Mechanistic Studies of the Biomimetic Epoxy Ester-Orthoester and Orthoester-Cyclic Ether Rearrangements

Article abstract of DOI:10.1021/jo034566s

The relative rates of acid-catalyzed rearrangements of epoxy esters to [3.2.1]bicyclic orthoesters, the subsequent rearrangements of these ortho esters to substituted tetrahydrofurans, and the rates of orthoester hydrolysis at pH 4.75 were measured in NMR kinetics experiments. The ease of formation and stabilities of these orthoesters compared favorably with the OBO-type [2.2.2]bicyclic orthoesters typically used as protecting groups of carboxylic acids. Studies with ¹³C NMR-detected ¹⁸O-labeling show that epoxy ester rearrangement takes place preferentially via 6-exo cyclization, although the 7-endo process competes when the distal center of the epoxide is disubstituted. The ortho ester-cyclic ether rearrangement was shown by ¹⁸O-labeling to occur exclusively via intermediacy of a five-membered dioxonium ion. The structures of the hydrolysis products also indicate the intermediacy of a dioxolanium ion during hydrolysis. The implications for a hypothetical biosynthesis of marine polyether toxins are discussed.

Full text of DOI:10.1021/jo034566s

Mech a n istic Stu d ies of th e Biom im etic Ep oxy Ester -Or th oester
a n d Or th oester -Cyclic Eth er Rea r r a n gem en ts†
J ose´-Luis Giner,*,‡ Xiaoyong Li,‡ and J oseph J . Mullins§
Department of Chemistry, State University of New York-ESF, Syracuse, New York 13210, and
Department of Chemistry, LeMoyne College, Syracuse, New York 13214
jlginer@syr.edu
Received April 30, 2003
The relative rates of acid-catalyzed rearrangements of epoxy esters to [3.2.1]bicyclic orthoesters,
the subsequent rearrangements of these ortho esters to substituted tetrahydrofurans, and the rates
of orthoester hydrolysis at pH 4.75 were measured in NMR kinetics experiments. The ease of
formation and stabilities of these orthoesters compared favorably with the OBO-type [2.2.2]bicyclic
orthoesters typically used as protecting groups of carboxylic acids. Studies with 13C NMR-detected
18O-labeling show that epoxy ester rearrangement takes place preferentially via 6-exo cyclization,
although the 7-endo process competes when the distal center of the epoxide is disubstituted. The
ortho ester-cyclic ether rearrangement was shown by 18O-labeling to occur exclusively via
intermediacy of a five-membered dioxonium ion. The structures of the hydrolysis products also
indicate the intermediacy of a dioxolanium ion during hydrolysis. The implications for a hypothetical
biosynthesis of marine polyether toxins are discussed.
In tr od u ction
ment in biomimetic syntheses of the natural orthoesters
ortho esterol B and petuniasteroid D.10,11 Unlike workers
who have used exotic Lewis acids to effect this transfor-
mation,6 we have found that this reaction occurs smoothly
in the presence of mild protic acids such as dilute
trifluoroacetic acid (TFA) in benzene or chloroform, and
we have taken advantage of the stereospecificity of this
reaction in a recent synthesis of 2-methylerythritol.12 The
orthoesters thus obtained are capable of subsequent
orthoester-cyclic ether rearrangement to give 3-acyl-
oxytetrahydrofurans under acidic conditions.4,6,8a,13 We
report herein experiments conducted to better under-
stand the formation and reactivity of bicyclic orthoesters
using simple model compounds.
Despite the potential synthetic utility of the epoxy
ester-orthoester rearrangement, few studies have ex-
plored its mechanism. Examples of the intramolecular
formation of orthoesters from epoxy esters, though few
in number, have long been known in the natural products
literature. Orthoesters have been isolated as unexpected
products during the chemical manipulation of labdane,1
neoclerodane,2 and taxane diterpenoids3 as well as during
the chemical syntheses of brevicomin,4 and muscarine.5
This reaction has also been observed in explorations of
zirconocene-catalyzed reactions.6 More deliberate in-
stances of the epoxy ester-orthoester rearrangement are
found in biomimetic studies of natural orthoesters oc-
curring in fungi,7 fragrances,8 and plants.9 We have
recently employed the epoxy ester-orthoester rearrange-
† Dedicated to Professor Carl Djerassi in celebration of his 80th
birthday.
‡ State University of New York-ESF.
§ LeMoyne College.
(1) (a) Demole, E.; Wuest, H. Helv. Chim. Acta 1967, 50, 1314-1327.
(b) Urones, J . G.; Marcos, I. S.; Basabe, P.; Sexmero, M. J .; Diez, D.;
Garrido, N. M.; Prieto, J . E. S. Tetrahedron 1990, 46, 2495-2502.
(2) Rodriguez, B.; de la Torre, M. C.; Perales, A.; Malakov, P. Y.;
Papanov, G. Y.; Simmonds, M. S. J .; Blaney, W. M. Tetrahedron 1994,
50, 5451-5468.
Resu lts
Measurement of the relative rates of the epoxy ester-
orthoester rearrangement and the orthoester-cyclic
(3) Hosoyama, H.; Shigemori, H.; Kobayashi, J . i. Perkin 1 2000,
449-451.
(8) (a) Kaiser, R. Devel. Food Sci. 1988, 18, 669-684. (b) Koepsel,
M.; Krempel, A.; Surburg, H., In Prog. Essent. Oil Res., Proc. Int. Symp.
Essent. Oils, 16th; Brunke, E.-J ., Ed.; de Gruyter: Berlin, 1986; pp
241-248.
(9) Elliger, C. A.; Benson, M.; Lundin, R. E.; Waiss, A. C., J r.
Phytochemistry 1988, 27, 3597-3603.
(10) Giner, J .-L.; Faraldos, J . A. J . Org. Chem. 2002, 67, 2717-2720.
(11) Faraldos, J . A.; Giner, J .-L. J . Org. Chem. 2002, 67, 4659-4666.
(12) Giner, J .-L.; William V. Ferris, J .; Mullins, J . J . J . Org. Chem.
2002, 67, 4856-4859.
(13) (a) Coxon, J . M.; Hartshorn, M. P.; Swallow, W. H. Chem.
Commun. 1973, 261-262. (b) Coxon, J . M.; Hartshorn, M. P.; Swallow,
W. H. J . Org. Chem. 1974, 39, 1142-1148.
(4) Wasserman, H. H.; Barber, E. H. J . Am. Chem. Soc. 1969, 91,
3674-3675.
(5) Chmielewski, M.; Guzik, P.; Hintze, B.; Daniewski, W. M. J . Org.
Chem. 1985, 50, 5360-5362.
