Catalysis with anion-π interactions

Article abstract of DOI:10.1002/anie.201305356

The conclusion is inevitable: Increasing stabilization of an anionic transition state with increasing π-acidity of the catalyst is observed; thus, anion-π interactions can contribute to catalysis.

Full text of DOI:10.1002/anie.201305356

Angewandte
Chemie
DOI: 10.1002/anie.201305356
Anion–p Interactions
Catalysis with Anion–p Interactions**
Yingjie Zhao, Yuya Domoto, Edvinas Orentas, Cꢀsar Beuchat, Daniel Emery, Jiri Mareda,
Naomi Sakai, and Stefan Matile*
The expansion of the number of intermolecular interactions
available to create molecular functional systems is of para-
mount importance. Quite recently, we have identified syn-
thetic transport systems as attractive tools to elaborate on
interactions that are otherwise difficult to detect.[1–4] Realized
examples include anion–p interactions,[1,2] halogen bonds,[2,3]
and anion–macrodipole interactions.[4] Intriguing results with
transport promised attractive applications to catalysis,
because evidence for anion binding in the ground state
implied that anionic transition states could be similarly
stabilized. Anion–p interactions[5–14] were particularly inter-
esting for this purpose because wonderful examples exist for
catalysis with complementary cation–p interactions,[15] reach-
ing from carbocation stabilization in terpenoid and steroid
cyclization to surprisingly rare and recent use in organo-
catalysis.[16] Anion–p interactions, however, have essen-
tially[5–7] not been used in catalysis.[5–14] This is understandable,
because experimental evidence for their functional relevance
appeared only recently,[1] and discussions concerning their
nature and significance continue.[5–14] The poor development
of the field presumably originates from the limited occur-
rence, availability, and diversity of the required p-acids, that is
Figure 1. Catalysis of the Kemp elimination with anion–p interactions.
A carboxylate is placed as general base near the p-acidic surface of
aromatic rings with strong enough electron-withdrawing
substituents to invert their usually negative quadrupole
moments into positive ones.
catalyst C to 1) couple deprotonation with the onset of anion–p
interactions for transition-state (TS) stabilization, and 2) protonate the
phenolate in the reactive intermediate (RI) to avoid product inhibition.
blue=electron deficient, red=electron rich, S=substrate, P=product,
CS=catalyst–substrate complex, CP=catalyst–product complex.
The Kemp elimination is an established tool to develop
conceptually innovative catalysts.[17–20] Useless with regard to
applications in organocatalysis, this reaction has served well
to elaborate on theoretically designed enzymes, catalytic
antibodies, promiscuous proteins, synthetic polymers, macro-
cyclic model systems, vesicles, micelles, and non-specific
medium effects.[17–20] The key step is the deprotonation of
a carbon in the benzisoxazole substrate S by a general base
(Figure 1). The reaction then proceeds with a single anionic
transition state to afford the nitrophenolate either as inter-
mediate or product, depending on conditions. There is general
agreement that catalysis in its most general sense occurs by
transition-state stabilization.[21] The anionic nature of the
transition state thus qualified the Kemp elimination as a valid
tool to identify contributions from anion–p interactions to
catalysis. Herein, we report that p-acidic naphthalenediimides
(NDIs)[1] with a covalently attached carboxylate base can
catalyze the Kemp elimination and, most importantly, that the
stabilization of the anionic transition state of this trans-
formation increases with increasing p-acidity of the new
catalysts.
The key to “anion–p catalysis” was to take the p-acidic
surface of an NDI (variable and strong), and to attach
a carboxylate base on one side and a solubilizing tail on the
other side (Figure 2). With this design, p-stacking between
substrate and catalyst should hold throughout the trans-
formation. The onset of anion–p interactions between the
compound in transformation and the catalyst C (Figure 1),
however, should coincide exactly with the key step, that is the
injection of a negative charge from the proximal carboxylate
into the substrate. The translocation of this negative charge
over five atoms (from the carboxylate oxygen to the
[*] Dr. Y. Zhao, Dr. Y. Domoto, Dr. E. Orentas, Dr. C. Beuchat,
Dr. D. Emery, Dr. J. Mareda, Dr. N. Sakai, Prof. S. Matile
Department of Organic Chemistry, University of Geneva
Geneva (Switzerland)
E-mail: stefan.matile@unige.ch
Homepage: http://www.unige.ch/sciences/chiorg/matile/
[**] We thank the NMR and MS platforms for services, the University of
Geneva, the European Research Council (ERC Advanced Investi-
gator), the National Centre of Competence in Research (NCCR)
Chemical Biology and the Swiss NSF for financial support, and the
Swiss National Supercomputing Center (CSCS) in Lugano-Cornar-
edo for CPU time. E.O. acknowledges a Sciex Fellowship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201305356.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Kemp elimination in the presence of the catalysts was
1
continuously followed by H NMR spectroscopy. In a typical
experiment, substrate S and catalysts C were dissolved in
CD3OD at different concentrations and ratios. The reaction
was initiated by partial deprotonation of C with 0.5 equiv-
alents of tetrabutylammonium hydroxide (TBAOH). Initial
velocities of product formation were measured first as
a function of catalyst concentration (Figure 3a). Product
Figure 2. Structure of the operational catalysts C1 and C2 together
with control molecules C3–C5.
