A dioxane template for highly selective epoxy alcohol cyclizations

Article abstract of DOI:10.1002/chem.201300845

Ladder polyether natural products are a class of natural products denoted by their high functional-group density and large number of well-defined stereocenters. They comprise the toxic component of harmful algal blooms (HABs), having significant negative economic and environmental ramifications. However, their mode of action, namely blocking various cellular ion channels, also denotes their promise as potential anticancer agents. Understanding their potential mode of biosynthesis will not only help with developing ways to limit the damage of HABs, but would also facilitate the synthesis of a range of analogs with interesting biological activity. 1,3-Dioxan-5-ol substrates display remarkable 'enhanced template effects' in water-promoted epoxide cyclization processes en route to the synthesis of these ladder polyether natural products. In many cases, they provide near complete endo-to-exo selectivity in the cyclization of epoxy alcohols, thereby strongly favoring the formation of tetrahydropyran (THP) over tetrahydrofuran (THF) rings. The effects of various Br?nsted and Lewis acidic and basic conditions are explored to demonstrate the superior selectivity of the template over the previously reported THP-based epoxy alcohols. In addition, the consideration of other synthetic routes are also considered with the goal of gaining rapid access to a plethora of potential starting materials applicable towards the synthesis of ladder polyethers. Finally, cascade sequences with polyepoxides are investigated, further demonstrating the versatility of this new reaction template. Reversed selectivity: 1,3-Dioxan-5-ol templated epoxy alcohols undergo a remarkably selective cyclization, which occurs with near complete endo selectivity that is opposite to that predicted by Baldwin's rules. This endo preference in water near pH 7.0 is an order of magnitude larger relative to that of a previously disclosed THP-templated cyclization (see scheme), suggesting that "templates" may be important in the biosynthesis of marine ladder polyethers. Copyright

Full text of DOI:10.1002/chem.201300845

DOI: 10.1002/chem.201300845
A Dioxane Template for Highly Selective Epoxy Alcohol Cyclizations
James J. Mousseau, Christopher J. Morten, and Timothy F. Jamison*[a]
Abstract: Ladder polyether natural
products are a class of natural products
denoted by their high functional-group
density and large number of well-de-
fined stereocenters. They comprise the
toxic component of harmful algal
blooms (HABs), having significant neg-
ative economic and environmental
ramifications. However, their mode of
action, namely blocking various cellu-
lar ion channels, also denotes their
promise as potential anticancer agents.
Understanding their potential mode of
biosynthesis will not only help with de-
veloping ways to limit the damage of
HABs, but would also facilitate the
synthesis of a range of analogs with in-
teresting biological activity. 1,3-
Dioxan-5-ol substrates display remark-
able ꢁenhanced template effectsꢀ in
water-promoted epoxide cyclization
processes en route to the synthesis of
these ladder polyether natural prod-
ucts. In many cases, they provide near
complete endo-to-exo selectivity in the
cyclization of epoxy alcohols, thereby
strongly favoring the formation of tet-
rahydropyran (THP) over tetrahydro-
furan (THF) rings. The effects of vari-
ous Brønsted and Lewis acidic and
basic conditions are explored to dem-
onstrate the superior selectivity of the
template over the previously reported
THP-based epoxy alcohols. In addition,
the consideration of other synthetic
routes are also considered with the
goal of gaining rapid access to a ple-
thora of potential starting materials ap-
plicable towards the synthesis of ladder
polyethers. Finally, cascade sequences
with polyepoxides are investigated, fur-
ther demonstrating the versatility of
this new reaction template.
Keywords: cyclization · epoxides ·
polyethers · regioselectivity · tem-
plate synthesis
Introduction
Developing new means to effect 6-endo cyclization process-
es remains a relevant synthetic challenge owing to an inher-
ent preference for the 5-exo adduct, as predicted by Bald-
winꢀs rules. Such transformations are particularly pertinent
in the construction of marine ladder polyethers. These prod-
ucts continue to garner much interest owing to their innate
structural complexity, their potential as anti-cancer and anti-
bacterial agents (Figure 1),[1] as well as their undesirable en-
vironmental and economic consequences, such as harmful
algal blooms.[2] It has been proposed by the research groups
of Nakanishi,[2b,3] Shimizu,[4] and Nicolaou[5] that Nature may
construct these elaborate structures through a ring-opening/
ring-closing sequence of a polyepoxide precursor. Laborato-
ry emulation of these proposed cascades has been hindered
by the fact that the 5-exo product (tetrahydrofuran, THF) is
generally favored over the desired 6-endo product (tetrahy-
dropyran, THP), as demonstrated by the research group of
Coxon in the 1970 s (Scheme 1, A).[6–8]
Figure 1. Selected examples of ladder-polyether-containing marine natu-
ral products.
rahydropyran (THP) ring and promoted by neutral water
(Scheme 1, B).[9,10] The presence of the oxygen atom in the
heterocyclic ring was vital for the desired selectivity. It has
also been suggested that enzymes may catalyze the forma-
tion of an initial THP ring, and that this THP then “tem-
plates” the cascade favoring an all-endo ring-closure se-
quence.[9,11] Given the lack of selectivity when no endocyclic
oxygen is present in the template (Scheme 1, C), we have
undertaken a systematic study of the structure–selectivity re-
lationships of these and related cascade templates. Toward
this end, we first explored a 1,3-dioxane ring in the context
of the synthesis of the HIJK rings of gymnocin A.[12] Howev-
In prior investigations, we observed highly endo-selective
epoxide-opening cyclizations templated by a preformed tet-
[a] Dr. J. J. Mousseau, Dr. C. J. Morten, Prof. Dr. T. F. Jamison
Department of Chemistry
Massachusetts Institute of Technology
77 Massachusetts Ave.
Cambridge, MA, 02139 (USA)
E-mail: tfj@mit.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201300845.
10004
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

FULL PAPER
Scheme 1. Templated cyclizations of epoxy alcohols.
er, the sensitivity to hydrolysis of this benzylidene acetal
under Lewis acidic and aqueous conditions precluded a
comparison with our previous system. We herein report that
methylene acetal (1) is robust towards epoxy alcohol cycli-
zation processes. Uniquely, this new template demonstrates
a striking selectivity profile (Scheme 1, D). In contrast to
the original THP template, where high (>10:1) selectivity
was observed in a narrow pH range, the methylene-acetal-
based 1,3-dioxane affords the desired endo cyclization prod-
uct for electronically unbiased, trans-disubstituted epoxides
independent of promoter used (acid, base, and water)—well
over 100:1 under certain reaction conditions. We believe
that these observations have important implications for the
mechanism of endo-selective epoxide-opening cyclizations
and that the stability of this template makes it well suited
for use in target-directed synthesis.
Scheme 2. Synthesis of epoxy alcohol 1. Reagents and reaction condi-
tions: a) Ph3P=CHCO2Et, THF, 708C; b) p-anisaldehyde dimethyl acetal,
CSA, CH2Cl2, RT, 70% over 2 steps; c) TBDPSCl, imidazole, DMF, RT;
d) pTSA, CH2Cl2, 708C, 86% over 2 steps; e) DMM, BF3·OEt2, CH2Cl2,
RT, 78 %; f) O3, CH2Cl2, À788C, 93%; g) CH3CHI2, CrCl2, THF, RT,
80%; h) Shi ketone, Oxone, K2CO3, nBu4NHSO4, Na2B4O7·10H2O,
Na2EDTA, 2:1 DMM/CH3CN, À58C to 08C, 72%; i) TBAF, THF, RT,
89%. CSA=camphorsulfonic acid, DMF=N,N-dimethylformamide,
DMM=dimethoxymethane, EDTA=ethylenediaminetetraacetic acid,
TBAF=tetrabutylammonium fluoride, TBDPS=tert-butyldiphenylsilyl,
pTSA=para-toluenesulfonic acid.
under ambient conditions, the endo/exo products were
formed in the ratio, 14:1 (Table 1, entry 1). Although this is
an improvement on the 10:1 selectivity previously found
with the THP ring,[9] the rate of reaction was significantly
Table 1. Cyclization of 1,3-dioxane alcohol 1.
Results and Discussion
Employing a chiron-based approach,[13] the synthesis of 1
began with a Wittig olefination of 2-deoxy-d-ribose (2) fol-
lowed by selective 1,3 protection to form PMP acetal 3
(Scheme 2).[14] The free secondary alcohol was masked as its
TBDPS ether and the PMP group removed to afford diol 4.
Installation of the methylene acetal and ozonolysis of the
corresponding enoate 5 revealed aldehyde 6. Subjection of
the aldehyde to a Takai olefination with 1,1-diiodoethane,[15]
followed by Shi epoxidation, to give 7, and removal of the
TBDPS group in presence of TBAF gave the desired epoxy
alcohol 1.
Entry[a]
Temp. [8C]
Time
8/9[b]
Yield [%][c]
1
2
3
4
5
6
22
40
55
70
100
125
35 days
4 days
2 days
16 h
12 h
4 h
14.1:1
14.6:1
19.9:1
19.4:1
19.8:1
34.8:1
62
49
54
57
62
66
We compared the selectivity of cyclization in H2O relative
to that previously disclosed for the THP template. When
cyclizations were performed in deionized water (DI H2O)
[a] All reactions were performed at 0.02m and carried to greater than
98% conversion. [b] Ratios were determined by GC using dodecane as
an external standard. [c] Yield of combined isolated products.
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10005

T. F. Jamison et al.
lower, requiring over a month to reach completion. Presum-
ably this is due to the presence of an additional inductively
withdrawing oxygen atom, thus decreasing the nucleophilici-
ty of the hydroxyl group. However, conducting the reaction
at a higher reaction temperatures not only led to a faster re-
action (Table 1, entries 2–6), but also afforded modest im-
provements in site selectivity for the endo product, the
ratios being 20:1 and 35:1 for the reactions conducted at
1008C and 1258C, respectively (Table 1, entry 5 and 6). This
phenomenon is unique to this template as such effects were
not observed with the previously investigated THP systems.
Both cyclization products derived from the dioxane are
stable to elevated temperatures, suggesting that it is unlikely
that degradation of the minor exo product accounts for the
increase in selectivity under these reaction conditions. Other
ramifications of the higher reaction temperature include a
lower pH value of the (unbuffered) water[16] and a higher
ionic strength of the medium. Either of these ramifications
could modulate selectivity by facilitating proton transfer
during cyclization, in addition to stabilizing charged transi-
tion-state intermediates. This combination may be more rel-
evant for this system. Lastly, reactions performed in poly-
propylene tubes at 708C (not shown) furnished a 24.1:1
preference of 8 thus demonstrating that SiO2 in the glass of
the reaction vessels was not responsible for promoting the
high endo selectivity.
trend of optimum site selectivity near neutral pH values.
Remarkably, even at low and high pH values (3.0 and 12.2),
the endo/exo levels of selectivity were 7.0:1 and 4.3:1, re-
spectively. This preference for endo cyclization processes
under such reaction conditions is unprecedented; several
other templates we have examined were deemed nonselec-
tive (ca. 1:1).[9]
To gain insight into this unique and near-perfect endo re-
gioselectivity, we prepared trisubstituted epoxy alcohols
from aldehyde 6 (Scheme 3). Epoxy alcohol 15a was pre-
pared through olefination of 6 by using a stabilized Wittig
reagent, followed by complete reduction of the ester (com-
pounds 11–13). Notably, Wittig olefination with isopropyltri-
phenylphosphonium iodide to furnish 13 directly from 6 led
to the product being isolated in poor yield and was thus not
amenable to scale up. Shi epoxidation of 13 and removal of
the TBDPS group furnished 15a. In parallel, methyl ketone
Accordingly, we next explored the effect of the pH value
on the selectivity of cyclization (Figure 2).[9] In potassium-
phosphate-based buffered solutions (0.1m) with a near neu-
tral pH value, the selectivity increased by nearly an order of
magnitude relative to that of the THP template and by two
orders of magnitude relative to that of the cyclohexanol-
based template (Scheme 1, C; Figure 2). The ratio, 8/9,
reached a maximum of 126:1 at pH 6.8, thus reprising a
Scheme 3. Synthesis of trisubstituted epoxides. Reagents and reaction
conditions: a) Ph3P=CMeCO2Et, CH2Cl2, RT, 89%; b) DIBAL-H, PhMe,
À788C to RT, 72%; c) MsCl, Et3N, Me3N·HCl, toluene, 08C, 78%;
d) LAH, Et2O, 0 8C, 49%; e) Shi ketone, Oxone, K2CO3, nBu4NHSO4,
Na2B4O7·10H2O, Na2EDTA, 2:1 DMM/CH3CN, À58C to 08C, 86%;
f) TBAF, THF, RT, 58%; g) MeMgBr, Et2O, 0 8C, then TPAP, NMO,
CH2Cl2, RT, 60% over 2 steps; h) Ph3PEt+BrÀ, nBuLi, THF, À788C to
RT, 71%; i) Shi ketone, Oxone, K2CO3, nBu4NHSO4, Na2B4O7·10H2O,
Na2EDTA, 2:1 DMM/CH3CN, À58C to 08C, 91%; j) TBAF, THF, ET,
90%. DIBAL=diisobutyl aluminum, LAH=lithium aluminum hydride,
MsCl=methanesulfonyl chloride, NMO=N-methylmorpholine N-oxide,
TPAP=tetrapropylammonium perruthenate.
Figure 2. Dependence of regioselectivity upon pH value.
10006
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