(6) (a) Wipf, P.; Xu, W. J . Org. Chem. 1993, 58, 5880-5882. (b) Wipf,
P.; Xu, W.; Kim, H.; Takahashi, H. Tetrahedron 1997, 53, 16575-
16596. (c) Wipf, P.; Tsuchimoto, T.; Takahashi, H. Pure Appl. Chem.
1999, 71, 415-421. (d) Wipf, P.; Aslan, D. C. J . Org. Chem. 2001, 66,
337-343.
(7) (a) Gill, M.; Steglich, W. Prog. Chem. Org. Nat. Prods. 1987, 51,
1-299. (b) Besl, H.; Bresinsky, A.; Geigenmueller, G.; Herrmann, R.;
Kilpert, C.; Steglich, W. Liebigs Ann. Chem. 1989, 803-810.
10.1021/jo034566s CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/03/2003
J . Org. Chem. 2003, 68, 10079-10086
10079

Giner et al.
TABLE 1. Rela tive Ra tes of Rea r r a n gem en t of Ep oxy
Ester s a n d Or th oester s (TF A/CDCl3,30 °C)a
F IGURE 1. Rates of epoxy ester rearrangements (TFA/CDCl3,
30 °C).
a
R ) PhCH2CH2.
ether rearrangement was carried out by NMR kinetics
in CDCl3 containing varying concentrations of TFA
(Table 1). Rates were measured for the rearrangements
of three simple epoxy esters (1-3), as well as for the
subsequent rearrangements of their resulting [3.2.1]-
bicyclic orthoesters (4-6) to 3-acyloxytetrahydrofurans
(7-9). For comparative purposes, the rearrangement of
the 3-oxetanemethyl ester 10 to the OBO [2.2.2]bicyclic
orthoester 11 was also measured. It proved experimen-
tally difficult to obtain consistent rates for these reac-
tions, but for comparative purposes, relative rates could
be obtained by simultaneous measurements of the rear-
rangements of these compounds in pairs. Thus, the rate
of rearrangement of the most reactive compound was
measured concurrently with that of the next most reac-
tive compound, and the rearrangement of that compound
was in turn measured together with the third most
reactive compound at a higher concentration of acid
(Figure 1). This method was complicated by the fact that
the least reactive epoxy ester 3 and most reactive
orthoester 6 stand in a precursor-product relationship.
The relative rates in this case were obtained by beginning
the measurement of the second reaction only after less
than 1% of epoxy ester 3 remained. In this way, it was
possible to obtain relative rates for a series of compounds
varying greatly in their reactivity (Table 1).
F IGURE 2. Rates of orthoester hydrolysis.
drolysis reactions (12-20) were identified, and the
product compositions were determined before and after
equilibration using 50 mM TFA/CDCl3. (Table 2).
The fates of the oxygen atoms during the epoxy ester-
orthoester and orthoester-cyclic ether rearrangements
was investigated with 18O-labeled epoxy esters and
orthoesters using 13C NMR spectrometry.14
Epoxy esters labeled in the carbonyl oxygen (1a -3a )
were prepared by carbodiimide condensation of 18O-
labeled hydrocinnamic acid with the appropriate homoal-
lylic alcohols, followed by peracid epoxidation (Scheme
1). All three epoxy esters showed labeling to the extent
of approximately 42% as measured by integration of the
13C NMR signals of the carbonyl group, with an upfield
∆δ of 38 ppb for the 18O-labeled 13C signals. In the
rearrangement of epoxy esters 1a and 3a to orthoesters
4a and 6a , the 18O-label was found exclusively in the
position connecting the orthoester carbon with the bridge-
head center (Scheme 2) as shown by upfield shifts in the
13C NMR signals for the bridgehead carbons (∆δ ) 32
and 27 ppb, respectively). Upon rearrangement of epoxy
ester 2a , orthoester 5 was found to have 18O-label in two
different positions, with 28% 18O-labeling of the bridge-
head carbon (5a ) and 14% labeling of the dimethyl
substituted carbon (5b) (∆δ ) 27 and 32 ppb, respec-
tively). The 18O-labeled signals of the orthoester carbons
showed a 21 ppb upfield shift in all three orthoesters.
To compare the relative stabilities of orthoesters 4-6,
measurements were made of their rates of hydrolysis
(Figure 2). Conditions were chosen that allowed the rates
to be conveniently measured using 1H NMR-50 mM
NaOAc buffer (pH 4.75)/acetone 1:2 using deuterated
solvents. The rates were measured pairwise as before,
although in this case the use of a buffer provided highly
consistent results. The products of the orthoester hy-
(14) (a) Risley, J . M.; Van Etten, R. L. J . Am. Chem. Soc. 1981, 103,
4389-4392. (b) Vederas, J . C. Can. J . Chem. 1982, 60, 1637-1642.
10080 J . Org. Chem., Vol. 68, No. 26, 2003

Studies of Biomimetic Epoxy Ester Rearrangements
TABLE 2. P r od u cts of Or th oester Hyd r olysisa
a
R ) PhCH2CH2.
SCHEME 1. 18O-La belin g of Ep oxy Ester 1
This was achieved by first labeling the homoallylic
alcohol with 18O through a Mitsunobu reaction with [18O]-
hydrocinnamic acid, followed by LAH reduction (Scheme
1). Re-esterification with unlabeled hydrocinnamic acid,
followed by epoxidation gave the epoxy ester 1b bearing
18O in the ester oxygen to the extent of 28%. The signal
of the -CH2O- carbon was shifted upfield by 30 ppb,
and that of the carbonyl group by 15 ppb. After rear-
rangement to orthoester 4b, an upfield shift of 20 ppb
was measured for the orthoester carbon, and 26 ppb for
the oxymethylene group of the longer bridge in the
bicyclic orthoester. Orthoester-cyclic ether rearrange-
ment led to 28% 18O-labeling in the ether oxygen of the
3-acyloxytetrahydrofuran (7b), as shown by upfield shifts
of 18 ppb for C-2, 23 ppb for C-5, and a two-bond shift of
6 ppb for C-3. No shift was seen of the carbonyl signal.
The 18O-labeled orthoesters thus obtained were rear-
ranged to 3-acyloxytetrahydrofurans (Scheme 2). Rear-
rangement of 4a and 6a led to products containing 18O
exclusively in the ester oxygen (7a and 9a ), as evident
from an upfield shift of the 13C NMR signal of the
carbonyl groups by 12 ppb, and of the THF 3-position by
39 and 33 ppb, respectively. Upon rearrangement of the
doubly 18O-labeled dimethyl orthoester 5 (5a and 5b), the
ester oxygen of 8 was found to be 28% labeled (8a ), and
the carbonyl oxygen 14% (8b). The 13C NMR signal of
the unlabeled carbonyl carbon was accompanied by two
new 13C NMR signals shifted upfield by 15 and 38 ppb
in a 58:28:14 ratio, respectively (Figure 3). The ester
carbon of the ring showed 28% of its signal shifted upfield
by 33 ppb and a smaller, poorly resolved shoulder (∆δ )
5 ppb) due to a two-bond 18O-shift. No 18O-labeling of the
ether oxygen was observed in 7a -9a .
