Selective monoterpene-like cyclization reactions achieved by water exclusion from reactive intermediates in a supramolecular catalyst

Article abstract of DOI:10.1021/ja308254k

A polyanionic supramolecular assembly (1) is shown to catalytically cyclize the monoterpene citronellal and two homologues. In contrast to cyclization in acidic aqueous solution, the hydrophobic interior of 1 prevents the capture of reactive intermediates by water. This effect was also observed in the gold-catalyzed cycloisomerization of an enyne. Due to the steric confinement of the catalyst's interior, Prins cyclizations in 1 proceed cleanly both for substrates containing and lacking gem-dimethyl substitution. Encapsulation in 1 consequently imposes a degree of mechanistic control that, similar to enzyme catalysis, is not observed in bulk aqueous solution.

Full text of DOI:10.1021/ja308254k

Communication
pubs.acs.org/JACS
Selective Monoterpene-like Cyclization Reactions Achieved by Water
Exclusion from Reactive Intermediates in a Supramolecular Catalyst
William M. Hart-Cooper, Kristen N. Clary, F. Dean Toste,* Robert G. Bergman,*
and Kenneth N. Raymond*
Chemical Sciences Division, Lawrence Berkeley National Laboratory, and Department of Chemistry, University of California,
Berkeley, California 94720, United States
S
* Supporting Information
Scheme 1. Biosynthesis of Limonene and α-Terpineol from
Geranyl Diphosphate
a
ABSTRACT: A polyanionic supramolecular assembly (1)
is shown to catalytically cyclize the monoterpene
citronellal and two homologues. In contrast to cyclization
in acidic aqueous solution, the hydrophobic interior of 1
prevents the capture of reactive intermediates by water.
This effect was also observed in the gold-catalyzed
cycloisomerization of an enyne. Due to the steric
confinement of the catalyst’s interior, Prins cyclizations
in 1 proceed cleanly both for substrates containing and
lacking gem-dimethyl substitution. Encapsulation in 1
consequently imposes a degree of mechanistic control
that, similar to enzyme catalysis, is not observed in bulk
aqueous solution.
a
PPi = diphosphate. While limonene is obtained through a
erpene synthases are enzymes that generate over 60 000
deprotonation route, capture of the α-terpinyl cation with water
1
Tsmall-molecule natural products from simple precursors.
affords α-terpineol.4−7
These enzymes catalyze cascading 1,5-diene cyclization
reactions that proceed through carbenium ion intermediates.2
Noncovalent interactions, such as cation−π stabilization and
steric repulsion, dictate the conformations of intermediates and
resulting product distributions.3,4 Although terpene synthases
can be highly selective, product distributions containing
multiple species are common. Contingent on the nature of
the enzyme’s active site, these intermediates may undergo
eventual deprotonation or nucleophilic capture (e.g., by water)
to furnish the final products.5−7 An example is the conversion
of geranyl diphosphate to limonene and α-terpineol via the α-
terpinyl cation, as illustrated in Scheme 1.
a
Scheme 2. Encapsulation of 2 by Host 1
Synthetic systems have modeled the selectivity and efficiency
of enzymes.8 Recent advances in supramolecular catalysis
demonstrate the potential for these systems to effect high rate
enhancements9−11 and a capacity for regulation12 reminiscent
of enzyme catalysis. The Raymond group has developed a
water-soluble, chiral metal−ligand assembly of K12Ga4L6
stoichiometry (L = N,N-bis(2,3-dihydroxybenzoyl)-1,5-diami-
nonaphthalene; polyanion 1 represented in Scheme 2).13
Bearing analogy to the active sites of terpene synthases,14,7
the constrictive steric interior of polyanion 1 is defined by
cation-stabilizing aromatic moieties. Combined with the
assembly’s high negative charge, this property has been
demonstrated to bring about pKa shifts for encapsulated
guests.15 Assembly 1 has consequently been shown to catalyze
proton-mediated processes in basic solution.16 Notably, 1
catalyzes the Nazarov cyclization of 1,3-pentadienols with rate
a
Spheres represent a Ga3+ center, and bisbidentate ligands are depicted
as lines.
accelerations on the order of 106 relative to background
reactivity, which has been attributed to transition-state binding
as well as substrate conjugate acid stabilization.17
Given the cation-stabilizing and hydrophobic properties of
both the interior of 1 and the active sites of terpene synthases,
we were eager to investigate a monoterpene cyclization in 1.
