Structure elucidation of host-guest complexes of tartaric and malic acids by quasi-racemic crystallography

Article abstract of DOI:10.1002/anie.201305364

Racemic on the outside, but not inside: Aromatic-foldamer hosts are enantiomers and as such prefer to co-crystallize even though the guests (e.g. L-tartaric acid, see picture) in each host are not present as enantiomers. This behavior allows a one-step structure elucidation of diastereomeric and quasi-racemic structures in the solid state. Copyright

Full text of DOI:10.1002/anie.201305364

Angewandte
Chemie
DOI: 10.1002/anie.201305364
Quasiracemates
Structure Elucidation of Host–Guest Complexes of Tartaric and Malic
Acids by Quasi-Racemic Crystallography**
Guillaume Lautrette, Brice Kauffmann, Yann Ferrand, Christophe Aube, Nagula Chandramouli,
Didier Dubreuil, and Ivan Huc*
The propensity of racemic solutions of organic molecules to
frequently produce racemic crystals and to only rarely resolve
into crystals that contain exclusively one enantiomer (con-
glomerates) has been known for a long time.[1] More than
a decade of foldamer chemistry has provided strong empirical
evidence that this observation also holds true for helical
aromatic foldamers. A vast number of racemic crystals that
contain both right-handed (P) and left-handed (M) helices
have been reported,[2,3] whereas conglomerates are extremely
rare.[4] When aromatic-foldamer helices are exclusively one-
handed through the introduction of chiral groups that control
helix handedness, our experience is that crystals do not grow
well and X-ray quality crystals that allow structure resolution
are rarely obtained.[5] In contrast, the corresponding race-
mates often crystallize readily. Thus, racemic mixtures have
purposely been used to assign absolute helix handedness of
aromatic foldamers induced by covalently appended chiral
residues[6] or by noncovalently bound chiral guests.[7] Racemic
crystallography has also been used to solve the structure of
some proteins that can be produced by chemical synthesis and
that were shown to crystallize more readily as a racemic pair
than as a single enantiomer.[8]
As an extension to racemic crystallography, quasi-race-
mates are sometimes found to co-crystallize, as was originally
described by Pasteur for malate and tartrate salts.[9] A quasi-
racemic crystal comprises a pair of molecules, the structures
of which are almost, but not exactly, mirror images.[9,10] Quasi-
racemic crystals of small proteins have recently been pro-
duced, allowing the resolution of the three-dimensional
structure of a slightly modified protein without having to
synthesize its enantiomer when the enantiomer of the
prototypical protein is available.[11] In helical aromatic
foldamers, quasi-racemic crystals have been encountered as
well, with a diastereomeric pair, a P helix and an M helix with
the same chiral terminal residue, co-crystallizing in an almost
perfectly centrosymmetric fashion.[6,12,13]
Herein, we describe the use of quasi-racemic crystallog-
raphy as an efficient method to simultaneously elucidate the
structures of diastereomeric complexes of tartaric and malic
acids, the same molecules that allowed the first description of
a quasi-racemic crystal by Pasteur 160 years ago,[9] encapsu-
lated into a helically folded host. We took advantage of the
high propensity of enantiomeric P and M helices to co-
crystallize, and demonstrated that co-crystals still form even
when the guests introduced in the cavities are not mirror
images.
We have previously shown that the aromatic oligoamide
sequence 1 (Figure 1) folds into a stable helix with a cavity
that binds to tartaric acid (2) with high affinity (Ka =
5300 LmolÀ1 in CDCl3/[D6]DMSO = 9:1 (v/v)) and high
diastereoselectivity: l-tartaric acid is preferentially bound in
M-1 and d-tartaric acid in P-1, with a diastereomeric excess
de > 99%.[7a] Upon mixing racemic tartaric acid with 1,
a racemic solution of P-1ꢀd-2 and M-1ꢀl-2 complexes was
[*] G. Lautrette, Dr. Y. Ferrand, Dr. N. Chandramouli, Dr. I. Huc
Universitꢀ de Bordeaux, CBMN, UMR5248
Institut Europꢀen de Chimie et Biologie
2 rue Escarpit, 33600 Pessac (France)
and
CNRS, CBMN, UMR5248 (France)
E-mail: i.huc@iecb.u-bordeaux.fr
Dr. B. Kauffmann
Universitꢀ de Bordeaux, UMS3033
Institut Europꢀen de Chimie et Biologie (IECB)
2 rue Escarpit, 33600 Pessac (France)
and
CNRS, IECB; UMS3033 (France)
and
INSERM, IECB, UMS3033 (France)
Dr. C. Aube, Prof. D. Dubreuil
Universitꢀ de Nantes, CEISAM, UMR6230
Facultꢀ des Sciences et des Techniques
2 rue de la Houssiniꢁre, BP 92208, 44322 Nantes Cedex 3 (France)
and
CNRS, CEISAM, UMR6230 (France)
[**] This work was supported by the Comitꢀ Interprofessionnel du Vin de
Bordeaux (CIVB) (predoctoral fellowship to G.L.) and by ANR grant
(ANR-09-BLAN-0082-01 and predoctoral fellowship to C.A.). We
thank Dr. Maurizot for providing terephthalic acid derivatives.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201305364.