Discu ssion
Ra tes of Rea r r a n gem en t. In our prior studies,10-12
the epoxy esters bore methyl substituents at the proximal
carbon of the epoxide and rearranged with inversion at
this center via 6-exo or 5-exo cyclization. Electron-
donating alkyl substitutents at the epoxide reaction
center are thought to promote acid-catalyzed nucleophilic
substitution by stabilizing a partial positive charge in a
“borderline SN2 mechanism”15 and have been used to
direct the regioselectivity of epoxide substitution.16-18 For
this reason, it was not surprising that unsubstituted
(15) Parker, R. E.; Isaacs, N. S. Chem. Rev. 1959, 59, 737-799.
(16) Behrens, C. H.; Ko, S. Y.; Sharpless, K. B.; Walker, J . Org.
Chem 1985, 50, 5687-96.
(17) (a) Nicolaou, K. C.; Prasad, C. V. C.; Somers, P. K.; Hwang, C.
K. J . Am. Chem. Soc. 1989, 111, 5330-5334. (b) Nicolaou, K. C.;
Prasad, C. V. C.; Somers, P. K.; Hwang, C. K. J . Am. Chem. Soc. 1989,
111, 5335-5340.
To account for the fates of all three oxygen atoms in
the rearrangement of 4 to 7, it was necessary to place
the 18O-label in another of the three possible positions.
J . Org. Chem, Vol. 68, No. 26, 2003 10081

Giner et al.
SCHEME 2. Resu lts of 18O-La belin g Exp er im en ts
F IGURE 3. 13C NMR signal (151 MHz) of the carbonyl carbon
of 18O-labeled 8.
substituted substrates showed the greatest difference
between the rearrangements of the monomethyl-substi-
tuted epoxy ester 1 and orthoester 4 (k1f4 vs k 4f7 ) 950:
1). In comparison, the relative rate was 83:1 for the
dimethyl substituted case (2 f 5 f 8) and 8.2:1 without
methyl substituents (3 f 6 f 9). Probably due to the
somewhat similar reactivities of 2 and 5, a previous study
using Cp2ZrCl2/AgClO4 as the Lewis acid catalyst failed
to detect orthoester 6 upon rearrangement of epoxide 3
and only found the acyl tetrahydrofuran 9.6d It is likely
that with certain substrates the orthoester-cyclic ether
rearrangement can become faster than the epoxy ester-
orthoester rearrangement.
Comparisons can also be made between the rates of
epoxy ester-orthoester rearrangements and the rate of
formation of the OBO [2.2.2]bicyclic orthoester 11 from
the 3-oxetanemethyl ester 10. This rearrangement is
commonly employed by synthetic chemists as a protecting
group strategy to convert carboxylate groups into
orthoesters which are resistant to strongly nucleophilic
reagents.19,20 Although OBO orthoester formation is
typically conducted using BF3-etherate catalysis, in this
study the milder TFA catalysis was used to permit
comparison with the formation of orthoester 4, which has
been proposed as an alternative to the OBO protecting
group.6 Our results show that under protic acid catalysis
(TFA/CDCl3) generation of the [3.2.1]bicyclic orthoester
4 is 2.2 × 104 times faster than the formation of the
[2.2.2]bicyclic orthoester 11 (Table 1). The facile forma-
tion of 4, combined with its superior stability to hydroly-
sis and to orthoester-cyclic ether rearrangement, indi-
cates that bridgehead substituted [3.2.1]bicyclic orthoesters
such as 4 might be preferable to OBO orthoesters (e.g.,
11) as protecting groups,6b,c,19,20 at least in cases
epoxy ester 3 was least reactive (krel ) 1) and the epoxy
ester bearing a methyl group on the proximal carbon of
the epoxide (1) was most reactive (krel ) 16). More
puzzling was the greater reactivity of the dimethyl epoxy
ester (2, krel ) 4.3) compared to the unsubstituted
compound (3, Table 1) since the substitution is at the
distal center of the epoxide. While the rate enhancement
may be due in part to conformational effects resulting
from methyl substitution, the electronic influence of the
substituents activates this molecule to an alternative
7-endo mode of cyclization (see the discussion of 18O
results below).
In contrast, the influence of alkyl substitution on the
orthoester-cyclic ether rearrangements appeared to be
predominantly one of steric hindrance. Here, the mono-
methyl orthoester (4) was least reactive (krel ) 1), followed
by the dimethyl orthoester (5, krel ) 3.0) and the unsub-
stituted orthoester (6, krel ) 7.1).
Comparison of the rates of epoxy ester-orthoester vs
orthoester-cyclic ether rearrangement for the differently
(19) Corey, E. J .; Raju, N. Tetrahedron Lett. 1983, 24, 5571-5574.
(20) (a) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; Wiley: New York, 1999. (b) Kocienski, P. J .
Protecting Groups; Thieme: New York, 2000.
(18) McDonald, F. E.; Bravo, F.; Wang, X.; Wei, X.; Toganoh, M.;
Rodriguez, J . R.; Do, B.; Neiwert, W. A.; Hardcastle, K. I. J . Org. Chem.
2002, 67, 2515-2523.
10082 J . Org. Chem., Vol. 68, No. 26, 2003

Studies of Biomimetic Epoxy Ester Rearrangements
SCHEME 3. P r op osed Hyd r olysis Mech a n ism for
[3.2.1]Bicyclic Or th oester s
where the intrinsic chirality of [3.2.1]bicyclic orthoesters
is not a problem.21
Or th oester Hyd r olysis. The hydrolysis of orthoesters
is one of the most thoroughly studied reactions in organic
chemistry.22 Mechanistic investigations of this reaction
have provided insight into the tetrahedral intermediates
in biochemically important reactions such as ester and
amide hydrolysis23,24 and have provided support for
stereoelectronic theory.25 Very few studies, however, have
been reported with [3.2.1]bicyclic orthoesters.26 Under the
conditions used in this study, the rate-limiting step of
hydrolysis is generally the formation of a dioxonium
ion.27,26b As in the rearrangement reactions discussed
above, no evidence was found in the NMR spectra for the
accumulation of any intermediates. The rates of hydroly-
sis ranged from t1/2 ) 6-30 min (Figure 2). The [3.2.1]-
bicyclic orthoesters 5 and 6 are roughly twice as suscep-
tible to hydrolysis as the OBO orthoester 11, but
orthoester 4 proved to be twice as resistant. The stabiliz-
ing influence of substitution at the bridgehead position
apparent in orthoester 4 vis-a`-vis 5 and 6 warrants
further investigation. A possible reason for this effect may
be that the bridgehead methyl group hinders the rotation
of the pendant 2-hydroxyethyl group away from the
newly formed dioxonium ion (21), thereby favoring
the back reaction or hindering the approach of water
(Scheme 3).
