTA-Dab-trisNTA (6) and NTA-Orn-trisNTA (7). Data were fitted by a Langmuir 1:1 isotherm. e) Schematic illustration of the high-
affinity His6-tag/trisNTA interaction pair. POI: protein of interest.
The respective kinetic parameters are summarized in Table 1. A
full dataset is given in Table S2. Interestingly, the reduced NTA
linker length of cyclam-Asp-trisNTA (2) and decreased flexibility
of the cyclic scaffold as in cyclen-Glu-trisNTA (3) did not improve
binding affinity and complex stability. In contrast, the off rates of
the linear and dendritic chelators are significantly different in
comparison to the cyclic chelator (1) and (3) (Table 1). Cyclam-
Glu-trisNTA (1) displayed the highest affinity and stability of all
trivalent NTA compounds and outperformed any linear or
dendritic trisNTA. Notably, the scaffold and linker length as well
as the structural arrangement (linear, cyclic, dendritic) have a
significant impact on the affinity.
This article is protected by copyright. All rights reserved.

10.1002/anie.201802746
Angewandte Chemie International Edition
COMMUNICATION
Table 1. Kinetic and thermodynamic parameters of Ni(II)-loaded MCHs
coordinating His6-tagged MBP analyzed by SPR spectroscopy.
The cyclam-Glu-trisNTA system (1) appears to be optimized with
regard to binding affinity and stability. A further option to increase
the binding affinity would be the expansion of the His-tag (e.g. ten
or twelve consecutive histidines). For MBP-H10 and MBP-H12, the
KD values obtained by SPR spectroscopy ranged from 0.5 to
4.0 nM (Figure S5 and S6). Noteworthy, SPR analyses with very
slow dissociation rates are difficult. Therefore, the MCH/His-tag
interaction was challenged by increasing concentrations of
imidazole (10 mM and 30 mM), which displaces the histidines and
hence weakens the MCH/His-tag interaction. The complex
stability with MBP-H12 was strongly increased for cyclam-Glu-
trisNTA (1) and cyclen-Glu-trisNTA (3), highlighting the important
role of the cyclic scaffold. For comparison, the dissociation phase
of MBP-H12 was normalized (Figure 2a-c) and the dissociation
rates were displayed as half-life times (Figure 2d; see Figure S9
for full dataset; see Figure S7 and S8 for full dataset of MBP-H6
and MBP-H10) In terms of the entropic freedom of the cyclic
scaffold, we concluded that the differences of the cyclam (flexible)
and the cyclen (more rigid) scaffold only have a minor influence
on the binding stability. In contrast, the NTA-linker arm has a
strong impact as illustrated by the comparison of (1), (2), and (3).
The long glutamate based NTA-linker arm in (1) and (3) is crucial
for a stable interaction. These results confirm the large difference
in the kinetic stability of the different trivalent chelator compounds.
Figure 2. Kinetic stability of cyclic, linear, and dendritic trisNTA. After complex
formation at 240 nM of MBP-H12, the dissociation phase was recorded in the
presence of 0 mM (grey), 10 mM (black), and 30 mM of imidazole (colored) for
Ni(II)-loaded a) cyclam-Glu-trisNTA (1), b) linear trisNTA (n=0) (4), and c)
NTA-Orn-trisNTA (7). All data were fitted by a Langmuir 1:1 isotherm. d)
Summary of complex stability of MCHs with 30 mM of imidazole as competitor.
Data presented as means ± s.d. of three independent experiments.
In conclusion, the cyclic trivalent chelator heads, cyclam- and
cyclen-Glu-trisNTA (1) and (3), displayed a significantly higher
kinetic stability compared to the linear or dendritic trisNTA, which
is also reflected in the SPR data. Taking the results of all trivalent
chelator heads into account, the cyclic scaffold as well as the
linker of cyclam-Glu-trisNTA (1) (cyclic trisNTA) is optimized for
high affinity and stable binding of His-tagged proteins.
