Asymmetric spiroacetalization catalysed by confined Bronsted acids

Article abstract of DOI:10.1038/nature10932

Acetals are molecular substructures that contain two oxygen-carbon single bonds at the same carbon atom, and are used in cells to construct carbohydrates and numerous other molecules. A distinctive subgroup are spiroacetals, acetals joining two rings, which occur in a broad range of biologically active compounds, including small insect pheromones and more complex macrocycles. Despite numerous methods for the catalytic asymmetric formation of other commonly occurring stereocentres, there are few approaches that exclusively target the chiral acetal centre and none for spiroacetals. Here we report the design and synthesis of confined Bronsted acids based on a C ₂-symmetric imidodiphosphoric acid motif, enabling a catalytic enantioselective spiroacetalization reaction. These rationally constructed Bronsted acids possess an extremely sterically demanding chiral microenvironment, with a single catalytically relevant and geometrically constrained bifunctional active site. Our catalyst design is expected to be of broad utility in catalytic asymmetric reactions involving small and structurally or functionally unbiased substrates. 2012 Macmillan Publishers Limited. All rights reserved.

Full text of DOI:10.1038/nature10932

LETTER
doi:10.1038/nature10932
Asymmetric spiroacetalization catalysed by confined
Brønsted acids
1
1
ˇ
´
Ilija Coric & Benjamin List
Acetals are molecular substructures that contain two oxygen– poor enantiofacial discrimination. Whereas large active sites can
carbon single bonds at the same carbon atom, and are used in cells
to construct carbohydrates and numerous other molecules. A dis-
tinctive subgroup are spiroacetals, acetals joining two rings, which
occur in a broad range of biologically active compounds, including
small insect pheromones and more complex macrocycles1,2. Despite
numerous methods for the catalytic asymmetric formation of other
commonly occurring stereocentres, there are few approaches that
exclusively target the chiral acetal centre and none for spiroacetals3,4.
Here we report the design and synthesis of confined Brønsted acids
based on a C2-symmetric imidodiphosphoric acid motif, enabling a
catalytic enantioselective spiroacetalization reaction. These ration-
ally constructed Brønsted acids possess an extremely sterically
demanding chiral microenvironment, with a single catalytically rel-
evant and geometrically constrained bifunctional active site. Our
catalyst design is expected to be of broad utility in catalytic asym-
metric reactions involving small and structurally or functionally
unbiased substrates.
Spiroacetal natural products are widely found in insects, plants, and
bacterial and marine sources, and display a diverse set of biological
activities1,2. Their spiroacetal subunit is not only essential for the bio-
activity but is also a privileged pharmacophore in drug discovery5.
Controlling the relative and absolute configuration of spiroacetals
can be extremely important, and syntheses of natural products dis-
playing the thermodynamically unfavoured spiroacetal configuration
represent daunting challenges2. Furthermore, natural products that
contain a spiroacetal as the only source of chirality are well known6.
Indeed, even the parent 6,6-spiroacetal (1a in Fig. 1a) is a natural
product, olean, the major female-produced sex pheromone of the olive
fruitfly(Bactrocera oleae). Althoughisolated asa racemate fromnatural
sources, olean’stwomirrorimagesdisplay the remarkableproperty that
one isomer, (R)-1a, is active on males, and the other one (S)-1a on
females7. Asymmetric syntheses of olean have been published but
invariably rely on chiral starting materials or reagents8. It has been
shown that enzymes are capable of stereospecific spiroacetal forma-
tion9, and organic chemists have developed several powerful synthetic
methods for the stereoselective formation of suitably substituted
spiroacetals2,10–12. However, so far direct access to spiroacetal motifs
in a catalytic enantioselective manner has not been achieved.