SCHEME 4. Alter n a te P a th w a ys of Ep oxy
Ester -Or th oester Rea r r a n gem en t
While the C3 symmetric [2.2.2]bicyclic OBO orthoester
11 yields only one product upon hydrolysis, there are
three possible products for each of the [3.2.1]bicyclic
orthoesters (Table 2, 12-20). The observed products are
best explained by the intermediacy of a five-membered
dioxonium ion (21) (Scheme 3). This is most evident in
the 3:2 ratio of 18 and 19 in the hydrolysis of orthoester
6. Even though tertiary esters (13 and 15) were not
detected in the hydrolysis reactions of orthoesters 4 and
5, the observed products (12 and 16) also fit this pattern.
In a previous study using the benzoate analog (R ) Ph)
of orthoester 6, the sole hydrolysis product was reported
to be the secondary ester 19 (R ) Ph).26c If this result is
correct, the different product preference may be due to
an electronic effect of the phenyl substitutent.28
Equilibration of the initial products of orthoester
hydrolysis by acid-catalyzed intramolecular transesteri-
fication led to more complex mixtures derived from 4 and
6 (Table 2). For synthetic applications of orthoesters,12,29
it may therefore be useful to carry out orthoester hy-
drolysis under buffered conditions where simpler product
mixtures are obtained. It is noteworthy that the second-
ary ester (16) obtained from the hydrolysis of 5 could be
equilibrated exclusively to the primary ester (17).
18O Stu d ies. The 18O-labeling experiments provide
insight into the course of the rearrangement reactions
(Scheme 2). The rearrangements of 18O-labeled epoxy
esters 1a and 3a to orthoesters 4a and 6a were found to
occur entirely via 6-exo cyclization to an intermediate six-
membered dioxonium ion. However, the rearrangement
product of 18O-labeled epoxy ester 2a showed 18O-label
in two positions (5a and 5b), indicating a 2:1 ratio of
6-exo and 7-endo cyclization pathways (Scheme 4). The
pathway via the six-membered dioxonium ion 22 remains
preferred; however, 7-endo cyclization via a seven-
membered dioxonium ion 23 is also observed, apparently
because alkyl substitution activates the distal center of
(21) (a) Wroblewski, A. E.; Applequist, J .; Takaya, A.; Honzatko,
R.; Kim, S. S.; J acobson, R. A.; Reitsma, B. H.; Yeung, E. S.; Verkade,
J . G. J . Am. Chem. Soc. 1988, 110, 4144-4150. (b) Davar, I.; Gedanken,
A.; Menge, W.; Verkade, J . G. J . Chem. Soc., Chem. Commun. 1993,
262-263.
(22) For reviews and leading references, see: (a) DeWolfe, R. H.
Carboxylic ortho acid derivatives; Academic Press: New York, 1970.
(b) Fife, T. H. Acc. Chem. Res. 1972, 5, 264-272. (c) Cordes, E. H.;
Bull, H. G. Chem. Rev. 1974, 74, 581-603. (d) Chiang, Y.; Kresge, A.
J .; Lahti, M. O.; Weeks, D. P. J . Am. Chem. Soc. 1983, 105, 6852-
6855. (e) McClelland, R. A.; Moreau, C. Can. J . Chem. 1985, 63, 2673-
2678. (f) Pindur, U.; Muller, J .; Flo, C.; Witzel, H. Chem. Soc. Rev.
1987, 16, 75-87. (g) Deslongchamps, P.; Dory, Y. L.; Li, S. Tetrahedron
2000, 56, 3553-3537.
(23) (a) Bruice, T. C.; Benkovic, S. J . Bioorganic Mechanisms;
Benjamin: New York, 1966; Vol. 1. (b) J encks, W. P. Catalysis in
Chemistry and Enzymology; McGraw-Hill: New York, 1969.
(24) (a) Capon, B.; Ghosh, A. K.; Grieve, D. M. A. Acc. Chem. Res.
1981, 14, 306-312. (b) McClelland, R. A.; Santry, L. J . Acc. Chem.
Res. 1983, 16, 6, 394-399.
(25) (a) Kirby, A. J . The anomeric effect and related stereoelectronic
effects at oxygen; Springer-Verlag: Berlin, 1983. (b) Deslongchamps,
P. Stereoelectronic effects in organic chemistry; Pergamon: Oxford,
1983.
(26) (a) Hall, H. K., J r.; DeBlauwe, F.; Pyriadi, T. J . Am. Chem.
Soc. 1975, 97, 3854. (b) Burt, R. A.; Chiang, Y.; Hall, H. K. J .; Kresge,
A. J . J . Am. Chem. Soc. 1982, 104, 3687-3690. (c) Bouab, O.; Lamaty,
G.; Moreau, C. Can. J . Chem. 1985, 63, 816-822.
(27) Ahmad, M.; Bergstrom, R. G.; Cashen, M. J .; Kresge, A. J .;
McClelland, R. A.; Powell, M. F. J . Am. Chem. Soc. 1977, 99, 4827-
4829.
(28) Chiang, Y.; Kresge, A. J .; Salomaa, P.; Young, C. I. J . Am.
Chem. Soc. 1974, 96, 4494-4499.
(29) (a) Corminboeuf, O.; Overman, L. E.; Pennington, L. D. Org.
Lett. 2003, 5, 1543-1546. (b) Corminboeuf, O.; Overman, L. E.;
Pennington, L. D. J . Am. Chem. Soc. 2003, 125, 6650-6652. (c)
Overman, L. E.; Pennington, L. D. J . Org. Chem. 2003, 68, 7143-
7157.
J . Org. Chem, Vol. 68, No. 26, 2003 10083

Giner et al.
SCHEME 5. P ossible Rou tes of Or th oester -Cyclic
Eth er Rea r r a n gem en t Sh ow in g th e F a tes of th e
Th r ee Or th oester Oxygen Atom s, As Dep icted for a
Gen er a l Un su bstitu ted [3.2.1]Bicyclic Or th oester a
SCHEME 6. Ster eoch em ica l In ter r ela tion sh ip s for
th e Hyp oth etica l Biosyn th esis of Ma r in e
P olyeth er s via Ep oxy Ester -Or th oester -Cyclic
Eth er Rea r r a n gem en ts
of the possible equilibration of orthoester 5 with epoxide
2, even a mechanism involving a common intermediate
in the forward and reverse reactions can be ruled out
since the rearrangement of epoxide 2 to orthoester 5
takes place by a mixture of 6-exo and 7-endo pathways,
and the ratio of 18O in doubly labeled orthoester 5 was
found unchanged in THF 8. Reversibilty in the formation
of the tetrahydrofurans (e.g., 30) is unlikely under our
reaction conditions, although it may be possible in the
presence of strong Lewis acids, based on acyl group
participation during TiCl4-catalyzed THF ring hydroly-
sis.30 Oxetanes (29) are expected to exhibit reactivities
intermediate between epoxides and tetrahydrofurans.