To further clarify the observed differences in affinity and kinetic
stability, we compared cyclic and linear trisNTA fluorophore
conjugates as probes for intracellular protein labeling. To
Figure 3. Labeling of intracellular His-tagged proteins by trivalent chelator probes. a) HeLa Kyoto cells were either treated by PFA or glyoxal and incubated with
cyclic trisNTAAlexa647 (40) (100 pM). Pearson’s coefficients (r) were determined by three distinct nuclei. Scale bar: 10 µm. b) Corresponding intensity profiles of the
indicated cross-sections. c) Labeling of transiently expressed, nuclear envelope located, His10-mEGFPLaminA by cyclic-trisNTAsCy5 (18) or linear trisNTAsCy5 (34)
(100 pM each). Mean normalized labeling signal of glyoxal-fixed HeLa Kyoto cells (n = 10) is displayed. After labeling, cells were incubated with PBS containing
10 mM imidazole (3 h).
This article is protected by copyright. All rights reserved.

10.1002/anie.201802746
Angewandte Chemie International Edition
COMMUNICATION
guarantee highest affinity, we additionally developed a novel
MCH labeling protocol in fixed and permeabilized cells. By
apprehending that consecutive histidines are prone to chemical
modification during chemical fixation, we examined the direct
labeling of intracellular His-tagged proteins by MCH fluorescent
probes using glyoxal[23]. For comparison, we chemically arrested
mammalian cells using conventional paraformaldehyde (PFA)[24]
or glyoxal and imaged labeling using cyclic trisNTA coupled to the
bright fluorophore Alexa647 (40) by confocal laser-scanning
microscopy (CLSM). HeLa Kyoto cells transiently expressing His10-
mEGFPLaminA, located at the nuclear envelope, revealed a strong
and precise labeling at picomolar probe concentrations after
glyoxal treatment, while PFA-treated cells showed no significant
labeling under similar conditions (Figure 3a/b). We therefore
hypothesized that the accessibility of the His-tag is largely
improved using our advanced MCH labeling protocol. For a final
statistic evaluation of the MCH scaffolds, we added 100 pM of
either cyclic trisNTAsCy5 (18) or linear trisNTAsCy5 (34) probes to
Keywords: minimal tag • super-chelators • biosensors • protein
labeling • target screening
[1]
a) I. Chen, A. Y. Ting, Curr. Opin. Biotechnol. 2005, 16, 35-40; b) E.
Toprak, P. R. Selvin, Annu. Rev. Biophys. Biomol. Struct. 2007, 36,
349-369; c) A. F. L. Schneider, C. P. R. Hackenberger, Curr. Opin.
Biotechnol. 2017, 48, 61-68; d) M. J. Hinner, K. Johnsson, Curr Opin
Biotechnol 2010, 21, 766-776; e) T. Schlichthaerle, M. T. Strauss,
F. Schueder, J. B. Woehrstein, R. Jungmann, Curr. Opin.
Biotechnol. 2016, 39, 41-47.
[2]
I. Nikic, T. Plass, O. Schraidt, J. Szymanski, J. A. Briggs, C. Schultz,
E. A. Lemke, Angew. Chem. Int. Ed. 2014, 53, 2245-2249.
X. Chen, Y.-W. Wu, Org. Biomol. Chem. 2016, 14, 5417-5439.
a) J. Porath, J. A. N. Carlsson, I. Olsson, G. Belfrage, Nature 1975,
258, 598-599; b) E. Hochuli, H. Döbeli, A. Schacher, J. Chrom. A
1987, 411, 177-184.
[3]
[4]
[5]
a) A. N. Kapanidis, Y. W. Ebright, R. H. Ebright, J. Am. Chem. Soc.
2001, 123, 12123-12125; b) E. G. Guignet, R. Hovius, H. Vogel, Nat.