Selective Epoxy Alcohol Cyclizations
FULL PAPER
Table 2. Effects of epoxide substitution on endo selectivity.
16 was subjected to Wittig conditions to provide alkene 17
in greater than 9:1 E/Z ratio. Again, this was followed by
epoxidation (giving 18) and protecting-group removal, thus
revealing epoxy alcohol 15b.
Alcohols 15a and 15b were subjected to various neutral,
acidic, and basic reaction conditions (Table 2). As noted
above, cyclization of alcohol 1 in neutral water led to a 20:1
preference for the endo product. Cyclization of 1 in the
presence of Cs2CO3 and CSA afforded 8 and 9 in ratios of
4.6:1 (70% yield) and 4.2:1 (87% yield), respectively; how-
ever, cyclization in the presence of BF3·OEt2 yielded only
ether 8 (82% yield). These three results are in stark contrast
to the selectivity of the cyclization of the THP template be-
cause water was not required to obtain good-to-complete
endo selectivity. Previously, the ratio of the two cyclic ethers
was found to be approximately 1:1, modestly favoring the
exo adduct in cases employing Cs2CO3 and CSA.[10b] Pre-
sumably, owing to enhanced stabilization of developing posi-
tive charge in the transition state, gem-dimethyl substituted
15a favored the formation of the endo product 19a in pres-
ence of DI H2O (64% yield), CSA (94% yield), and
BF3·OEt2 (88% yield); CSA offered significant improve-
ment relative to the THP template (>20:1 versus 5.8:1
endo/exo).[10b,17] Although, exo product 20a was favored
when the reaction was conducted in the presence of Cs2CO3,
presumably owing to steric effects, the magnitude of selec-
tivity was decreased, suggesting
Reaction conditions
1
15a
15b
[g]
X=O
X=CH2
X=O
X=CH2
X=O
X=CH2
X=O
19.4:1[e]
10.0:1[f]
>20:1[f]
>20:1[f]
>20:1[f]
>20:1[f]
>20:1[f]
5.8:1[f]
–
DI H2O[a]
4.9:1[f]
<1:20[f]
1:11.0[f]
<1:20[f]
1:5.2[f]
[i]
[i]
[i]
[i]
>200:1[e]
1.4:1[f]
4.2:1[e]
1:1.2[f]
[b]
BF3
CSA[c]
4.6:1[e]
1:2.7[f]
1:2.9[f]
1:17.0[f]
–
[h]
[d]
Cs2CO3
X=CH2
3.0:1[f]
[a] Performed at 0.02m, 708C, 16 h. [b] 0.25 equiv BF3·OEt2, CH2Cl2,
0.02m, À788C, 0.5 h. [c] 1.0 equiv CSA, CH2Cl2, 0.02m, RT, 4 h.
[d] 30 equiv Cs2CO3, MeOH, 0.02m, RT, 16 h. [e] Ratios were determined
by GC using dodecane as an external standard. [f] Ratios determined by
1H NMR spectroscopy. [g] A 7.8:1 mixture of 15b/20b was obtained.
endo product (19b) not observed. [h] A 2.7:1 mixture of 15b/20b was ob-
tained. endo product (19b) not observed. [i] See ref. [11b].
a directing counter effect of the
dioxane ring. The THP variant
of epoxy alcohol 15b provided
endo product in water and in
the presence of Cs2CO3;[10b]
however, in the case of the di-
oxane template, no such prod-
uct was observed. Instead, start-
ing material was recovered in
both cases. Under acidic condi-
tions, 20b was isolated in 80%
yield both in the presence of
BF3·OEt2 and CSA. Terminal
epoxides 23 and 26 were also
prepared by using
a similar
process involving aldehyde
6
and ketone 16 (Scheme 4). The
use of these alcohols led to ex-
clusive formation of the corre-
sponding 5-exo products under
all reaction conditions; neither
hydrolysis nor methanolysis of
the epoxide was observed.
Having confirmed the stabili-
ty of methylene acetal group
under the cyclization condition,
in addition to the understand-
ing of the selectivity criteria, we
next investigated the feasibility
Scheme 4. Synthesis and cyclization of terminal epoxides. Reagents and reaction conditions: a) Ph3PMe+BrÀ,
nBuLi, THF, À788C to RT, 94%; b) Shi ketone, Oxone, K2CO3, nBu4NHSO4, Na2B4O7·10H2O, Na2EDTA, 2:1
DMM/CH3CN, À58C to 08C, 89%; c) TBAF, THF, RT, 95%; d) Ph3PMe+BrÀ, nBuLi, THF, À788C to RT,
87%; e) Shi ketone, Oxone, K2CO3, nBu4NHSO4, Na2B4O7·10H2O, Na2EDTA, 2:1 DMM/CH3CN, À58C to
of this template in a cascade 08C, 87%; f) TBAF, THF, RT, 69%.
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10007

T. F. Jamison et al.
drolysis of the epoxides. This yield could be improved to
54% when using water buffered to pH 7.0 with less hydroly-
sis byproducts being observed (Table 3, entry 2). This result
confirms that the 1,3-dioxane-5-ol template initiates the first
cyclization with near complete endo selectivity. Further-
more, in this case, the second cyclization proved equally se-
lective as predicted by our previous results with fused THP
rings (Scheme 6).[10c] Curious to see if these trends persisted
under other reaction conditions, we subjected the epoxide to
Scheme 5. Synthesis of diepoxide 35. Reagents and reaction conditions:
a) DIBAL-H, CH2Cl2, À788C, 65%; b) 1-phenyltetrazole-5-thiol, DIAD,
PPh3, THF, RT, then H2O2, (NH4)6Mo7O24, CH3CH2OH, RT, 93%; c) 6,
KHMDS, DMPU, THF, À788C to RT, 61% (85% brsm); d) Shi ketone,
Oxone, K2CO3, nBu4NHSO4, Na2B4O7·10H2O, Na2EDTA, 2:1 DMM/
CH3CN, À58C to 08C, 88%; e) TBAF, THF, RT, 78%. DIAD=diiso-
propyl azodicarboxylate, DIBAL=diisobutylaluminium, DMPU=N,N’-
dimethylpropylene urea, HMDS=hexamethyldisilazide.
process. Sulfone 32 was prepared in three steps from availa-
ble 3-pentenoic acid and subjected to Julia/Barbier condi-
tions with aldehyde 6 (Scheme 5). Shi epoxidation of 33 (to
give diepoxide 34) proved sluggish as the rate of epoxida-
tion of the alkene closest to the template proving to be slow.
Gratifyingly, under prolonged reaction times, exhaustive ep-
oxidation was possible, and removal of the TBDPS group
with TBAF furnished diepoxide 35.
Subjecting diepoxy alcohol 35 to cyclization at 708C in de-
ionized water led to the isolation of the 6,6,6 adduct 36 in
51% yield with no 6,6,5 adduct 37 isolated (Table 3,
entry 1).[17] The balance of the material was the result of hy-
Scheme 6. Proposed order of selectivity for cascade cyclizations.
basic conditions and noted a 1.4:1 ratio of 32/33 (Table 3,
entry 3). No 6,5,5 ring products were observed, indicating
that the first cyclization is highly endo controlled, as gov-
erned by the dioxane ring. Again, the poor endo/exo selec-
tivity for the second ring closure may be predicted should
this be determined by the THP ring (Table 2).[10b] A similar
trend was noted when the reaction was performed in the
presence of CSA, a 2.5:1 ratio of 32/33 being observed
(Table 3, entry 4). In both cases, a higher yield of the isolat-
ed products were observed owing to increased rates of reac-
tion mitigating destruction of the epoxide rings.
Table 3. Cyclization ratios for diepoxide 35.
We propose that the selectivity is determined by several
cooperative factors, the two most important being, generally,
the conformation of the template ring and the stabilization
of developing positive charge in the transition state. It has
been documented that the preferred conformation of 1,3-di-
oxane-5-ol is characterized by an axially orientated hydroxy
moiety that forms two hydrogen bonds with the endocyclic
oxygen atoms (Scheme 7).[18] This diaxial conformation
places the reactive sites too far apart to interact in a produc-
tive manner, thus explaining the enhanced stability of the
Entry
Promoter
36
37
Yield [%][a]
1
2
3
4
DI H2O[b]
1
1
1.4
2.5
0
0
1
1
50
54
75
85
pH 7.0 buffer[c]
[d]
Cs2CO3
CSA[e]
[a] Yield of isolated product. [b] Reaction performed in DI H2O (0.02m)
at 708C for 5 days. [c] Reaction performed in potassium phosphate based
buffer at pH 7.0 (0.02m) at 708C for 5 days. [d] Reaction performed in
MeOH at RT for 3 days. [e] Reaction performed in CH2Cl2 at RT for
3 days.
Scheme 7. Proposed selectivity model of the cyclization of epoxy alcohol
1.
10008
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

Selective Epoxy Alcohol Cyclizations
FULL PAPER
epoxy alcohol starting material. We previously reasoned
that the regioselectivity of the cyclization of the THP-tem-
plated substrate could be explained by the involvement of a
reactive twist conformation,[10a] which aligns the attacking
hydroxy group at the optimal 1008 trajectory to the distal
carbon atom of the epoxide.[19] A plausible explanation for
the enhanced selectivity, therefore, is the tendency for 1,3-
dioxane rings to even more readily adopt such a twist con-
formation.[20] As reported by Eliel and Sister over 40 years
ago, 1,3-dioxanes often adopt more puckered conformations
Experimental Section
General considerations: All reactions were run under aerobic conditions
(air) with flame-dried glassware using standard techniques for manipulat-
ing air-sensitive compounds unless otherwise stated. Anhydrous solvents
were obtained either by filtration through drying columns or by distilla-
tion over sodium and calcium hydride. Flash column chromatography
was performed using 230–400 mesh silica with the indicated solvent
system according to standard techniques. Analytical thin-layer chroma-
tography (TLC) was performed on precoated, glass-backed silica gel
plates. Visualization of the developed chromatogram was performed by
using either UV absorbance (254 nm), aqueous potassium permanganate,
vanillin, p-anisaldehyde, or a combination of these. Preparative high-per-
formance liquid chromatography (HPLC) was performed using normal-
phase elution on a system equipped with simultaneous diode-array UV
detection. Data are reported as follows: (column type, eluent, flow rate;
retention time (tr)). Automated flash chromatography was performed
using Biotage Isolera One Instruments. Nuclear magnetic resonance
spectra were recorded either on 300, 400, 500, or 600 MHz Bruker or
Varian spectrometers. Chemical shifts for 1H NMR spectra are recorded
in parts per million from tetramethylsilane with the solvent resonance as
the internal standard (chloroform, d=7.27 ppm or benzene, d=7.15;
app=apparent, m=multiplet and br=broad). Chemical shifts for
13C NMR spectra were recorded in parts per million from tetramethylsi-
lane using the central peak of deuterochloroform (77.36 ppm), or C6D6
(128 ppm) as the internal standard. Optical rotations were measured on a
Jasco Model 1010 polarimeter at 589 nm. High-resolution mass spectra
(HRMS) were obtained on a Bruker Daltronics APEXIV 4.7 Tesla Four-
ier-transform ion cyclotron resonance mass spectrometer. Infrared spec-
tra were recorded using an Agilent Cary 630 FTIR. Gas chromatography
(GC) data were collected on a Varian CP-3800 GC using an Agilent
CHIRALDEX U-TA column (30 mꢃ0.25 mm).
À
owing to the presence of shortened endocyclic C O bonds
that in turn distort the ring bond angles.[18c,d] This result was
further corroborated by Rychnovski et al., who in 1993
demonstrated, through various means, that substitution at
either the 4- or 6-positions orients the ring into the twist
conformation to minimize steric interaction with C2 sub-
À
stituents, interactions that are augmented by the short C O
bonds.[20a] These effects are absent in the THP system be-
cause the ring is only slightly distorted by the single endocy-
clic oxygen atom. Notably, particularly with unbiased alco-
hol 1, there is increased selectivity under acidic conditions
(Figure 2, Table 2, Scheme 4), possibly owing to the develop-
ment of positive charge at the distal site, resulting in a con-
certed conformational/electronic effect reinforcing reactivity
at the endo position. This phenomenon is further highlighted
by alcohol 15a (Table 2), where in presence of CSA, the se-
lectivity was very high for the dioxane-templated substrate,
yet only 6:1 for the THP-templated substrate.[10b] The pres-
ence of a proximal methyl group (15b and 23) serves to sta-
bilize developing positive charge in the transition state, lead-
ing to cyclization at the hindered position. Moreover, that
19b is not observed (see Table 2) may be a consequence of
the decreased nucleophilicity of the hydroxyl group resulting
from the presence of an additional endocyclic oxygen atom
and a sterically congested transition state.
AHCTUNGTERG(NNUN 4S,5R)-4-(((2R,3R)-3-Methyloxiran-2-yl)methyl)-1,3-dioxan-5-ol (1): To
a solution of silyl ether 7 (0.102 g, 0.248 mmol) in THF (0.400 mL) was
added a 1m solution of TBAF in THF (0.497 mL, 0.497 mmol). The reac-
tion solution was stirred at RT for 20 min, then applied directly to a
column of SiO2 (gradient: 30–100% EtOAc in hexanes) to yield 1 as a
colorless oil (0.038 g, 89%). Chiraldex G-TA: 30 mꢃ0.25 mmꢃ0.12 mm
film thickness, 1258C 10 min, then 28CminÀ1 to 1508C, then 108CminÀ1
to 1808C, then hold 10 min, tR =20.09 min. Rf =0.70 (EtOAc); [a]2D0 = +
2.1 (c=1.00 in CH2Cl2); 1H NMR (500 MHz, C6D6): d=4.90 (d, J=
6.0 Hz, 1H), 4.25 (d, J=6.0 Hz, 1H), 4.05 (dd, J=10.6, 5.1 Hz, 1H), 3.61
(app td, J=9.5, 5.2 Hz, 1H), 3.24 (ddd, J=9.3, 5.6, 4.0 Hz, 1H), 3.15 (app
t, J=10.3 Hz, 1H), 2.82 (ddd, J=6.5, 3.8, 2.3 Hz, 1H), 2.48 (qd, J=5.2,
2.2 Hz, 1H), 2.43 (br s, 1H), 1.93 (app dt, J=14.9, 3.9 Hz, 1H), 1.71
(ddd, J=14.9, 6.7, 5.8 Hz, 1H), 0.98 ppm (d, J=5.2 Hz, 3H); 13C NMR
(125 MHz, C6D6): d=93.6, 80.2, 71.6, 65.9, 56.5, 54.3, 34.8, 17.8 ppm; IR
(thin film, NaCl): n˜ =3423, 2964, 2923, 2853, 2774, 1653, 1457, 1438, 1381,
1257, 1225, 1175, 1150, 1073, 1028 cmÀ1; HRMS (DART): m/z calcd for
C8H14O4: 175.0965 [M+H]+; found 175.0973.
Conclusion
In summary, we have demonstrated the application of a
powerful new 1,3-dioxane template that provides very high
endo selectivity in epoxy alcohol cyclization reactions, selec-
tivity that is opposite to that very commonly observed and
predicted by the Baldwin rules for ring closure. It is now
clear that the number of oxygen atoms in the template ring
plays a critical role in epoxide-opening regioselectivity. This
high level of selectivity also extends to cascade processes
under a variety of reaction conditions. Not only is this tem-
plate highly selective, it offers the possibility of subsequent
removal of the methylene acetal, thus revealing two chemi-
cally different hydroxyl groups.[21] Thus, we anticipate that
this new scaffold will enable the synthesis of ladder polyeth-
er natural products. Such investigations of this motif are un-
derway in our laboratories and will be reported in due
course.
(E)-Ethyl
4-((2R,4S,5R)-5-hydroxy-2-(4-methoxyphenyl)-1,3-dioxan-4-
yl)but-2-enoate (3): To a slurry of 2-deoxyribose 2 (17.6 g, 131 mmol) in
THF (300 mL) was added (carbethoxymethylene)triphenylphosphorane
(Ph3P=CHCO2Et, 50 g, 143 mmol). Upon heating at 758C for 3 h, the sol-
ution became
a homogenous golden yellow. The reaction was then
cooled to RT and concentrated under reduced pressure to afford the
crude enoate as a heavy, orange-brown oil, which was used in the subse-
quent PMP acetal protection step without further purification. The crude
diol was dissolved in CH2Cl2 (250 mL), to which was added first p-anisal-
dehyde dimethyl acetal (50 mL, 299 mmol) and then (+/À)-camphorsul-
fonic acid (CSA; 6.0 g, 25 mmol). Upon addition of CSA, the reaction
solution immediately turned paler, becoming a light orange brown. After
stirring the mixture for 16 h at RT, the reaction was quenched with sat.
aq. NH4Cl (ca. 200 mL) and the resulting reaction mixture was extracted
with CH2Cl2 (3ꢃ200 mL), dried with Na2SO4, and concentrated under re-
duced pressure. The residue was purified by column chromatography
(gradient: 20–50% EtOAc in hexanes) to afford PMP acetal 3 as a
yellow oil (36.1 g, 86%). Rf =0.46 (1:1 EtOAc/hexanes); [a]2D2 =À26.1
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10009