Because other oxetanyl esters have been shown to
rearrange to [2.2.2]- and [2.2.1]bicyclic orthoesters,19,31
the possible equilibrium between orthoester 24 and
oxetane 29 will most probably lie on the side of the [3.2.1]-
bicyclic orthoester (24).
Biosyn th etic Im p lica tion s. To what extent does the
epoxy ester-orthoester-cyclic ether rearrangement op-
erate in nature? We are aware of only one instance in
which a natural cyclic ether has been proposed to arise
by this route,8a although the ubiquity of epoxides, esters,
and acid catalysis suggests that this pathway should be
of biosynthetic relevance. We have recently proposed that
the oxetane ring of Taxol may originate in this way31 and
would like to speculate here that an iterative version of
this mechanism may play a role in the formation of
marine polyethers such as brevetoxin. This proposal
differs in its stereochemical requirements from the
Cane-Westley proposal32 insofar as the tandem epoxy
ester-orthoester-cyclic ether rearrangements proceed
with stereochemical inversion at both centers of the
epoxide (Scheme 6). Accordingly, the biosynthetic poly-
epoxy precursor of an all-trans fused polycylic polyether
would be required to have an all-Z-conguration, as is also
required for the polyolefin precursor if the cyclization
occurs according to the Townsend proposal.33
a
Bold arrows indicate the major reaction pathways observed
in these studies.
the intermediate protonated epoxide by stabilizing a
partial positive charge.15-18
The possible mechanisms of orthoester-cyclic ether
rearrangement are somewhat more complicated. To
demonstrate the mechanistic course of orthoester-cyclic
ether rearrangement, it was insufficient to follow the
oxygen atom in the short bridge of a [3.2.1]bicyclic
orthoester (24, Scheme 5), since its location in the product
(30a or 30b) would be the same whether the reaction
went through a five-membered (25) or a six-membered
dioxonium ion (26). The fates of all three oxygen atoms,
however, could be unambiguously determined by follow-
ing two of the three oxygen atoms. Thus, rearrangement
of the doubly labeled dimethyl orthoester (5a and 5b) can
be shown to proceed via the intermediacy of a five-
membered dioxonium ion. The reaction follows the same
course for the monomethyl orthoester, as shown by the
rearrangements of 4a and 4b. This evidence contradicts
Wipf’s depiction of a six-membered dioxonium ion
intermediate6a-c and agrees with Coxon’s interpreta-
tions.13
In principle, a [3.2.1]bicyclic ortho ester in the presence
of acid could be in equilibrium with three different
dioxonium ions (25-27, Scheme 5), and each of these can
possibly undergo reversible intramolecular displacement
reactions leading to the formation of three different cyclic
etherssan epoxide, an oxetane, or a tetrahydrofuran
(28-30). On the basis of 18O-labeling, no evidence was
found for equilibration. Equilibration of orthoester 24
with epoxide 28 could only occur without scrambling of
the label if the forward and reverse reactions pass
through the same intermediate dioxonium ion (e.g., 24
f 26 f 28b f 26 f 24). This rules out the transient
formation of epoxide 1 via a seven-membered dioxonium
ion during the rearrangement of orthoester 4. In the case
(30) Harwood: L. M.; J ackson, B.; Prout, K.; Witt, F. J . Tetrahedron
Lett. 1990, 31, 1885-1888.
(31) Giner, J .-L.; Faraldos, J . A. Helv. Chim. Acta 2003, 86, 3613-
3622.
(32) Cane, D. E.; Celmer, W. D.; Westley, J . W. J . Am. Chem. Soc.
1983, 105, 3594-3600.
(33) (a) Townsend, C. A.; Basak, A. Tetrahedron 1991, 47, 2591-
2602. (b) McDonald, F. E.; Towne, T. B. J . Am. Chem. Soc. 1994, 116,
7921-79212.
10084 J . Org. Chem., Vol. 68, No. 26, 2003

Studies of Biomimetic Epoxy Ester Rearrangements
product. 1H NMR: 7.31-7.26 (2H, m), 7.22-7.18 (3H, m),
4.27-4.18 (2H, m), 2.96 (2H, t, J ) 7.8 Hz), 2.75 (1H, t, J )
6.2 Hz), 2.65 (2H, t, J ) 7.8 Hz), 1.91-1.83 (1H, m), 1.83-
1.77 (1H, m), 1.30 (3H, s), 1.26 (3H, s). 13C NMR: 172.7, 140.3,
128.4, 128.2, 126.2, 61.7, 61.1, 58.0, 35.7, 30.8, 28.4, 24.6, 18.7.
Anal. Calcd for C15H20O3: C, 72.56; H, 8.12. Found: C, 72.31;
H, 7.98.
Ep oxy Ester s Bea r in g 18O in th e Ca r bon yl Gr ou p (1a -
3a ). To a solution of hydrocinnamoyl chloride (168 mg, 1 mmol)
in 2 mL of dry dichloromethane was added 18 µL of H2O (1
mmol, 97 atom % 18O), followed by 418 µL of dry triethylamine
(3 mmol) and 4-(dimethylamino)pyridine (12 mg, 0.1 mmol).
After standing overnight, 1-(3-dimethylaminopropyl)-3-ethyl-
carbodiimide hydrochloride (192 mg, 1 mmol) and 128 µL of
4-methyl-3-penten-1-ol (1.1 mmol) were added, and the mix-
ture allowed to react for 7 h. The mixture was partitioned
between ether and water, and the ether layer was dried with
Na2SO4 and concentrated under a stream of nitrogen. The
crude product was purified by silica gel chromatography
(hexane/EtOAc 19:1) to yield 4-methyl-3-pentenyl hydrocin-
namate (78.3 mg, 34%) bearing 18O in the carbonyl group to
the extent of 42%. Epoxidation as described above gave 18O-
labeled epoxy ester 2a . Epoxy esters 1a and 3a were prepared
in the same way.
3,4-Ep oxy-3-m eth ylbu tyl Hyd r ocin n a m a te Bea r in g 18O
in th e Ester Oxygen (1b). To a solution of 18O-labeled
hydrocinnamic acid prepared as above (42 mg, 0.28 mmol), 200
µL of 3-methyl-3-buten-1-ol (2 mmol), and triphenylphosphine
(132 mg, 0.5 mmol) in 2 mL of dry THF, was slowly added a
solution of diisopropylazodicarboxylate (100 µL, 0.51 mmol)
in 1 mL of dry THF. After standing overnight, the reaction
mixture was purified by silica gel chromatography (hexane/
EtOAc 19:1) to yield 3-methyl-3-butenyl hydrocinnamate (47.2
mg, 77%) bearing 18O in both positions of the carboxyl group.