Biotechnol. 2004, 22, 440-444.
[6]
[7]
S. Lata, A. Reichel, R. Brock, R. Tampe, J. Piehler, J. Am. Chem.
Soc. 2005, 127, 10205-10215.
a) A. Tinazli, J. Tang, R. Valiokas, S. Picuric, S. Lata, J. Piehler, B.
Liedberg, R. Tampé, Chemistry 2005, 11, 5249-5259; b) A.
Kollmannsperger, A. Sharei, A. Raulf, M. Heilemann, R. Langer, K.
F. Jensen, R. Wieneke, R. Tampe, Nat. Commun. 2016, 7, 10372;
c) K. Gatterdam, E. F. Joest, M. S. Dietz, M. Heilemann, R. Tampé,
Angew Chem Int Ed Engl 2018.
His10-
glyoxal-fixed HeLa Kyoto cells transiently expressing
mEGFPLaminA. Hereby, we mimicked SPR conditions and live cell
environment by subsequently adding 10 mM imidazole (Figure
S10). Afterwards, we imaged the cells by CLSM and analyzed the
ratio of bound probe relative to the His10-mEGFPLaminA by
densitometry (Figure 3c). Strikingly, we observed approximately
3-4 times more cyclic trisNTAsCy5 (18) bound to His10-mEGFPLaminA
in contrast to the linear trisNTAsCy5 (34). We further delivered the
cyclic trisNTA probe into the living cells and observed a highly
specific labeling of His10-mEGFPLaminA (Figure S11).
[8]
[9]
S. Lata, M. Gavutis, R. Tampe, J. Piehler, J. Am. Chem. Soc. 2006,
128, 2365-2372.
V. Roullier, S. Clarke, C. You, F. Pinaud, G. G. Gouzer, D. Schaible,
V. Marchi-Artzner, J. Piehler, M. Dahan, Nano. Lett. 2009, 9, 1228-
1234.
[10]
[11]
A. Reichel, D. Schaible, N. Al Furoukh, M. Cohen, G. Schreiber, J.
Piehler, Anal. Chem. 2007, 79, 8590-8600.
a) C. L. van Broekhoven, J. G. Altin, Biochim. Biophys. Acta 2005,
1716, 104-116; b) S. Lata, M. Gavutis, J. Piehler, J. Am. Chem. Soc.
2006, 128, 6-7.
[12]
V. Platt, Z. Huang, L. Cao, M. Tiffany, K. Riviere, F. C. Szoka, Jr.,
Bioconjug. Chem. 2010, 21, 892-902.
In conclusion, we systematically worked out the importance of the
scaffold design of trivalent MCHs affecting affinity and kinetic
stability. Corresponding fundamental distinctions were visualized
by a comprehensive SPR study of all MCHs, differing in their basic
structural arrangement: linear, cyclic, dendritic. Furthermore, we
directly contrasted cyclic trisNTA and linear trisNTA fluorophore
conjugates for the tracing of intracellular His-tagged proteins. We
therefore assigned a new glyoxal-fixation method for the first time
to MCH labeling and thereby enabled the use of picomolar probe
concentrations. Ultimately, we defined cyclic trisNTA as superior
for in-vitro and cell applications, compared to the other described
[13]
[14]
S. Lata, J. Piehler, Anal. Chem 2005, 77, 1096-1105.
R. Valiokas, G. Klenkar, A. Tinazli, A. Reichel, R. Tampé, J. Piehler,
B. Liedberg, Langmuir 2008, 24, 4959-4967.
[15]
[16]
[17]
[18]
J. S. Kim, C. A. Valencia, R. Liu, W. Lin, Bioconjug. Chem. 2007,
18, 333-341.
M. Braner, A. Kollmannsperger, R. Wieneke, R. Tampé, Chem. Sci.
2016, 7, 2646-2652.