T. F. Jamison et al.
(c=1.23 in CDCl3); 1H NMR (400 MHz, CDCl3): d=7.40 (d, J=8.5 Hz,
2H), 7.08 (dt, J=15.7, 7.2 Hz, 1H), 6.89 (d, J=8.8 Hz, 2H), 5.96 (dt, J=
15.7, 1.4 Hz, 1H), 5.44 (s, 1H), 4.24 (dd, J=10.2, 4.2 Hz, 1H), 4.20 (q, J=
7.1 Hz, 2H), 3.80 (s, 3H), 3.72–3.53 (m, 3H), 2.80 (dddd, J=15.1, 6.9, 3.2,
1.6 Hz, 1H), 2.56 (app dtd, J=15.1, 7.5, 1.4 Hz, 1H), 2.19 (d, J=5.2 Hz,
1H), 1.30 ppm (t, J=7.1 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=
166.7, 160.2, 144.8, 130.2, 127.6, 123.9, 113.8, 101.1, 80.5, 71.4, 65.5, 60.6,
55.5, 34.8, 14.4 ppm; IR (thin film): n˜ =3462, 2979, 2935, 1716, 1655,
1615, 1251, 1174, 1080, 1034 cmÀ1; HRMS (ESI): m/z calcd for C17H22O6:
323.1489 [M+H]+; found: 323.1482.
sidual ozone, and then PPh3 (664 mg, 2.53 mmol) was added, at which
point the cold bath was removed and the reaction was allowed to warm
to RT over 1.5 h. The solution was then concentrated under reduced pres-
sure and purified by column chromatography (gradient: 5–20% EtOAc
in hexanes) to provide 6 as a colorless oil (788 mg, 93%). Rf =0.44 (1:4
EtOAc/hexanes); [a]D22 =À6.5 (c=1.26 in CH2Cl2); 1H NMR (400 MHz,
CDCl3): d=9.71 (dd, J=2.6, 1.4 Hz,1H), 7.68–7.61 (m, 4H), 7.51–7.38
(m, 6H), 4.87 (d, J=6.2 Hz, 1H), 4.58 (d, J=6.2 Hz, 1H), 4.00 (app td,
J=9.0, 2.8 Hz, 1H), 3.90 (dd, J=10.7, 4.9 Hz, 1H), 3.57 (app td, J=9.4,
5.0 Hz, 1H), 3.40 (app t, J=10.3 Hz, 1H), 2.80 (ddd, J=16.6, 2.8, 1.4 Hz,
1H), 2.41 (ddd, J=16.6, 9.1, 2.7 Hz, 1H), 1.06 ppm (s, 9H); 13C NMR
(100 MHz, CDCl3): d=200.7, 135.9, 135.9, 133.3, 132.7, 130.5, 130.4,
128.2, 128.0, 93.3, 77.5, 71.6, 67.1, 45.8, 27.1, 19.4 ppm; IR (thin film): n˜ =
3072, 2932, 2858, 1729, 1473, 1217, 1171, 1066, 1035 cmÀ1; HRMS (ESI):
m/z calcd for C22H28O4Si: 407.1649 [M+Na]+; found: 406.1660.[22]
A
6-((tert-butyldiphenylsilyl)oxy)-5,7-dihydroxyhept-2-
enoate (4): To a solution of alcohol 3 (17.6 g, 53 mmol) in DMF (25 mL)
was added first imidazole (5.4 g, 80 mmol) and then TBDPSCl (16.3 mL,
17.5 g, 64 mmol). The resulting viscous solution was stirred at RT for
16 h, then quenched by addition of sat. aq. NH4Cl (ca. 50 mL). The aque-
ous layer was separated and extracted with EtOAc (3ꢃ150 mL) and the
combined organic layers were washed with brine (ca. 50 mL), dried over
MgSO4, filtered, and concentrated in vacuo to afford the crude silyl ether
as a pale yellow oil (Rf =0.52; 1:4 EtOAc/hexanes). This crude material
was carried forward into acetal hydrolysis without further purification.
To a solution of one half of this crude material in a 12:3:1 v/v/v mixture
of MeOH/THF/H2O (212 mL) was added p-toluenesulfonic acid monohy-
drate (1.0 g, 5.3 mmol). The solution was heated to 708C for 2 h. The sol-
ution was cooled in an ice bath, quenched with Et3N (1.1 mL, 810 mg,
8.0 mmol), and concentrated under reduced pressure to a clear, golden
yellow oil, which was purified by column chromatography (gradient: 20–
50% EtOAc in hexanes) to provide 4 as a heavy yellow oil (10.1 g, 86%
over 2 steps). Rf =0.67 (1:1 EtOAc/hexanes); [a]2D2 = +21.6 (c=4.9 in
CDCl3); 1H NMR (500 MHz, CDCl3): d=7.71–7.66 (m, 4H), 7.49–7.44
(m, 2H), 7.44–7.38 (m, 4H), 6.88 (app dt, J=15.6, 7.3 Hz, 1H), 5.82 (app
dt, J=15.7, 1.5 Hz, 1H), 4.18 (q, J=7.1 Hz, 2H), 3.85 (app dq, J=8.8,
4.4 Hz, 1H), 3.73 (ddd, J=11.1, 6.4, 3.7 Hz, 1H), 3.68–3.59 (m, 2H), 2.58
(d, J=4.5 Hz, 1H), 2.44 (dddd, J=14.8, 7.1, 3.8, 1.5 Hz, 1H), 2.28 (dddd,
J=14.8, 8.9, 7.4, 1.4 Hz, 1H), 2.08 (app t, J=6.0 Hz, 1H), 1.29 (t, J=
7.1 Hz, 3H), 1.10 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=166.4,
145.2, 136.0, 135.9, 133.6, 133.0, 130.4, 130.3, 128.1, 128.1, 124.0, 75.7,
73.0, 63.9, 60.5, 36.1, 27.2, 19.6, 14.5 ppm; IR (thin film): n˜ =3448, 3072,
2932, 2858, 1718, 1654, 1473, 1428, 1271, 1166, 1112, 1045 cmÀ1; HRMS
(ESI): m/z calcd for C25H34O5Si: 465.2068 [M+Na]+; found: 465.2049.
tert-ButylACTHNUTRGNEU(GN ((4S,5R)-4-(((2R,3R)-3-methyloxiran-2-yl)methyl)-1,3-dioxan-
5-yl)oxy)diphenylsilane (7): A dry round-bottom flask was charged with
CrCl2 (1.92 g, 15.6 mmol), to which was added dry THF (20 mL) to
afford a pale-green slurry. Meanwhile, 6 (0.500 g, 1.30 mmol) was added
to a dry round-bottom flask, and the flask was evacuated using high
vacuum and backfilled with argon. This procedure was repeated, and
then 6 was diluted in dry THF (3 mL). 1,1-Diiodoethane S1 (1.1 g,
3.9 mmol) was added, and this mixture was added dropwise to the CrCl2
slurry. The flask contained 6 and MeCHI2 was washed out with a further
portion of dry THF (3 mL). The mixture was stirred at RT for 3.5 h, over
which time the color changed from pale green to chocolate brown. The
reaction was then quenched by pouring the mixture into brine (ca.
50 mL). The aqueous layer was separated and extracted with Et2O (4ꢃ
100 mL), and the combined organic layers were washed with brine, dried
over MgSO4, filtered, and concentrated in vacuo to afford the crude
alkene as an orange-brown oil. 1H NMR analysis revealed this material
to be of ca. 91:9 E/Z stereopurity, and contained some unreacted 1,1-diio-
doethane, which does not adversely affect the next step. This crude mate-
rial was carried forward into Shi epoxidation without further purification.
Rf of alkene is 0.64 (1:9 EtOAc/hexanes).
To a solution of this crude alkene in 2:1 v/v DMM/MeCN (48 mL) was
added
a m Na2EDTA
0.05m solution of Na2B4O7·10H2O in 4ꢃ10À4
(24 mL), nBu4HSO4 (0.098 g, 0.066 mmol), and Shi ketone (0.455 g,
1.76 mmol). This biphasic mixture was stirred vigorously at 08C. To this
mixture was added, simultaneously over 40 min through a syringe pump,
a solution of Oxone (4.32 g, 7.02 mmol) in 4ꢃ10À4 m Na2EDTA (15.8 mL)
and a 0.89m solution of K2CO3 (15.8 mL, 14.0 mmol). After the K2CO3
and Oxone solutions had been added, the resulting mixture was stirred
an additional 60 min, at which point it was diluted with water (ca.
100 mL). The aqueous layer was separated and extracted with EtOAc
(3ꢃ100 mL), and the combined organic layers were washed with brine,
dried over MgSO4, filtered, and concentrated under reduced. 1H NMR
analysis at this point indicated incomplete conversion, so the crude mate-
rial was subjected again to identical epoxidation conditions and worked
up as before. The crude epoxide 7 was purified by column chromatogra-
phy (gradient: 3–25% EtOAc in hexanes) to provide 7, a colorless oil, as
an inseparable mixture of diastereomers (0.408 g of a less than 10:1 mix-
ture of diastereomers, 76% over 2 steps). Epoxide 7 could be purified
slightly further by using preparative HPLC (Supelco SUPELCOSIL LC-
SI, 20 mm diameter achiral SiO2 column, 5 mm particle size, 25 cm length;
0.4% iPrOH in hexanes, 20 mLminÀ1; tR of major and minor diastereom-
ers is 7.3 min; collect only the first quarter of the peak to obtain material
in high d.r.) to afford 7 in 10:1–15:1 d.r. The compound was obtained in
higher purity by using Biotage high-performance silica-gel column (Biot-
age SNAP HP-Sil, gradient: 0–20% Et2O in benzene). The two diaster-
eomers appear as one spot by TLC, but early fractions are enriched in
the minor diastereomer and late fractions contain 7 in high purity (>20:1
d.r., as determined using 1H NMR analysis). Rf =0.56 (1:4 EtOAc/hex-
anes); [a]2D2 =À6.6 (c=0.81 in CDCl3); 1H NMR (500 MHz, CDCl3): d=
7.68–7.61 (m, 4H), 7.48–7.43 (m, 2H), 7.43–7.37 (m, 4H), 4.90 (d, J=
6.1 Hz, 1H), 4.55 (d, J=6.1 Hz, 1H), 3.84 (dd, J=10.7, 3.6 Hz, 1H),
3.58–3.51 (m, 2H), 3.37–3.30 (m, 1H), 2.77–2.70 (m, 2H), 1.96 (ddd, J=
14.5, 5.8, 1.8 Hz, 1H), 1.72 (ddd, J=13.8, 8.3, 5.0 Hz, 1H), 1.28 (d, J=
5.1 Hz, 3H), 1.04 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=136.0,
(E)-Ethyl 4-((4S,5R)-5-((tert-butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)but-
2-enoate (5): To a solution of diol 4 (5.5 g, 12.0 mmol) in CH2Cl2 (50 mL)
was added first dimethoxymethane (1.6 mL, 18.0 mmol) and then
BF3·OEt2 (2.2 mL, 18.0 mmol). The solution, which turned yellow upon
addition of BF3·OEt2, was stirred 1.25 h. at RT, then quenched with sat.
aq. NaHCO3 (ca. 50 mL). The aqueous layer was separated, diluted with
brine (ca. 50 mL), and extracted with CH2Cl2 (3ꢃ50 mL). The combined
organic layers were washed with brine (ca. 50 mL), dried over MgSO4,
filtered, and concentrated under reduced pressure to afford crude 5 as a
colorless oil, which was purified by automated chromatography (Biotage
SNAP HP-Sil, gradient: 0–30% EtOAc in hexanes) to afford 5 as a color-
less oil (5.05 g, 90%). Rf =0.55 (1:4 EtOAc/hexanes); [a]2D2 = +14.2 (c=
2.6 in CDCl3); 1H NMR (500 MHz, CDCl3): d=7.68–7.61 (m, 4H), 7.49–
7.44 (m, 2H), 7.44–7.38 (m, 4H), 6.91 (ddd, J=15.7, 7.5, 6.7 Hz, 1H),
5.81 (app dt, J=15.7, 1.5 Hz, 1H), 4.89 (d, J=6.1 Hz, 1H), 4.53 (d, J=
6.1 Hz, 1H), 4.19 (q, J=7.1 Hz, 2H), 3.90–3.85 (m, 1H), 3.57–3.49 (m,
2H), 3.35 (dd, J=10.7, 9.5 Hz, 1H), 2.73 (app ddt, J=15.1, 6.7, 1.9 Hz,
1H), 2.18 (dddd, J=15.3, 8.6, 7.6, 1.4 Hz, 1H), 1.30 (t, J=7.1 Hz, 3H),
1.06 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=166.5, 144.8, 136.0,
135.9, 133.5, 132.9, 130.4, 130.3, 128.1, 128.0, 123.9, 93.4, 81.0, 71.6, 67.3,
60.4, 34.5, 27.1, 19.4, 14.5 ppm; IR (thin film, NaCl): n˜ =3072, 2932, 2858,
1720, 1657, 1473, 1428, 1267, 1184, 1111, 1033 cmÀ1; HRMS (ESI): m/z
calcd for C26H34O5Si: 477.2068 [M+Na]+; found: 477.2080.
2-((4S,5R)-5-((tert-Butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)acetaldehyde
(6): Enoate 5 (1.00 g, 2.20 mmol) was dissolved in CH2Cl2 (22 mL), and
the resulting solution cooled to À788C. A stream of ozone was bubbled
through until a pale blue color evolved (about 2.5 h). Conversion of 5
may also be monitored by TLC (Rf of 5, 0.55; Rf of ozonide, 0.50; 1:4
EtOAc/hexanes). Argon was bubbled through the solution to remove re-
10010
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