The monoterpene ( )-citronellal (2) has been shown to
cyclize in the presence of Brønsted acids and is a relevant
Received: August 20, 2012
Published: October 15, 2012
© 2012 American Chemical Society
17873
dx.doi.org/10.1021/ja308254k | J. Am. Chem. Soc. 2012, 134, 17873−17876

Journal of the American Chemical Society
Communication
a
Scheme 3. Proton-Mediated Cyclization of 2 to Products 3−5
a
Under catalysis by acidic solution, 3−5 are observed with 4 as the major product by Path B. Catalysis with 1 affords 3a−d as the major class of
products, demonstrating that Path A is instead favorable.
industrial intermediate in the manufacture of menthol.18 We
Scheme 4. Selectivity of Alkene Products from the
hypothesized that 1 would stabilize the conjugate acid of
encapsulated 2, driving protonation at the aldehyde oxygen and
subsequent cyclization, the latter process being accelerated by
the constrictive interior of 1. Herein we report our studies of a
catalytic cyclization of 2 and two homologues (6a, 7a) in a
water-soluble supramolecular assembly at moderate temper-
atures and physiological pH.
Cyclization of 2 by 1
It has been reported that three classes of products are formed
when 2 is treated with buffered acidic solution, as depicted in
Scheme 3.19−21 We confirmed this experimentally: in addition
to minor products isopulegol (3a), neoisopulegol (3b),
isoisopulegol (3c), and neoisoisopulegol (3d), a mixture of
four stereoisomeric p-menthane-3,8-diols (4) is observed as the
major class of products. Once formed, 4 may undergo
condensation with 2 to generate p-menthane-3,8-diol citronellal
acetal stereoisomers (5).20 We observed 4 to be composed of
predominantly cis isomers (cis:trans, 3:2); minor components
3a−d contained mostly trans products (cis:trans, 3:7). These
distributions are consistent with a mechanistic divergence
occurring at or before ring closing of 2. Treating 3a−d with a
buffered solution of sufficient acidity to induce cyclization (pH
3.20, 60 °C, 2 h)21 yielded a product mixture identical to that
formed from the starting material, demonstrating that 3a−d are
persistent in an acidic aqueous environment at this pH and do
not convert to corresponding p-menthane-3,8-diols. On the
basis of these experiments, we propose that the cyclization of 2
in acidic solution involves the two pathways leading to 3a−d
and 4 illustrated in Scheme 3.
Trace amounts of the stereoisomeric mixture 4 were also
observed, the presence of which can be accounted for by
background reactivity.23 Isolated 4 did not undergo any
transformation (e.g., dehydration) when treated with 1. Thus,
in contrast to cyclization in acidic solution, alkene products
3a−d form with high selectivity upon treatment of 2 with 1
(Table 1).
Table 1. Cyclization of 2 to 3−5 by 1 and Buffered Acidic
a
Solution
selectivity (%)
entry
catalyst
pH
conv. (%)
3a−d
4
5
b
d
1
1
7.50
3.20
71
91
97
9
3
<1
<1
c
2
KH2PO4
91
We assessed whether 1 could cyclize 2 under stoichiometric
conditions. Encapsulation of 2 by 1 was confirmed through 1H
NMR analysis of an aqueous mixture containing 2 and 1, which
exhibited a set of broad resonances shifted upfield by 1−3 ppm.