accommodate various transition state geometries leading to different
isomers, a confined space could limit this freedom and thereby
increase selectivity. On the basis of this hypothesis, we initiated
attempts towards modelling sterically demanding acid catalysts for
the envisaged spiroacetalization reaction. We realized that a solution
to the spiroacetalization problem could potentially also tackle an
important issue of broader significance in current Brønsted acid
catalysis. Although this field has acquired wide popularity and import-
ance in recent years, reactions of small aliphatic substrates that do not
possess sterically demanding protecting groups, large aromatic/planar
surfaces, or bulky substituents are still extremely rare13–16,19. We
reasoned that this is due to the inability of current synthetic Brønsted
acid catalysts to provide a truly compact chiral microenvironment. We
speculated that such an environment could effectively restrict molecu-
lar motion, leadingtoefficient enantiodiscriminationevenofsubstrates
or intermediates that lack spatially defined interactions, such as hydro-
gen bonding with the catalyst (Fig. 1c). Here we present the design of
simple and readily available synthetic Brønsted acid catalysts
featuring a sterically highly demanding and rigid chiral microenviron-
ment around their active site. The design is inspired by enzymes that
a
O
O
(R)-1a
(S)-1a
O
O
(R)-Olean
(S)-Olean
active on males active on females
b
Chiral Brønsted acid
O
OH
O
(HX*)
O
–
HO
X*
2a
(S)-1a
+
via:
O
c
Encouraged by our studies of catalytic asymmetric acetalizations3,
we targeted chiral Brønsted acid catalysed spiroacetalizations of readily
availablehydroxyenolethers, suchas2a(Fig. 1b, SupplementaryFig. 1).
Although the envisaged asymmetric hydroetherification of enol ethers
is unknown with chiral Brønsted acid catalysts4, achiral acids are
routinely used to perform the non-asymmetric reaction. However, an
extensive screening of a wide range of known chiral Brønsted acids13–18
gave only disappointing results in the generation of both the 5,5- and
the 6,6-spiroacetal (the best enantiomeric ratio was 29.5:70.5;
Supplementary Tables 1 and 2).
Previous chiral acid catalysts:
Present design:
Small substrate
O
O
Low
e.r.
High
e.r.
Loosely bound
substrate/
HO
OH
intermediate
Sterically constrained
active site
Open active site
We hypothesized that the absence of a compact chiral environment
for the small oxocarbenium ion intermediate, which lacks specific and
sterically well defined interactions with its chiral anion, resulted in
Figure 1
|
Towards catalytic asymmetric synthesis of olean. a, The two
enantiomers of olean from olive fruitfly. b, Proposed catalytic asymmetric
spiroacetalization. c, Catalyst design; see text for details. e.r., enantiomericratio.
1Max-Planck-Institut fu¨r Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mu¨lheim an der Ruhr, Germany.
1 5 M A R C H 2 0 1 2
| V O L 4 8 3 | N A T U R E | 3 1 5
Macmillan Publishers Limited. All rights reserved
©2012

RESEARCH LETTER
frequentlydisplay deep active site pockets constructed bythe folding of
On the basis of these considerations, we designed a novel class of
the protein, and also by the architectures provided by artificial catalytic Brønsted acids, using a C2-symmetric imidodiphosphate anion
cavities20,21
.
(Fig. 2c). In principle, such a C2-symmetric imidodiphosphate moiety
has two distinct Brønsted basic sites, O and N. The corresponding acid
should have a flexible relative positioning of acid/basic pairs owing to
free P–N rotation, and would not satisfy the above mentioned criteria
for the design of a general chiral Brønsted acid. However, our catalyst
design aimed at restricting the imidodiphosphate moiety to a single
O,O-syn conformation between two identical BINOL subunits with
bulky 3,39-substituents (Fig. 2d). We hypothesized that the inclusion
of two BINOLsubunits would result in their interlocking due to sterically
demanding 3,39-substituents (Fig. 2d). As a direct consequence, the
BINOL subunits are unable to freely rotate and the resulting molecular
structure possesses a very high rigidity. Importantly, such an arrange-
ment also resulted in the steric blocking of the undesirable alternative
Brønsted basic N-site. As the two BINOL subunits are identical,
anion 6 (Fig. 2d) is C2-symmmetric, and has therefore only a single
type of catalytically relevant Brønsted basic site. Consequently, the
corresponding Brønsted acid possesses a single catalytically active
bifunctional acid/base pair with a fixed geometry (Fig. 2e). The inter-
locking of BINOL-subunits could in principle also result in the con-
formational locking of the imidodiphosphate moiety in the O,O-anti
conformation to give the O,O-anti-isomer of 6 (Fig. 2c). However, on
the basis of our modelling studies we expected that the formation of the
corresponding O,O-anti-isomer will be disfavoured for steric reasons
due to the bulky 3,39-substituents on the backbone.