Selective Epoxy Alcohol Cyclizations
FULL PAPER
135.9, 133.7, 132.9, 130.3, 130.2, 128.1, 127.9, 93.3, 80.3, 71.6, 67.4, 56.8,
54.2, 34.1, 27.1, 19.5, 17.8 ppm; IR (thin film): n˜ =3072, 2931, 2857, 1473,
1428, 1176, 1112 cmÀ1; HRMS (ESI): m/z calcd for C24H32O4Si: 435.1962
[M+Na]+; found: 435.1975.
ing solution was allowed to warm to RT over 16 h and was then diluted
with water (30 mL). The resulting product was extracted with Et2O (3ꢃ
50 mL) and the combined extracts were dried with MgSO4. Purification
by automated chromatography (Biotage SNAP HP-Sil, gradient: 2–20%
EtOAc in hexanes) gave 13 as a colorless oil (0.158 g, 49%). Rf =0.55
(1:9 EtOAc/hexanes); [a]2D2 =À3.5 (c=0.36 in CH2Cl2); 1H NMR
(600 MHz, CDCl3): d=7.72–7.60 (m, 4H), 7.49–7.40 (m, 6H), 5.19 (app
dt, J=6.8, 1.3 Hz 1H), 4.93 (d, J=6.0 Hz, 1H), 4.55 (d, J=6.1 Hz, 1H),
3.86 (app dd, J=10.7, 5.1 Hz, 1H), 3.59–3.56 (m, 1H), 3.45 (app dt, J=
8.8, 4.4 Hz, 1H), 3.35 (t, J=10.3 Hz, 1H), 2.62 (dd, J=15.3, 6.8 Hz, 1H),
2.01 (app dt, J=15.3, 6.7 Hz, 1H), 1.72 (s, 3H), 1.58 (s, 3H), 1.09 ppm (s,
9H); 13C NMR (150 MHz, CDCl3): d=135.84, 135.74, 133.66, 133.60,
132.8, 129.96, 129.90, 127.75, 127.63, 119.7, 93.3, 82.4, 71.4, 67.3, 30.2,
26.9, 25.8, 19.2, 17.9 ppm; IR (thin film): n˜ =2930, 2856, 1472, 1428, 1109,
1039 cmÀ1. HRMS (ESI): m/z calcd for C25H34O3Si: 433.2169 [M+Na]+;
found: 433.2175.
(E)-Ethyl 4-((4S,5R)-5-((tert-butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)-2-
methylbut-2-enoate (10): To a dry flask with stir bar was added (carbe-
thoxyethylidene)tri-phenylphosphorane (0.941 g, 2.6 mmol) and CH2Cl2
(5 mL). Aldehyde 6 was diluted in CH2Cl2 (1 mL) and added by syringe.
The resulting solution was stirred for 16 h, concentrated under reduced
pressure, and purified by automated chromatography (Biotage SNAP
KP-Sil, gradient: 5–40% EtOAc in hexanes) to furnish 10 as a colorless
oil (0.957 g, 89%). Rf =0.51 (1:4 EtOAc/hexanes); [a]2D2 = +10.7 (c=4.95
in CH2Cl2); 1H NMR (500 MHz, CDCl3): d=7.71–7.65 (m, 4H), 7.48–
7.39 (m, 6H), 6.82 (td, J=6.9, 1.4 Hz, 1H), 4.90 (d, J=6.1 Hz, 1H), 4.54
(d, J=6.1 Hz, 1H), 4.20 (q, J=7.1 Hz, 2H), 3.92 (dd, J=10.7, 4.9 Hz,
1H), 3.63–3.53 (m, 2H), 3.38 (t, J=10.2 Hz, 1H), 2.74–2.69 (m, 1H), 2.17
(dd, J=16.1, 7.7 Hz, 1H), 1.77 (app d, J=1.2 Hz, 3H), 1.30 (t, J=7.1 Hz,
3H), 1.08 ppm (s, 9H); 13C NMR (125 MHz, CDCl3): d=167.6, 137.3,
135.63, 135.59, 133.2, 132.6, 129.96, 129.91, 129.5, 127.72, 127.60, 93.1,
81.0, 71.2, 67.2, 60.3, 30.9, 26.8, 19.1, 14.2, 12.5 ppm; IR (thin film): n˜ =
2959, 2932, 2857, 1710, 1428, 1219, 1111, 948, 821 cmÀ1; HRMS (ESI): m/
z calcd for C27H36O5Si: 491.2224 [M+Na]+; found: 491.2220.
tert-Butyl
yl)oxy)diphenylsilane (14): To a solution of 13 (0.170 g. 0.410 mmol) in
2:1 v/v DMM/MeCN (15 mL) was added 0.05m solution of
ACHTUNGERTN(NUNG ((4S,5R)-4-(((R)-3,3-dimethyloxiran-2-yl)methyl)-1,3-dioxan-5-
a
Na2B4O7·10H2O in 4ꢃ10À4 m Na2EDTA (7.7 mL), nBu4HSO4 (0.029 g,
0.010 mmol), and Shi ketone (0.170 g, 0.656 mmol). This biphasic mixture
was stirred vigorously at 08C. To this mixture was added, simultaneously
over 40 min by syringe pump, a solution of Oxone (1.45 g, 4.72 mmol) in
4ꢃ10À4 m Na2EDTA (5.5 mL) and a 0.89m solution of K2CO3 (5.5 mL,
5.4 mmol). After the K2CO3 and Oxone solutions had been added, the re-
sulting mixture was stirred an additional 60 min, at which point it was di-
luted with water (ca. 100 mL). The aqueous layer was separated and ex-
tracted with EtOAc (3ꢃ100 mL), and the combined organic layers were
washed with brine, dried over MgSO4, filtered, and concentrated under
reduced pressure. The crude epoxide 14 was purified by automated chro-
matography (Biotage SNAP HP-Sil, gradient: 0–40% EtOAc in hexanes)
to provide 14, a colorless oil, as an inseparable mixture of diastereomers
(0.151 g of an approximate 6:1 mixture of diastereomers, 86%). Epoxide
14 could be purified further by using a Biotage high-performance silica-
gel column using benzene as the mobile phase (Biotage SNAP HP-Sil,
gradient: 0–20% Et2O in benzene). The two diastereomers appeared as
one spot by TLC, but early fractions are enriched in the minor diaster-
eomer and late fractions contain 14 in high purity (>20:1 d.r. by
1H NMR analysis). Rf =0.57 (1:4 EtOAc/hexanes); [a]2D2 =À10.6 (c=1.47
in CH2Cl2); 1H NMR (600 MHz, CDCl3): d=7.68–7.61 (m, 4H), 7.47–
7.38 (m, 6H), 4.93 (d, J=6.0 Hz, 1H), 4.56 (d, J=6.1 Hz, 1H), 3.86 (app
dd, J=10.8, 4.8 Hz, 1H), 3.62–3.54 (m, 2H), 3.34 (t, J=10.2 Hz, 1H),
2.86 (t, J=6.1 Hz, 1H), 2.03 (ddd, J=14.5, 6.3, 2.6 Hz, 1H), 1.64 (ddd,
J=14.6, 8.7, 5.9 Hz, 1H), 1.31 (s, 3H), 1.25 (s, 3H), 1.07 ppm (s, 9H);
13C NMR (150 MHz, CDCl3): d=135.76, 135.67, 133.4, 132.5, 130.08,
129.96, 127.82, 127.66, 93.1, 80.8, 71.3, 67.4, 61.1, 57.4, 31.3, 26.9, 24.7,
19.2, 18.8 ppm; IR (thin film, NaCl): n˜ =2928, 2856, 1462, 1399, 2856,
(E)-4-((4S,5R)-5-((tert-Butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)-2-methyl-
but-2-en-1-ol (11): Ester 10 (0.783 g, 1.67 mmol) was diluted in toluene
(10 mL) and cooled to À788C. A solution of DIBAL-H in toluene (1m,
3.8 mL, 3.8 mmol) was added by syringe and the reaction mixture was
stirred for 30 min, and then warmed to RT for 1 h. After this time the
solution was cooled to 08C, methanol was added (20 mL), followed by a
saturated solution of Rochelleꢀs salt. After extraction with EtOAc (3ꢃ
100 mL), drying with Na2SO4, and concentration under reduced pressure,
10 was purified by automated chromatography (Biotage SNAP HP-Sil,
gradient: 25–100% EtOAc in hexanes) to give a colorless oil. (0.510 g,
72%). Rf =0.11 (1:4 EtOAc/hexanes); [a]2D2 =À1.4 (c=9.36 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=7.72–7.66 (m, 4H), 7.48–7.40 (m, 6H),
5.46 (t, J=6.8 Hz, 1H), 4.90 (d, J=6.0 Hz, 1H), 4.53 (d, J=6.10 Hz, 1H),
3.97 (s, 2H), 3.90 (app dd, J=10.7, 5.0 Hz, 1H), 3.62–3.58 (m, 1H), 3.48
(app td, J=8.8, 1.8 Hz, 1H), 3.37 (t, J=10.3 Hz, 1H), 2.65 (dd, J=15.3,
6.9 Hz, 1H), 2.07–2.01 (m, 2H), 1.61 (s, 3H), 1.10 ppm (s, 9H); 13C NMR
(150 MHz, CDCl3): d=136.8, 135.73, 135.64, 133.4, 132.7, 129.93, 129.88,
127.70, 127.58, 120.9, 93.1, 82.0, 71.2, 68.4, 67.2, 29.7, 26.8, 19.1, 13.7 ppm;
IR (thin film): n˜ =3430, 2931, 2856, 1427, 1103, 1033 cmÀ1; HRMS (ESI):
m/z calcd for C25H34O4Si: 449.2119 [M+Na]+; found: 449.2190.
(E)-4-((4S,5R)-5-((tert-Butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)-2-methyl-
but-2-en-1-yl methanesulfonate (12): Alcohol 11 (0.460 g, 1.1 mmol) was
added to a round-bottomed flask with stir bar. To this was added toluene
(1.1 mL), Et3N (0.307 mL, 2.2 mmol), and trimethylamine hydrochloride
salt (0.011 g, 0.11 mmol). The resulting mixture was cooled to 08C after
which MsCl (0.124 mL, 1.6 mmol) in toluene (1.6 mL) was added drop-
wise. After stirring at this temperature for 1 h N,N-dimethylethylenedia-
mine (0.150 mL) was added and the solution stirred for an additional
20 min. The reaction was diluted with water (20 mL), extracted with
CH2Cl2 (3ꢃ50 mL), washed with brine (20 mL), and dried with Na2SO4.
Purification by column chromatography (20% EtOAc in hexanes) fur-
nished 12 as a pale yellow oil (0.431 g, 78%). Rf =0.23 (1:4 EtOAc/hex-
1172, 1109 cmÀ1
[M+Na]+; found: 449.2123.
AHCTUNGTERG(NNUN 4S,5R)-4-(((R)-3,3-Dimethyloxiran-2-yl)methyl)-1,3-dioxan-5-ol (15a):
; HRMS (ESI): m/z calcd for C25H34O4Si: 449.2119
To a solution of silyl ether 14 (0.129 g, 0.302 mmol) in THF (0.500 mL)
was added a 1m solution of TBAF in THF (0.5 mL, 0.500 mmol). The re-
action solution was stirred at RT for 20 min, then applied directly to a
column of SiO2 (eluted with a gradient 30–100% EtOAc in hexanes) to
yield 15a as a colorless oil (0.033 g, 58%). Rf =0.18 (40% EtOAc/hex-
anes); [a]2D2 =À6.6 (c=0.59 in CH2Cl2); 1H NMR (300 MHz, C6D6): d=
4.93 (d, J=6.1 Hz, 1H), 4.31 (d, J=6.1 Hz, 1H), 4.10 (dd, J=10.8,
5.4 Hz, 1H), 3.71 (app td, J=9.2, 5.1 Hz, 1H), 3.36 (m, 1H), 3.22 (t, J=
10.3 Hz, 2H), 3.08 (dd, J=7.3, 4.5 Hz, 1H), 1.95–1.86 (m, 2H), 1.07 ppm
(s, 6H); 13C NMR (125 MHz, C6D6): d=93.4, 80.5, 71.2, 60.3, 57.5, 31.4,
24.7, 18.7 ppm; IR (thin film): n˜ =3423, 2964, 2923, 2853, 2774, 1653,
1457, 1438, 1381, 1257, 1225, 1175, 1150, 1073, 1028 cmÀ1; HRMS (ESI):
m/z calcd for C9H16O4: 211.0941 [M+Na]+; found: 211.0955.
1
anes); [a]2D2 = +1.1 (c=0.46 in CH2Cl2); H NMR (400 MHz, CDCl3): d=
7.70–7.63 (m, 4H), 7.48–7.40 (m, 6H), 5.61 (d, J=1.0 Hz, 1H), 4.87 (d,
J=6.1 Hz, 1H), 4.58 (s, 2H), 4.52 (d, J=6.1 Hz, 1H), 3.89 (app dd, J=
10.7, 5.0 Hz, 1H), 3.56–3.53 (m, 1H), 3.47 (t, J=4.4 Hz, 1H), 3.35 (t, J=
10.2 Hz, 1H), 2.98 (s, 3H), 2.06 (s, 1H), 1.65 (s, 3H), 1.07 ppm (s, 9H);
13C NMR (100 MHz, CDCl3): d=135.75, 135.69, 133.3, 132.7, 130.4,
130.06, 130.01, 128.5, 127.81, 127.69, 93.1, 81.5, 76.1, 71.3, 67.1, 37.9, 30.0,
26.9, 19.2, 13.9 ppm; IR (thin film): n˜ =3073, 2934, 2858, 1356, 1174,
1100, 924 cmÀ1
[M+Na]+; found: 527.1903.
tert-Butyl(((4S,5R)-4-(3-methylbut-2-en-1-yl)-1,3-dioxan-5-yl)oxy)diphe-
; HRMS (ESI): m/z calcd for C26H36O6SSi: 527.1894
1-((4S,5R)-5-((tert-Butyldiphenylsilyl)oxy)-1,3-dioxan-4-yl)propan-2-one
(16): To a dry flask equipped with a stir bar was added aldehyde 6
(1.004 g, 2.6 mmol) and Et2O (13 mL). The resulting solution was cooled
to 08C after which was added MeMgBr (3m, 1.6 mL, 3.9 mmol). After
stirring for 45 min at this temperature the reaction was quenched with
ACHTUNGTRENNUNG
nylsilane (13): To a dried flask equipped with a stir bar was added LAH
(0.012 g, 0.32 mmol) and Et2O (8 mL). The mixture was cooled to 08C,
after which was added 12 (0.400 g, 0.79 mmol) in 4 mL Et2O. The result-
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10011