Compound 2 was treated with an equivalent of 1 and the
mixture heated for 18 h. Upon extraction and 1H NMR
analysis, we observed the quantitative consumption of 2
accompanied by new resonances corresponding to 3a−d.
a
Conversion and selectivity assessed by 1H NMR. Selectivity
determined as a proportion of the identified product. Aqueous
b
solutions contained 50 mM phosphate buffer for both trials. 10 mol
c
d
% 1, 60 °C, 28 h. 50 °C, 8 h. Product not observed by 1H NMR or
GC-MS.
When incorporated in the backbone of acyclic substrates
capable of undergoing cyclization, gem-dimethyl substitution
has been shown to bring about increased cyclization rates (the
gem-dimethyl effect) and, in some cases, product selectivity.24 It
has been hypothesized that this effect arises from conforma-
tionally destabilized ground states of gem-dimethyl-substituted
substrates compared to those lacking substitution.25 Similarly,
the efficiency of certain enzyme-catalyzed cyclizations has been
attributed in part to conformational control of the bound
substrate by the enzyme active site.26 For example, limonene
synthase is thought to bind intermediate linalyl diphosphate in
+
Adding a slight excess of PEt4 , which is strongly encapsulated
by 1, halted this conversion. It was thus clear not only that the
unblocked cavity of 1 was necessary for stoichiometric reactivity
with 2, but also that trace quantities of free ligand or Ga3+ could
not be responsible for the observed transformation.22 Treating
2 with 10 mol % of 1 resulted in the catalytic conversion of 2 to
stereoisomeric products 3a−d (Scheme 4). Due to the low
solubility of 2, the reaction mixture was heterogeneous and
stirred vigorously. The ratio of cis to trans product did not differ
appreciably from that observed for 3a−d in acidic solution.
17874
dx.doi.org/10.1021/ja308254k | J. Am. Chem. Soc. 2012, 134, 17873−17876

Journal of the American Chemical Society
Communication
a cisoid conformation, facilitating electrocyclization to an α-
terpinyl cation.7
selectivity toward trans-alkene products, regardless of whether
gem-dimethyl substitution is present at the β-position.
Given the conceptual similarity between the gem-dimethyl
effect and certain instances of enzyme catalysis, we next
examined the effect of gem-dimethyl substitution on product
selectivity. The structure of 2 was varied by replacing −Me and
−H β-substituents with dihydro or dimethyl substitution,
affording achiral homologues 6a and 7a (Scheme 5). The effect
Having established enzyme-like selectivity in the Prins
cyclizations of 2, 6a, and 7a, we then investigated whether 1
would impart similar selectivity during transition-metal-
mediated transformations. We have recently reported the
gold(I) host−guest complex Me3PAu+⊂1 (where ⊂ denotes
encapsulation) to be a viable catalyst for the hydroalkoxylation
of allenes in water.28 Gold-catalyzed cycloisomerizations of 1,6-
enynes have been well documented to result in different
products depending on reaction conditions and substituent
effects.29 In the absence of assembly 1, Me3PAuCl catalyzed the
cycloisomerization of 8 to 10, which was obtained in 85% yield
(Scheme 6).30 When the cavity of 1 was blocked by strongly
Scheme 5. Proton-Mediated Cyclization of 6a and 7a to
Products 6b−e and 7b−e
Scheme 6. Gold-Catalyzed Cycloisomerization of Enyne 8 to
Products 9 and 10
of these substitutions on product selectivity became apparent
when 6a and 7a were treated with acidic solution. Like 2, 6a
cyclized to predominantly stereoisomeric diols, 6d and 6e.
However, in the absence of gem-dimethyl substitution, 7a
formed a complex mixture of products.27 In contrast, when
treated with 1, both 6a and 7a cyclized to predominantly trans-
alkene products (6b and 7b, respectively), demonstrating that
encapsulation in 1 affords conformational control during
cyclization (Table 2).
