Silver-promoted desilylation catalyzed by ortho- and allosteric cucurbiturils

Article abstract of DOI:10.1021/ol100667z

Cucurbit[6]-, [7]-, and [8]uril can catalyze the silver(I)-promoted desilylation of trimethylsilylalkynyl-containing pseudorotaxanes by stabilizing a key π-alkynyl silver intermediate through favorable interactions between the metallic cation and the cavi

Full text of DOI:10.1021/ol100667z

ORGANIC
LETTERS
2010
Vol. 12, No. 10
2310-2313
Silver-Promoted Desilylation Catalyzed
by Ortho- and Allosteric Cucurbiturils
Xiaoyong Lu and Eric Masson*
Department of Chemistry and Biochemistry, Ohio UniVersity, Athens, Ohio 45701
masson@ohio.edu
Received March 21, 2010
ABSTRACT
Cucurbit[6]-, [7]-, and [8]uril can catalyze the silver(I)-promoted desilylation of trimethylsilylalkynyl-containing pseudorotaxanes by stabilizing
a key π-alkynyl silver intermediate through favorable interactions between the metallic cation and the cavitand carbonylated portal.
Recently, the scientific community celebrated the 30th
anniversary of the discovery of the famous p53 protein, a
tumor suppressor critical to the body’s anticancer defense.
An issue1 of Nature ReViews Cancer was dedicated to the
occasion, which revisited the p53 structure,1a its numerous
targets,1b its effects on apoptosis regulation,1c its role in
metabolism,1d and the various strategies to design p53-
binding drugs.1e Impressed by the phenomenal efficiency
of this 393 amino acid-long choreographer in the cellular
ballet, we sought to design a small organic guest, which
could interact with different targets (in analogy to p53
interacting with the MDM2 regulator, DNA, etc.) at
specific binding sites (similar to p53 transactivation and
DNA-binding domains, for example) and which could
undergo chemical alteration by a third partner upon such
an interaction (in analogy to the interaction with MDM2,
which promotes the ubiquitylation of p53,1a,b thereby
triggering its proteasomal degradation).
Cucurbit[6]-, [7]-, and [8]uril cavitands2a (CB[n], n ) 6
7, 8) were selected as our targets. These macrocycles share
a common pumpkin-shaped structure and contain 6, 7, or 8
glycoluril motifs, respectively, linked by methylene bridges.2
The high degree of similarity between the targets was
deliberately intended and adds a level of complexity to the
recognition process.
The isobutylammonium cation has been shown to interact
strongly with CB[6];2f trimethylsilyl groups have proven to
fit tightly into the cavity of CB[7] (affinity up to 8.9 × 108
M-1),2d and CB[8] can encompass larger or multiple guests
due to its wider diameter. In most cases, a positive charge,
such as an ammonium or a pyridinium group located close
to the CB[n] portal, is necessary for strong host-guest
affinity through ion-dipole interaction.2 Consequently, we
designed guest 1a, which was found to display (1) perfect
site selectivity toward CB[6] and CB[7] as well as good site
(2) (a) Freeman, W. A.; Mock, W. L.; Shih, N. Y. J. Am. Chem. Soc.
1981, 103, 7367. (b) Isaacs, L. Chem. Commun. 2009, 6, 619. (c) Lagona,
J.; Mukhopadhyay, P.; Chakrabarti, S.; Isaacs, L. Angew. Chem., Int. Ed.
2005, 44, 4844. (d) Liu, S.; Ruspic, C.; Mukhopadhyay, P.; Chakrabarti,
S.; Zavalij, P. Y.; Isaacs, L. J. Am. Chem. Soc. 2005, 127, 15959. (e) Kim,
J.; Jung, I. S.; Kim, S. Y.; Lee, E.; Kang, J. K.; Sakamoto, S.; Yamaguchi,
K.; Kim, K. J. Am. Chem. Soc. 2000, 122, 540. (f) Masson, E.; Lu, X.;
Ling, X.; Patchell, D. L. Org. Lett. 2009, 11, 3798.
