Total synthesis of (±)-dihydrokawain-5-ol. Regioselective monoprotection of vicinal syn-diols derived from the iodocyclofunctionalization of α-allenic alcohols

Article abstract of DOI:10.1021/jo961653u

The synthesis of (±)-dihydrokawain-5-ol (1) from the vinyl iodo syn-vicinal diol 7a is described. This diol was prepared in a highly diastereoselective fashion via the iodocyclofunctionalization reaction of the N-tosyl carbamate derivative of the corresponding α-allenic alcohol (6a). A key to the synthesis of 1 involved the differentiation of the alcohol groups of the diol moiety in 7a. Application of Yamamoto's monoprotection protocol for the introduction of MOM ethers in vicinal diols provided 8a in a highly regioselective fashion (8a:9a > 30:1) from 7a. This regioselective monoprotection was found to be general for vinyl iodo and acetylenic vicinal diols 7 and 10, placing the MOM protecting group on the homoallylic and homopropargylic alcohol of the diol moieties, respectively. Alternatively, a highly regioselective (11.5:1) monosilylation of the homoallylic alcohol in 7a followed by etherification (MOM) of the allylic alcohol provided the differentially protected diol 25. Further manipulation of the vinyl iodide function in 25 (dehydroiodination, carbonylation) followed by desilylation generated the γ-alkoxy-δ-hydroxy-α,β-acetylenic ester 28. Cyclization of 28 produced the unique 4,5-dialkoxypyran-2-one moiety present in (±)-1. This latter transformation involved the interesting acid-catalyzed and thermodynamically driven isomerization of the intermediate β-alkoxy-α,β-unsaturated ester (Z)-29 to the corresponding E-isomer.

Full text of DOI:10.1021/jo961653u

J . Org. Chem. 1996, 61, 9103-9110
9103
Tota l Syn th esis of (()-Dih yd r ok a w a in -5-ol. Regioselective
Mon op r otection of Vicin a l Syn -Diols Der ived fr om th e
Iod ocyclofu n ction a liza tion of r-Allen ic Alcoh ols
Richard W. Friesen* and Christopher Vanderwal
Department of Medicinal Chemistry, Merck Frosst Centre for Therapeutic Research, P.O. Box 1005,
Pointe Claire-Dorval, Que´bec, Canada H9R 4P8
Received August 27, 1996X
The synthesis of (()-dihydrokawain-5-ol (1) from the vinyl iodo syn-vicinal diol 7a is described.
This diol was prepared in a highly diastereoselective fashion via the iodocyclofunctionaliztion
reaction of the N-tosyl carbamate derivative of the corresponding R-allenic alcohol (6a ). A key to
the synthesis of 1 involved the differentiation of the alcohol groups of the diol moiety in 7a .
Application of Yamamoto’s monoprotection protocol for the introduction of MOM ethers in vicinal
diols provided 8a in a highly regioselective fashion (8a :9a > 30:1) from 7a . This regioselective
monoprotection was found to be general for vinyl iodo and acetylenic vicinal diols 7 and 10, placing
the MOM protecting group on the homoallylic and homopropargylic alcohol of the diol moieties,
respectively. Alternatively, a highly regioselective (11.5:1) monosilylation of the homoallylic alcohol
in 7a followed by etherification (MOM) of the allylic alcohol provided the differentially protected
diol 25. Further manipulation of the vinyl iodide function in 25 (dehydroiodination, carbonylation)
followed by desilylation generated the γ-alkoxy-δ-hydroxy-R,â-acetylenic ester 28. Cyclization of
28 produced the unique 4,5-dialkoxypyran-2-one moiety present in (()-1. This latter transformation
involved the interesting acid-catalyzed and thermodynamically driven isomerization of the
intermediate â-alkoxy-R,â-unsaturated ester (Z)-29 to the corresponding E-isomer.
In tr od u ction
diol moiety at C5/C6, make 1 an attractive and challeng-
ing synthetic target.8
(+)-Dihydrokawain-5-ol (1), a unique 6-alkyl-5-hy-
droxy-5,6-dihydropyran-2-one, was isolated in 1970 from
the methanol extracts of the kava plant (Piper Methys-
ticum Forst.), a Polynesian shrub of the pepper family.1
Extracts of the roots and stem of this plant are utilized
in folk medicine and in the preparation of a traditional
ceremonial beverage.2 The dihydropyran-2-one moiety
is a frequently encountered substructure in natural
products.3 Oxygen substitution at either the C4 or C5
position is fairly common, as exemplified by (+)-kawain
(2)4 and (+)-asperlin (3),5 respectively. However, the
oxygenation pattern in 1 is, as far as we are aware,
unknown in any other dihydropyranone-containing natu-
ral product except for a more highly oxygenated member
of the same family (4).6 Moreover, the presence of the
C5 hydroxyl in 1 renders the molecule susceptible to
rearrangement as illustrated by the facile conversion of
1, under basic conditions (KOH-MeOH), to the corre-
sponding furanone 5.7 These unique structural features,
including the presence of oxygen substituents at both the
C4 and C5 positions on the pyranone scaffold, and the
threo relationship between the oxygens that comprise the
We have been interested in the synthesis of natural
products using our recently described method for the
highly diastereoselective conversion of the N-tosyl car-
bamate derivatives of secondary R-allenic alcohols 6 into
vinyl iodo syn-vicinal diols 7 (eq 1).9-11 This procedure
involves initial treatment of the allene O-carbamate with
I2 to produce an intermediate diiodo carbamate11 which
X Abstract published in Advance ACS Abstracts, December 1, 1996.