The ester was treated with 25 mg of LAH (0.66 mmol) in 5
mL of dry ether. After 3 min, the reaction was quenched with
a few drops of brine and filtered through a short column of
silica gel. The ether solution was diluted with 5 mL of
dichloromethane, and unlabeled hydrocinnamic acid (155 mg,
1.03 mmol), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (200 mg, 1.04 mmol), and 4-(dimethylamino)-
pyridine (24 mg, 0.20 mmol) were added. After standing
overnight, the mixture was partitioned between ether and
dilute HCl, dilute NaOH, and brine. The ether layer was dried
with Na2SO4 and concentrated under reduced pressure. Pu-
rification by silica gel chromatography (hexane/EtOAc 19:1)
gave 3-methyl-3-butenyl hydrocinnamate (42.0 mg, 89%) bear-
ing 18O in the ester position to the extent of 28%. Epoxidation
as described above gave 18O-labeled epoxy ester 1b.
Su m m a r y
The epoxy ester-[3.2.1]bicyclic orthoester rearrange-
ment was shown by 18O-labeling to proceed by predomi-
nantly via 6-exo cyclization. Competing cyclization by a
7-endo route was noted in a case where methyl substitu-
tion activates the distal epoxide reaction center in
accordance with a “borderline SN2 mechanism”.15 Also in
accordance with this effect, kinetics measurements showed
that the presence of methyl substitution at the proximal
center of epoxide 1 leads to a 16-fold acceleration in epoxy
ester rearrangement compared to the unsubstituted
case. On the other hand, the product of this reaction
(orthoester 4) is stablized by methyl substitution of the
bridgehead position, resulting in a greater stability
toward acidic hydrolysis. Since the rate of formation of
ortho ester 4 was 22 000 times faster than that of the
OBO [2.2.2]bicyclic orthoester 11, and the stability of 4
toward aqueous acid is double that of 11, the “ABO”
bicyclic orthoester (e.g., 4)6 appears to be an attractive
alternative to Corey’s OBO protecting group (e.g., 11)13s
at least in cases where its intrinsic chirality is not a
problem. The use of exotic zirconocene Lewis acids,6
however, seems unnecessary, since low concentrations of
Bro¨nsted acids (e.g., TFA in benzene or chloroform)
provide effective catalysis.
The orthoester-cyclic ether rearrangement was shown
by 18O-labeling to proceed entirely via a five-membered
cyclic dioxonium ion. The net stereochemical consequence
of tandem epoxy ester-orthoester-cyclic ether rear-
rangement is therefore generally inversion of configura-
tion at both centers of the epoxide ring. A five-membered
cyclic dioxonium ion was also indicated in the hydrolysis
of the [3.2.1]bicyclic orthoesters.
Exp er im en ta l Section
1H NMR spectra were acquired at 600 MHz and 13C NMR
spectra at 151 MHz using CDCl3 as the solvent, unless
otherwise specified. NMR assignments are based on DEPT,
HMBC, and HSQC experiments. The known epoxy esters 1
and 3, orthoester 4, and tetrahydrofuran 96d were prepared
as described for epoxy ester 2, orthoester 5, and tetrahydro-
furan 8. The NMR data were in agreement with the published
values.
4-Meth yl-1,3,4-p en ta n etr iol 3-P h en ylor th op r op ion a te
(5). To a dry NMR tube was added 21.9 mg (0.09 mmol) of 2,
diluted to 665 µL with CDCl3 (filtered through neutral alumina
and dried over 3 Å molecular sieves), and treated with 35 µL
of 1.0 M TFA in CDCl3. The tube was shaken vigorously and
monitored by 1H NMR. Upon completion, the reaction was
quenched by the addition of ca. 50 µL of TEA. The reaction
mixture was purified by preparative TLC (1% TEA in hexane/
3,4-Ep oxy-4-m eth ylp en tyl Hyd r ocin n a m a te (2). To a
stirred solution of 4-methyl-3-penten-1-ol (1.0 g, 10 mmol) in
15 mL of CH2Cl2 containing 2.0 mL of pyridine was added
dihydrocinnamoyl chloride (2.0 mL, 13 mmol). After 1 h at rt,
the reaction mixture was diluted with 30 mL of hexane/EtOAc
2:1 and extracted sequentially with 15 mL portions of satu-
rated aqueous NaHCO3, 5% hydrochloric acid, saturated
aqueous NaHCO3, and brine. After drying over Na2SO4, the
solvents were evaporated under reduced pressure to give 2.07
g (8.9 mmol, 89% yield) of 4-methyl-3-pentenyl hydrocin-
namate. 1H NMR (300 MHz): 7.32-7.16 (5H, m), 5.08 (1H, br
t, J ) 7.8 Hz), 4.04 (2H, t, J ) 7.1 Hz), 2.96 (2H, t, J ) 7.8
Hz), 2.63 (2H, t, J ) 7.8 Hz), 2.31 (2H, br q, J ) 7.1 Hz), 1.71
(3H, s), 1.62 (3H, s). 13C NMR (75 MHz): 172.9, 140.6, 134.6,
128.4, 128.3, 126.2, 119.2, 64.1, 35.9, 31.0, 27.6, 25.7, 17.7.
A stirred solution of 4-methyl-3-pentenyl hydrocinnamate
(0.30 g, 1.29 mmol) in 15 mL of dichloromethane was treated
with 470 mg (1.9 mmol) of m-CPBA (75%) at rt. After 45 min,
the reaction was quenched with 5 mL of 5% NaOH and
extracted with 20 mL hexane/EtOAc 2:1. The organic layer
was extracted with 15 mL of 5% NaOH and, 20 mL of brine
and dried over Na2SO4. The solvents were removed under
reduced pressure to give 311 mg (1.25 mmol, 97% yield) of
1
EtOAc 4:1) to yield 20.3 mg of 5 (0.08 mmol, 93%). H NMR:
7.29-7.25 (2H, m, Ph), 7.21 (2H, d, J ) 7.4 Hz, Ph), 7.17 (1H,
t, J ) 7.3 Hz, Ph), 4.21 (1H, dt, J ) 4.7, 11.9 Hz, C-1), 4.05
(1H, d, J ) 3.7 Hz, C-3), 3.86 (1H, dd, J ) 7.5, 11.3 Hz, C-1),
2.81 (2H, dd, J ) 6.9, 10.8 Hz, PhCH2), 2.32-2.24 (1H, m, C-2),
2.17-2.06 (2H, m, CH2CO3), 1.60 (1H, dd, J ) 4.7, 14.0 Hz,
C-2), 1.50 (3H, s, Me), 1.34 (3H, s, Me). 13C NMR: 141.8 (Ph),
128.3 (Ph), 128.3 (Ph), 125.8 (Ph), 119.6 (CO3), 80.8 (C-4), 79.2
(C-3), 58.7 (C-1), 38.1 (CH2CO3), 29.7 (PhCH2), 27.4 (Me), 24.9
(C-2), 20.3 (Me). Anal. Calcd for C15H20O3: C, 72.56; H, 8.12.