(1) (a) Whibley, C.; Pharoah, P. D. P.; Hollstein, M. Nat. ReV. Cancer
2009, 9, 90. (b) Menendez, D.; Inga, A.; Resnick, M. A. Nat. ReV. Cancer
2009, 9, 724. (c) Levine, A. J.; Oren, M. Nat. ReV. Cancer 2009, 9, 749.
(d) Vousden, K. H.; Ryan, K. M. Nat. ReV. Cancer 2009, 9, 691. (e) Brown,
C. J.; Lain, S.; Verma, C. S.; Fersht, A. R.; Lane, D. P. Nat. ReV. Cancer
2009, 9, 862.
10.1021/ol100667z  2010 American Chemical Society
Published on Web 04/26/2010

Figure 1. Consecutive site- and host-selective encapsulation of guest
1a with CB[8], CB[6], and CB[7].
selectivity toward CB[8], (2) strict exclusivity, since only
one CB unit at a time can interact with the host, and (3)
excellent discrimination between targets: CB[8] can be
ejected from guest 1a upon addition of CB[6], which in turn
can be expelled by adding CB[7] (see Figure 1).
Figure 2
.
1H NMR spectra of (a) free guest 1a, (b) 1a⊂CB[6], (c)
1a⊂CB[7], and (d) 1a⊂CB[8] (500 MHz; 1.0 mM in D2O, in the
presence of 20 mM sodium nitrate).
1
Selective recognition could be readily assessed using H
nuclear magnetic resonance spectroscopy (1H NMR), since
(1) hydrogen atoms located near the center of the CB[n]
cavity undergo a strong upfield shift (up to 1.6 ppm);3a (2)
decentered hydrogens are affected by a moderate upfield
shift, which becomes weaker as hydrogen atoms get closer
to the CB[n] portals (1.5 f 0.1 ppm);3b and (3) hydrogens
outside the cavity undergo a significant downfield shift (up
to 0.7 ppm) that weakens as the distance between the
hydrogens and the portal increases.3c The isobutyl unit of
guest 1a interacts strongly with CB[6], since hydrogen atoms
at positions a, b, and c undergo a strong upfield shift (0.75,
1.08, and 0.45 ppm, respectively), and hydrogens at position
d are moderately shifted downfield (0.24 ppm, see Figure 2,
spectrum b). Hydrogen atoms e are too far from CB[6] and
are not affected. The aromatic hydrogen at position f, the
closest to the CB[6] rim, is significantly shifted downfield
(0.54 ppm), and the resolution of g-i signals improves from
a multiplet in the absence of CB[6] to three distinctive signals
with readable multiplicities (i appears as a doublet and g
and h show two triplet-like signals).
CB[7] targets the trimethylsilyl unit of guest 1a selectively
(see Figure 2, spectrum c), and hydrogen atoms at position
e undergo a strong upfield shift (0.77 ppm). While the
moderate downfield shift of hydrogens d is expected (0.25
ppm), the similar behavior of hydrogens b (0.26 ppm) is due
to the particular conformation of assembly 1a⊂CB[7], where
hydrogens b apparently sit at a short distance of the CB[7]
rim. In the aromatic region of the spectrum, hydrogen atom
i interacts with the CB[7] portal and its doublet undergoes
an exceptionally high downfield shift (0.90 ppm), while
hydrogens f-h remain unaltered.
up into its cavity.4d Here we show that CB[8] encapsulates
both the aryl and the isobutyl units of guest 1a, since
hydrogens a-d and f-i are all shifted upfield (0.15-0.31
ppm). Trimethylsilyl hydrogens e are barely affected by
CB[8], indicating that they are most probably located at a
significant distance from the cavitand. Since (1) upfield shifts
are mild and (2) signals undergo significant broadening, we
suspect a weak interaction and possibly the presence of a
mixture of rapidly equilibrating interlocked assemblies. While
CB[6] and CB[7] were found to undergo slow exchange with
guest 1a, in accordance with previous reports,2d,f CB[8] and
guest 1a exchange quickly on the NMR time scale.