(1) Achenbach, H.; Wittman, G. Tetrahedron Lett. 1970, 3259.
(2) The kava plant is a Polynesian shrub of the pepper family.
Extracts of its roots and stem are utilized in folk medicine and in the
preparation of a traditional ceremonial beverage. Ko¨ppel, C.; Tenczer,
J . J . Chromatogr. 1991, 562, 207.
(3) Dickinson, J . M. Nat. Prod. Rep. 1993, 88.
(4) Ha¨nsel, R.; Beiersdorf, H. U. Arzneim.-Forsch. 1959, 9, 581.
Kawain (2), one of the principal compounds isolated from the kava
shrub is marketed in racemic form (Neuronika (Klinge Pharma,
Munich)) because of its anxiolytic and analgesic properties.2
(5) Argoudelis, A. D.; Zieserl, J . F. Tetrahedron Lett. 1966, 1969.
(6) Ha¨nsel, R.; Pelter, A.; Schulz, J .; Hille, C. Chem. Ber. 1976, 109,
1617.
(7) Ha¨nsel, R.; Schulz, J .; Pelter, A.; Ayoub, M. T.; Reinhardt, R. Z.
Naturforsch., B: Anorg. Chem., Org. Chem. 1978, 33B, 1020 (Chem.
Abstr. 90, 22705u).
(8) Two identical syntheses of 1 have been described involving
nonstereoselective allylic oxidation (SeO2) and hydrogenation of kawain
(2). (a) Achenbach, H.; Huth, H. Tetrahedron Lett. 1974, 119. (b)
Ha¨nsel, R.; Schulz, J . Chem. Ber. 1973, 106, 570.
(9) Friesen, R. W.; Bissada, S. Tetrahedron Lett. 1994, 35, 5615.
(10) (a) Friesen, R. W.; Giroux, A. Tetrahedron Lett. 1993, 34, 119.
(b) Friesen, R. W.; Giroux, A. Can. J . Chem. 1994, 72, 1857.
(11) Friesen, R. W.; Bayly, C. I.; Fogg, J . A. J . Org. Chem. 1995, 60,
448.
S0022-3263(96)01653-2 CCC: $12.00 © 1996 American Chemical Society

9104 J . Org. Chem., Vol. 61, No. 26, 1996
Friesen and Vanderwal
Sch em e 1
undergoes cyclization upon exposure to Ag2CO3.9,10 We
felt that the implementation of this iodocyclofunctional-
ization strategy would allow us to address the structur-
ally interesting features in 1 (Scheme 1). Most impor-
tantly, the relative stereochemistry between the C5 and
C6 hydroxy and alkoxy moieties in 1 would be readily
established upon conversion of allenic alcohol 6a to diol
7a . The vinyl iodide moiety would serve as a synthetic
precursor for subsequent transformation to the R,â-
acetylenic ester B via a dehydroiodination-carbonylation
sequence. Several methods have been described for the
conversion of such unsaturated esters into 4-alkoxydi-
hydropyran-2-ones as required for the synthesis of 1.12,13
Finally, because of the aforementioned base lability of
1, it was imperative that the protecting group chosen for
the C5 hydroxyl would be removable under acidic or
neutral conditions in the final step of the synthesis. In
keeping with this plan, we felt that the optimum choices
for PG1 and PG2 would be a silyl ether and an acetal
protecting group (MOM, THP), respectively. Therefore,
another key feature of the synthesis would be the
preparation of A by the initial regioselective introduction
of either of these groups onto the requisite hydroxyl of
the diol moiety in the syn-vicinal diol 7a . We felt that
the difference in the electronic and steric environment
surrounding each alcohol in the vicinal diol, due to the
presence of the vinyl iodide moiety, would enable us to
readily differentiate the two alcohol groups and introduce
protecting groups in a sequential and regioselective
fashion.