Found: C, 72.34; H, 8.04.
1,2,4-Bu ta n etr iol 3-P h en ylor th op r op ion a te (6). The
reaction of 27 mg of epoxy ester 3 (0.12 mmol) in 0.7 mL of
0.05 M TFA in CDCl3 was followed by 1H NMR. When
approximately one-third of 3 remained, the reaction was
J . Org. Chem, Vol. 68, No. 26, 2003 10085

Giner et al.
quenched by the addition of 3 drops of TEA and purified by
preparative TLC (hexane/EtOAc 2:1) to give 8 mg of recovered
3 and 18 mg of 6 (0.08 mmol, 66%). H NMR: 7.29-7.24 (2H,
(3H, m), 4.17 (2H, s), 3.49 (2H, dd, J ) 3.2, 11.5 Hz), 3.41 (2H,
dd, J ) 4.2, 11.5 Hz), 2.97 (2H, t, J ) 7.6 Hz), 2.70 (2H, t, J )
7.6 Hz), 2.59 (2H, br s), 0.77 (3H, s). 13C NMR: 173.9, 140.1,
128.6, 128.2, 126.4, 67.6, 66.6, 40.7, 35.8, 31.0, 16.7.
2-Meth ylbu ta n e-1,2,4-tr iol 1-Hyd r ocin n a m a te (12). Hy-
drolysis of orthoester 4 as above gave 12. 1H NMR: 7.32-7.27
(2H, m), 7.23-7.19 (3H, m), 4.01 (1H, d, J ) 11.2 Hz), 4.00
(1H, d, J ) 11.2 Hz), 3.91-3.85 (1H, m), 3.83-3.78 (1H, m),
2.98 (2H, t, J ) 7.6 Hz), 2.81 (1H, br s), 2.71 (2H, t, J ) 7.6
Hz), 2.41 (1H, br s), 1.77 (1H, ddd, J ) 4.5, 7.8, 14.7 Hz), 1.62
(1H, ddd, J ) 4.1, 6.7, 14.7 Hz), 1.20 (3H, s). 13C NMR: 172.9,
140.2, 128.6, 128.2, 126.4, 72.4, 71.0, 59.3, 39.0, 35.7, 30.9, 24.3.
1
m, Ph), 7.21 (2H, d, J ) 7.3 Hz, Ph), 7.17 (1H, t, J ) 7.4 Hz,
Ph), 4.68-4.66 (1H, br m, C-2), 4.13 (1H, dt, J ) 4.3, 11.8 Hz,
C-4), 4.10 (1H, d, J ) 7.4 Hz, C-1), 3.97-3.93 (1H, m, C-1),
3.87 (1H, dd, J ) 6.7, 11.4 Hz, C-4), 2.80 (2H, dd, J ) 6.3,
11.0 Hz, PhCH2), 2.35-2.28 (1H, br m, C-3), 2.18-2.13 (2H,
m, CH2CO3), 1.41 (1H, ddd, J ) 2.3, 3.8, 13.5 Hz, C-3). 13C
NMR: 141.7 (Ph), 128.3 (Ph), 128.3 (Ph), 125.8 (Ph), 119.8
(CO3), 73.0 (C-2), 69.0 (C-1), 58.6 (C-4), 37.1 (CH2CO3), 29.6
(PhCH2), 28.3 (C-3). Anal. Calcd for C13H16O3: C, 70.90; H,
7.32. Found: C, 70.68; H, 7.19.
4-Met h ylp en t a n e-1,3,4-t r iol 3-H yd r ocin n a m a t e (16).
3-Met h yl-3-t et r a h yd r ofu r a n yl H yd r ocin n a m a t e (7).
Acid-catalyzed isomerization of 4 was carried out as described
1
Hydrolysis of orthoester 5 as above gave 16. H NMR: 7.32-
7.27 (2H, m), 7.23-7.19 (3H, m), 4.87 (1H, dd, J ) 2.9, 9.9
Hz), 3.59 (1H, dt, J ) 11.7, 4.7 Hz), 3.34 (1H, ddd, J ) 3.4,
10.0, 11.6 Hz), 2.99 (2H, t, J ) 7.6 Hz), 2.75 (2H, t, J ) 7.6
Hz), 1.95-1.87 (1H, m), 1.69-1.62 (1H, m), 1.17 (3H, s), 1.15
(3H, s). 13C NMR: 173.6, 140.0, 128.6, 128.2, 126.4, 77.2, 71.3,
58.3, 35.7, 32.2, 30.9, 26.1, 25.4.
1
for 5 using 0.2 M TFA in CDCl3. H NMR: 7.31-7.27 (2H, m,
Ph), 7.23-7.18 (3H, m, Ph), 3.99 (1H, d, J ) 10.0 Hz, C-2),
3.88-3.79 (2H, m, C-5), 3.71 (1H, d, J ) 10.0 Hz, C-2), 2.92
(2H, t, J ) 7.8 Hz, PhCH2), 2.60 (2H, t, J ) 7.8 Hz, CH2CO2),
2.33 (1H, ddd, J ) 1.9, 7.0, 13.5 Hz, C-4), 1.97 (1H, dt, J )
13.5, 8.0 Hz, C-4), 1.57 (3H, s, Me). 13C NMR: 172.4 (CdO),
140.3 (Ph), 128.4 (Ph), 128.3 (Ph), 126.2 (Ph), 86.4 (C-3), 77.4
(C-2), 67.2 (C-5), 39.3 (C-4), 36.5 (CH2CO2), 30.9 (PhCH2), 22.0
(Me). Anal. Calcd for C14H18O3: C, 71.78; H, 7.74. Found: C,
71.86; H, 7.69.
4-Met h ylp en t a n e-1,3,4-t r iol 1-H yd r ocin n a m a t e (17).
Treatment of 16 for 21 h at room temperature with 50 mM
TFA/CDCL3 yielded 17 in quantitative yield. 1H NMR: 7.31-
7.27 (2H, m), 7.22-7.18 (3H, m), 4.37 (1H, ddd, J ) 4.9, 9.3,
11.0 Hz), 4.18 (1H, ddd, J ) 4.8, 6.0, 11.1 Hz), 3.32 (1H, d, J
) 10.7), 2.96 (2H, t, J ) 7.7 Hz), 2.66 (2H, t, J ) 7.7 Hz), 2.54
(1H, br s), 1.93 (1H, br s), 1.82-1.76 (1H, m), 1.62-1.54 (1H,
m), 1.18 (3H, s), 1.14 (3H, s). 13C NMR: 173.4, 140.3, 128.5,
128.4, 126.4, 74.6, 72.6, 62.0, 35.8, 30.9, 30.8, 26.3, 23.4.