The trans product selectivity of 6a in acidic solution is
presumably the result of a 1,3-diaxial repulsion between the
aldehyde oxygen and axially oriented β-methyl group in the
transition state leading to cis product. The complex product
mixture observed upon treating 7a with acidic buffer
demonstrates that in acidic solution, alternate reaction
pathways are competitive with cyclization when gem-dimethyl
substitution is absent at the β-position. In light of the very
different product selectivity observed between 6a and 7a
following acidic solution treatment, the tendency for these
substrates to stereoselectively form alkene products in 1 is
surprising, given that both bulk solution and cluster catalysis are
proton-mediated processes. While the presence of gem-
dimethyl substitution vastly improves the product selectivity
obtained from acidic solution catalysis, this discrepancy is
eliminated with 1. In the latter case, overriding steric repulsion
experienced by the guest during encapsulation confers high
+
bound NEt4 , compound 10 was likewise observed as the sole
product. Use of Me3PAuBr as a catalyst resulted in a lower yield
of 10, presumably due to the relatively strong gold−bromide
bond. However, following treatment of 8 with Me3PAu+⊂1, 9
was instead produced as the major product. Preparing the
encapsulation complex Me3PAu+⊂1 from Me3PAuBr instead of
Me3PAuCl did not have a significant effect on the selectivity of
this process, and again 60:40 mixtures of products 9:10
resulted. The tendency for 1 to exclude water from reactive
intermediates was thus demonstrated for a gold-catalyzed
cycloisomerization of enyne 8.
In conclusion, we report the first example of a terpene
cyclization by a water-soluble supramolecular catalyst at
physiological pH. In analogy with the active sites of many
terpene synthases, 1 directs the cyclization of monoterpene 2
toward deprotonation instead of nucleophilic capture by
water.31 The generality of this property was demonstrated in
the gold-catalyzed cycloisomerization of enyne 8. We attribute
this effect to the hydrophobic environment of the assembly’s
cavity, which prevents water from capturing carbenium ion
intermediates during catalysis. Identification of 3a−d is of
interest, as these compounds are frequently used in the
asymmetric synthesis of complex natural products.32 The
synthesis of 3a from 2 is conventionally accomplished using
organic solvents and Lewis acids, where dehydration and
a
Table 2. Cyclization of 6a and 7a by 1 and Buffered Acidic Solution
selectivity (%)
entry
substrate
catalyst
pH
conv. (%)
6b/7b
6c/7c
14
6d/7d
6e/7e
b
e
1
6a (R = Me)
6a (R = Me)
7a (R = H)
7a (R = H)
1 (10 mol %)
KH2PO4
7.50
3.20
7.50
3.20
91
80
83
3
3
<1
c
e
2
<1
11
75
22
2
b
e
3
1 (10 mol %)
KH2PO4
60
87
<1
d
f
f
f
f
4
>95
nd
nd
nd
nd
a
Conversion and selectivity assessed by 1H NMR. Selectivity determined as a proportion of the identified product. Aqueous solutions contained 50
b
c
d
e
f
1
mM phosphate buffer for all trials. 60 °C, 28 h. 60 °C, 18 h. 60 °C, 20 h. Product not observed by H NMR or GC-MS. Complex product
mixture obtained; selectivity was not determined.
17875
dx.doi.org/10.1021/ja308254k | J. Am. Chem. Soc. 2012, 134, 17873−17876

Journal of the American Chemical Society
Communication
dimerization products are often observed.33,34 In contrast,
catalysis by 1 provides an environmentally benign method to
afford products of synthetic and economic utility without the
byproducts often observed from Lewis acid treatment.35 Also,
in contrast to cyclization in acidic solution, assembly 1 affords
product selectivity in both the presence and the absence of gem-
dimethyl substitution. This effect attests to the high degree of
substrate conformational control provided by 1. Both
conformational control and the exclusion of water from reactive
intermediates are characteristic properties of terpene synthases,
to which the activity of 1 presented here bears analogy.