synthetic protocols are well established. The regioselec-
tive monoprotection of acyclic vicinal diols in which both
alcohols are secondary alcohols is less straightforward
and more challenging. Typical strategies involve the
utilization of neighboring functionality, such as an
ester,18 to protect one of the alcohols intramolecularly
while directing a second reagent to the remaining alcohol
moiety. Alternatively, the presence of functional groups
immediately adjacent to the diol moiety can result in a
differential reactivity of the hydroxyls such that a selec-
tive monoprotection strategy is possible. Recent studies
have described, for example, the regioselective sulfony-
lation of threo-2,3-dihydroxy esters19 and the monosily-
lation of 2,3-dihydroxyphosphonates.20 There are a few
reports of the regioselective introduction of an acetal
protecting group, such as MOM or THP, in a polyhy-
droxylated substrate.15,21
Regioselective In tr od u ction of th e MOM P r otect-
in g Gr ou p . As described above, the synthetic plan called
for the differentiation of the alcohols in 7a and for the
installation of an acid labile protecting group on the
allylic alcohol (PG2). Yamamoto has described a proce-
dure for introducing a MOM group onto the more steri-
cally hindered alcohol of a vicinal diol (secondary vs
primary alcohol) via the regioselective (5-13:1) reductive
cleavage (DIBAL, CH2Cl2, -78 to 0 °C) of the intermedi-
ate orthoester prepared in situ from the diol and tri-
methyl orthoformate (eq 2).21a By analogy, it was an-
ticipated that application of this procedure to the vicinal
In this paper we describe the general and highly
regioselective introduction of the MOM protecting group
onto vicinal diols 7 and their acetylene derivatives 10.
In addition, the regioselective silylation (TBDMS) of the
diols 7 is reported, as well as the application of this
protecting group strategy in the synthesis of the novel
dihydropyran-2-one (()-dihydrokawain-5-ol (1).
Resu lts a n d Discu ssion
The regioselective monoprotection of polyhydroxylated
compounds is of continuing interest in organic synthe-
sis.14 Regioselectivity is readily achieved when the
alcohol moieties are in drastically different electronic or
steric environments, such as the selective protection of
a primary alcohol in the presence of a secondary alcohol.15
Other examples include the regioselective silylation
reactions of nucleosides16 and carbohydrates17 for which
diol 7a might lead to the the desired monoprotected diol
9a since the allylic alcohol adjacent to the vinyl iodide
function appeared to be the more sterically encumbered
alcohol (eq 3). Surprisingly, exposure of diol 7a to this
one-pot reaction sequence led solely to the protected diol
8a (8a :9a > 30:1; 88% isolated yield) in which the MOM
(12) Henbest, H.; J ones, E. R. J . Chem. Soc. 1950, 3628.
(13) (a) Carlson, R. M.; Oyler, A. R. Tetrahedron Lett. 1974, 2615.
(b) Carlson, R. M.; Oyler, A. R.; Peterson, J . R. J . Org. Chem. 1975,
40, 1610.
(14) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; J ohn Wiley & Sons: New York, 1991; pp 68-87,
137-140.
(15) For example, see: (a) Ishihara, K.; Kurihara, H.; Yamamoto,
H. J . Org. Chem. 1993, 58, 3791. (b) Oikawa, M.; Wada, A.; Okazaki,
F.; Kusumoto, S. J . Org. Chem. 1996, 61, 4469. (c) Bailey, W. F.;
Zarcone, L. M. J .; Rivera, A. D. J . Org. Chem. 1995, 60, 2532.
(16) (a) Ogilvie, K. K.; Hakimelahi, G. H.; Proba, Z. A.; McGee, D.
P. C. Tetrahedron Lett. 1982, 23, 1997. (b) Ogilvie, K. K.; McGee, D.
P. C.; Boisvert, S. M.; Hakimelahi, G. H.; Proba, Z. A. Can. J . Chem.
1983, 61, 1204.
(17) For example, see: Bhatt, R. K.; Chauhan, K.; Wheelan, P.;
Murphy, R. C.; Falck, J . R. J . Am. Chem. Soc. 1994, 116, 5050.
(18) (a) Keinan, E.; Sinha, S. C.; Sinha-Bagchi, A.; Wang, Z.; Zhang,
X.; Sharpless, K. B. Tetrahedron Lett. 1992, 33, 6411. (b) Walsh, P. J .;
Bennani, Y. L.; Sharpless, K. B. Tetrahedron Lett. 1993, 34, 5545.
(19) Fleming, P. R.; Sharpless, K. B. J . Org. Chem. 1991, 56, 2869.
(20) Yokomatsu, T.; Suemune, K.; Yamagishi, T.; Shibuya, S. Synlett
1995, 847.
(21) (a) Takasu, M.; Naruse, Y.; Yamamoto, H. Tetrahedron Lett.
1988, 29, 1947. (b) Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K.
Chem. Lett. 1983, 1593.