Our design was based on the analysis of highly successful chiral
phosphoric acids, in particular those based on the BINOL (1,19-
binaphthalene-2,29-diol) backbone13–15, their derivatives, N-triflyl
phosphoramides16, and the more recently developed N-phosphinyl
phosphoramides and C1-symmetric imidodiphosphoric acids22 (Fig. 2a).
The active site of a BINOL-derived phosphoric acid is composed of
one Brønsted acidic (–OH) and one Brønsted basic site (5O). The
remarkable success of phosphoric acids in asymmetric catalysis is
widely attributed to bifunctional activation, in which bothof these sites
are involved in the stabilization of a transition state15 (Fig. 2b).
Although bulky phosphates 3 (Fig. 2a) have found wide application
in asymmetric catalysis, it is challenging to further modify their steric
environment because 3,39-substituents on BINOL radiate away from
the active site. Recently, significant synthetic efforts have been under-
taken by a number of groups to design alternative backbones that
would narrow the chiral environment of the phosphoric acid23,24
.
C1-Symmetric anions 4 and particularly 5 (Fig. 2a) seem to provide a
simple alternativeway of introducing additional steric demand close to
the active site. However, both of these possess several different
Brønsted basic sites potentially giving a large number of geometrically
distinct acid/base pairs compared to only one such pair present in the
phosphoric acid (Fig. 2a, b). This effectively leads to an ensemble of
catalytically active species, which can stabilize different transition
states, resulting in a low selectivity. Furthermore, rotational freedom
Following this design, wehave synthesized severalimidodiphosphoric
around P–N and N–S bonds renders flexible the relative positioning of acids based on anion 6, including 6a with 2,4,6-Et3C6H2 substituents
acidic and basic sites, and therefore the entire chiral environment. on the 3,39-positions of the two BINOL backbones (Supplementary
Indeed, although anions 4 and 5 display remarkable efficiency in a Table 3). It is noteworthy that imidodiphosphoric acids such as 6a
specific set of reactions, both of these structures lack the generality of are easily accessible. Their synthesis requires only a single additional
phosphoric acids. We reasoned that only one distinct and geometrically synthetic step compared to the corresponding phosphoric acids (Sup-
fixed acid/base pair, translating into a single type of basic site in the plementary Figs 2 and 3). Indeed, the crystal structure of imidodi-
corresponding anion is a prerequisite for the construction of a success- phosphoric acid 6a reveals a confined active site deeply buried
ful and broadly applicable Brønsted acid catalyst.
within a highly sterically demanding chiral environment (Fig. 3a,
a
b
R
R
O
O
R
R
Electrophile
Nucleophile
O 3
S
O3
P
O1
O1
O1
O1
O
O
R
O
O
R
O4
H
Ar
Ar
P
P
P
H
Acidic
site
N
N
CF3
2
2
Basic site
O
O
P
O
O
3
4
5
c
d
e
R
R
O1
O1
P
Electrophile
Nucleophile
O
P
O
O,O-syn
O,O-anti
O1
P
O1
N
O
O
O
O
O
O
H
2
P
Acidic
site
H
N
O
Basic site
O
O
P
(shielded)
O 3
P
O
O
P
O
O
R
R
N
P
O
O
O
N
O
1
2
6
Figure 2
|
Development of C2-symmetric imidodiphosphoric acids.
anion (showing O,O-syn- and O,O-anti isomers). d, Conformationally locked
O,O-syn-imidodiphosphosphate anion with C2-symmetry (showing in green
bulky 3,39-substituents responsible for locking). e, Bifunctional activation of
reactants with Brønsted acid/base pair in imidodiphosphoric acid catalysis.