In order to fulfill the p53 protein analogy, one of the
binding sites of guest 1a must be able to undergo chemical
alteration, and the rate of the alteration must be affected by
the neighboring binding sites and hosts. The trimethylsily-
lalkynyl group satisfies these requirements, since it can be
desilylated to afford phenylacetylene derivative 2a. In order
to prevent deprotonation of the ammonium unit, we opted
for a silver nitrate-promoted desilylation under acidic condi-
tions, as recently described by Pale et al.5 The following
mechanism has been proposed: (1) the silver cation interacts
with the trimethylsilylalkynyl unit to form a π-complex
intermediate; (2) a σ-alkynyl silver intermediate is obtained
upon displacement of the trimethylsilyl group by a nucleo-
phile (such as methanol, subsequently liberating a proton);
(3) the alkynyl silver intermediate is hydrolyzed under acidic
conditions to afford the desilylated product, and the free
silver cation is regenerated.5a
(4) For a set of very recent publications, see: (a) Andersson, S.; Zou,
D.; Zhang, R.; Sun, S.; Aakermark, B.; Sun, L. Eur. J. Org. Chem. 2009,
8, 1163. (b) Hwang, I.; Ziganshina, A. Y.; Ko, Y. H.; Yun, G.; Kim, K.
Chem. Commun. 2009, 4, 416. (c) Rajgariah, P.; Urbach, A. R. J. Incl.
Phenom. Macrocyclic Chem. 2008, 62, 251. (d) Ko, Y. H.; Kim, H.; Kim,
Y.; Kim, K. Angew. Chem., Int. Ed. 2008, 47, 4106.
As described on numerous occasions, CB[8] can bind to
multiple guests4a-c or can incite long alkyl chains to curl
(3) (a) Moon, K.; Kaifer, A. E. Org. Lett. 2004, 6, 185. (b) Mock, W. L.;
Shih, N. Y. J. Org. Chem. 1986, 51, 4440. (c) Buschmann, H. J.; Wego,
A.; Zielesny, A.; Schollmeyer, E. J. Inclusion Phenom. Macrocyclic Chem.
2006, 54, 241.
(5) (a) Halbes-Letinois, U.; Weibel, J. M.; Pale, P. Chem. Soc. ReV.
2007, 36, 759. (b) Viterisi, A.; Orsini, A.; Weibel, J. M.; Pale, P.
Tetrahedron Lett. 2006, 47, 2779.
Org. Lett., Vol. 12, No. 10, 2010
2311

The effect of CB[6], CB[7], and CB[8] on the desilylation
of guest 1a in deuterated water was consequently studied.
While desilylation in the presence of a catalytic amount of
silver nitrate has been reported,5 we opted for a large excess
of the metallic cation (5.0 equiv) in order to avoid an
additional level of complexity when evaluating the effect of
CB[n] units on the desilylation kinetics. The reaction was
found to be quantitative, and the concentration of byproduct
trimethylsilanol-d1 (5; see Figure 4) could be readily
1
monitored by H NMR in the presence of a known amount
of an inert reference compound for quantification (N,N-
dimethylformamide). Due to the bulkiness of the CB[n]
cavitands, we expected that CB[6] and CB[7] would shield
the alkynyl unit from silver attack and that the remote CB[8]
would barely prevent desilylation. Amazingly, the exact
opposite effect was observed (see Figure 3a), and these
Figure 4. Plausible cycle for CB[n]-catalyzed desilylations in the
presence of Ag(I) salts in aqueous medium.
the trimethylsilyl unit (water probably crosses the CB[7]
portal and reaches the interior of the cavity before displacing
the trimethylsilyl group); products of the substitution are
CB[7]-bound trimethylsilanol-d1 5⊂CB[7], D+ cations and
presumably alkynylsilver 4, in analogy with the mechanism
proposed by Pale et al.;5 (4) alkynylsilver 4 is hydrolyzed
in the presence of D+ and phenylacetylene derivative 2a is
obtained quantitatively, while CB[7] liberates trimethylsil-
anol-d1 (5) and can interact with guest 1a as a new cycle
starts (the affinity of guest 1a toward CB[7] was found to
be at least 103 times higher than silanol 5).