Figure 1. Formula of hosts 1 and 3 together with structures of the
isomers of tartaric and malic acids.
Angew. Chem. Int. Ed. 2013, 52, 11517 –11520
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11517

.
Angewandte
Communications
formed. This racemate crystallized in the centrosymmetric
P21/n monoclinic space group, and its structure was solved,
showing an array of hydrogen bonds between the acid and
hydroxy functions of 2 and the inner wall of 1, consistent with
data obtained in solution.[7a] In another experiment, enantio-
merically pure tartaric acid l-2 was mixed with 1 and two
diastereomeric complexes P-1ꢀl-2 and M-1ꢀl-2 were formed
in solution.[14] Thus, the less favored complex P-1ꢀl-2 still
shows enough stability to form quantitatively at mm concen-
tration, as a kinetic supramolecular product (see below). We
called the P-1ꢀl-2 complex the mismatching complex as
opposed to the M-1ꢀl-2 complex, which has matching host–
guest stereochemistry. Intrigued by this phenomenon, we
sought for structural information about the mismatching
complex to shed light on the better complementarity and
higher stability of the matching complex.
The P-1ꢀl-2 plus M-1ꢀl-2 mixture may be considered to
be a quasi-racemate inasmuch as the two complexes are
expected to be mirror images for most part (the large helical
host) at the exclusion of the stereochemistry and orientation
of the small tartaric acid guest. As such, this mixture
represented a candidate to grow quasi-racemic crystals.
Unfortunately, the mismatching complex is a transient spe-
cies. As the P and M helices equilibrate through an unfold-
ing–refolding process, the mismatching complex converts into
the more stable matching complex.[14] Attempts to grow
quasi-racemic crystals before the mismatching complex dis-
appeared failed in our hands.
Figure 2. Schematic representation of the encapsulation of a) two
molecules of the l-enantiomer of tartaric acid and b) a racemic mixture
of d- and l-tartaric acid (left equilibrium) and a mixture of l-tartaric
acid and l-malic acid (right equilibrium) in host 3, which possesses
two cavities of opposite handedness. P and M handedness are
denoted in blue and red, respectively. The inversion center controlling
the stereochemistry of 3 is colored in light grey. The golden balls stand
for the hydroxy groups of the guests. Red-in-blue and blue-in-red
complexes are mismatching complexes.
In order to capture the mismatching complex, we
considered an original approach to lock helix handedness,
making use of an earlier design that allows to introduce
a
helix reversal within
a
helical aromatic-foldamer
sequence.[15] Oligomer 3 consists of two segments, each
derived from the sequence of host 1 connected at their
termini by a terephthalic acid unit (Figure 1, see also
Figure 2). As described in detail in the Supporting Informa-
tion, the preparation of 3 required the development of a new
synthetic approach of the host sequences to enable desym-
metrization and distinct functionalization at each end. The
intramolecular hydrogen-bonding motif at the central ter-
ephthalamide is such that the two host segments on each side
of the central unit have opposite handedness.[15] The overall
structure of 3 thus possesses two independent binding sites for
tartaric acid, one in a P helix, the other in an M helix. On
average, 3 possesses an inversion center, as reflected by the
multiplicity of 1H NMR signals (13 amide resonances, Fig-
ure 3a and Figure S1). The presence of both P and M seg-
ments were expected to facilitate its crystallization. This was
indeed the case and the structure of 3 was solved in the P1
space group with a single molecule in the unit cell (Figure 4a).