a, Phosphate anion and its derivatives (catalytic moiety shown in blue),
showing numbering of different Brønsted basic sites (one in green and
additional ones in red). b, Bifunctional activation of reactants with Brønsted
acid/base pair in phosphoric acid catalysis. c, General imidodiphosphosphate
3 1 6
| N A T U R E | V O L 4 8 3 | 1 5 M A R C H 2 0 1 2
Macmillan Publishers Limited. All rights reserved
©2012

LETTER RESEARCH
Supplementary Fig. 4). Viewing the catalytic moiety from the top configuration. Gratifyingly, with our confined Brønsted acid catalyst this
illustrates how the two BINOL-substituents surround the active site goal could be achieved. A variety of non-thermodynamic spiroacetals
from above (Fig. 3b). Viewed from the bottom, the catalyst demon- (1g, 1i, 1k in Table 1) could be formed with diastereomeric ratios ranging
strates a tight arrangement of the two remaining BINOL subunits, from 5:1 to 23:1, against a strong thermodynamic preference of up to
which completely block access to the imidodiphosphate (Fig. 3c). 1:124 (entries 8, 10 and 12 in Table 1). Moreover, when the catalyst
For comparison, the crystal structure of one of the most popular and control matches the inherent thermodynamic preference, spiroacetals
sterically demanding phosphoric acids, TRIP (3a, that is, protonated 3 can be obtained with excellent diastereoselectivities that can even
with R 5 2,4,6-iPr3C6H2), is given in Fig. 3d and 3e25. The active site of exceed thermodynamic ratios (entries 9, 11, 13). In this way,
TRIP is placed inside a broad open cone, whereas that of the imidodi- thermodynamic spiroacetals can be obtained under very mild, kinetic
phosphate can better be described as a chiral pocket. Furthermore, the conditions, as compared to the relatively strongly acidic conditions
geometrical parameters of the imidodiphosphoric acid, which co-crys- typically required for thermodynamic equilibration. It is important to
tallized with a molecule of water, are consistent with a hydronium ion note that kinetic spiroacetalizations of substrates 2g–2m (Table 1),
bridgingthetwoimidodiphosphateoxygen atoms(SupplementaryFig. catalysed with a simple achiral acid, favoured the thermodynamic
4, Supplementary Tables 4–6). This supports proton location on oxy- product with only poor diastereoselectivity ranging from 1:1 to 1:3.
gen rather than on nitrogen, as shown in Fig. 2e.
Having produced a Brønsted acid confined within an extremely powerful capability of the confined Brønsted acid 6a. An efficient
kinetic resolution occurred, delivering both bisacetal 1n and enolacetal
Spiroacetalization of substituted enol ether 2m revealed another
sterically demanding chiral cavity, we turned to the spiroacetalization
problem (Supplementary Table 3). Our catalyst design was quickly (R)-2m with excellent enantioselectivity (entry 14). Asymmetric
rewarded with the first (to our knowledge) highly enantioselective synthesis of these sensitive motifs by any other method would present
catalytic spiroacetalization reaction, obtaining (S)-olean from alcohol a considerable challenge. Finally, to further explore the behaviour of
2a with an excellent enantiomeric ratio of 98:2 (Table 1, entry 1). The our catalyst, we also examined large substrate 2o, an open form of the
(R)-enantiomer of olean was easily obtained by using the enantiomer steroidal sapogenin, diosgenin (entry 15). As expected, the reactivity
of the catalyst (entry 2). Imidodiphosphoric acid catalyst 6a proved was rather low using either catalyst enantiomer, although only a single
quite general, and various other spiroacetals were obtained with high diastereomer (the known non-thermodynamic isomer 20-epi-
enantioselectivity. Only 0.1 mol% of catalyst 6a was required in the diosgenin) was obtained26. This probably results from the inability of
construction of the small 5,5-spiroacetal (entry 3). The 5,6-spiroacetal our confined acid to efficiently accommodate such large substrates.
motif could be accessed equally successful by cyclization of either five
or six-membered rings (entries 4 and 5). Different enol ether ring sizes
are similarly tolerated, as demonstrated by highly enantioselective
reactions to form the corresponding 7,6- and 7,5-spiroacetals (entries
6 and 7). Notably, all these small spiroacetals are core structures of
many natural products1,2.
Whereas recentdevelopments in Brønsted acid catalysishave enabled
an array of transformations to be performed asymmetrically13–16,27, sub-
strates have typically been sterically or electronically biased. We have
shown here that confined Brønsted acids with extreme steric demand
and chiral pockets reminiscent of those found in enzymes can
overcome this limitation and solve an important problem in organic
When otherstereocentres arepresent in the targetmolecule, accessing synthesis. Our spiroacetalization reaction offers a straightforward
the spiroacetal stereocentre in the most stable thermodynamic con- catalytic enantioselective access to spiroacetal motifs, and is expected
figuration is usually easy. However, the formation of fragile non- to find applications in the synthesis of diverse natural products and
thermodynamic spiroacetals, having the less stable configuration of biologically active molecules. We believe that the concepts described
the spiroacetal stereocentre, presents a formidable challenge for organic here open the door to the development of asymmetric reactions that
synthesis2,10,11. An ideal solution would be provided by a catalyst that include small aliphatic and/or loosely bound molecules, and will be
could override thermodynamic preference and control the spiroacetal widely applicable, possibly even beyond Brønsted acid catalysis28–30
.