benzisoxazole oxygen) on the p-acidic surface is a powerful
expression of operational anion–p interactions in the tran-
sition state. Stabilization by anion–p interactions should
continue with the similarly anionic phenolate in the reactive
intermediate RI and vanish only with the neutral phenol in
CP. Acidified by intramolecular anion–p interactions with the
NDI surface, the carboxylic acid in catalyst C should be strong
enough to protonate the weakly basic nitrophenolate in
RI,[12,20] less acidic ammonium cations, pyridinium cations,
thiols, or phenols would fail to do so.
To elaborate on possible contributions of anion–p inter-
actions to catalysis, the collection of candidates and controls
C1–C5 was considered (Figure 2). Based on established
procedures, their synthesis was very straightforward. Details
can be found in the Supporting Information. NDIs C1–C3
were selected to explore anion–p catalysis because their p-
acidity is very high.[1] Unsubstituted NDIs with peripheral
phenyl substituents already have a quadrupole moment
(Qzz =+ 19 B) that is in the range of p-acids such as the
explosive trinitrotoluene (TNT). Analogous NDIs with two
cyano groups in the core, as in catalyst C2 (Qzz =+ 39 B), are
probably the strongest organic p-acids known today.[1] This
increase in Qzz naturally coincides with a decreasing LUMO
energy of À4.31 eV for unsubstituted NDIs such as C1 to
À4.78 eV for dicyano NDIs such as C2.
Figure 3. a) Initial velocity of product formation as a function of the
*
*
concentration of C1 ( ), C3 (&) and C4 ( ); S (13 mm), TBAOH
(0.5 equiv), CD3OD, room temperature. b) Initial velocity of product
formation as a function of the concentration of substrate S in the
&
*
*
presence of 8.3 mm C1 ( ), C2 ( ) and C5 ( ); TBAOH (5.0 mm),
*
CD3OD/CDCl3 1:1, room temperature; with linear ( ) or Michaelis–
&
*
Menten ( , ) curve fit.
Pyrenebutyrate C4 (Qzz = À14 B) was selected as a p-
basic control. It is almost as p-basic as the native NDI C1 is p-
acidic, and operational cation–p interactions on the p-basic
surface have been suspected in the context of cell-penetrating
peptides.[22] This was an important choice because theore-
tically designed enzymes, catalytic antibodies, and synthetic
model systems all contain p-basic groups in their active
site.[17–20] Extensive computational studies have suggested that
these p-bases could serve to stabilize the transition state.[17–20]
This conclusion is surprising because, from p-bases, one would
expect ground-state stabilization of catalyst–substrate or
catalyst–product complexes, whereas interactions in the
anionic transition state should be repulsive with p-bases and
attractive with p-acids (Figure 1). Control C3 features a fully
contracted and rigidified bridge between NDI and carboxyl-
ate, control C5 contains all structural motifs of C1 and C2
except for the p-acidic naphthalenes.
formation in the presence of the p-acidic NDI C1 was clearly
faster than in the presence of controls C3 and C4. C3 is thus
too rigid to attain the optimum geometry in the substrate–
catalyst complex, and transition-state stabilization with the p-
basic C4 is as ineffective as expected. Turnovers were
followed for up to 13 substrates per catalyst C1, more
should be possible without any problems.
The dependence on the substrate concentration at con-
stant catalyst concentration in 1:1 CD3OD/CDCl3 revealed
saturation behavior for the p-acidic catalyst C1, but not for
*
*
the close control C5 (Figure 3b,
vs. ). This finding
was important; it demonstrated the formation of catalyst–
substrate complex CS1, whereas CS5 is too weak to be
detected under the same conditions. A Km of 82.5 Æ 7.9 mm
was determined by Michaelis–Menten analysis, which trans-
lated into a ground-state stabilization of DDGGS = 6.2 Æ
0.2 kJmolÀ1 for CS1 (Figures 1 and 4).
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Figure 5. Molecular model of the transition state. Optimized geometry
(M06L/6-311G**) for TS2 is shown in the same orientation as in
Figures 1 and 4. Electrostatic potential surface (blue=positive, red=
negative; À262.6/+170.1 kJmolÀ1) computed at the MP2/6-311G**//
M06 L/6-311G** level highlights the electron-transfer pathway on the
p-acidic surface of the NDI (the branched alkyl substituent is replaced
by a methyl group).
