Two new chiral EDTA-based metal chelates for weak alignment of proteins in solution

Article abstract of DOI:10.1021/ol053049o

A short synthesis of EDTA-based metal chelates that can be attached to the cysteine residue of a protein via a bisulfide bond is described. The complexes were used after coordination of lanthanides to align trigger factor and apo-calmodulin in solution to

Full text of DOI:10.1021/ol053049o

ORGANIC
LETTERS
2006
Vol. 8, No. 7
1275-1278
Two New Chiral EDTA-Based Metal
Chelates for Weak Alignment of
Proteins in Solution
Peter Haberz, Fernando Rodriguez-Castan˜eda, Jochen Junker, Stefan Becker,
Andrei Leonov,* and Christian Griesinger*
Max-Planck Institute for Biophysical Chemistry, Department of NMR-Based Structural
Biology, Am Fassberg 11, 37077 Go¨ttingen, Germany
anle@nmr.mpibpc.mpg.de; cigr@nmr.mpibpc.mpg.de
Received December 16, 2005
ABSTRACT
A short synthesis of EDTA-based metal chelates that can be attached to the cysteine residue of a protein via a disulfide bond is described.
The complexes were used after coordination of lanthanides to align trigger factor and apo-calmodulin in solution to yield residual dipolar
couplings and pseudocontact shifts. Alignment tensors for the new tags are linearly independent compared to those of previously published
tags.
Residual dipolar couplings (RDCs) improve the speed and
accuracy of structure determination for biomolecules1 by
solution NMR. In addition, dynamics within globular pro-
teins2 or between protein domains3 can be measured using
RDCs.
Successful measurement of RDCs requires the partial
orientation of samples in a strong magnetic field using either
external alignment media or paramagnetic alignment of
diamagnetic proteins.
We and other groups have therefore developed paramag-
netic tags that can be attached to diamagnetic proteins either
as fusion proteins4 or via a covalently linked tag5 (Figure 1)
attached to cysteine. Therefore, we describe in this com-
munication the synthesis of another set of paramagnetic tags
whose alignment tensors differ from previously described
tags. Similar to the external alignment media, structures can
be defined more precisely, if more than one alignment tensor
(2) (a) Prestegard, J. H. Nat. Struct. Biol. 1998, 5, 517. (b) Lakomek, N.
A.; Carlomagno, T.; Becker S.; Meiler, J.; Griesinger, C. J. Biomol. NMR,
in press. (c) Lakomek, N. A.; Fare`s, C.; Becker, S.; Carlomagno, T.; Meiler,
J.; Griesinger, C. Angew. Chem., Int. Ed. 2005, 44, 7776. (d) Meiler, J.;
Prompers, J.; Peti, W.; Griesinger, C.; Bru¨schweiler, R. J. Am. Chem. Soc.
2001, 123, 6098. (e) Peti, W.; Meiler, J.; Bru¨schweiler, R.; Griesinger, C.
J. Am. Chem. Soc. 2002, 124, 5822.
(1) For reviews, see: (a) Blackledge, M. Prog. Nucl. Magn. Reson.
Spectrosc. 2005, 46, 23. (b) Prestegard, J. H.; Bougault, C. M.; Kishore,
A. I. Chem. ReV. 2004, 104, 3519. (c) Tolman, J. R. Curr. Opin. Struct.
Biol. 2001, 11, 532. (d) Bax, A.; Kontaxis, G.; Tjandra, N. Methods Enzymol.
2001, 339, 127. (e) Prestegard, J. H.; al-Hashimi, H. M.; Tolman, J. R. Q.
ReV. Biophys. 2000, 33, 371.
10.1021/ol053049o CCC: $33.50
© 2006 American Chemical Society
Published on Web 03/09/2006