(22) The vinyl iodo syn-diols 7a -e were prepared from the R-allenic
alcohols 6a -e using our previously reported iodocyclofunctionalization
procedure (eq 1).10

Total Synthesis of (()-Dihydrokawain-5-ol
J . Org. Chem., Vol. 61, No. 26, 1996 9105
Ta ble 1. Mon op r otection (MOM) of Syn -Vicin a l Diols 7
noteworthy that the R group in these two examples is
an unsaturated moiety (R ) Ph) whereas the other
examples all have an alkyl substituent at this site. In
three examples containing sterically bulky R groups (7c
R ) c-hex and 7d /10d R ) t-Bu; Table 1, entries 3, 4,
and 9), a cyclic acetal (13c/d and 14d ) is also observed.24
a n d 10
ratio 8:9 or
11:12
yield 8 or
11 (%)b
entry
substrate (R)
18a
1
2
3
4
5
6
7
8
9
7a PhCH2CH2 1.6:1
>30:1c
>30:1
>30:1
>30:1
83:17
>30:1
>30:1
>30:1
>30:1
85:15
8a 88
8b 90
7b i-Bu
7c C6H11
7d t-Bu
7e Ph
1.8:1
1.3:1
1.4:1
1.6:1
8c 87d
8d 10 (67)e
8e/9e 90f
11a 88
10a PhCH2CH2 1.3:1
10b i-Bu
10c C6H11
10d t-Bu
10e Ph
1.2:1
1.3:1
1.6:1
1.1:1
11b 92
11c 96
11d 40g
10
11e/12e 93h
a
Ratio of diastereomers of unidentified relative stereochemistry
measured from the integrated 1H NMR spectrum of the crude
Interestingly, the anti-vicinal diol 1525 also provided
the MOM ether 16 as the sole regioisomer (>30:1, 94%
isolated yield) when subjected to the Yamamoto reaction
conditions (eq 5). Although additional experiments are
warranted, it appears that the relative stereochemistry
b
reaction mixture. Yields of pure isolated isomers. c For ratios
reported as >30:1, the minor isomer could not be observed in the
d
1H NMR spectrum. Although the acetal 13c was observed in the
1H NMR spectrum of the crude reaction mixture (8c:13c ∼15:1),
its volatility and/or instability precluded its isolation. e Value in
brackets refers to isolated yield of the acetal 13d . f Yield of
g
combined, inseparable isomers. Although the acetal 14d was
observed in the 1H NMR spectrum of the crude reaction mixture,
its high volatility precluded the accurate measurement of the ratio
h
of 11d :14d or its isolation. Yield of combined isomers. A small
sample of 11e was isolated for characterization.
moiety had been introduced onto the homoallylic alcohol
(eq 3, Table 1, entry 1).
between the alcohol moieties also has little effect upon
the reaction regioselectivity.
The regioisomers in each series were readily assigned
1
to the structures indicated on the basis of their H NMR
spectra. Thus, in isomers 8, 11, and 16, the methine
proton Ha adjacent to the site of unsaturation couples to
both the hydroxyl proton and either the vinylic or
acetylenic proton.
The degree of regioselectivity observed in this example
was sufficiently interesting that we wanted to explore
the scope of this monoprotection protocol. Thus, the vinyl
iodo syn-diols 7b-e22 and the acetylenic syn-diols 10a -e
were prepared in order to explore the effects of (1) the
nature of the R group (R ) primary to tertiary alkyl,
phenyl) and (2) the type of unsaturation (vinyl iodide or
alkyne) on the reaction regioselectivity. The acetylenic
diols 10 were readily obtained by dehydroiodination
(NaN(TMS)2) of the vinyl iodide moiety, either on the
parent diol or one of several readily available deriva-
tives.23 Each diol was then subjected to the standard
Yamamoto monoprotection protocol.20a As can be seen
from the results presented in Table 1 (entries 2-10), the
reaction is extremely regioselective for both vinyl iodo
and acetylenic syn-diols, resulting in the preferential
formation of MOM ethers 8 and 11 by the introduction
of the MOM moiety on the homoallylic and homopro-
pargylic alcohols, respectively (eqs 3 and 4). Except for
Although this monoprotection reaction is not useful in
the context of the synthesis of dihydrokawain-5-ol (1) as
outlined above, the mechanistic basis for the regioselec-
tivity that is observed is still intriguing. The mechanism
of the DIBAL-H-mediated reduction has been proposed
to proceed via the initial and potentially reversible
coordination of the oxophilic aluminum atom to the
orthoester 18 (X ) vinyl iodide or acetylene), in this case
forming species such as 19 or 20.20,26 Cleavage of the
orthoester with participation of the remaining oxygen
atoms provides the ring opened oxonium ions 21 and 22.
However, inter- or intramolecular transfer of hydride
could take place from either 19/20 or 21/22 to produce
the MOM-monoprotected diols. If the reduction takes
place directly from 19/20, one might expect that the
regiochemical outcome of the reaction would be affected
by the configuration of the OMe moiety relative to the R
and X groups. The crude reaction mixtures containing
the orthoester 18 were isolated prior to reduction with
DIBAL-H, and inspection of the resulting 1H NMR
spectra indicated the presence of mixtures of diastereo-
mers, with ratios ranging from 1.1:1 to 1.6:1 (unidentified
(24) Although the acetals 13c (8c:13c ∼15:1) and 14d were observed
in the 1H NMR spectra of the crude reaction mixtures, their volatility
and/or instability precluded their isolation.
(25) Anti diol 15 was prepared from 8b via an unoptimized three-
step sequence involving (1) Mitsunobu inversion, (2) hydrolysis, and
(3) deprotection. See Experimental Section for details.
(26) The regioselective reductive cleavage of benzylidene acetals and
other related systems has also been explained by a mechanism in which
C-O bond cleavage takes place at the site of metal complexation. See,
for example: (a) Garegg, P. J .; Hultberg, H.; Wallin, S. Carbohydr.