a
b
c
Top view
HO
O
Bottom view
d
e
Top view
O
OH
Figure 3
|
Comparison of crystal structures of anions of 6a (6 with
d, Crystal structure of TRIP anion (main figure) and schematic representation
R 5 2,4,6-Et3C6H2) andTRIP. a, Crystal structure of the anion of 6a, side view (inset). e, TRIP, top view at the active site, space filling model. Colour coding:
(main figure) and schematic representation (inset). b, 6a, top view at the active dark grey, carbon; light grey, hydrogen; red, oxygen; orange, phosphorus;
site, space filling model. c, 6a, bottom view at the active site, space filling model. purple, nitrogen.
1 5 M A R C H 2 0 1 2
| V O L 4 8 3 | N A T U R E | 3 1 7
Macmillan Publishers Limited. All rights reserved
©2012

RESEARCH LETTER
Table 1
|
Catalytic asymmetric spiroacetalization
Entry
Substrate
Reaction details
Product
Yield
e.r. or d.r.
e.r. 98:2
O
O
O
OH
OH
O
O
O
1
2
3
6a (5 mol%), –25 uC, MTBE
77%
2a
2a
(S)-1a
(R)-1a
1b
O
O
ent-6a (5 mol%), –25 uC,
70%
62%
e.r. 97.5:2.5
e.r. 96:4
MTBE
O
6a (0.1 mol%), –55 uC,
CH2Cl2
OH
2b
2c
O
O
O
4
5
6a (1 mol%), –35 uC, DCE
6a (1 mol%), –35 uC, MTBE
O
81%
69%
e.r. 95.5:4.5
e.r. 96:4
OH
1c
1c
O
O
OH
2d
OH
6
7
8
9
6a (1 mol%), –25 uC, MTBE
6a (1 mol%), –35 uC, MTBE
O
O
78%
88%
83%
e.r. 96:4
O
1e
1f
2e
2f
O
e.r. 98.5:1.5
O
OH
O
OH
d.r. 5:1, non-thermodynamic
spiroacetal (thermodynamic
d.r. 1:124)
O
O
Me
O
O
(S)-2g
O
O
1g
6a (5 mol%), –35 uC, MTBE
OH
89%
70%
d.r. 65:1
Me
(R)-2g
1h
10
11
d.r. 23:1, non-thermodynamic
spiroacetal (thermodynamic
d.r. 1:9)
O
OH
O
O
(R)-2i
6a (5 mol%), –35 uC, MTBE
Me
1i
O
OH
O
O
70%
76%
d.r. 50:1
O
(S)-2i
Me
1j
OH
12
13
d.r. 7:1, non-thermodynamic
spiroacetal (thermodynamic
d.r. 1:60)
O
Me
O
(R)-2k
1k
OH
O
O
6a (10 mol%), –35 uC,
86%
70%
d.r. 100:1
O
(S)-2k
MTBE
Me
1l
MeO
14
Kinetic resolution
e.r. 97.5:2.5 (d.r. .50:1)
MeO
MeO
OH
O
O
O
(S)-2m
1n
6a (5 mol%), –25 uC, MTBE
+
(50% conv.)
OH
MeO
86%
88%
e.r. 97:3
OH
O
O
(R)-2m
(R)-2m
HO
O
15
6a or ent-6a (1 mol%),
20 uC, DCE
d.r. .20:1:1:1
O
20
H
22
H
O
H
H
H
H
H
H
H
HO
H
1o
HO
2o
Reactions were run in the presence of molecular sieves. Substrates in entries 8 to 14 were examined simultaneously as racemic mixtures. The thermodynamic diastereomeric ratio (d.r.) was measured after
treating substrates with 3 M HCl(aq.) at room temperature. e.r., enantiomeric ratio; MTBE, t-butylmethyl ether; DCE, 1,2-dichloroethane.