condensation reagent to give the compound 5 in 66% yield.
The iodine oxidative sulfenylation6 of the sodium methane-
sulfinate with disulfide 5a afforded the protected thiosul-
fonate, which was deprotected without purification with
formic acid followed by HPLC purification to give the target
tag 6. In a similar manner, the enantiomeric acid ent-4 gave
the thiosulfonate ent-6.
The new tags were attached to the proteins apo-calmodulin
(apo-CaM)7 and trigger factor.8 An 15N,1H HSQC spectrum
of trigger factor is shown in Figure 2 for tag 6 loaded with
Figure 1. EDTA-based metal chelates.
can be implemented for the protein under investigation. In
addition, for the study of domain motions, it is essential to
align one domain by the paramagnetic tag and to study the
induced alignment on the others.3
The enantioselective synthesis of two novel tags is
described in Scheme 1. The optically active (R)-2,3-bis[di-
Scheme 1. Preparation of Tags 6
Figure 2. 15N,1H HSQC of the 15N-labeled S100C mutant of trigger
factor tagged with 6. The full loading of the tag with Dy3+ is
obvious from the absence of any peaks from the isotropic spectrum
of trigger factor. The inset shows the isotropic (red) and anisotropic
(blue) ω1-coupled resonances of A52.
Dy3+. The overview spectrum indicates the high quality of
the sample. No isotropic peaks were detected. NH resonances
even in close proximity to the tag site (C100) can be
observed, as for example Q99. This is because the metal is
farther away from the protein backbone for 6 and ent-6 (16
Å) than for the tags 3a and 3b (13 Å). Inlays show
expansions of the isotropic (red) and anisotropic (blue) ω1-
(3) (a) Tu¨chelmann, A.; Schwalbe, H.; Griesinger, C. Poster at the
Meeting on Stable Isotope Aided NMR of Biomolecules, Third European
Conference, Oxford 1998. (b) Bertini, I.; Del Bianco, C.; Gelis, I.; Katsaros,
N.; Luchinat, C.; Parigi, G.; Peana, M.; Provenzani, A.; Zoroddu, M. A.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 6841.
(4) (a) Feeney, J.; Birdsall, B.; Bradbury, A. F.; Biekofsky, R. R.; Bayley,
P. M. J. Biomol. NMR 2001, 21, 41. (b) Ma, C.; Opella, S. J. J. Magn.
Reson. 2000, 146, 381. (c) Wo¨hnert, J.; Franz, K. J.; Nitz, M.; Imperiali,
B.; Schwalbe, H. J. Am. Chem. Soc. 2003, 125, 13338.
(5) (a) Gaponenko, V.; Sarma, S. P.; Altieri, A. S.; Horita, D. A.; Li, J.;
Byrd, R. A. J. Biomol. NMR 2004, 28, 205. (b) Gaponenko, V.; Altieri, A.
S.; Li, J.; Byrd, R. A. J. Biomol. NMR 2002, 24, 143. (c) Ikegami, T.;
Verdier, L.; Sakhaii, P.; Grimme, S.; Pescatore, B.; Saxena, K.; Fiebig, K.
M.; Griesinger, C. J. Biomol. NMR 2004, 29, 339. (d) Leonov, A.; Voigt,
B.; Rodriguez-Castan˜eda, F.; Sakhaii, P.; Verdier, L.; Griesinger, C. Chem.-
Eur. J. 2005, 11, 3342.
(6) Fujiki, K.; Tanifuji, N.; Sasaki, Y.; Yokoyama, T. Synthesis 2002,
343.
(7) Watterson, D. M.; Sharief, F.; Vanaman, T. F. J. Biol. Chem. 1980,
255, 962.
(8) Crooke, E.; Wickner, W. Proc. Natl. Acad. Sci. U.S.A. 1987, 84,
5216.
(tert-butyloxycarbonylmethyl)amino]propionic acid 45d was
coupled to 4-aminophenyl disulfide, using HATU as the
1276
Org. Lett., Vol. 8, No. 7, 2006