Res. 1982, 108, 97. (b) Takano, S.; Akiyama, M.; Sato, S.; Ogasawara,
K. Chem. Lett. 1983, 1593. (c) J ohansson, R.; Samuelsson, B. J . Chem.
Soc., Perkin Trans 1 1984, 2371. (d) Ishihara, K.; Mori, A.; Yamamoto,
H. Tetrahedron Lett. 1987, 28, 6613. (e) Mikami, T.; Asano, H.;
Mitsunobu, O. Chem. Lett. 1987, 2033.
two examples involving the monoprotection of diols 7e
and 10e (R ) Ph, Table 1, entries 5 and 10), there was
no evidence (1H NMR spectra) in any of the crude
reaction mixtures for the formation of the alternative
regioisomers 9 or 12 (>30:1 regioselectivity). It is
(23) Diol 10a was prepared in two steps (NaN(TMS)2; TBAF) from
the TBDMS diol 23a . Diols 10b and 10c were prepared directly from
the diols 7b and 7c, respectively. Diols 10d and 10e were prepared
via the acetonide derivatives of diols 7d and 7e, respectively. See the
Experimental Section for details.

9106 J . Org. Chem., Vol. 61, No. 26, 1996
Friesen and Vanderwal
Ta ble 2. Mon osilyla tion of Syn -Vicin a l Diol 7a
ratio
yield
(%)c
entry
base, solvent, temperaturea
23a :24a b
1
2
3
4
5
6
7
8
9
10
11
12
13
imidazole, DMF, rtd
1.2:1
1.2:1
1:2.6
1:1.3
1.2:1
1.3:1
2:1
4.5:1
3.5:1
8.5:1
9.5:1
9.5:1
11.5:1
87
73
98
92
71
55
98
90
pyridine, AgNO3, THF, rtd
DBU, CH2Cl2, -78 °C
Et3N, CH2Cl2, -78 °C
pyridine, CH2Cl2, -78 to 0 °C
DBMP,e CH2Cl2, -78 °C
2,6-lutidine, CH2Cl2, -78 °C
i-Pr2NEt, CH2Cl2, -78 °C
i-Pr2NEt, CH2Cl2, 0 °C
2,6-lutidine, DMF, 0 °C
i-Pr2NEt, DMF, rt
96
90
96
i-Pr2NEt, DMF, 0 °C
i-Pr2NEt, DMF, -45 °C
92
93 (89)f
a
In general, [substrate] approximately 100 mg/mL, 2.5 equiv
b
of base, 1.5-2.5 equiv of TBDMSOTf. Measured from the inte-
grated H NMR spectrum of the crude reaction mixture. c Purified
1
d
combined yield of 23a /24a . Reaction carried out with TBDMSCl.
e DBMP ) 2,6-di-tert-butyl-4-methylpyridine. f Yield in brackets
refers to isolated yield of pure 23a . For detailed reaction condi-
tions, see Experimental Section.
Regioselective Mon osilyla tion . The alternative
strategy for differentiating the alcohols in 7a called for
the initial introduction of PG1 (SiR3) on the homoallylic
alcohol of 7a , followed by the installation of PG2 (MOM)
on the allylic alcohol. A representative sampling of our
results in attempting the regioselective monosilylation
of vinyl iodo syn-diol 7a (eq 6) is found in Table 2.
relative stereochemistry, see Table 1).27 Thus, it appears
that the configuration of the OMe moiety relative to R
and X in 18 has little bearing on the regiochemical
outcome of the reduction process as would be expected if
cleavage of the orthoester preceded the rate-determining
reduction step. Moreover, the site of unsaturation must
play a key role in the reaction regioselectivity. Support
for this hypothesis is seen in the reaction of diols 7e and
10e in which the alcohols are each flanked by unsatur-
ated substituents, a phenyl group and either a vinyl
iodide or acetylene moiety, respectively. In these ex-
amples, both MOM regioisomers are formed (Table 1,
entries 5 and 10). Although suggestive, further study is
required to clarify the mechanistic details regarding the
manner in which the site of unsaturation dictates the
regioselective reduction of the intermediate orthoester
18.28 Nonetheless, this potentially useful regioselective
transformation should find other applications since enan-
tiomerically enriched acetylenic diols such as 10 are
readily available using Sharpless dihydroxylation tech-
nology on the corresponding enynes.29
Typical silylating conditions14 using TBDMSCl and
imidazole in DMF, or reaction conditions that prove to
be very effective in differentiating the C2 and C3 second-
ary alcohols of nucleosides (TBDMSCl, pyridine and
AgNO3),16 were nonregioselective (Table 2, entries 1 and
2). However, more interesting results were obtained
upon treatment of diol 7a with TBDMSOTf under a
variety of reaction conditions. The isomeric, and readily
separable, monosilylated diols 23a and 24a were obtained
in the ratios and combined yields indicated in Table 2,
entries 3-13. It is clear from these results that base,
solvent and temperature all have an effect on the
observed regioselectivity. When the reaction was con-
ducted in CH2Cl2 at -78 °C, the regioselectivity of
silylation was strongly influenced by the base (Table 2,
entries 3-8). Silyl ether 24a was formed preferentially
using DBU and Et3N, but the desired silyl ether 23a was
obtained in moderate regioselectivity using both 2,6-
lutidine and i-Pr2NEt. These latter observations encour-
aged us to explore the effect of temperature and solvent
using these two bases (Table 2, entries 9-13). While the
temperature effect was relatively small (for example,
compare entries 8 and 9), a more significant effect was
brought about by changing the solvent. Using i-Pr2NEt,
the regioselectivity in favor of 23a increased from 4.5:1
in CH2Cl2 to 11.5:1 in DMF (Table 2, entries 8, 13).