(0.0125 mmol, 5 mol%) in MTBE (1 ml) was added dropwise and the mixture
METHODS SUMMARY
was stirred at –25 uC for 24 h. Et3N (50 ml) was added at the same temperature,
the mixture concentrated, and the product isolated by column chromatography.
For complete experimental details and characterization of compounds, see
Catalyticasymmetricsynthesis of (S)-oleanwas performedas follows. MTBE (tert-
˚
butylmethyl ether) (7 ml) and 4 A molecular sieves (50 mg) were cooled to –25 uC
in a vial equipped with a septum. A solutionof 2a (0.25 mmol) in MTBE (2 ml) was
added and the mixture stirred for 5–10 min. A solution of catalyst 6a Supplementary Information.
3 1 8
| N A T U R E | V O L 4 8 3 | 1 5 M A R C H 2 0 1 2
Macmillan Publishers Limited. All rights reserved
©2012

LETTER RESEARCH
´
20. Shenoy, S. R., Crisostomo, F. R. P., Iwasawa, T. & Rebek, J. Jr. Organocatalysis in a
syntheticreceptorwithaninwardly directed carboxylicacid. J. Am. Chem. Soc. 130,
5658–5659 (2008).
Received 14 November 2011; accepted 10 February 2012.
1. Perron,F.&Albizati,K.F.Chemistryofspiroketals.Chem.Rev.89,1617–1661(1989).
2. Aho, J. E., Pihko, P. M. & Rissa, T. K. Nonanomeric spiroketals in natural products:
structures, sources, and synthetic strategies. Chem. Rev. 105, 4406–4440 (2005).
21. Hastings, C. J., Pluth, M. D., Bergman, R. G. & Raymond, K. N. Enzymelike catalysis
of the Nazarov cyclization by supramolecular encapsulation. J. Am. Chem. Soc.
132, 6938–6940 (2010).
22. Vellalath, S., Coric, I. & List, B. N-Phosphinyl phosphoramide—a chiral Brønsted
acid motif for the direct asymmetric N,O-acetalization of aldehydes. Angew. Chem.
Int. Edn 49, 9749–9752 (2010).
23. Xu, F. et al. SPINOL-Derived phosphoric acids: synthesis and application in
enantioselective Friedel-Crafts reaction of indoles with imines. J. Org. Chem. 75,
8677–8680 (2010).
24. Coric, I., Mu¨ller, S. & List, B. Kinetic resolution of homoaldols via catalytic
asymmetric transacetalization. J. Am. Chem. Soc. 132, 17370–17373 (2010).
25. Klussmann, M. et al. Synthesis of TRIP and analysis of phosphate salt impurities.
Synlett 2189–2192 (2010).
26. Wall, M. E., Eddy, C. R. & Serota, S. Steroidal sapogenins. XIX. Stereochemistry of
sapogenins and cholesterol at carbon 20. J. Am. Chem. Soc. 76, 2849–2850
(1954).
27. Shapiro, N. D., Rauniyar, V., Hamilton, G. L., Wu, J. & Toste, F. D. Asymmetric
additions to dienes catalysed by a dithiophosphoric acid. Nature 470, 245–249
(2011).
28. Mayer, S. & List, B. Asymmetric counteranion-directed catalysis. Angew. Chem. Int.
Edn 45, 4193–4195 (2006).
29. Hamilton, G. L., Kang, E. J., Mba, M. & Toste, F. D. A powerful chiral counterion
strategy for asymmetric transition metal catalysis. Science 317, 496–499 (2007).
30. Mukherjee, S. & List, B. Chiral counteranions in asymmetric transition-metal
catalysis: highly enantioselective Pd/Brønsted acid-catalyzed direct a-allylation of
aldehydes. J. Am. Chem. Soc. 129, 11336–11337 (2007).
ˇ
´
3. Coric, I., Vellalath, S. & List, B. Catalytic asymmetric transacetalization. J. Am. Chem.
Soc. 132, 8536–8537 (2010).
ˇ
´
4. Nagano, H. & Katsuki, T. Stereocontrolled OH protection: asymmetric
tetrahydrofuranylation. Chem. Lett. 31, 782–783 (2002).
5. Zinzalla, G., Milroy, L.-G. & Ley, S. V. Chemical variation of natural product-like
scaffolds: design and synthesis of spiroketal derivatives. Org. Biomol. Chem. 4,
1977–2002 (2006).