It is worth noting that while the two enantiomeric P- and M-
helical segments are important for crystallization, the result-
ing structure does not necessarily have a crystallographic
inversion center or plane of symmetry, as was also observed in
some racemic protein crystals.[16] The propensity of the P and
M helices to co-crystallize may not be directly related to the
availability of centrosymmetric space groups.
Figure 3. Parts of the 1H NMR spectra (700 MHz) of 3 (0.5 mm in
CDCl3:[D6]DMSO=99:1 (v/v)) at 298 K in the presence of a) 0 equiv of
guest; b) 2 equiv of d/l-2; c) 0.5 equiv; d) 1 equiv; e) 1.5 equiv; and
f) 2 equiv of l-2; g) 1 equiv of l-2 and 3 equiv of l-4. For 3ꢀ(2-d;2-l)
(b) signals of the carboxylic acid protons of 2 in the matching helices
are denoted with full black circles. For 3ꢀ(2-l)2 (c–g) signals of the
carboxylic acid protons of l-2 in the M helix (matching) are marked
with half-black circles when the P helix is empty, and with black circles
when the P helix also hosts a guest; signals of the carboxylic acid
protons of l-2 in the P helix (mismatching) are marked with a black
square. It should be noted that the two mismatching acid resonances
are at coalescence at this temperature. For 3ꢀ(2-l;4-l), signals of the
carboxylic acid protons of l-2 in the M helix (matching) are marked
with black circles and carboxylic acid protons of l-4 in the P helix
(matching) are denoted with black triangles.
1
All H NMR titrations of 3 were carried out in CDCl3,
using [D6]DMSO (1%) to dissolve the guest in the stock
11518 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11517 –11520

Angewandte
Chemie
Figure 4. a,b,e) Side view of the solid-state structure in the tube representation of a) 3, b) 3ꢀ(2-l)2, and e) 3ꢀ(2-l;4-l). c,d) Slice views of 3ꢀ(2-
l)2 with c) the left-handed cavity (M) showing guest 2-l with its hydroxy groups pointing toward the naphthyridines, and d) the right-handed cavity
(P) showing guest 2-l with its hydroxy groups pointing toward the pyridazine. f,g) Slice views of 3ꢀ(2-l;4-l) with f) the left-handed cavity (M)
showing guest 2-l, and g) the right-handed cavity (P) showing guest 4-l with its hydroxy group pointing toward the naphthyridines. The position
of malic acid in the P helix is disordered, leading to two partially occupied positions of its hydroxy oxygen atom (e,f: yellow). P and M helices are
denoted in blue and red, respectively. The central terephthalic acid unit is colored in grey. Tartaric and malic acid are represented as CPK (b,e) or
tube (c,d,f,g) structures. Isobutyl groups and included solvent molecules are not shown for clarity.
solution. As observed previously for 1,[7] binding and release
of the guest was slow on the NMR time scale at 298 K. A
titration of 3 using d/l-2 led to the emergence of a new set of
signals corresponding to a centrosymmetric 1:2 complex
(3ꢀ(2-d;2-l)) in which both cavities of 3 are filled with the
matching guest (Figure 2b, left). Evidence for this assignment
stemmed from the symmetry of the final complex (Figure S1,
13 amide resonances), which is the same as that of the empty
host, from the stoichiometry assigned by integrating the host
and guest signals in the complex, and from the chemical shift
values of the carboxylic acid protons of the guest at 13.5–
14 ppm, which are characteristic of a matching complex.[7]
During the titration, the formation of this 1:2 complex is
preceded by a 1:1 matching complex (i.e. 3ꢀ2-d or 3ꢀ2-l)
having one empty cavity and lacking any symmetry (26 amide
signals, Figure S2). Interestingly, NMR signals of the guest in
both 1:1 and 1:2 complexes show that its own symmetry is
broken upon encapsulation, regardless of the symmetry of the
overall structure. Distinct signals are observed for each
carboxylic acid proton because each cavity of 3 is itself
dissymmetric.
of the mismatching 2-l in the P helix of 3 (Figure 3, squares)
to be broad and to show less upfield-shifted resonances at
10.20 and 10.13 ppm at 273 K (Figure S4), thus indicating
weaker hydrogen bonding than in matching complexes.