Figure 4. Energy diagram for the Kemp elimination catalyzed with
anion–p interactions. Ground-state stabilization DDGGS (Km) and
transition-state stabilization DDGTS (KTS) for C1 and C2 obtained from
Michaelis–Menten analysis (compare with Figure 3b).
The transition-state stabilization by anion–p catalyst C1
tional analysis of the quite complex situation is ongoing and
will be reported in due course.
was estimated from Michaelis–Menten analysis.[21] A rate
enhancement of kcat/knon = 7606, a catalytic efficiency of kcat
Km = 6.5 ꢀ 10À3 mÀ1 sÀ1 and a catalytic proficiency of (kcat/Km)/
non = 9.2 ꢀ 104 mÀ1 were obtained en route to a transition-state
/
This study provides experimental evidence for contribu-
tions of anion–p interactions to catalysis. The presence of a p-
acidic surface in the catalyst is shown to stabilize the anionic
transition state of the selected reaction. Most importantly,
increasing p-acidity of the catalyst increases the stabilization
of the anionic transition state. This finding demonstrates that
anion–p interactions contribute to catalysis, the exact mode of
anion binding is irrelevant for the validity of this conclusion.
Naturally delocalized and enhanced by p–p interactions,
these interactions are necessarily beyond the strict definition
of pure anion–p interactions. They encourage continuation of
the thoughts formulated concerning nitrate recognition[1,13] by
possible contributions from p–p interactions and complement
evolution in the perception of cation–p interactions, partic-
ularly when applied to catalysis.[15,16]
In conclusion, the herein reported experimental evidence
for contributions of anion–p interactions to catalysis enriches
our understanding of organocatalysis and will lead to
conceptually innovative design strategies to stabilize anionic
transition states. Ongoing studies with modified, sulfur-
containing NDI catalysts[14] confirm the general validity of
increasing transition-state stabilization with increasing p-
acidity with regard to the Kemp elimination. Preliminary
results further indicate that anion–p interactions will become
applicable to the stabilization of the anionic tetrahedral
intermediates of addition and substitution reactions on
carbonyl groups. Enolate chemistry is particularly appealing
for anion–p catalysis, also because their importance in
polyketide biosynthesis is nicely complementary to carbocat-
ion chemistry in terpenoid and steroid biosynthesis. However,
in sharp contrast to the cation–p interactions contributing to
the latter, catalysis with anion–p interactions is new and
clearly moves beyond the grand principles operating in
nature.[6]
k
stabilization
K
TS = 10.9 Æ 1.6 mm, which translated into
DDGTS = 28.3 Æ 0.4 kJmolÀ1 for TS1 (Figure 4).
To further elaborate on the relevance of anion–p inter-
actions for catalysis with C1, we decided to synthesize catalyst
C2 with maximized p-acidity. As with C1, the catalysis of the
Kemp elimination with C2 showed saturation behavior
&
(Figure 3b, ). According to Michaelis–Menten analysis,
ground-state stabilization increased by 0.9 kJmolÀ1 to
DDGGS = 7.1 Æ 0.3 kJmolÀ1. With increasing p-acidity of the
catalyst, transition-state stabilization increased more than
twice as much (2.0 kJmolÀ1; Figure 4). This is a very reason-
able value.[5–14] The DDGTS = 30.3 Æ 0.4 kJmolÀ1 corresponds
to a transition-state recognition by the most p-acidic catalyst
C2 with an apparent dissociation constant of KTS = 4.9 Æ
0.8 mm. The catalytic proficiency more than doubled from
(kcat/Km)/knon = 9.2 ꢀ 104 mÀ1 for C1 to (kcat/Km)/knon = 2.0 ꢀ
105 mÀ1 for C2.
Molecular models of the anionic transition state TS2 were
computed using the M06-2X/6-311G**//M06L/6-311G** level
of theory.[23,24] In all convincing structures, the electron flow
from the carboxylate to the benzisoxazole oxygen occurs on
the p-acidic surface (Figure 5). In agreement with operational
anion–p interactions, the distance between the electron-
transfer cascade and the p-acidic surface decreases from
3.347 ꢁ in CS2 to 3.290 ꢁ in TS2 and finally to 3.247 ꢁ in RI2.
Structure analysis suggests that TS2 is an early transition state
with C···H and O···H distances of 1.223 and 1.430 ꢁ,
respectively. The carboxylate base of the catalyst is found
on top of the electron-deficient area of the pyridinedione
heterocycle. One oxygen atom is on the way to accept the
proton from the isoxazole ring of the substrate, the other
forms an O···H-C interaction with the phenyl ring of the
substrate. The formation of the carbanion in the isoxazole
ring, the critical step of this reaction,[17–20] is stabilized on top
of one aromatic ring of the naphthalene. A detailed computa-
Received: June 21, 2013
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[10] L. M. Salonen, M. Ellermann, F. Diederich, Angew. Chem. 2011,
123, 4908 – 4944; Angew. Chem. Int. Ed. 2011, 50, 4808 – 4842.