ˇ
´
¨
6. Brasholz, M., Sorgel, S., Azap, C. & Reißig, H.-U. Rubromycins: structurally
intriguing, biologically valuable, synthetically challenging antitumour antibiotics.
Eur. J. Org. Chem. 3801–3814 (2007).
7. Haniotakis, G., Francke, W., Mori, K., Redlich, H. & Schurig, V. Sex-specific activity of
(R)-(2)- and (S)-(1)-1,7-dioxaspiro[5.5]undecane, the majorpheromone of Dacus
oleae. J. Chem. Ecol. 12, 1559–1568 (1986).
8. Redlich, H. & Francke, W. Synthesis of enantiomerically pure 1,7-
dioxaspiro[5.5]undecanes, pheromone components of the olive fly (Dacus oleae).
Angew. Chem. Int. Ed. 23, 519–520 (1984).
9. Takahashi, S. et al. Reveromycin A biosynthesis uses RevG and RevJ for
stereospecific spiroacetal formation. Nature Chem. Biol. 7, 461–468 (2011).
10. Takaoka, L. R., Buckmelter, A. J., LaCruz, T. E. & Rychnovsky, S. D. Rationalsynthesis
of contra-thermodynamic spiroacetals by reductive cyclizations. J. Am. Chem. Soc.
127, 528–529 (2005).
11. Moilanen, S. B., Potuzak, J. S. & Tan, D. S. Stereocontrolled synthesis of spiroketals
via Ti(Oi-Pr)4-mediated kinetic spirocyclization of glycal epoxides with retention of
configuration. J. Am. Chem. Soc. 128, 1792–1793 (2006).
12. Audrain, H., Thorhauge, J., Hazell, R. G. & Jørgensen, K. A. A novel catalytic and
highly enantioselective approach for the synthesis of optically active carbohydrate
derivatives. J. Org. Chem. 65, 4487–4497 (2000).
13. Akiyama, T., Itoh, J., Yokota, K. & Fuchibe, K. Enantioselective Mannich-type
reaction catalyzed by a chiral Brønsted acid. Angew. Chem. Int. Edn 43,
1566–1568 (2004).
14. Uraguchi, D. & Terada, M. Chiral Brønsted acid-catalyzed directMannich reactions
via electrophilic activation. J. Am. Chem. Soc. 126, 5356–5357 (2004).
15. Akiyama, T. Stronger Brønsted acids. Chem. Rev. 107, 5744–5758 (2007).
16. Nakashima, D. & Yamamoto, H. Design of chiral N-triflyl phosphoramide as a
strong chiral Brønsted acid and its application to asymmetric Diels-Alder reaction.
J. Am. Chem. Soc. 128, 9626–9627 (2006).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank A. Dreier and R. Goddard for crystal structure analysis of
catalyst 6a. S. Vellalath and S. Mu¨ller are acknowledged for donating several previously
described catalysts, and N. Wippich and S. Dehn for technical assistance. We gratefully
acknowledge support from the Max Planck Sociey and the European Research Council.
ˇ
Author Contributions I.C. and B.L. jointly designed and developed C2-symmetric
imidodiphosphoric acids, developed the spiroacetalization reaction, and wrote the
ˇ
manuscript. I.C. conducted the laboratory experiments. B.L. initiated and oversaw the
project.
17. Doyle, A. G. & Jacobsen, E. N. Small-molecule H-bond donors in asymmetric
catalysis. Chem. Rev. 107, 5713–5743 (2007).
18. Huang, Y., Unni, K. A., Thadani, A. N. & Rawal, V. H. Singleenantiomers froma chiral-
alcohol catalyst. Nature 424, 146 (2003).
19. Zhang, Q.-W. et al. Brønsted acid catalyzed enantioselective semipinacol
rearrangement for the synthesis of chiral spiroethers. Angew. Chem. Int. Edn 48,
8572–8574 (2009).
Author Information X-ray crystallographic data have been deposited in the Cambridge
Crystallographic Data Centre database (http://www.ccdc.cam.ac.uk/) under accession
code CCDC 864762. Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of this article at
www.nature.com/nature. Correspondence and requests for materials should be
addressed to B.L. (list@kofo.mpg.de).
1 5 M A R C H 2 0 1 2
| V O L 4 8 3 | N A T U R E | 3 1 9
Macmillan Publishers Limited. All rights reserved
©2012