Unlike 3ꢀ(2-d;2-l), 3ꢀ(2-l)2 is not centrosymmetric. Yet
these two 1:2 complexes differ only by the configuration of
one of the guests inside one of the two helical cavities of 3.
Thus, 3ꢀ(2-l)2 appeared to be a good candidate for a quasi-
racemic crystallographic analysis that would finally allow the
elucidation of the structure of a mismatching complex.
Indeed, single crystals of 3ꢀ(2-l)2 grew from the slow
diffusion of hexane into chloroform solutions and the
structure was solved in the P1 space group. In both the P
and M cavities, 2-l is positioned perpendicular to the helical
axis and adopts a conformation with the two acid and the two
hydroxy groups in trans and gauche position, respectively
(Figure 4b–d). In both mismatching and matching complexes,
carboxylic acid groups are doubly hydrogen bonded to the
neighboring 7-aminonaphthyridine (P helix dOH···N = 2.67 ꢀ,
d
NH···OC = 3.11 ꢀ; M helix dOH···N = 2.79 ꢀ, dNH···OC = 2.95 ꢀ)
and also hydrogen bonded to the aromatic pyridazine protons
A second titration experiment consisted in the addition of
a single enantiomer of tartaric acid 2-l to host 3 (Figure 2a).
The addition of 0.5 equivalents of 2-l led to the appearance of
a new set of signals with twice the number of resonances
compared to 3 alone, corresponding to a 1:1 complex in which
2-l was fully encapsulated in the M cavity and not in the
P cavity. Again, the two sharp signals at 13.96 and 13.63 ppm
are those of the hydrogen-bonded acid protons of the guest in
a matching configuration and surrounded by a dissymmetric
environment (Figure 3c–e, half-circles). Upon further addi-
tion of 2-l, resonances corresponding to 3ꢀ2-l were replaced
by another set of signals corresponding to 3ꢀ(2-l)2 with both
a matching and a mismatching complex. Using 13C and 15N-
heteronuclear single-quantum coherence (HSQC) spectros-
copy, we identified the signals of the carboxylic acid protons
(P helix dCH···OC = 3.33 ꢀ; M helix dCH···OC = 3.11 ꢀ). Tartaric
= À À
acid adopts an anti arrangement of the O C C OH bonds in
the matching configuration, whereas the syn conformer is
observed in the mismatching configuration. Overall, the guest
2-l in its mismatching configuration has to shift its position in
the cavity, moving away from the pyridazine central unit in
order to accommodate its hydroxy groups in between. In
contrast, in the matching complex, the CH groups of the guest
point toward the pyridazine.
A step further in our study consisted in exploiting the
centrosymmetric nature of 3 to serve the crystallographic
analysis of the complexes it may form with two different
guests. Malic acid (4; Figure 1) also binds to 1, though with
lower affinity (Ka = 70 LmolÀ1 in CDCl3/[D6]DMSO = 9:1 (v/
v)) and poorer diastereoselectivity (de = 52%) than tartaric
Angew. Chem. Int. Ed. 2013, 52, 11517 –11520
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11519

.
Angewandte
Communications
2006, 71, 1131 – 1138; f) H.-Y. Hu, J.-F. Xiang, Y. Yang, C.-F.
Chen, Org. Lett. 2008, 10, 69 – 72; g) A. J. Blake, P. A. Cooke,
K. J. Doyle, S. Gair, N. S. Simpkins, Tetrahedron Lett. 1998, 39,
9093 – 9096.
acid. A consequence of the poor diastereoselectivity is the
coexistence of matching and mismatching complexes, even
when using racemic malic acid, which has prevented crystal-
lization and structural analysis until now. l-malic acid (l-4)
differs from l-tartaric acid (l-2) by the lack of one hydroxy
group and by the absolute configuration of its only stereo-
genic center. The two acids are thus quasi-enantiomers and
their successful combination in quasi-racemic crystals was
demonstrated by Pasteur long ago[9a] and recently fully
characterized.[9b] We thus endeavored to crystallize both
acids in complex with 3.
The addition of l-2 (1 equiv) and l-4 (3 equiv) to 3
resulted in the almost quantitative formation of a single
species 3ꢀ(2-l;4-l) (Figure 3g, see also Figure 2b, right). This
hetero-trimolecular complex features four low-field resonan-
ces that correspond to the carboxylic acid protons of
encapsulated l-2 and l-4, both in a matching complex.