A similar mechanism can take place when CB[7] is
replaced with CB[6] or CB[8] at allosteric positions, since
in both cases the carbonylated rim of the cavitand can
stabilize the ternary guest/CB[n]/silver complex which is
responsible for the desilylation rate enhancement. In both
cases, trimethylsilanol-d1 does not interact with the cavitands,
which become available for the next catalytic cycle im-
mediately after desilylation. One should also note that an
additional series of concomitant host-guest interactions
dramatically increase the complexity of these catalytic
processes: upon desilylation in the presence of CB[6], the
latter may subsequently interact with the isobutyl unit of
guest 1a and participate in a next round of catalysis, but it
will also bind to the isobutyl unit of product 2a, which thus
acts as a catalyst poison. This effect is particularly damaging
at higher concentrations of product 2a, and it may explain
its rapid formation during the first 100-300 s, followed by
a noticeable retardation (see Figure 3a). A similar effect could
be observed with CB[8], which can encapsulate the pheny-
lacetylene unit of product 2a, and to a far lesser extent with
CB[7], which displays a low affinity toward the isobutyl unit
of product 2a.2f
Figure 3. Yield of the Ag(I)-promoted desilylation of (a) guest 1a
and (b) guest 1b, as a function of time (s), in the absence of CB[n]
cavitands (black dot), as well as in the presence of 10% CB[6]
(red dot), 50% CB[6] (red open dot), 10% CB[7] (blue dot), and
10% CB[8] (green dot). The yield is derived from the amount of
trimethylsilanol-d1 formed during the reaction.
experiments constitute the first rationalized examples of an
organometallic reaction catalyzed by CB[n] cavitands.6 A
10% catalytic amount of CB[7] was found to increase the
rate of desilylation by 4.3 times (half-life of guest 1a in
the absence of CB[n] t1/2 ) 1.5 × 103 and 3.4 × 102 s in the
presence of 10% equiv CB[7]), allosteric CB[6] increased it
by a factor of 1.5 (t1/2 ) 1.0 × 103 s) when 10% CB[6] was
used, and by a factor of 6.6 in the presence of 50% CB[6]
(t1/2 ) 2.2 × 102 s); a similar enhancement is observed when
CB[6] is replaced by CB[8] (t1/2 ) 1.1 × 103 s when 10%
CB[8] is present, a 1.4-time rate increase).
We propose the following mechanism for the catalytic
process in the presence of CB[7] (see Figure 4): (1) a fraction
of guest 1a interacts with CB[7]; (2) silver cations form
π-complex 3⊂CB[7] with assembly 1⊂CB[7]; favorable
interactions between silver and the oxygen lone pairs of the
CB[n] portal stabilize the complex; (3) assembly 3⊂CB[7]
undergoes a nucleophilic substitution, when water displaces
The results described above indicate undoubtedly that
CB[n] plays a crucial role during the formation of the
alkynylsilver π-complexes. However, one may question the
existence of assembly 3⊂CB[n] as the key component of
(6) Catalytic hydrogenation of 1-hexanal and 1-octanal in the presence
of CB[6] and ruthenium(II) has been reported; however, the role of CB[6]
has not been described. See: Karakhanov, E. A.; Karapetyan, L. M.;
Kardasheva, Y. S.; Maksimov, A. L.; Runova, E. A.; Terenina, M. V.;
Filippova, T. Y. Macromol. Symp. 2008, 270, 106.
2312
Org. Lett., Vol. 12, No. 10, 2010

the catalytic cycle and argue that the free fraction of CB[n]
cavitands, although present at extremely low concentration
and possibly coordinated with silver cations, may be
responsible for the rate enhancements (i.e., CB[n] would not
have to interact with guest 1a in order to catalyze its
desilylation).
In order to discriminate between both mechanisms, we
prepared guest 1b, which lacks the CB[6] binding site, and
we subsequently determined the effect of CB[6], CB[7], and
CB[8] on its desilylation rate. As expected, CB[6] does not
interact with guest 1b (see Figure 5, spectrum b), and CB[7]
aromatic hydrogens undergo mild upfield and downfield
shifts, and their location cannot be attributed with any degree
of certainty.
While desilylation of guest 1a was enhanced by a factor
of 6.6 with 50% CB[6], no effect was observed in the
desilylation of guest 1b (t1/2 ) (1.0-1.1) × 103 in the absence
or presence of CB[6]), thereby indicating that CB[6] must
be bound to the guest in order to catalyze its desilylation.