T. F. Jamison et al.
sat. aq. NH4Cl, extracted with Et2O (3ꢃ50 mL), and dried with MgSO4
to furnish the intermediate alcohol as a yellow oil (0.844 g, 2.1 mmol)
that was carried forward to the oxidation.
10.3 Hz, 1H), 2.87 (t, J=5.5 Hz, 1H), 2.28 (dd, J=14.2, 1.8 Hz, 1H), 1.40
(s, 3H), 1.38–1.34 (m, 1H), 1.24 (d, J=5.6 Hz, 3H), 1.09 ppm (s, 9H);
13C NMR (150 MHz, CDCl3): d=135.77, 135.69, 132.6, 130.06, 129.95,
128.3, 127.82, 127.66, 93.0, 79.9, 71.4, 67.7, 61.0, 59.0, 34.2, 26.9, 23.0, 19.2,
14.6 ppm; IR (thin film, NaCl): n˜ =2931, 2857, 1471, 1427, 1170, 1103,
1035 cmÀ1; HRMS (ESI): m/z calcd for C25H34O4Si: 449.2119 [M+Na]+;
found: 449.2113.
To this oil was added CH2Cl2 (15 mL), MS 4 ꢄ (3.5 g), 4-methylmorpho-
line N-oxide (0.370 g, 3.2 mmol), tetrapropylammonium perruthenate
(0.037 g, 0.1 mmol); the resulting mixture was stirred for 3 h. The mixture
was filtered on Celite and purified by automated chromatography (Biot-
age SNAP HP-Sil, gradient: 10–50% EtOAc in hexanes) to give 16 as a
colorless oil (0.602 g, 60% over 2 steps). Rf =0.48 (3:7 EtOAc/hexanes);
[a]2D2 =À8.3 (c=0.61 in CH2Cl2); 1H NMR (600 MHz, CDCl3): d=7.67–
7.63 (m, 4H), 7.49–7.40 (m, 6H), 4.85 (d, J=6.1 Hz, 1H), 4.57 (d, J=
6.2 Hz, 1H), 3.97 (td, J=9.4, 1.9 Hz, 1H), 3.90 (dd, J=10.7, 5.0 Hz, 1H),
3.52 (td, J=9.5, 4.9 Hz, 1H), 3.40 (t, J=10.4 Hz, 1H), 2.77 (dd, J=16.1,
2.1 Hz, 1H), 2.37 (m, 1H), 2.12 (s, 3H), 1.04 ppm (s, 9H);13C NMR
(150 MHz, CDCl3): d=206.4, 135.68, 135.65, 133.1, 132.7, 130.09, 130.05,
127.83, 127.73, 93.0, 78.2, 71.3, 66.9, 45.6, 30.9, 26.8, 19.2 ppm; IR (thin
film): n˜ =2933, 2858, 1721, 1428, 1108, 1034 cmÀ1; HRMS (ESI): m/z
calcd for C23H30O4Si: 421.1806 [M+Na]+; found: 421.1808.
AHCTUNGTRENNUNG
a
0.700 mmol). The reaction solution was stirred at RT for 20 min, then ap-
plied directly to a column of SiO2 (eluted with a gradient 30–100%
EtOAc in hexanes) to yield 15b as a colorless oil (0.079 g, 90%). Rf =
0.28 (40% EtOAc/hexanes); [a]2D2 =À18.4 (c=2.47 in CH2Cl2); 1H NMR
(300 MHz, C6D6): d=4.86 (d, J=6.1 Hz, 1H), 4.25 (d, J=6.0 Hz, 1H),
4.10 (app dd, J=10.7, 5.0 Hz, 1H), 3.42 (app td, J=8.1, 5.2 Hz, 1H),
3.36–3.32 (m, 1H), 3.21 (t, J=10.4 Hz, 1H), 2.96 (t, J=3.6 Hz, 1H), 2.58
(q, J=5.6 Hz, 1H), 1.94 (dd, J=14.4, 5.1 Hz, 1H), 1.72 (dd, J=14.4,
7.3 Hz, 1H), 1.20 (s, 3H), 1.04 ppm (d, J=5.6 Hz, 3H); 13C NMR
(125 MHz, C6D6): d=93.2, 79.5, 71.1, 67.1, 61.8, 59.1, 35.9, 22.9,
14.3 ppm; IR (thin film): n˜ =3405, 2968, 2856, 1455, 1379, 1168, 1059,
1025, 938 cmÀ1; HRMS (ESI): m/z calcd for C8H14O4: 211.0941 [M+Na]+;
found: 211.0957.
tert-ButylACHTUNGTRENNUNG(((4S,5R)-4-((E)-2-methylbut-2-en-1-yl)-1,3-dioxan-5-yl)oxy)di-
phenylsilane (17): To a dry flask equipped with a stir bar was added eth-
yltriphenylphosphonium bromide (0.459 g, 1.24 mmol) and THF (17 mL).
The resuling slurry was cooled to 08C followed by the addition of nBuLi
(2.65m, 0.396 mL, 1.05 mmol). After stirring at this temperature for
40 min the orange/red solution was cooled to À788C. Ketone 16 (0.379 g,
0.952 mmol) in THF (1.5 mL) was added dropwise by syringe. The
ketone was azeotroped with toluene prior to use. The reaction mixture
was allowed to warm to RT over 16 h after which it was quenched with
sat. aq. NH4Cl (ca. 50 mL), extracted with EtOAc (3ꢃ50 mL), and dried
over MgSO4. Purification by automated chromatography (Biotage SNAP
HP-Sil, gradient: 1–40% EtOAc in hexanes) gave 17 as a colorless oil
(0.273 g, 71%,). The alkene was isolated as an inseparable 2.7:1 mixture
of E/Z isomers, and used without further purification. Rf =0.79 (1:4
EtOAc/hexanes); [a]D22 =À9.3 (c=0.40 in CH2Cl2); 1H NMR (600 MHz,
CDCl3): d=7.72–7.65 (m, 6H), 7.49–7.40 (m, 9H), 5.41–5.39 (m, 1H),
4.92 (m, 1H), 4.53 (m, 1H), 3.88 (dd, J=10.5, 4.1 Hz, 1H), 3.58–3.54 (m,
3H), 3.38–3.35 (m, 1H), 2.56 (d, J=14.0 Hz, 1H), 2.10–2.06 (m, 1H),
1.78 (s, 3H), 1.58 (d, J=6.7 Hz 3H), 1.08 ppm (m, 14H);13C NMR
(150 MHz, CDCl3): d=136.13, 136.04, 133.8, 133.12, 132.93, 130.28,
130.23, 128.05, 127.94, 121.8, 93.5, 81.5, 77.5, 77.3, 77.1, 71.6, 68.4, 34.1,
27.2, 24.2, 19.5, 13.7 ppm; IR (thin film): n˜ =2930, 2857, 1471, 1428, 1107,
tert-ButylACTHNUTRGNEU(GN ((4S,5R)-4-(2-methylallyl)-1,3-dioxan-5-yl)oxy)diphenylsilane
(21): To a dry flask equipped with a stir bar was added methyltriphenyl-
phosphonium bromide (0.260 g, 0.728 mmol) and THF (8.5 mL). The re-
sulting slurry was cooled to 08C followed by the addition of nBuLi
(2.32m, 0.241 mL, 0.56 mmol). After stirring at this temperature for
40 min the orange/red solution was cooled to À788C. Ketone 16 (0.225 g,
0.560 mmol) in THF (1.0 mL) was added dropwise by syringe. The
ketone was azeotroped with toluene prior to use. The reaction mixture
was allowed to warm to RT over 16 h after which it was quenched with
sat. aq. NH4Cl (ca. 50 mL), extracted with EtOAc (3ꢃ50 mL), and dried
over MgSO4. Purification by automated chromatography (Biotage SNAP
HP-Sil, gradient: 0–20% EtOAc in hexanes) gave 21 as a colorless oil
(0.209 g, 94%,). Rf =0.59 (1:9 EtOAc/hexanes); [a]2D2 =À16.3 (c=0.93 in
CH2Cl2); 1H NMR (600 MHz, CDCl3): d=7.70–7.65 (m, 4H), 7.48–7.40
(m, 6H), 4.93 (d, J=6.1 Hz, 1H), 4.85 (s, 1H), 4.77 (s, 1H), 4.56 (d, J=
6.1 Hz, 1H), 3.90 (app dd, J=10.7, 5.0 Hz, 1H), 3.60 (d, J=8.5 Hz, 1H),
3.52 (d, J=5.0 Hz, 1H), 3.39 (t, J=10.3 Hz, 1H), 2.66 (d, J=14.5 Hz,
1H), 1.95 (app dd, J=14.5, 10.2 Hz, 1H), 1.79 (s, 3H), 1.10 ppm (s,
9H);13C NMR (150 MHz, CDCl3): d=142.6, 136.10, 136.03, 133.7, 133.1,
130.30, 130.26, 128.06, 127.98, 112.9, 93.5, 80.8, 71.7, 67.9, 40.3, 27.2, 22.8,
1035, 946 cmÀ1
[M+Na]+; found: 433.2183.
tert-Butyl(((4S,5R)-4-(((2R,3R)-2,3-dimethyloxiran-2-yl)methyl)-1,3-
dioxan-5-yl)oxy)diphenylsilane (18): To solution of 17 (0.270 g,
; HRMS (ESI): m/z calcd for C25H34O3Si: 433.2169
ACHTUNGTRENNUNG
19.5 ppm; IR (thin film): n˜ =2931, 2852, 1471, 1428, 1107, 1036, 948 cmÀ1
;
a
HRMS (ESI): m/z calcd for C24H32O3Si: 419.2013 [M+Na]+; found:
0.659 mmol) in 2:1 v/v DMM/MeCN (25 mL) was added a 0.05m solution
of Na2B4O7·10H2O in 4ꢃ10À4 m Na2EDTA (12 mL), nBu4HSO4 (0.029 g,
0.010 mmol), and Shi ketone (0.272 g, 1.05 mmol). This biphasic mixture
was stirred vigorously at 08C. To this mixture was added, simultaneously
over 40 min by syringe pump, a solution of Oxone (2.32 g, 7.57 mmol) in
4ꢃ10À4 m Na2EDTA (8.5 mL) and a 0.89m solution of K2CO3 (8.5 mL,
7.57 mmol). After the K2CO3 and Oxone solutions had been added, the
resulting mixture was stirred an additional 60 min, at which point it was
diluted with water (ca. 100 mL). The aqueous layer was separated and
extracted with EtOAc (3ꢃ100 mL), and the combined organic layers
were washed with brine, dried over MgSO4, filtered, and concentrated
under reduced. The crude epoxide 18 was purified by automated chroma-
tography (Biotage SNAP HP-Sil, gradient: 0–40% EtOAc in hexanes) to
provide 18, a colorless oil, as an inseparable mixture of diastereomers
(0.255 g of an approximate 3:1 mixture of diastereomers, 91%). Epoxide
18 could be purified further by a Biotage high-performance silica-gel
column using benzene as the mobile phase (Biotage SNAP HP-Sil, gradi-
ent: 0–20% Et2O in benzene). The two diastereomers appear as a single
spot by TLC, but early fractions are enriched in the minor diastereomer
419.2012.
tert-Butyl
yl)oxy)diphenylsilane (22): To a solution of 21 (0.208 g, 0.524 mmol) in
2:1 v/v DMM/MeCN (20 mL) was added 0.05m solution of
ACHTUNGERTN(NUNG ((4S,5R)-4-(((R)-2-methyloxiran-2-yl)methyl)-1,3-dioxan-5-
a
Na2B4O7·10H2O in 4ꢃ10À4 m Na2EDTA (10 mL), nBu4HSO4 (0.029 g,
0.010 mmol), and Shi ketone (0.240 g, 0.839 mmol). This biphasic mixture
was stirred vigorously at 08C. To this mixture was added, simultaneously
over 40 min by syringe pump, a solution of Oxone (1.85 g, 6.026 mmol) in
4ꢃ10À4 m Na2EDTA (6.9 mL) and a 0.89m solution of K2CO3 (6.9 mL,
6.14 mmol). After the K2CO3 and Oxone solutions had been added, the
resulting mixture was stirred an additional 60 min, at which point it was
diluted with water (ca. 100 mL). The aqueous layer was separated and
extracted with EtOAc (3ꢃ100 mL), and the combined organic layers
were washed with brine, dried over MgSO4, filtered, and concentrated
under reduced pressure. The crude epoxide 22 was purified by automated
chromatography (Biotage SNAP HP-Sil, gradient: 1–30% EtOAc in hex-
anes) to provide 22, a colorless oil, as an inseparable mixture of diaster-
eomers (0.196 g of an approximate 6:1 mixture of diastereomers, 89%).
Epoxide 22 could be purified further by using a Biotage high-perform-
ance silica-gel column using benzene as the mobile phase (Biotage SNAP
HP-Sil, gradient: 0–20% Et2O in benzene). The two diastereomers
appear as one spot by TLC, but early fractions are enriched in the minor
diastereomer and late fractions contain 22 in high purity (>14:1 d.r. by
1
and late fractions contain 18 in high purity (>9:1 d.r. by H NMR analy-
sis). Rf =0.58 (1:4 EtOAc/hexanes); [a]2D2 =À14.3 (c=1.69 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=7.71–7.66 (m, 4H), 7.50–7.42 (m, 6H),
4.91 (d, J=6.0 Hz, 1H), 4.58 (d, J=6.1 Hz, 1H), 3.87 (dd, J=10.7,
5.0 Hz, 1H), 3.64–3.62 (m, 1H), 3.51 (td, J=9.4, 5.0 Hz, 1H), 3.38 (t, J=
10012
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