[11] D.-X. Wang, M.-X. Wang, J. Am. Chem. Soc. 2013, 135, 892 – 897.
[12] C. J. Cadman, A. K. Croft, Beilstein J. Org. Chem. 2011, 7, 320 –
328.
[13] L. Adriaenssens, C. Estarellas, A. Vargas Jentzsch, M. Martinez
Belmonte, S. Matile, P. Ballester, J. Am. Chem. Soc. 2013, 135,
8324 – 8330.
Keywords: kinetics · molecular recognition ·
noncovalent interactions · organocatalysis · transition states
.
[1] R. E. Dawson, A. Hennig, D. P. Weimann, D. Emery, V.
Ravikumar, J. Montenegro, T. Takeuchi, S. Gabutti, M. Mayor,
J. Mareda, C. A. Schalley, S. Matile, Nat. Chem. 2010, 2, 533 –
538.
[2] A. Vargas Jentzsch, D. Emery, J. Mareda, P. Metrangolo, G.
Resnati, S. Matile, Angew. Chem. 2011, 123, 11879 – 11882;
Angew. Chem. Int. Ed. 2011, 50, 11675 – 11678.
[3] A. Vargas Jentzsch, D. Emery, J. Mareda, S. K. Nayak, P.
Metrangolo, G. Resnati, N. Sakai, S. Matile, Nat. Commun.
2012, 3, 905.
[4] A. Hennig, L. Fischer, G. Guichard, S. Matile, J. Am. Chem. Soc.
2009, 131, 16889 – 16895.
[5] C. M. Rojas, J. Rebek, Jr., J. Am. Chem. Soc. 1998, 120, 5120 –
5121.
[6] C. Estarellas, A. Frontera, D. QuiÇonero, P. M. Deyꢂ, Angew.
Chem. 2011, 123, 435 – 438; Angew. Chem. Int. Ed. 2011, 50, 415 –
418.
[7] P. Phuengphai, S. Youngme, I. Mutikainen, J. Reedijk, Inorg.
Chem. Commun. 2012, 24, 129 – 133.
ˇ
[14] J. Mꢃsek, A. Vargas Jentzsch, S. Sakurai, D. Emery, J. Mareda, S.
Matile, Angew. Chem. 2010, 122, 7846 – 7849; Angew. Chem. Int.
Ed. 2010, 49, 7680 – 7683.
[15] D. A. Dougherty, Acc. Chem. Res. 2013, 46, 885 – 893.
[16] S. Lin, E. N. Jacobsen, Nat. Chem. 2012, 4, 817 – 824.
[17] D. Rçthlisberger, O. Khersonsky, A. M. Wollacott, L. Jiang, J.
DeChancie, J. Betker, J. L. Gallaher, E. A. Althoff, A. Zanghel-
lini, O. Dym, S. Albeck, K. N. Houk, D. S. Tawfik, D. Baker,
Nature 2008, 453, 190 – 195.
[18] Y. Hu, K. N. Houk, K. Kikuchi, K. Hotta, D. Hilvert, J. Am.
Chem. Soc. 2004, 126, 8197 – 8205.
[19] A. J. Kennan, H. W. Whitlock, J. Am. Chem. Soc. 1996, 118,
3027 – 3028.
[20] F. Hollfelder, A. J. Kirby, D. S. Tawfik, J. Org. Chem. 2001, 66,
5866 – 5874.
[21] R. Wolfenden, M. J. Snider, Acc. Chem. Res. 2001, 34, 938 – 945.
[22] N. Sakai, S. Matile, J. Am. Chem. Soc. 2003, 125, 14348 – 14356.
[23] Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 194101.
[24] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215 – 241.
[8] A. Frontera, P. Gamez, M. Mascal, T. J. Mooibroek, J. Reedijk,
Angew. Chem. 2011, 123, 9736 – 9756; Angew. Chem. Int. Ed.
2011, 50, 9564 – 9583.
[9] H. T. Chifotides, K. R. Dunbar, Acc. Chem. Res. 2013, 46, 894 –
906.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Anion–p Interactions
The conclusion is inevitable: Increasing
stabilization of an anionic transition state
with increasing p-acidity of the catalyst is
observed; thus, anion–p interactions can
contribute to catalysis.
Y. Zhao, Y. Domoto, E. Orentas,
C. Beuchat, D. Emery, J. Mareda, N. Sakai,
S. Matile*
&&&& — &&&&
Catalysis with Anion–p Interactions
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!