As expected, 10% CB[7] catalyzes the reaction (by a factor
of 3.1; t1/2 ) 3.3 × 102 s vs 1.0 × 103 s in the absence of
cavitand). CB[8] was found to have almost no effect on the
desilylation rate of guest 1b (t1/2 ) 8.5 × 102 s in the
presence of 10% CB[8]); since several 1b⊂CB[8] assemblies
are present in the reaction mixture, we suspect that some of
them favor the desilylation reaction, and others inhibit it,
especially if the alkynyl unit is trapped inside the cavitand.
The structure of the key intermediates 3⊂CB[n] is also
supported by the fact that the asymmetric coordination of
silver to the CB[6] portal has been reported once.7 In
addition, CB[n] cavitands do interact in a similar way with
other metallic cations, including cesium,8a calcium,8b zinc,8b
strontium,8b cadmium,8c and some lanthanides.8d-f While this
study was intended as a proof of concept, we think that it
will trigger the development of new CB[n]-containing
catalytic systems, in particular for metal-catalyzed coupling
reactions in aqueous medium.
Acknowledgment. This work was supported by the
Department of Chemistry and Biochemistry, the College of
Arts and Sciences and the Vice President for Research at
Ohio University, Athens, OH.
Figure 5
.
1H NMR spectra of (a) free guest 1b, (b) 1b⊂CB[6], (c)
1b⊂CB[7], and (d) 1b⊂CB[8] (500 MHz; 1.0 mM in D2O, in the
presence of 20 mM sodium nitrate).
Supporting Information Available: Preparation and
characterization of CB[n] guests 1a, 1b, 2a, and 2b, detailed
characterization of interlocked assemblies, and procedures
for kinetic studies. This material is available free of charge
via the Internet at http://pubs.acs.org.
displays strong affinity toward its trimethylsilyl unit; hydro-
gens e are shifted upfield by 0.77 ppm, and hydrogens c, d
and i, close to the CB[7] rim, undergo downfield shifts (0.06,
0.20, and 0.79 ppm, respectively). The interaction between
CB[8] and guest 1b is ill-defined; the trimethylsilyl singlet
e of free guest 1b splits into two shielded and broad signals
upon interaction with CB[8] (upfield shifts of 0.61 and 0.88
ppm; 8:1 ratio), indicating the presence of at least two slowly
interconverting assemblies, in which the trimethylsilyl unit
is encapsulated; hydrogens c undergo a similarly distributed
split, and a very weak upfield shift. Hydrogens d are shifted
upfield by 0.06 ppm (an unusual value, which suggests
neither encapsulation, nor localization at the CB[n] portal,
or possibly a rapidly equilibrating mixture at both locations);
OL100667Z
(7) Wang, Z. B.; Zhao, M.; Li, Y. Z.; Chen, H. L. Supramol. Chem.
2008, 20, 689.
(8) (a) Whang, D.; Heo, J.; Park, J. H.; Kim, K. Angew. Chem., Int. Ed.
1998, 37, 78. (b) Zhang, Y. Q.; Zhen, L. M.; Yu, D. H.; Zhao, Y. J.; Xue,
S. F.; Zhu, Q. J.; Tao, Z. J. Mol. Struct. 2008, 875, 435. (c) Feng, X.; Lu,
X. J.; Xue, S. F.; Zhang, Y. Q.; Tao, Z.; Zhu, Q. J. Inorg. Chem. Commun.
2009, 12, 849. (d) Mainicheva, E. A.; Tripolskaya, A. A.; Gerasko, O. A.;
Naumov, D. Y.; Fedin, V. P. Russ. Chem. Bull. 2006, 55, 1566. (e) Thue´ry,
P. Inorg. Chem. 2009, 48, 4497. (f) Gerasko, O. A.; Mainicheva, E. A.;
Naumova, M. I.; Yurjeva, O. P.; Alberola, A.; Vicent, C.; Llusar, R.; Fedin,
V. P. Eur. J. Inorg. Chem. 2008, 416.
Org. Lett., Vol. 12, No. 10, 2010
2313