Selective Epoxy Alcohol Cyclizations
FULL PAPER
1
1H NMR analysis). Rf =0.48 (1:4 EtOAc/hexanes); [a]2D2 =À14.3 (c=1.69
in CH2Cl2); 1H NMR (600 MHz, CDCl3): d=7.65 (m, 4H), 7.48–7.39 (m,
6H), 4.90 (d, J=6.1 Hz, 1H), 4.57 (d, J=6.1 Hz, 1H), 3.87 (app dd, J=
10.5, 4.8 Hz, 1H), 3.62 (app dt, J=8.7, 1.6 Hz, 1H), 3.44–3.43 (m, 1H),
3.37 (t, J=10.2 Hz, 1H), 2.61 (app dd, J=13.9, 4.9 Hz, 2H), 2.30 (d, J=
14.0 Hz, 1H), 1.38 (s, 3H), 1.26 (dd, J=14.0, 10.5 Hz, 1H), 1.06 ppm (s,
9H); 13C NMR (150 MHz, CDCl3) d 136.04, 135.98, 133.5, 132.9, 130.32,
130.26, 128.07, 127.98, 93.2, 80.1, 71.7, 67.6, 55.3, 39.4, 27.2, 22.0, 21.4,
19.5 ppm; IR (thin film, NaCl): n˜ =2931, 2857, 1471, 1427, 1170, 1103,
1035 cmÀ1; HRMS (ESI): m/z calcd for C24H32O4Si: 435.1962 [M+Na]+;
found: 435.1961.
tions contain 25 in high purity (>14:1 d.r. by H NMR analysis). Rf =0.46
(1:4 EtOAc/hexanes); [a]2D2 =À22.5 (c=1.72 in CH2Cl2); 1H NMR
(600 MHz, CDCl3): d=7.65 (m, 4H), 7.48–7.38 (m, 6H), 4.91 (d, J=
6.1 Hz, 1H), 4.58 (d, J=6.1 Hz, 1H), 3.87 (dd, J=10.7, 5.0 Hz, 1H), 3.66
(app dt, J=9.4, 1.8 Hz, 1H), 3.48 (m, 1H), 3.38 (t, J=10.4 Hz, 1H), 3.04
(m, 1H), 2.77 (app dd, J=5.1, 4.0 Hz, 1H), 2.41 (app dd, J=5.1, 2.8 Hz,
1H), 1.91 (app ddd, J=14.3, 7.4, 2.3 HZ, 1H), 1.52 (m, 1H), 1.04 ppm (s,
9H); 13C NMR (150 MHz, CDCl3): d=136.06, 135.98, 133.6, 133.0,
130.30, 130.27, 128.06, 127.97, 93.4, 79.9, 71.6, 67.7, 49.2, 48.0, 35.3, 27.1,
19.5 ppm; IR (thin film): n˜ =3049, 2931, 2857, 1472, 1428, 1107, 1035,
947 cmÀ1; HRMS (ESI): m/z calcd for C23H30O4Si: 421.1806 [M+Na]+;
found: 421.1811.
ACHTUNGTRENNUNG(4S,5R)-4-(((R)-2-Methyloxiran-2-yl)methyl)-1,3-dioxan-5-ol (23): To a
(E)-Pent-3-en-1-ol (31): To a flame-dried flask with stir bar was added
CH2Cl2 (60 mL) and 3-pentenoic acid (1.50 mL, 14.8 mmol). The result-
ing solution was cooled to À788C after which DIBAL-H (60 mL, 1.0m in
CH2Cl2). After stirring at À788C for 30 min the reaction mixture was
warmed to RT for 1 h after which a saturated solution of Rochelleꢀs salt
was added and stirred for 12 h. The organic layer was separated and the
aqueous phase extracted with 3ꢃ50 mL of CH2Cl2, dried with Na2SO4
and concentrated under reduced pressure. The crude alcohol 31 was puri-
fied by automated chromatography (Biotage SNAP HP-Sil, gradient: 30–
100% EtOAc in hexanes) to provide 31, a colorless oil (0.816 g, 65%).
Rf =0.44 (1:1 EtOAc/hexanes); 1H NMR (500 MHz; CDCl3): d=5.61–
5.55 (m, 1H), 5.44–5.38 (m, 1H), 3.65–3.62 (t, J=6.3 Hz, 2H), 2.29–2.24
(q, J=6.4 Hz, 2H), 1.70 ppm (app dquintet, J=6.4, 1.3 Hz, 3H); spec-
trum matches commercial standard.
solution of silyl ether 22 (0.155 g, 0.376 mmol) in THF (0.564 mL) was
added a 1m solution of TBAF in THF (0.564 mL, 0.564 mmol) at 08C.
The reaction solution was stirred at 08C for 20 min, then applied directly
to a column of SiO2 pretreated with 5% Et3N in EtOAc (eluted with a
gradient 20–100% EtOAc in hexanes) to yield 23 as a colorless oil
(0.065 g, 95%). Rf =0.68 (100% EtOAc). [a]2D2 =À27.5 (c=0.70 in
1
CH2Cl2); H NMR (300 MHz, C6D6): d=4.85 (d, J=6.1, 1H), 4.24 (d, J=
6.1 Hz, 1H), 4.06 (dd, J=10.4, 4.5 Hz, 1H), 3.41–3.27 (m, 2H), 3.20 (q,
J=11.0 Hz, 1H), 2.78 (br s, 1H), 2.37 (d, J=4.9 Hz, 1H), 2.25 (d, J=
4.9 Hz, 1H), 1.86–1.67 (m, 2H), 1.17 ppm (d, J=6.7 Hz, 3H); 13C NMR
(100 MHz, C6D6): d=92.9, 79.3, 70.9, 66.6, 55.5, 55.0, 40.5, 20.8 ppm; IR
(thin film): n˜ =3425, 2924, 2860, 1393, 1165, 1061, 1027, 941 cmÀ1; HRMS
(ESI): m/z calcd for C8H14O4: 197.0784 [M+Na]+; found: 197.0790.
ACHTUNGTRENNUNG(((4S,5R)-4-Allyl-1,3-dioxan-5-yl)oxy)(tert-butyl)diphenylsilane (24): To a
(E)-5-(Pent-3-en-1-ylsulfonyl)-1-phenyl-1H-tetrazole (32): Alcohol 31
(0.50 g, 5.8 mmol), 1-phenyltetrazole-5-thiol (2.1 g, 11.6 mmol), PPh3
(2.28 g, 8.7 mmol) were added to a flame-dried flask with THF (50 mL).
The resulting solution was cooled to 08C under an argon after which
DIAD (2 mL, 10.4 mmol) was added over 5 min by syringe pump. The
solution was stirred at 08C for 5 min, then at RT for 3 h and then
quenched with sat. aq. NaCl (ca. 100 mL). After extraction with Et2O
(3ꢃ100 mL), drying with Na2SO4, and concentrating under reduced pres-
sure, the crude sulfide was purified by automated chromatography (Biot-
age SNAP HP-Sil, gradient: 7–60% EtOAc in hexanes) to provide the
dry flask equipped with a stir bar was added methyltriphenylphosphoni-
um bromide (0.260 g, 0.728 mmol) and THF (8.5 mL). The resuling slurry
was cooled to 08C followed by the addition of nBuLi (2.32m, 0.241 mL,
0.56 mmol). After stirring at this temperature for 40 min the orange/red
solution was cooled to À788C. Aldehyde 6 (0.220 g, 0.560 mmol) in THF
(1.0 mL) was added dropwise by syringe. The aldehyde was azeotroped
with toluene prior to use. The reaction mixture was allowed to warm to
RT over 16 h after which it was quenched with sat. aq. NH4Cl (ca.
50 mL), extracted with EtOAc (3ꢃ50 mL), and dried over MgSO4. Purifi-
cation by automated chromatography (Biotage SNAP HP-Sil, gradient:
0–20% EtOAc in hexanes) gave 24 as a colorless oil (0.187 g, 87%,).
Rf =0.59 (1:9 EtOAc/hexanes); [a]2D2 =À8.4 (c=2.77 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=7.69–7.64 (m, 4H), 7.48–7.39 (m, 6H),
5.85–5.78 (m, 1H), 5.07–5.04 (m, 2H), 4.92 (d, J=6.0 Hz, 1H), 4.54 (d,
J=6.1 Hz, 1H), 3.84 (app dd, J=10.7, 4.9 Hz, 1H), 3.54–3.49 (m, 2H),
3.34 (t, J=10.2 Hz, 1H), 2.68–2.64 (m, 1H), 2.13–2.10 (m, 1H), 1.08 ppm
(s, 9H);13C NMR (150 MHz, CDCl3): d=136.07, 135.98, 134.6, 133.7,
133.0, 130.28, 130.21, 128.04, 127.93, 117.4, 93.5, 82.0, 71.6, 67.3, 36.2,
27.2, 19.5 ppm; IR (thin film): n˜ =3072, 2932, 2856, 1472, 1428, 1107,
sulfide,
a colorless oil (1.4 g, 98%). Rf =0.54 (3:7 EtOAc/hexanes);
1H NMR (400 MHz; CDCl3): d=7.60–7.54 (m, 5H), 5.59–5.53 (m, 1H),
5.48–5.42 (m, 1H), 3.43 (t, J=7.2 Hz, 2H), 2.54–2.49 (q, J=7.2 Hz, 2H),
1.67 ppm (app dq, J=6.2, 1.3 Hz, 3H).
This sulfide was diluted in EtOH (80 mL) and cooled to 08C after which
H2O2 (2.6 g of 30% wt H2O, 24.0 mmol) and (NH4)6Mo7O24 (0.71 g,
0.58 mmol) were added. The reaction mixture was allowed to reach RT
over 16 h quenched with sat. aq. NaCl (ca. 100 mL). After extraction
with Et2O (3ꢃ100 mL), drying with Na2SO4, and concentrating under re-
duced pressure, crude 32 was purified by automated chromatography (Bi-
otage SNAP HP-Sil, gradient: 5–60% EtOAc in hexanes) to provide sul-
fone 32, a colorless oil (1.47 g, 93%). Rf =0.48 (3:7 EtOAc/hexanes);
1H NMR (400 MHz; CDCl3): d=7.72 (m, 2H), 7.68–7.60 (m, 3H), 5.66–
5.63 (m, 1H), 5.46–5.42 (m, 1H), 3.81–3.79 (m, 2H), 2.67 (q, J=7.6 Hz,
2H), 1.69 ppm (dd, J=6.4, 0.6 Hz, 3H); spectrum was consistent with re-
ported values.[23]
1038, 703 cmÀ1
[M+Na]+; found: 435.1869.
tert-Butyl(((4S,5R)-4-((R)-oxiran-2-ylmethyl)-1,3-dioxan-5-yl)oxy)diphe-
; HRMS (ESI): m/z calcd for C23H30O3Si: 405.1856
ACHTUNGTRENNUNG
nylsilane (25): To a solution of 24 (0.236 g, 0.612 mmol) in 2:1 v/v DMM/
MeCN (22 mL) was added a 0.05m solution of Na2B4O7·10H2O in 4ꢃ
10À4 m Na2EDTA (11.5 mL), nBu4HSO4 (0.029 g, 0.010 mmol), and Shi
ketone (0.279 g, 0.979 mmol). This biphasic mixture was stirred vigorous-
ly at 08C. To this mixture was added, simultaneously over 40 min by sy-
tert-ButylACTHUNRTGNEUNG(((4S,5R)-4-((2E,5E)-hepta-2,5-dien-1-yl)-1,3-dioxan-5-yl)oxy)-
ringe pump,
a m
solution of Oxone (2.16 g, 7.04 mmol) in 4ꢃ10À4
diphenylsilane (33): To a dry round-bottomed flask with stir bar was
added 32 (0.500 g, 1.83 mmol). To this was added THF (9 mL) and the re-
sulting solution was cooled to À788C. To this solution was added a fresh-
ly prepared solution of KHMDS in THF over 5 min by syringe pump
(2.10 mL, 1m, 2.10 mmol) and then stirred at this temperature for 1 h. Al-
dehyde 6 (0.806 g, 2.10 mmol) dried by azeotroping with toluene was di-
luted in THF (5 mL) and added over 5 min by syringe pump. The reac-
tion was left to warm to room temperature over 16 h, quenched with sat.
aq. NaCl (ca. 100 mL). After extraction with Et2O (3ꢃ100 mL), drying
with MgSO4, and concentrating under reduced pressure, crude 33 was pu-
rified by automated chromatography (Biotage SNAP HP-Sil, gradient: 5–
60% EtOAc in hexanes) to provide 33, a colorless oil (0.487 g, 61%).
Rf =0.81 (3:7 EtOAc/hexanes); [a]2D2 = +3.2 (c=0.67 in CH2Cl2);
1H NMR (400 MHz; CDCl3): d=7.68–7.63 (m, 4H), 7.47–7.44 (m, 2H),
7.41–7.38 (m, 4H), 5.45–5.39 (m, 4H), 4.91 (d, J=6.0 Hz, 1H), 4.53 (d,
Na2EDTA (8.0 mL) and a 0.89m solution of K2CO3 (8.0 mL, 7.12 mmol).
After the K2CO3 and Oxone solutions had been added, the resulting mix-
ture was stirred an additional 60 min, at which point it was diluted with
water (ca. 100 mL). The aqueous layer was separated and extracted with
EtOAc (3ꢃ100 mL), and the combined organic layers were washed with
brine, dried over MgSO4, filtered, and concentrated under reduced. The
crude epoxide 25 was purified by automated chromatography (Biotage
SNAP HP-Sil, gradient: 0–40% EtOAc in hexanes) to provide 25, a col-
orless oil, as an inseparable mixture of diastereomers (0.2135 g of an ap-
proximate 2:1 mixture of diastereomers, 87%). Epoxide 25 could be puri-
fied further by using a Biotage high-performance silica-gel column using
benzene as the mobile phase (Biotage SNAP HP-Sil, gradient: 1–20%
Et2O in benzene). The two diastereomers appear as one spot by TLC,
but early fractions are enriched in the minor diastereomer and late frac-
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10013