Somewhat surprisingly, the formation of disilylated
compounds was not observed (<5%) under any of the
reaction conditions, suggesting that the reactivity of the
remaining unprotected alcohol in either 23a or 24a is
being attenuated by either the bulky vinyl iodide moiety
(in the case of 23a ) or the adjacent silyloxy group (in the
case of 24a ). Therefore the use of excess silylating
(27) We have had limited success in reducing these orthoesters after
workup and isolation. In most cases, these reductions are extremely
messy compared to the reduction of the orthoesters prepared in situ
and used without isolation. These observations are most likely due to
the instability of the labile orthoester moiety as noted by Yamamoto.21a
(28) One possibility is that the regiochemical outcome is a reflection
of the relative stabilites of the intermediates 21 and 22. The interme-
diate 22 would be expected to be destabilized relative to 21 due to the
requirement for forming an oxomium ion with the already electron
deficient allylic (or propargylic) oxygen. As a result, reduction of 21
may be more facile than that of 22. In addition, the predominant
formation of the acetals 13d and 14d (R ) t-Bu) in the reduction of
the orthoesters derived from diols 7d and 10d could be explained by
complexation of DIBAL-H to the OMe oxygen atom of the orthoester
18 rather than the oxygen atoms of the five-membered ring as in 19
or 20. This alternative pathway would be favored in these latter
examples since the increased steric bulk of the tert-butyl group would
make complexation to the oxygen atoms of the five-membered ring in
orthoester 18 less facile than when R is a sterically less demanding
substituent.
(29) (a) J eong, K.-S.; Sjo¨, P.; Sharpless, K. B. Tetrahedron Lett. 1992,
33, 3833. (b) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. Rev. 1994, 94, 2483.

Total Synthesis of (()-Dihydrokawain-5-ol
J . Org. Chem., Vol. 61, No. 26, 1996 9107
Ta ble 3. Mon osilyla tion of Syn -Vicin a l Diols 7c-ea
Sch em e 2
entry
substrate R
ratio 23:24b
yield (%)c
1
2
3
7c c-hex
7d t-Bu
7e Ph
1.8:1
1:4.6
9.5:1
23c 65,
23d 15,
23e 82,
24c 34
24d 60
24e 9
a
These reactions were carried out under the conditions de-
scribed in Table 2, entry 13. Measured from the integrated 1H
b
NMR spectrum of the crude reaction mixture. c Yields of pure
regioisomers.
reagent was not problematic and enabled us to achieve
essentially complete conversion of 7a to monosilylated
products. Thus, the silylation of 7a under the optimized
conditions indicated in Table 2, entry 13, takes place with
excellent regioselectivity (11.5:1) on the homoallylic
alcohol group to afford 23a in 89% isolated yield.
It should be noted that the TBDMS groups in 23a and
24a are not prone to undergo 1,2-O,O migration30 with
the reaction conditions that are employed. Exposure of
each to 2.5 equiv of i-Pr2NEt in CH2Cl2 or DMF at room
temperature for 24 h led to undetectable amounts of the
alternative regioisomer. Therefore, the reactions sum-
marized in Table 2 are being conducted under kinetically
controlled conditions and reflect the kinetic reactivity of
the alcohol groups. The regioselectivity of silylation
using alternative bulky silylating reagents such as TIP-
SCl, TIPSOTf, TBDPSCl, and TESCl was also explored
in this reaction but with much less success. The site of
silylation in 23a and 24a was readily determined from
inspection of their 1H NMR spectra and homonuclear
decoupling experiments. For example, the allylic me-
thine proton Ha in 23a is coupled to both the hydroxyl
proton and the vinylic protons. In contrast, the homoal-
lylic methine proton Hb in 24a is coupled to the hydroxyl
proton while the allylic methine proton Hc is coupled to
one of the vinylic protons.
problematic migration of the TBDMS group to the
adjacent oxyen atom30 was not observed under these more
forcing basic reaction conditions. Dehydrohalogenation
of 25 was readily effected using NaN(TMS)2 in DMF10a
to provide the alkyne 26 in 94% yield. Deprotonation of
26 (MeLi, THF) followed by trapping of the resulting
lithium acetylide anion with methyl chloroformate gave
the R,â-acetylenic ester 27. Desilylation (TBAF, THF)
yielded the hydroxy ester 28 (2 steps, 53%), the key
substrate for cyclization to the dihydropyranone.