coupled peaks of residue A52. The alignment tensors for
the previously published 3a and 3b as well as the new tags
6 and ent-6 are compared by determining their intertensor
angles, which are shown in Table 1 for both proteins trigger
Table 1. Axial (Da-HN) and Rhombic (Rhomb) Components
of the Alignment Tensors and Angles between Them Achieved
with the Four Paramagnetic Tags for Trigger Factora
tag1/tag2 deg angle
tag
Hz Da-HN rhomb Å distance
3a/3b
3a/6
69
147
157
120
119
8
3a
3b
6
ent-6
6
ent-6
4.1
4.4
4.3
4.1
3.2b
3.4b
0.56
0.38
0.35
0.47
0.48b
0.32b
13
13
16
16
17b
18b
Figure 4. Titration of 15N-labeled S17C mutant of apo-CaM tagged
with 6 and loaded with Tb3+. As is obvious from the titration, there
is no general broadening of the lines because the tag binds the
lanthanide quantitatively. The Ca2+ binding sites of apo-CaM
especially do not show any indication of binding.
3a/ent-6
3b/6
3b/ent-6
6a/ent-6
6a/ent-6
24b
a Distances from the tagged sulfur atom to the metal positions are given.
b These data belong to apo-CaM.
observed,9 indicating that all the metal is bound to the tag
and none to the protein. This is due to affinity of the tag to
lanthanides in the 10-18 M range, while CaM binds them
only with 10-9 M. High-quality ω1-coupled 15N,1H HSQC
spectra were recorded, and RDCs were extracted for 26
residues for 6 and 30 for ent-6. The RDCs of helix A were
excluded due to their poor fit to the alignment tensor derived
from the rest of the residues in the N-terminal domain. One
probable explanation is the dynamics of helix A with respect
to the rest of the domain that has been detected by
fluorescence anisotropy measurements.10 Further investiga-
tions of the dynamics of apo-CaM are under way. The metal
positions could be determined as shown in the stereoplot for
the tags 6 and ent-6 in Figure 5 from the pseudocontact shifts
factor and apo-CaM. The distances of the metal position from
the sulfur atom of the tagged cysteine are also listed.
Structure calculations have been performed in the case of
trigger factor with the information from all four alignment
tensors and are compared to the structure calculation
performed with only one alignment tensor (Figure 3). The
Figure 3. (a) The 10 lowest energy structures of trigger factor
calculated using the RDCs induced by the tag 6 loaded with Dy3+
as compared to (b) the 10 lowest energy structures of trigger factor
in which all four tags were used and loaded with Dy3+. The increase
of the precision of the structures is clearly visible and reflected in
the rmsd to the mean of 0.353 Å for (a) and 0.202 Å for (b).
Figure 5. Stereoplot of apo-CaM with the metals and the principal
axes of the alignment tensors indicated by axes in green, blue, and
red. The green dot indicates the position of the metal for 6, and
the orange for ent-6. The distance to the sulfur of cysteine (yellow)
is approximatley 17 Å.
structure from four alignment tensors is more precise (rmsd
to mean: 0.202 Å) than that calculated from only one
alignment tensor (rmsd to mean: 0.353 Å for 6) as can be
inferred from the rmsd of the well-folded regions for the
backbone atoms.
The second protein that was tagged with the two com-
pounds 6 and ent-6 is apo-CaM. A titration with the
lanthanide was done up to a concentration of 60% of the
apo-CaM (Figure 4). No shifts due to metalated CaM are
by fitting them to a previously solved NMR structure.11
The correlation plot has a Q-factor of 0.206 and a slope
of 0.938, which indicate the quality of the data as well as of
the derived alignment tensor (Figure 6).
(9) (a) Tjandra, N.; Kuboniwa, H.; Ren, H.; Bax, A. Eur. J. Biochem.
1995, 230, 1014. (b) Seeholzer, S. H.; Wand, A. J. Biochemistry 1989, 28,
4011.
Org. Lett., Vol. 8, No. 7, 2006
1277

single cysteine site that can be incorporated at any given
protein position. For trigger factor the additional alignment
tensors allowed us to improve the precision of the structure.
The extremely large affinity of the tag to lanthanides allows
investigation of proteins that have tight metal binding sites
such as apo-CaM and is therefore a versatile tag for all kinds
of proteins.
Acknowledgment. This work was supported by the MPG,
the DFG (SFB 492 Z-project), and the Fonds of the Chemical
Industry. We thank Kerstin Overkamp and Gerhard Wolf
for performing the HPLC purification and ESI-MS analyses.
We thank Karin Giller for site-directed mutagenisis. We
acknowledge measurement of ESI-HRMS spectra by Dr.
Holm Frauendorf, University of Go¨ttingen. Frank Hambloch,
University of Go¨ttingen, is acknowledged for conducting the
elemental analyses, and Evelyn Pfeil, University of Go¨ttin-
gen, for measurement of the specific rotations. We also want
to thank Dr. Christoph Fare`s, Dr. Karel Kubicek, and Dr.
Laurent Verdier for useful discussions.
Figure 6. Correlation plot of the pseudocontact shifts for the S17C
mutant of apo-CaM tagged with ent-6 loaded with Tb3+. The slope
is 0.938, and the Q-factor 0.206. The position of the metals was
determined by minimizing the Q-factor against a previously solved
NMR structure.11
In conclusion, we developed two new tags, inducing new
alignments that are linearly independent from those induced
by previously published tags. The tags are introduced at a
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(10) Chen, B.; Mayer, M. U.; Markillie, L. E.; Stenoien, D. L.; Squier,
T. C. Biochemistry 2005, 44, 905.
(11) Chou, J.; Li, S.; Bax, A. J. Biomol. NMR 2001, 18, 217.
OL053049O
1278
Org. Lett., Vol. 8, No. 7, 2006