T. F. Jamison et al.
J=6.1 Hz, 1H), 3.82 (app dd, J=10.7, 5.0 Hz, 1H), 3.52 (app dt, J=9.3,
4.7 Hz, 1H), 3.45 (app td, J=8.5, 2.5 Hz, 1H), 3.32 (t, J=10.3 Hz, 1H),
2.67 (d, J=4.5 Hz, 2H), 2.61–2.58 (m, 1H), 2.09 (app dd, J=9.8, 3.8 Hz,
1H), 1.68–1.67 (m, 3H), 1.06 ppm (d, J=5.4 Hz, 9H); 13C NMR
(150 MHz, CDCl3): d=136.08, 135.98, 133.8, 133.1, 131.7, 130.23, 130.17,
129.7, 128.0, 127.9, 126.2, 125.8, 93.5, 82.3, 71.6, 67.2, 35.9, 34.8, 27.2, 19.5,
then washed out of the reaction vial with a large volume of MeOH (typi-
cally 6–8 washes of ca. 2 mL each) and concentrated under reduced pres-
sure (2 torr, 408C). The combined yield was found the endo/exo ratio of
products was determined by either 1H NMR analysis or GC-FID follow-
ing isolation of both products together after column chromatography
(30–50% EtOAc in hexanes). Then the product mixture was again chro-
matographed to separate the endo product, the 6,6-fused product, from
the exo product, the 6,5-fused product. In cases of difficult separation,
acetylation of the free alcohol facilitated purification of the product mix-
tures.
18.2 ppm; IR (thin film): n˜ =2930, 2854, 1427, 1167, 1102, 960 cmÀ1
;
HRMS (ESI): m/z calcd for C27H36O3Si: 454.2772 [M+NH4]+; found:
454.2778.
tert-ButylACHTUNGTRENNUNG(((4S,5R)-4-(((2R,3R)-3-(((2R,3R)-3-methyloxiran-2-yl)meth-
yl)oxiran-2-yl)methyl)-1,3-dioxan-5-yl)oxy)diphenylsilane (34): To a solu-
Representative procedure for reaction promoted by Cs2CO3: A sample of
epoxy alcohol (5–10 mg, 0.03–0.06 mmol, >15:1 d.r.) was dissolved in a
solution of Cs2CO3 (30 equiv) in anhydrous MeOH to 0.02m in a glass
vial. The threads of the vial were lined with Teflon tape, the cap was
sealed and covered with parafilm, and the solution was stirred under air
at RT for 1–2 days. The solution was then diluted with Et2O, quenched
with sat. aq. NH4Cl, and the aqueous layer was extracted with Et2O. The
combined organics were dried with MgSO4 and concentrated under re-
duced pressure. The combined yield was found and the endo/exo ratio of
products was determined by either 1H NMR analysis or GC-FID follow-
ing isolation of both products together after column chromatography
(30–50% EtOAc in hexanes). Then the product mixture was again chro-
matographed to separate the endo product, the 6,6-fused product, from
the exo product, the 6,5-fused product. In cases of difficult separation,
acetylation of the free alcohol facilitated purification of the product mix-
tures.
tion of 33 (0.450 g, 1.03 mmol) in 2:1 v/v DMM/MeCN (38 mL) was
added
a m Na2EDTA
0.05m solution of Na2B4O7·10H2O in 4ꢃ10À4
(20 mL), nBu4HSO4 (0.062 g, 0.021 mmol), and Shi ketone (0.387 g,
1.50 mmol). This biphasic mixture was stirred vigorously at 08C. To this
mixture was added, simultaneously over 60 min by syringe pump, a solu-
tion of Oxone (3.81 g, 12.40 mmol) in 4ꢃ10À4 m Na2EDTA (14.0 mL) and
a 0.89m solution of K2CO3 (14.0 mL, 12.40 mmol). After the K2CO3 and
Oxone solutions had been added, the resulting mixture was stirred an ad-
ditional 4 h, at which point it was diluted with water (ca. 200 mL). The
aqueous layer was separated and extracted with EtOAc (3ꢃ200 mL), and
the combined organic layers were washed with brine, dried over MgSO4,
filtered, and concentrated under reduced. Analysis of the crude mixture
determined that the epoxidation was incomplete and the crude epoxy
alkene was resubjected to the identical reaction conditions. Following the
second work up the crude epoxide 34 was purified by automated chroma-
tography (Biotageꢅ SNAP HP-Sil, gradient: 1–30% EtOAc in hexanes)
to provide 34, a colorless oil, as an inseparable mixture of diastereomers
(0.424 g of an approximate 6:1 mixture of diastereomers, 88%). Epoxide
34 could be purified further by using a Biotage high-performance silica-
gel column using benzene as the mobile phase (Biotage SNAP HP-Sil,
gradient: 0–20% Et2O in benzene). The two diastereomers appear as one
spot by TLC, but early fractions are enriched in the minor diastereomer
Representative procedure for reaction promoted by CSA: A sample of
epoxy alcohol (5–10 mg, 0.03–0.06 mmol, >15:1 d.r.) was dissolved in
CH2Cl2 to 0.02m in an oven-dried round-bottom flask. To this was added
(+/À)-CSA (1 equiv), and the solution was stirred under argon at RT for
4–15 h. The solution was then quenched with sat. aq. NaHCO3, and the
aqueous layer was extracted with Et2O and dried with MgSO4. The com-
bined organics were concentrated under reduced pressure. The combined
yield was found and the endo/exo ratio of products was determined by
either 1H NMR or GC-FID following isolation of both products together
after column chromatography (30–50% EtOAc in hexanes). Then the
product mixture was again chromatographed to separate the endo prod-
uct, the 6,6-fused product, from the exo product, the 6,5-fused product.
In cases of difficult separation, acetylation of the free alcohol facilitated
purification of the product mixtures.
1
and late fractions contain 34 in high purity (>10:1 d.r. by H NMR analy-
sis). Rf =0.44 (1:4 EtOAc/hexanes); [a]2D2 =6.5 (c=1.35 in CH2Cl2);
1H NMR (500 MHz, CDCl3): d=7.64 (m, 4H), 7.46–7.37 (m, 6H), 4.89
(d, J=6.0 Hz, 1H), 4.54 (d, J=6.1 Hz, 1H), 3.84 (app dd, J=10.7, 3.5 Hz,
1H), 3.54 (app dd, J=5.7, 3.8 Hz, 2H), 3.33 (d, J=10.5 Hz, 1H), 2.83–
2.76 (m, 4H), 1.95 (app dd, J=5.9, 1.9 Hz, 1H), 1.74 (app dt, J=6.0,
4.4 Hz, 2H), 1.67 (app dd, J=7.0, 4.7 Hz, 1H), 1.32 (d, J=5.0 Hz, 3H),
1.04 ppm (s, 9H); 13C NMR (150 MHz, CDCl3): d=136.05, 135.95, 133.6,
132.8, 130.37, 130.25, 128.10, 127.95, 93.3, 80.2, 71.6, 67.3, 56.8, 55.8, 55.0,
54.7, 35.3, 33.9, 27.2, 19.5, 17.8 ppm; IR (thin film): n˜ =2927, 2855, 1470,
1174, 1104, 1033, 945 cmÀ1; HRMS (ESI): m/z calcd for C27H36O5Si:
486.2670 [M+NH4]+; found: 486.2684.
Representative procedure for reaction promoted by BF3·OEt2: A sample
of epoxy alcohol (5–10 mg, 0.03–0.06 mmol, >15:1 d.r.) was dissolved in
CH2Cl2 to 0.02m in an oven-dried round-bottom flask and cooled to
À788C. To this was added, dropwise, a 0.1m solution of BF3·OEt2 in
CH2Cl2 (0.25 equiv), and the solution was stirred at À788C under argon
for 30 min. The solution was then allowed to warm gradually to RT over
5 min. and quenched with sat. aq. NaHCO3. The aqueous layer was ex-
tracted with Et2O and dried with MgSO4. The combined organics were
concentrated under reduced pressure. The combined yield was found and
the endo/exo ratio of products was determined by either 1H NMR analy-
sis or GC-FID following isolation of both products together after column
chromatography (30–50% EtOAc in hexanes). Then the product mixture
was again chromatographed to separate the endo product, the 6,6-fused
product, from the exo product, the 6,5-fused product. In cases of difficult
separation, acetylation of the free alcohol facilitated purification of the
product mixtures.
ACHTUNGTRENNUNG(4S,5R)-4-(((2R,3R)-3-(((2R,3R)-3-Methyloxiran-2-yl)methyl)oxiran-2-
yl)methyl)-1,3-dioxan-5-ol (35): To a solution of silyl ether 34 (0.270 g,
0.620 mmol) in THF (1.00 mL) was added a 1m solution of TBAF in
THF (1.00 mL, 1.00 mmol) at 08C. The reaction solution was stirred at
08C for 30 min, then applied directly to a column of SiO2 pretreated with
5% Et3N in EtOAc (eluted with a gradient 20–100% EtOAc in hexanes)
to yield 35 as a colorless oil (0.112 g, 78%). Rf =0.65 (100% EtOAc);
[a]2D2 = +23.2 (c=0.75 in CH2Cl2); 1H NMR (500 MHz, C6D6): d=4.99
(d, J=6.1 Hz, 1H), 4.57 (d, J=6.2 Hz, 1H), 4.15 (app dd, J=10.8, 5.1 Hz,
1H), 3.68 (m, 1H), 3.45 (m, 1H), 3.32 (t, J=10.4 Hz, 1H), 3.00 (m, 1H),
2.93–2.87 (m, 2H), 2.84–2.80 (m, 2H), 2.12 (dt, J=15.0, 4.0 Hz, 1H),
1.30 ppm (d, J=9.0 Hz, 3H); 13C NMR (126 MHz, C6D6): d=93.4, 79.8,
71.1, 65.4, 56.8, 55.65, 55.52, 55.0, 35.1, 34.4, 17.7 ppm; IR (thin film): n˜ =
3413, 2923, 2855, 1437, 1174, 1073, 1027, 940 cmÀ1; HRMS (ESI): m/z
calcd for C11H18O5: 231.1227 [M+H]+; found: 231.1232.
(4aR,6S,7R,8aS)-6-Methylhexahydropyrano
G
G
(8):
Chiraldex G-TA 30 mꢃ0.25 mmꢃ0.12 mm film thickness, 1258C for
10 min, then 28CminÀ1 to 1508C, then 108CminÀ1 to 1808C, then hold
10 min, tR =17.59 min; Rf =0.65 (100% EtOAc); [a]2D2 =À8.8 (c=0.15 in
CH2Cl2); 1H NMR (500 MHz, CDCl3): d=5.02 (d, J=6.2 Hz, 1H), 4.62
(d, J=6.2 Hz, 1H), 4.18 (dd, J=10.4, 4.2 Hz, 1H), 3.44–3.36 (m, 2H),
3.34–3.23 (m, 3H), 2.43 (app dt, J=11.4, 4.3 Hz, 1H), 1.61 (app q, J=
11.2 Hz, 1H), 1.56 (s, 1H), 1.30 ppm (s, 3H); 13C NMR (125 MHz,
CDCl3): d=93.9, 78.8, 76.8, 73.5, 71.6, 69.4, 38.3, 18.0 ppm; IR (thin film,
NaCl): n˜ =3365, 2963, 2923, 2862, 1463, 1401, 1349, 1279, 1211, 1161,
1118, 1075, 1059, 1024 cmÀ1; HRMS (DART): m/z calcd for C8H14O4:
175.0965 [M+H]+; found: 175.0967.
General cyclization protocols[24]
Representative procedure for reaction in water or buffered water: A
sample of epoxy alcohol (5–10 mg, 0.03–0.06 mmol, >15:1 d.r.) was dis-
solved in deionized water to 0.02m in a glass vial. The threads of the vial
were lined with Teflon tape, the cap was sealed and covered with paraf-
ilm, and the solution was stirred at the desired temperature and time. In
the case of buffered reactions, 0.1m phosphate buffer was used to control
the pH value of the reaction mixture. Upon completion, the solution was
10014
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016