If the sole determinant in the regioselective silylation
of 7a is related to the relative steric bulk of the vinyl
iodide moiety with respect to the R group, then changing
the nature of the R group (primary to tertiary) should
have an observable effect on the outcome. We have
explored this issue using the vinyl iodo vicinal syn-diols
7c,d (eq 6). Silylations were conducted under the
optimized reaction conditions found for syn-diol 7a (Table
2, entry 13). As summarized in Table 3, the reaction
regioselectivity is indeed affected by the steric bulk of
the R group. As the steric size of R increases, the
regioselectivity in favor of isomer 23 changes until, using
diol 7d (R ) t-Bu), the alternative regioisomer 24d
predominates. From these observations, it appears that
the regioselectivity of tert-butyldimethylsilylation (using
TBDMSOTf) of vinyl iodo syn-vicinal diols 7 bearing alkyl
R groups is significantly influenced by the steric bulk of
this moiety. However, monosilylation of diol 7e (R ) Ph)
again favors isomer 23 (Table 3, entry 3). In this latter
example, an electronic component to the regioselectivity
cannot be ruled out. Thus, the regioselective monosily-
lation of vinyl iodo syn-vicinal diols 7 can be a useful
synthetic process when R is a primary alkyl (7a ) or
phenyl (7e) substituent.
Two distinct strategies have been described for the
formation of 4-alkoxydihydropyran-2-ones from δ-hy-
droxy-R,â-usaturated esters, employing either acid12 or
base13 catalysis (eq 7). However, in these reports, the
C5 position is invariably unsubstituted and we are not
aware of any examples in which an alkoxy group is
situated at the C5 position as is required for the synthesis
of 1. Treatment of 28 with HgO and BF3-Et2O in MeOH,
conditions described by J ones and Henbest,12 resulted in
partial cyclization to 30, but the acidic reaction conditions
also brought about competing deprotection of the MOM
group. As a result, the final reaction mixture was fairly
complex and contained 1 as well as material tentatively
identified as the corresponding furanone, the latter
presumably being formed by cyclization of the fully
deprotected diol. Employing the alternative base-cata-
lyzed protocol of Carlson and Oyler,13 we exposed 28 to
catalytic NaOMe in MeOH to provide a ∼2:1 mixture of
(Z)-29 (58% isolated yield) and (E)-29, the latter undergo-
ing lactonization in the reaction mixture to the desired
pyranone 30 (33% isolated yield).31 Using purified (Z)-
29, it was determined that (Z)-29 is not converted to (E)-
29 using these reaction conditions while decomposition
Syn th esis of (()-Dih yd r ok a w a in -5-ol. The monosi-
lylated diol 23a was surprisingly resistant to reaction
with MOM-Cl and ultimately afforded 25 in good yield
(89%) only at elevated temperature (85 °C) using i-Pr2-
NEt as solvent (Scheme 2). Fortunately, the potentially
(30) (a) J ones, S. S.; Reese, C. B. J . Chem. Soc., Perkin Trans. 1
1979, 2763. (b) Mulzer, J .; Scheo¨llhorn, B. Angew. Chem., Int. Ed. Engl.
1990, 29, 431.

9108 J . Org. Chem., Vol. 61, No. 26, 1996
Friesen and Vanderwal
of (Z)-29 occurs under more forcing reaction conditions
(up to 2 equiv of NaOMe in refluxing MeOH).
A fortuitous observation was made when it was found
that a pure sample of (Z)-29 underwent isomerization to
(E)-29 while acquiring a 13C NMR spectrum in CDCl3. It
was rationalized that this isomerization was being cata-
lyzed by trace amounts of HCl or DCl in the solvent,
resulting in the formation of the apparently thermody-
namically more stable olefin (E)-29. No isomerization
takes place when the commercial solvent CDCl3 was
treated with solid NaHCO3 immediately prior to dissolu-
tion of (Z)-29, confirming that the isomerization is indeed
acid catalyzed.32 On a preparative scale, the optimum
reaction conditions involved simply stirring a solution of
(Z)-29 in untreated CDCl3 (Isotech) for 1 h. After this
time, a clean conversion of (Z)-29 to (E)-29 was observed
the coupling constant between H5 and H6 was found to
be 2.5 Hz (lit.1 J 5,6 ) 2.3 Hz), confirming the relative
stereochemistry between these two substituents.
Su m m a r y
The introduction of a MOM protecting group onto
vicinal diols 7 and 10 using Yamamoto’s procedure is
highly selective and proceeds, in a one pot process,
through the formation and regioselective DIBAL-H re-
duction of an intermediate orthoester. This reaction
introduces the MOM group preferentially on the homoal-
lylic and homopropargylic alcohols, respectively, regard-
less of the nature of the R group. The monosilylation
(TBDMS) of the diols 7 is much more dependant upon
the size of the R group, and useful regioselectivites are
obtained only when R ) primary alkyl or Ph. The
synthesis of (()-dihydrokawain-5-ol (1) has been ac-
complished from vinyl iodo syn-diol 7a in 28% overall
yield. The key steps of the synthesis, involving regiose-
lective monsilylation chemistry and manipulation of the
vinyl iodide moiety, exemplify the synthetic utility of
these functionalized diols that are derived from R-allenic
alcohols.