(15) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc.
2007, 129, 11459.
(16) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Acc. Chem. Res.
2009, 42, 1650 and references therein.
(17) (a) Hastings, C. J.; Pluth, M. D.; Bergman, R. G.; Raymond, K.
N. J. Am. Chem. Soc. 2010, 132, 6938. (b) Hastings, C. J.; Backlund, M.
P.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2011, 50, 1.
(18) Lenardao, E. J.; Botteselle, G. V.; de Azambuja, F.; Perin, G.;
Jacob, R. G. Tetrahedron 2007, 63, 6671.
(19) Clark, B. C., Jr.; Chamblee, T. S.; Iacobucci, G. A. J. Org. Chem.
1984, 49, 4557.
(20) Yuasa, Y.; Tsuruta, H.; Yuasa, Y. Org. Process Res. Dev. 2000, 4,
159.
(21) Cheng, H.; Meng, X; Liu, R.; Hao, Y.; Yu, Y.; Cai, S.; Zhao, F.
ASSOCIATED CONTENT
■
Green Chem. 2009, 11, 1227.
(22) Treating 2 with excess ligand or Ga3+ also did not afford 3a−d;
see SI for details.
S
* Supporting Information
Experimental procedures and 1H NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
(23) Conditions in the absence of 1: 28 h, 60 °C, pH 7.50; only 4%
yield of 4 was observed as product from 2; see SI for details.
(24) Jung, M. E.; Piizzi, G. Chem. Rev. 2005, 105, 1735 and references
therein.
AUTHOR INFORMATION
■
Corresponding Author
fdtoste@berkeley.edu; rbergman@berkeley.edu; raymond@
socrates.berkeley.edu
(25) Jung, M. E. Synlett 1990, 4, 186−190.
(26) (a) Page, M. J.; Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A. 1971,
68, 1678. (b) Bruice, T. C.; Pandit, U. K. Proc. Natl. Acad. Sci. U.S.A.
1960, 46, 402.
Notes
(27) An insoluble white precipitate formed over the course of 18 h;
see SI for details.
The authors declare no competing financial interest.
(28) Wang, Z. J.; Brown, C. J.; Bergman, R. G.; Raymond, K. N.;
Toste, F. D. J. Am. Chem. Soc. 2011, 133, 7358.
(29) For reviews see: (a) Watson, I. D. G.; Toste, F. D. Chem. Sci
ACKNOWLEDGMENTS
■
This research was supported by the Director, Office of Science,
Office of Basic Energy Sciences, and the Division of Chemical
Sciences, Geosciences, and Biosciences of the U.S. Department
of Energy at LBNL (DE-AC02-05CH11231). The authors are
grateful to Drs. Casey Brown, Jason Nichols, John Curley,
Jerome Volkman, and Courtney Hastings for helpful dis-
cussions.
2012, 3, 2899. (b) Nunez-Jimen
2008, 108, 3326. (c) Michelet, V.; Toullec, P. Y.; Genet, J. P. Angew.
Chem., Int. Ed. 2008, 47, 4268. (d) Gorin, D. J.; Toste, F. D. Nature
2007, 446, 395. (e) Zhang, L.; Sun, J.; Kozmin, S. A. Adv. Synth. Catal.
2006, 348, 2271.
́
ez, E.; Echavarren, A. M. Chem. Rev.
́
̃
̂
(30) As described in the SI, 10 has previously been observed under
aqueous conditions (catalyst = Hg(OTf)2), while 9 is observed when
water is absent (catalyst = AgSbF6).
(31) For examples of dehydration reactions driven by supramolecular
hosts, see: Murase, T.; Nishijima, Y.; Fujita, M. J. Am. Chem. Soc 2012,
134, 162 and references17a and b.