Upon cooling to 243 K, each of the two signals (at 13.63 and
13.36 ppm) assigned to carboxylic acid protons of l-4
broadened and then split into two (Figure S7), consistent
with two different orientations of l-4 within the P helix of 3.
Again, quasi-racemic crystallography proved successful to
analyze the structure of encapsulated l-4, an otherwise
elusive complex. X-ray quality crystals of 3ꢀ(2-l;4-l) were
obtained and its structure solved in the P1 space group. Both
2-l and 4-l were found to adopt a matching anti conformation
and to be similarly hydrogen bonded to the 7-aminonaph-
thyridine (2-l in M helix dOH···N = 2.80 ꢀ, dNH···OC = 3.08 ꢀ; 4-
l in P helix dOH···N = 2.78 ꢀ, dNH···OC = 3.08 ꢀ) and to the
pyridazine protons (dCH···OC = 3.25 ꢀ and dCH···OC = 3.26 ꢀ,
respectively). As observed in solution, 4-l exists in two
disordered positions in the solid state, arising from its
dissymmetric structure. This disorder relates to the disorder
of malic acid in its quasi-racemic crystals with tartaric acid.[9b]
Our investigation thus adds to very recent reports that
quasi-racemic crystallography can be used deliberately to
analyze X-ray structures that would be difficult to elucidate
otherwise, including those of host–guest complexes.[11]
[4] To the best of our knowledge, only two examples have been
reported: a) E. Ohta, H. Sato, S. Ando, A. Kosaka, T. Fukush-
ima, D. Hashizume, M. Yamasaki, K. Hasegawa, A. Muraoka, H.
Ushiyama, K. Yamashita, T. Aida, Nat. Chem. 2011, 3, 68 – 73;
b) H. Zhao, W. Qiang Ong, F. Zhou, X. Fang, X. Chen, S. F. Y.
Li, H. Su, N.-J. Cho, H. Zeng, Chem. Sci. 2012, 3, 2042 – 2046.
[5] a) A. M. Kendhale, L. Poniman, Z. Dong, K. Laxmi-Reddy, B.
Kauffmann, Y. Ferrand, I. Huc, J. Org. Chem. 2011, 76, 195 – 200;
b) H. Masu, M. Sakai, K. Kishikawa, M. Yamamoto, K.
Yamaguchi, S. Kohmoto, J. Org. Chem. 2005, 70, 1423 – 1431;
c) Y. Tanaka, H. Katagiri, Y. Furusho, E. Yashima, Angew.
Chem. 2005, 117, 3935 – 3938; Angew. Chem. Int. Ed. 2005, 44,
3867 – 3870.
[6] C. Dolain, H. Jiang, J.-M. Lꢁger, P. Guionneau, I. Huc, J. Am.
Chem. Soc. 2005, 127, 12943 – 12951.
[7] a) Y. Ferrand, A. M. Kendhale, B. Kauffmann, A. Grꢁlard, C.
Marie, V. Blot, M. Pipelier, D. Dubreuil, I. Huc, J. Am. Chem.
Soc. 2010, 132, 7858 – 7859; b) Q. Gan, Y. Ferrand, N. Chan-
dramouli, B. Kauffmann, C. Aube, D. Dubreuil, I. Huc, J. Am.
Chem. Soc. 2012, 134, 15656 – 15659.
[8] a) L. E. Zawadzke, J. M. Berg, Proteins Struct. Funct. Genet.
1993, 16, 301 – 305; b) C. Toniolo, C. Peggion, M. Crisma, F.
Formaggio, X. Shui, D. S. Eggleston, Nat. Struct. Biol. 1994, 1,
908 – 914; c) B. L. Pentelute, Z. P. Gates, V. Tereshko, J. L.
Dashnau, J. M. Vanderkooi, A. Kossiakoff, S. B. H. Kent, J. Am.
Chem. Soc. 2008, 130, 9695 – 9701.
[9] a) L. Pasteur, Ann. Chim. Phys. 1853, 28, 437 – 483; b) K. A.
Wheeler, R. C. Grove, R. E. Davis, W. S. Kassel, Angew. Chem.
2008, 120, 84 – 87; Angew. Chem. Int. Ed. 2008, 47, 78 – 81.
[10] a) I. L. Karle, J. Karle, J. Am. Chem. Soc. 1966, 88, 24 – 27; b) A.