Selective Epoxy Alcohol Cyclizations
FULL PAPER
(R)-1-((4aR,6S,7aS)-Tetrahydro-4H-furo
[3,2-d]
E
zations performed on 15 mg of alcohol starting material. Rf =0.48 (100%
1
(9): Chiraldex G-TA 30 mꢃ0.25 mmꢃ0.12 mm film thickness, 1258C for
10 min, then 28CminÀ1 to 1508C, then 108CminÀ1 to 1808C, then hold
10 min, tR =18.86 min; Rf =0.53 (100% EtOAc); [a]2D2 =À17.9 (c=0.045
in CH2Cl2); 1H NMR (500 MHz, CDCl3): d=5.11 (d, J=6.2 Hz, 1H),
4.63 (d, J=6.2 Hz, 1H), 4.40 (dd, J=9.7, 3.7 Hz, 1H), 4.07–4.00 (m, 2H),
3.58 (app t, J=9.6 Hz, 1H), 3.48–3.40 (m, 2H), 2.23 (app dt, J=11.1,
5.8 Hz, 1H), 2.09 (app dt, J=11.1, 9.2 Hz, 1H), 1.86 (d, J=3.8 Hz, 1H),
1.18 ppm (d, J=6.2 Hz, 3H); 13C NMR (125 MHz, CDCl3): d=94.2, 81.8,
80.7, 73.8, 72.0, 69.1, 29.3, 18.2 ppm; IR (thin film, NaCl): n˜ =3450, 2921,
2851, 1463, 1256, 1155, 1126, 1053 cmÀ1; HRMS (DART): m/z calcd for
C8H14O4: 197.0784 [M+Na]+; found: 197.0794.
EtOAc); [a]2D2 =À6.1 (c=0.30, CH2Cl2); H NMR (600 MHz, CDCl3): d=
5.02 (d, J=6.2 Hz, 1H), 4.63 (d, J=6.2 Hz, 1H), 4.18 (app dd, J=10.4,
4.3 Hz, 1H), 3.43 (t, J=10.0 Hz, 1H), 3.40–3.11 (m, 6H), 2.39 (app ddt,
J=11.8, 8.4, 3.9 Hz, 2H), 1.63 (dd, J=22.1, 11.0 Hz, 2H), 1.31 ppm (d,
J=6.1 Hz, 3H); 13C NMR (150 MHz, CDCl3): d=94.0, 78.7, 77.19, 77.07,
76.5, 73.9, 71.6, 69.2, 38.5, 35.0, 18.0 ppm; IR (thin film): n˜ =3379, 2928,
2872, 1452, 1164, 1116, 1079, 1031 cmÀ1; HRMS (ESI): m/z calcd for
C11H18O5: 231.1227 [M+H]+; found: 231.1234; chemical shifts for the di-
oxane methylene protons, as well at the presence of a terminal CH3
group indicates an all-6,6,6 system.
(R)-1-((4aR,5aS,7S,8aR,9aS)-Octahydrofuro[2’,3’:5,6]pyranoACTHUNTGRENNG[U 3,2-d]-
A
4aR,7R,8aS)-6,6-DimethylhexahydropyranoACTHNUTRGENN[UG 3,2-d]ACHUTNGTREN[NGUN 1,3]dioxin-7-ol (19a):
0.30 in CH2Cl2); 1H NMR (600 MHz, CDCl3): d=5.04 (d, J=6.2 Hz,
1H), 4.61 (d, J=8.1 Hz, 1H), 4.23 (app dd, J=10.2, 4.3 Hz, 1H), 4.12–
4.09 (m, 1H), 4.02 (td, J=6.2, 3.1 Hz, 1H), 3.51–3.42 (m, 3H), 3.34–3.30
(m, 1H), 2.55 (dt, J=10.6, 3.9 Hz, 1H), 2.12 (dt, J=11.5, 5.9 Hz, 1H),
2.00 (t, J=10.4 Hz, 1H), 1.92 (d, J=3.2 Hz, 1H), 1.70 (q, J=11.1 Hz,
1H), 1.15 ppm (d, J=6.5 Hz, 3H); 13C NMR (150 MHz, CDCl3): d=94.2,
82.1, 81.5, 77.7, 75.0, 73.7, 69.37, 69.21, 35.2, 28.7, 18.1 ppm; IR (thin
film): n˜ =3436, 2927, 2876, 1449, 1163, 1108, 1076, 1020 cmÀ1; HRMS
(ESI): m/z calcd for C11H18O5: 231.1227 [M+H]+; found: 231.1231; chem-
ical shifts for the dioxane methylene protons, as well at the presence of a
terminal CH3 group indicates an all-6,6,5 system.
Rf =0.67 (100% EtOAc); [a]2D2 =18.6 (c=0.85 in CH2Cl2); 1H NMR
(600 MHz, CDCl3): d=5.01 (d, J=6.2 Hz, 1H), 4.62 (d, J=6.2 Hz, 1H),
4.10 (app dd, J=10.0, 4.5 Hz, 1H), 3.56 (app dt, J=11.5, 5.2 Hz, 1H),
3.44 (app td, J=9.5, 4.6 Hz, 1H), 3.38 (t, J=10.0 Hz, 1H), 3.24 (app ddd,
J=11.8, 9.0, 4.3 Hz, 1H), 2.20 (app dt, J=11.6, 4.5 Hz, 1H), 1.73 (m,
2H), 1.26 ppm (d, J=20.5 Hz, 6H); 13C NMR (150 MHz, CDCl3): d=
94.0, 77.7, 76.7, 73.3, 70.1, 66.9, 34.3, 27.7, 16.9 ppm; IR (thin film): n˜ =
3481, 2980, 2943, 2862, 1461, 1369, 1166, 1101, 1074, 1022, 941 cmÀ1
;
HRMS (ESI): m/z calcd for C9H16O4: 211.0941 [M+Na]+; found:
211.0947.
2-((4aR,6S,7aS)-Tetrahydro-4H-furoACTHNUTRGENN[UG 3,2-d]ACHUTNGTREN[NUNG 1,3]dioxin-6-yl)propan-2-ol
(20a): Rf =0.59 (100% EtOAc); [a]2D2 =À13.1 (c=0.21 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=5.10 (d, J=6.2 Hz, 1H), 4.62 (d, J=
6.3 Hz, 1H), 4.41 (app dd, J=9.7, 3.7 Hz, 1H), 3.93 (app dd, J=9.8,
6.2 Hz, 1H), 3.59 (t, J=9.6 Hz, 1H), 3.46–3.41 (m, 2H), 2.26–2.22 (m,
1H), 2.05–2.00 (m, 1H), 1.91 (s, 1H), 1.25 (s, 3H), 1.19 ppm (s, 3H);
13C NMR (150 MHz, CDCl3): d=94.2, 83.9, 80.6, 73.9, 72.4, 72.1, 30.6,
26.6, 24.2 ppm; IR (thin film): n˜ =3479, 2976, 2924, 2872, 1376, 1157,
1128, 1063, 981 cmÀ1; HRMS (ESI): m/z calcd for C9H16O4: 211.0941
[M+Na]+; found: 211.0953.[25]
Acknowledgements
We thank the NIGMS (GM72566) for support of this work and both
NSERC and FQRNT for postdoctoral fellowships (J.J.M). We are grate-
ful to Dr. A. R. van Dyke for preliminary results and to Dr. J. A. Byers,
K. W. Armbrust, J. Tanuwidjaja, Dr. Yuan Zhang, and Dr. A. Kleinke for
insightful discussions and input during the preparation of this manuscript.
Thanks is also given to Eric Standley for HRMS data.
(R)-1-((4aR,6S,7aS)-6-Methyltetrahydro-4H-furoACTHNUGTRENNUG[3,2-d]ACHTUNGTRNE[NUGN 1,3]dioxin-6-yl)e-
thanol (20b): Rf =0.55 (100% EtOAc); [a]2D2 =À17.9 (c=0.42 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=5.08 (d, J=6.2 Hz, 1H), 4.61 (d, J=
6.2 Hz, 1H), 4.41 (app dd, J=9.6, 4.2 Hz, 1H), 3.63–3.59 (m, 2H), 3.50
(app tdd, J=9.5, 4.3, 1.0 Hz, 1H), 3.36 (app td, J=9.5, 7.9 Hz, 1H), 2.26
(app ddd, J=11.7, 7.6, 0.8 Hz, 1H), 2.16 (br s, 1H), 1.74 (t, J=
10.9 Hz,1H), 1.31 (s, 3H), 1.18–1.16 ppm (d, J=6.5 Hz, 3H); 13C NMR
(150 MHz, CDCl3): d=94.2, 85.1, 80.7, 73.3, 72.4, 71.8, 38.6, 22.0, 17.7; IR
(thin film): n˜ =3473, 2978, 2874, 1452, 1374, 1270, 1129, 1061, 1000,
904 cmÀ1; HRMS (ESI): m/z calcd for C9H16O4: 209.0784 [M+Na]+;
found: 209.0791.
[1] a) K. G. Sellner, G. J. Doucette, G. J. Kirkpatrick, J. Ind. Microbiol.
Biotechnol. 2003, 30, 383; b) G. Ronzitti, F. Callegani, C. Malaguti,
G. P. Rossini, Br. J. Cancer 2004, 90, 1100.
[2] a) M. Schrope, Nature 2008, 452, 24; b) K. Nakanishi, Toxicon 1985,
23, 473.
[3] M. S. Lee, G. Qin, K. Nakanishi, M. G. Zagorski, J. Am. Chem. Soc.
1989, 111, 6234.
[4] H. N. Chou, Y. Shimizu, J. Am. Chem. Soc. 1987, 109, 2184.
[5] a) K. C. Nicolaou, C. V. C. Prasad, P. K. Somers, C. K. Hwang, J.
Am. Chem. Soc. 1989, 111, 5330; b) K. C. Nicolaou, M. O. Frederick,
R. J. Aversa, Angew. Chem. 2008, 120, 7292; Angew. Chem. Int. Ed.
2008, 47, 7182.
[6] J. E. Baldwin, J. Chem. Soc. Chem. Commun. 1976, 734.
[7] J. M. Coxon, M. P. Hartshorn, W. H. Swallow, Aust. J. Chem. 1973,
26, 2521.
[8] For selected reviews on ladder polyether synthesis, see: a) A. R.
Gallimore, Nat. Prod. Rep. 2009, 26, 266; b) C. J. Morten, J. A.
Byers, A. R. Van Dyke, I. Vilotijevic, T. F. Jamison, Chem. Soc. Rev.
2009, 38, 3175; c) I. Vilotijevic, T. F. Jamison, Angew. Chem. 2009,
121, 5352; Angew. Chem. Int. Ed. 2009, 48, 5250; d) I. Vilotijevic,
T. F. Jamison, Mar. Drugs 2010, 8, 763; e) E. A. Anderson, Org.
Biomol. Chem. 2011, 9, 3997.
[9] I. Vilotijevic, T. F. Jamison, Science 2007, 317, 1189.
[10] a) J. A. Byers, T. F. Jamison, J. Am. Chem. Soc. 2009, 131, 6383;
b) C. J. Morten, T. F. Jamison, J. Am. Chem. Soc. 2009, 131, 6678;
c) C. J. Morten, J. A. Byers, T. F. Jamison, J. Am. Chem. Soc. 2011,
133, 1902.
((4aR,6S,7aS)-6-Methyltetrahydro-4H-furoACTHNUTRGENN[UG 3,2-d]ACHUTNGTREN[NNGU 1,3]dioxin-6-yl)metha-
nol (28): Rf =0.56 (100% EtOAc); [a]2D2 =À33.0 (c=0.36 in CH2Cl2);
1H NMR (600 MHz, CDCl3): d=5.09 (d, J=6.3 Hz, 1H), 4.62 (d, J=
6.3 Hz, 1H), 4.40 (dd, J=9.6, 4.2 Hz, 1H), 3.62 (t, J=9.8 Hz, 1H), 3.53–
3.49 (m, 2H), 3.46–3.41 (m, 2H), 2.34 (dd, J=11.5, 7.5 Hz, 1H), 1.89 (dd,
J=7.8, 5.4 Hz, 1H), 1.80 (t, J=11.1 Hz 1H), 1.34 ppm (s, 3H); 13C NMR
(150 MHz, CDCl3): d=94.3, 83.1, 80.9, 72.7, 71.7, 69.6, 38.2, 24.4 ppm; IR
(thin film): n˜ =3388, 2972, 2871, 1453, 1373, 1277, 1151, 1129, 1050 cmÀ1
;
HRMS (DART): m/z calcd for C8H14O4: 197.0784 [M+Na]+; found:
197.0786.
((4aR,6S,7aS)-Tetrahydro-4H-furoACTHUNTGRENNUG[3,2-d]ACHTUNTGREN[NUGN 1,3]dioxin-6-yl)methanol (30):
Rf =0.53 (100% EtOAc); [a]2D2 =À40.8 (c=0.36 in CH2Cl2); 1H NMR
(600 MHz, CDCl3): d=5.10 (d, J=6.3 Hz, 1H), 4.63 (d, J=6.3 Hz, 1H),
4.43 (app dd, J=9.7, 4.0 Hz, 1H), 4.30 (app ddt, J=9.6, 4.9, 3.2 Hz, 1H),
3.77 (app ddd, J=11.9, 5.6, 3.1 Hz, 1H), 3.65–3.62 (m, 1H), 3.57 (app
ddd, J=12.0, 7.0, 5.0 Hz, 1H), 3.42–3.35 (m, 2H), 2.16–2.07 (m, 2H),
1.81 ppm (app dd, J=6.9, 5.8 Hz, 1H); 13C NMR (150 MHz, CDCl3): d=
94.3, 80.2, 77.8, 74.3, 71.7, 65.3, 30.8 ppm; IR (thin film): n˜ =3435, 2871,
1259, 1158, 1126, 1044, 932, 901 cmÀ1; HRMS (ESI): m/z calcd for
C7H12O4: 183.0628 [M+Na]+; found: 183.0638.
[11] K. Hotta, X. Chen, R. S. Paton, A. Minami, H. Li, K. Swaminathan,
I. I. Mathews, K. Watanabe, H. Oikawa, K. N. Houk, C.-Y. Kim,
Nature 2012, 483, 355.
(4aR,5aS,7R,8S,9aR,10aS)-8-Methyloctahydro-4H-
pyrano[2’,3’:5,6]pyranoACHTUNGTRENNUNG[3,2-d]AHCTUNGTREN[NGUN 1,3]dioxin-7-ol (36): As per the general
procedure in either buffered or unbuffered water except that the mixture
was left to stir for five days at 708C to ensure complete cyclization. Cycli-
[12] A. R. Van Dyke, T. F. Jamison, Angew. Chem. 2009, 121, 4494;
Angew. Chem. Int. Ed. 2009, 48, 4430.
Chem. Eur. J. 2013, 19, 10004 – 10016
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10015

T. F. Jamison et al.
[13] S. Hanessian, Total Synthesis of Natural Products: The “Chiron” Ap-
proach, Pergamon Press, New York, 1983.
Soc. 1993, 115, 8453; c) J. M. Coxon, A. J. Thorpe, J. Org. Chem.
1999, 64, 5530; d) S. Wan, H. Gunaydin, K. N. Houk, P. E. Florean-
cig, J. Am. Chem. Soc. 2007, 129, 7915.
[14] For similar approaches, see: a) K. C. Nicolaou, D. A. Nugiel, C.
Couladouros, C. K. Hwang, Tetrahedron 1990, 46, 4517; b) H. Fuwa,
M. Sakaki, K. Tachibana, Tetrahedron 2001, 57, 3019; c) M. Inoue, J.
Wang, G.-X. Wang, Y. Ogasawara, M. Hirama, Tetrahedron 2003, 59,
5645.
[15] J. J. Mousseau, J. A. Bull, A. B. Charette, Angew. Chem. 2010, 122,
1133; Angew. Chem. Int. Ed. 2010, 49, 1115.
[16] CRC Handbook of Chemistry and Physics, 92nd ed. (Ed.: W. M.
Haynes), CRC Press, 2011.
[17] All yields reported are based on isolated material.
[18] a) N. Baggett, J. S. Brimacombe, A. B. Foster, M. Stacey, D. H. Whif-
fen, J. Chem. Soc. 1960, 2574; b) N. Baggett, M. A. Bukhari, A. B.
Foster, J. Lehmann, J. M. Webber, J. Chem. Soc. 1963, 4157; c) E. L.
Eliel, M. C. K. Sister, J. Am. Chem. Soc. 1968, 90, 3444; d) E. L.
Eliel, Acc. Chem. Res. 1970, 3, 1.
[20] a) S. D. Rychnovsky, G. Yang, J. P. Powers, J. Org. Chem. 1993, 58,
5251; b) E. Juaristi, G. A. Rosquete-Pina, A. J. Mota, Pure Appl.
Chem. 2003, 75, 589.
[21] For facile hydrolysis of the methylene acetal, see: W. F. Bailey,
L. M. J. Zarcone, A. D. Rivera, J. Org. Chem. 1995, 60, 2532.
[22] For a similar method employing a dihydroxylation/oxidative cleav-
age route, see the Supporting Information.
[23] Y. Zhang, L. Deng, G. Zhao, Org. Biomol. Chem. 2011, 9, 4518.
[24] Acetylation of the alcohol and COSY analysis were performed
when needed to confirm the structure of the cyclization adduct; see
the Supporting Information for details.
[25] An attempt to oxidize 20a with TPAP (5 mol%) and NMO
(1.5 equiv) led to recovered starting material.
[19] a) G. Stork, J. F. Cohen, J. Am. Chem. Soc. 1974, 96, 5270; b) J. Na,
K. N. Houk, C. G. Shevlin, K. D. Janda, R. A. Lerner, J. Am. Chem.
Received: March 4, 2013
Published online: June 17, 2013
10016
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10004 – 10016