R. Wei, V. Gatterdam, R. Wieneke, R. Tampé, U. Rant, Nat.
Nanotechnol. 2012, 7, 257.
a) G. Giannone, E. Hosy, F. Levet, A. Constals, K. Schulze, A. I.
Sobolevsky, M. P. Rosconi, E. Gouaux, R. Tampe, D. Choquet, L.
Cognet, Biophys. J. 2010, 99, 1303-1310; b) O. Rossier, V. Octeau,
J. B. Sibarita, C. Leduc, B. Tessier, D. Nair, V. Gatterdam, O.
Destaing, C. Albiges-Rizo, R. Tampe, L. Cognet, D. Choquet, B.
Lounis, G. Giannone, Nat. Cell Biol. 2012, 14, 1057-1067.
a) R. Wombacher, M. Heidbreder, S. van de Linde, M. P. Sheetz,
M. Heilemann, V. W. Cornish, M. Sauer, Nat. Methods 2010, 7, 717-
719; b) R. Wieneke, A. Raulf, A. Kollmannsperger, M. Heilemann,
R. Tampe, Angew. Chem. Int. Ed. 2015, 54, 10216-10219.
Z. Huang, P. Hwang, D. S. Watson, L. Cao, F. C. Szoka, Jr.,
Bioconjug. Chem. 2009, 20, 1667-1672.
A. E. Backmark, N. Olivier, A. Snijder, E. Gordon, N. Dekker, A. D.
Ferguson, Protein Sci. 2013, 22, 1124-1132.
D. J. O'Shannessy, M. Brigham-Burke, K. K. Soneson, P. Hensley,
I. Brooks, Anal. Biochem. 1993, 212, 457-468.
K. N. Richter, N. H. Revelo, K. J. Seitz, M. S. Helm, D. Sarkar, R. S.
Saleeb, E. D'Este, J. Eberle, E. Wagner, C. Vogl, D. F. Lazaro, F.
Richter, J. Coy-Vergara, G. Coceano, E. S. Boyden, R. R. Duncan,
S. W. Hell, M. A. Lauterbach, S. E. Lehnart, T. Moser, T. F. Outeiro,
P. Rehling, B. Schwappach, I. Testa, B. Zapiec, S. O. Rizzoli, EMBO
J 2018, 37, 139-159.
trivalent chelator heads. Thus, we suggest
a
detailed
[19]
nomenclature for MCH molecules or to henceforth limit the simple
term trisNTA to cyclam-Glu-trisNTA (cyclic trisNTA) in order to
avoid obscurities. We hypothesize that, in combination with
established MCH probe live-cell delivery techniques, like cell-
squeezing or genetically encoded nanopores, these structural
details and optimized trivalent MCHs can potentially enable
chemically induced dimerization. Finally, these nanotools will
open up new perspectives for ultra-small, non-disturbing and
high-affinity labeling to gain deeper insights into dynamic cellular
structures and processes.
[20]
[21]
[22]
[23]
[24]
A. Kollmannsperger, A. Sharei, A. Raulf, M. Heilemann, R. Langer,
K. F. Jensen, R. Wieneke, R. Tampé, Nat Commun 2016, 7, 10372.
Acknowledgements
We thank Katrin Schanner for technical assistance, Drs. Ralph
Wieneke, Markus Braner, and the entire lab for the helpful
discussions.
This article is protected by copyright. All rights reserved.

10.1002/anie.201802746
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
And the winner is:
cyclic trisNTA
Karl Gatterdam, Eike F.
Joest, Volker Gatterdam,
and Robert Tampé*
The scaffold design
controls the interaction of
minimalistic trivalent NTA
chelators for in-vitro and
in-cell labeling of His-
tagged proteins.
Page No. – Page No.
Scaffold design of
trivalent chelator heads
dictates high-affinity and
stable His-tagged protein
labeling in vitro and in
cellulo
This article is protected by copyright. All rights reserved.