Fredga, Bull. Soc. Chim. Fr. 1973, 1, 173 – 182; c) A. M. Line-
berry, E. T. Benjamin, R. E. Davis, W. S. Kassel, K. A. Wheeler,
Cryst. Growth Des. 2008, 8, 612 – 619; d) M. E. Breen, S. L.
Tameze, W. G. Dougherty, W. S. Kassel, K. A. Wheeler, Cryst.
Growth Des. 2008, 8, 3863 – 3870; e) R. C. Grove, S. H. Male-
horn, M. E. Breen, K. A. Wheeler, Chem. Commun. 2010, 46,
7322 – 7324; f) B. Calmuschi, M. Alesib, U. Englert, Dalton
Trans. 2004, 1852 – 1857.
[11] a) D. E. Mortenson, K. A. Satyshur, I. A. Guzei, K. T. Forest,
S. H. Gellman, J. Am. Chem. Soc. 2012, 134, 2473 – 2476; b) K.
Mandal, B. L. Pentelute, D. Bang, Z. P. Gates, V. Yu. Torbeev,
S. B. H. Kent, Angew. Chem. 2012, 124, 1510 – 1515; Angew.
Chem. Int. Ed. 2012, 51, 1481 – 1486.
Received: June 21, 2013
Published online: September 6, 2013
[12] H. Jiang, C. Dolain, J.-M. Lꢁger, H. Gornitzka, I. Huc, J. Am.
Chem. Soc. 2004, 126, 1034 – 1035; see also the asymmetric unit
of compound 3b in Ref. [5b].
Keywords: crystallography · foldamers · helical structures ·
host–guest systems · quasi-racemic
.
[13] These quasi-racemic structures actually showed better statistics
ꢀ
when resolved in the centrosymmetric space group P1, in which
the number of parameters to refine is twice as small as in the
correct space group P1.
[1] J. Jacques, A. Collet, S. H. Wilen in Enantiomers, racemates and
resolutions, 3rd ed., Krieger, Malabar, 1994.
[2] For reviews: D.-W. Zhang, X. Zhao, J.-L. Hou, Z.-T. Li, Chem.
Rev. 2012, 112, 5271 – 5316; I. Huc, Eur. J. Org. Chem. 2004, 17 –
29.
[3] a) V. Berl, I. Huc, R. Khoury, M. J. Krische, J.-M. Lehn, Nature
2000, 407, 720 – 723; b) H. Jiang, J.-M. Lꢁger, I. Huc, J. Am.
Chem. Soc. 2003, 125, 3448 – 3449; c) N. Delsuc, S. Massip, J.-M.
Lꢁger, B. Kauffmann, I. Huc, J. Am. Chem. Soc. 2011, 133, 3165 –
3172; d) B. Gong, H. Zeng, J. Zhu, L. Yuan, Y. Han, S. Cheng, M.
Furukawa, R. D. Parra, A. Y. Kovalevsky, J. L. Mills, E.
Skrzypczak-Jankun, S. Martinovic, R. D. Smith, C. Zheng, T.
Szyperski, X. C. Zeng, Proc. Natl. Acad. Sci. USA 2002, 99,
11583 – 11588; e) Z.-Q. Hu, H.-Y. Hu, C.-F. Chen, J. Org. Chem.
[14] Y. Ferrand, N. Chandramouli, A. M. Kendhale, C. Aube, B.
Kauffmann, A. Grꢁlard, M. Laguerre, D. Dubreuil, I. Huc, J.
Am. Chem. Soc. 2012, 134, 11282 – 11288.
[15] V. Maurizot, C. Dolain, Y. Leydet, J.-M. Lꢁger, P. Guionneau, I.
Huc, J. Am. Chem. Soc. 2004, 126, 10049 – 10052.
[16] a) K. Mandal, B. L. Pentelute, V. Tereshko, V. Thammavongsa,
O. Schneewind, A. A. Kossiakoff, S. B. H. Kent, Protein Sci.
2009, 18, 1146 – 1154; b) W. R. Patterson, D. H. Anderson, W. F.
DeGrado, D. Cascio, D. Eisenberg, Protein Sci. 1999, 8, 1410 –
1422; c) L. W. Hung, M. Kohmura, Y. Ariyoshi, S. H. Kim, J.
Mol. Biol. 1999, 285, 311 – 321.
11520 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 11517 –11520