(32) (a) Ferraz, H. M. C.; Grazini, M. V. A.; Ribeiro, C. M. R.;
Brocksom, U.; Brocksom, T. J. J. Org. Chem. 2000, 65, 2606. (b) Gill,
S.; Kocienski, P.; Kohler, A.; Pontiroli, A.; Qun, L. Chem. Commun.
1996, 1743.
REFERENCES
■
(1) Koksal, M.; Hu, H.; Coates, R. M.; Peters, R. J.; Christianson, D.
̈
W. Nat. Chem. Biol. 2011, 7, 431.
(2) Christianson, D. W. Chem. Rev. 2006, 106, 3412.
(3) Faraldos, J. A.; Antonczak, A. K.; Gonzal
Tippmann, E. M.; Allemann, R. K. J. Am. Chem. Soc. 2011, 133, 13906.
(4) Degenhardt, J.; Kollner, T. G.; Gershenzon, J. Phytochemistry
́
ez, V.; Fullerton, R.;
̈
(33) (a) Negoi, A.; Wuttke, S.; Kemnitz, E.; Macovei, D.; Parvulescu,
V. I.; Teodorescu, C. M.; Coman, S. M. Angew. Chem., Int. Ed. 2010,
49, 8134. (b) Silva, K. A.; Robles-Dutenhefner, P. A.; Sousa, E. M. B.;
Kozhevnikova, E. F.; Kozhevnikov, I. V.; Gusevskaya, E. V. Catal.
Commun. 2004, 5, 425.
2009, 70, 1621.
(5) Wheeler, C. J.; Croteau, R. J. Biol. Chem. 1987, 17, 8213.
(6) Martin, D. M.; Bohlmann, J. Phytochemistry 2004, 65, 1223.
(7) Hyatt, D. C.; Youn, B.; Zhao, Y.; Santhamma, B.; Coates, R. M.;
Croteau, R. B.; Kang, C. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 5360.
(8) For reviews, see: (a) Avram, L.; Cohen, Y.; Rebek, J., Jr. Chem.
Commun. 2011, 47, 5368. (b) Purse, B. W.; Rebek, J., Jr. Proc. Natl.
Acad. Sci. U.S.A. 2005, 102, 10777. (c) Yoshizawa, M.; Klosterman, J.
K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418. (d) Breiner, B.;
Clegg, J. K.; Nitschke, J. R. Chem. Sci 2011, 2, 51. (e) Weister, M. J.;
Ulmann, P. A.; Mirkin, C. A. Angew. Chem., Int. Ed. 2010, 50, 114.
(f) Meeuwissen, J.; Reek, J. N. H. Nat. Chem 2010, 2, 615 and
references therein.
(34) No dimerization or dehydration products were observed (by 1H
NMR, GC-MS) from treatment of 2 with 1.
(35) For an alternative approach to controlling the outcome of
terpene cyclization reactions, see: Pronin, V. S.; Shenvi, R. A. Nat.
Chem. 2012, 1458.
(9) Hastings, C. J.; Pluth, M. D.; Bergman, R. G.; Raymond, K. N. J.
Am. Chem. Soc. 2010, 132, 6938.
(10) Mock, W. L.; Irra, T. A.; James P. Wepsiec, J. P.; Adhya, M. J.
Org. Chem. 1989, 54, 5302.
(11) Marinescu, L. G.; Bols, M. Angew. Chem., Int. Ed. 2006, 45, 4590.
(12) Yoon, H. J.; Kuwabara, J.; Kim, J.-H.; Mirkin, C. A. Science 2010,
330, 66.
(13) Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N.
Angew. Chem., Int. Ed. 1998, 37, 1840.
(14) Reinert, D. J.; Balliano, G.; Schulz, G. E. Chem. Biol. 2004, 11,
121.
17876
dx.doi.org/10.1021/ja308254k | J. Am. Chem. Soc. 2012, 134, 17873−17876