2,2-Dim eth yl-3-tetr ah ydr ofu r an yl Hydr ocin n am ate (8).
Acid-catalyzed isomerization of 5 was carried out using 0.1 M
TFA in CDCl3. Evaporation of the solvent gave 8 in quantita-
tive yield. 1H NMR: 7.31-7.26 (2H, m, Ph), 7.22-7.17 (3H,
m, Ph), 4.99 (1H, dd, J ) 2.7, 6.3 Hz, C-3), 3.91 (1H, dt, J )
8.2, 7.7 Hz, C-5), 3.85 (1H, dt, J ) 4.8, 8.6 Hz, C-5), 2.96 (2H,
t, J ) 7.7 Hz, PhCH2), 2.67 (2H, t, J ) 7.7 Hz, CH2CO2), 2.39-
2.33 (1H, m C-4), 1.87-1.81 (1H, m C-4), 1.19 (3H, s, Me), 1.14
(3H, s, Me). 13C NMR: 172.4 (CdO), 140.3 (Ph), 128.5 (Ph),
128.3 (Ph), 126.3 (Ph), 81.8 (C-2), 79.1 (C-3), 64.5 (C-5), 35.9
(CH2CO2), 32.4 (C-4), 30.9 (PhCH2), 26.0 (Me), 21.8 (Me). Anal.
Calcd for C15H20O3: C, 72.56; H, 8.12. Found: C, 72.39; H,
8.06.
Bu ta n e-1,2,4-tr iol 1-Hyd r ocin n a m a te (18). Hydrolysis of
orthoester 6 as above gave 18-20. Only partial separation
could be achieved by TLC. Their NMR data were obtained from
a combination of difference spectra and 2D NMR methods
1
(HMBC and HSQC). H NMR: 7.32-7.19 (5H, m), 4.13 (1H,
d, J ) 8.0 Hz), 4.04-4.00 (2H, m), 3.88-3.79 (2H, m), 2.97
(2H, t, J ) 7.7 Hz), 2.69 (2H, t, J ) 7.7 Hz), 2.57 (1H, br s),
2.10 (1H, br s), 1.69-1.64 (2H, m). 13C NMR: 173, 140, 128,
128, 126, 69.6, 68.5, 60.9, 34.7, 36, 31.
3-Meth yl-3-oxeta n ylm eth yl h yd r ocin n a m a te (10).34 Es-
terification of 3-methyl-3-oxetanemethanol with dihydrocin-
Bu ta n e-1,2,4-tr iol 2-Hyd r ocin n a m a te (19). 1H NMR:
7.32-7.19 (5H, m), 5.07-5.02 (1H, m), 3.70-3.60 (3H, m), 3.47
(1H, ddd, J ) 4.0, 9.3, 11.6 Hz), 2.98 (2H, t, J ) 7.6 Hz), 2.71
(2H, t, J ) 7.6 Hz), 2.18 (1H, br s), 1.91 (1H, br s), 1.86-1.79
(1H, m), 1.77-1.70 (1H, m). 13C NMR: 173, 140, 128, 128, 126,
72.9, 64.6, 58.3, 36, 33.7, 31.
Bu ta n e-1,2,4-tr iol 4-Hyd r ocin n a m a te (20). 1H NMR:
7.32-7.19 (5H, m), 4.36 (1H, ddd, J ) 4.9, 8.9, 11.3 Hz), 4.15
(1H, dt, J ) 11.3, 5.4 Hz), 3.66-3.58 (2H, m), 3.46-3.40 (1H,
m), 2.96 (2H, t, J ) 7.8 Hz), 2.66 (2H, t, J ) 7.8 Hz), 2.51 (1H,
br s), 1.88 (1H, br s), 1.75-1.63 (2H, m). 13C NMR: 173, 140,
128, 128, 126, 68.9, 66.6, 61.2, 36, 32.3, 31.
1
namoyl chloride was carried out as described for 2. H NMR:
7.31-7.26 (2H, m), 7.22-7.18 (3H, m), 4.45 (2H, d, J ) 5.9
Hz), 4.34 (2H, d, J ) 5.9 Hz), 4.15 (2H, s), 2.97 (2H, d, J ) 7.7
Hz), 2.69 (2H, d, J ) 7.7 Hz), 1.28 (3H, s). 13C NMR (75 MHz):
172.9, 140.3, 128.5, 128.2, 126.3, 79.5, 68.7, 39.0, 21.1.
2-Hyd r oxym eth yl-2-m eth yl-1,3-p r op a n ed iol 3-p h en yl-
or th op r op ion a te (11).34,35 Acid-catalyzed isomerization of 10
was carried out as described for 5 using 0.5 M TFA in CDCl3.
1H NMR: 7.28-7.24 (2H, m), 7.21-7.14 (3H, m), 3.93 (6H, s),
2.79-2.74 (2H, m), 2.00-1.96 (2H, m), 0.82 (3H, s). 13C NMR
(75 MHz): 141.9, 128.4, 128.3, 125.7, 108.7, 72.6, 38.5, 30.3,
29.6, 14.5.
3-H yd r oxy-2-h yd r oxym et h yl-2-m et h ylp r op yl h yd r o-
cin n a m a te. Hydrolysis of orthoester 10 (8.6 mg, 0.04 mmol)
was carried out in a 2:1 mixture of acetone-d6 and 50 mM
NaOAc/D2O pH 4.75 buffer. When 1H NMR analysis showed
the reaction to be complete, the acetone was removed with a
stream of N2, and the mixture was partitioned beween
saturated aqueous NaHCO3 and Et2O. The crude product was
purified by preparative TLC (hexane/EtOAc 1:2) to yield 9.0
mg (0.04 mmol, 97%). 1H NMR: 7.32-7.27 (2H, m), 7.23-7.19
Ack n ow led gm en t. We thank the Petroleum Re-
search Fund, administered by the American Chemical
Society, for financial support, and David Kiemle for
technical NMR assistance.
Su p p or tin g In for m a tion Ava ila ble: General procedures
for NMR kinetics. Graphs for the rearrangements of 4-6 and
10. Spectra of selected 13C NMR signals for 18O-labeled
compounds 2a , 5a ,b, 7a ,b, 8a ,b, and 9a . 2D NMR spectra
(HMBC) for 5-8. This material is available free of charge via
the Internet at http://pubs.acs.org.
(34) J ohnson, R. E.; Schlegel, D. C.; Ezrin, A. M. Eur. Pat. Appl.
93203115, 1994; Chem. Abstr. 1994, 121, 280683.
(35) Charette, A. B.; Chua, P. Tetrahedron Lett 1997, 38, 8499-
8502.
J O034566S
10086 J . Org. Chem., Vol. 68, No. 26, 2003