1
by H NMR spectroscopy (>30:1). Characteristic in the
1
monitoring of this isomerization reaction by H NMR is
the change in the chemical shift of the allylic methine
proton resonance. This resonance shifts from δ 3.85 ppm
in the spectrum of (Z)-29 to δ 5.50 ppm in the spectrum
of (E)-29 due to the deshielding effect of the ester moiety.
After isomerization, the crude reaction mixture was
subjected to the original conditions of NaOMe/MeOH in
order to effect conversion of (E)-29 to the pyranone 30
(79% overall for the conversion of 28 to 30).
The relative heats of formation of the model compounds
31 and 32 were calculated33 in an attempt to rationalize
the apparent thermodynamic stability of (E)-29 with
respect to (Z)-29. Indeed, the minimum energy confor-
mation of the E-isomer 32 was found to be at least 4.3
kcal/mol more stable than that of the (Z)-isomer 31.
From the calculations, it appears that the major desta-
bilizing force in 31 is an electrostatic interaction between
the electronegative methoxy and ester groups that are
cis with respect to each other on the olefin bond. This
contributor to the overall heat of formation accounts for
almost all of the energy differential between the two
isomers.34
Exp er im en ta l Section
Gen er a l. 1H NMR spectra were recorded at 300 or 400
MHz in acetone-d6 or CDCl3 as specified. Broad-band proton-
decoupled 13C NMR spectra were recorded at 100 MHz in
CDCl3. IR spectra were recorded on neat samples unless
stated otherwise. Workup procedures involving the drying of
organics was done with MgSO4. Flash column chromatogra-
phy (referred to as chromatography) was performed with 230-
400 mesh silica gel, eluting with the solvents indicated (v/v).
The vinyl iodo syn-diol 7a has been prepared previously.10 The
vinyl iodo syn-diols 7b-e were prepared according to our
previously published procedure, and characterization data can
be found in the supporting information.10
Removal of the MOM protecting group in 30 was
accomplished under acidic conditions to provide (()-
dihydrokawain-5-ol (1) (90%) as a white solid (mp 121-
122 °C) that exhibited spectra (1H NMR, IR, LRMS) in
accord with that reported for (+)-1.1 Most importantly,
a n ti-3,4-Dih yd r oxy-6-m eth yl-2-iod o-1-h ep ten e (15). To
a solution of alcohol 8b (1.15 g, 3.66 mmol), p-nitrobenzoic acid
(1.84 g, 3 equiv), and Ph3P (2.88 g, 3 equiv) in benzene (40
mL) at 0 °C was added di-tert-butyl azodicarboxylate (2.53 g,
3 equiv), and the mixture was allowed to warm to rt. After
15 h, another 2 equiv of each reagent was added, and the
mixture was stirred a further 15 h. The mixture was diluted
with ether and washed with brine, dried, and concentrated.
Chromatography (5:2 to 2:1 hexane/ether) provided the Mit-
sunobu product (610 mg, 36%) which was dissolved in MeOH
(10 mL) and treated with solid K2CO3 (360 mg, 2 equiv). After
30 min, the solvent was evaporated, the residue was dissolved
in ethyl acetate, and 1 N HCl was added. The organics were
removed, dried, and concentrated. The residue was dissolved
in MeOH (8 mL), treated with BF3-OEt2 (150 µL, 1.1 equiv)
and stirred for 24 h at rt. The solvent was evaporated, and
the residue was dissolved in ethyl acetate, washed with brine,
dried, and concentrated. Chromatography (5:2 hexane/ether)
provided the diol 15 (245 mg, 25% for three steps) as a white
solid: mp 51-52 °C; IR (KBr) 3400, 1610 cm-1; 1H NMR (300
MHz, acetone-d6) δ 0.88 (d, 3H, J ) 6.6 Hz), 0.91 (d, 3H, J )
6.7 Hz), 1.37 (m, 2H), 1.87 (m, 1H), 3.45 (d, 1H, J ) 6.8 Hz),
3.69 (m, 1H), 3.78 (m, 1H), 4.46 (d, 1H, J ) 4.4 Hz), 5.90 (dd,
1H, J ) 0.8, 1.3 Hz), 6.47 (t, 1H, J ) 1.3 Hz); 13C NMR δ 21.4,
23.8, 24.4, 39.7, 70.8, 80.2, 111.7, 127.4. Anal. Calcd for
C8H15O2I: C, 35.57; H, 5.60. Found: C, 35.80; H, 5.45.
syn -3,4-Dih yd r oxy-6-p h en yl-1-h exyn e (10a ). To a solu-
tion of vinyl iodide 23a (2.00 g, 4.62 mmol) in DMF (30 mL)
(31) The base-mediated addition of alcohols (ROH) to R,â-acetylenic
esters usually provides the Z-isomer as the predominant reaction
product under kinetically controlled conditions. (a) Winterfeldt, E.;
Preuss, H. Chem. Ber. 1966, 99, 450. (b) Winterfeldt, E.; Preuss, H.
Angew. Chem., Int. Ed. Engl. 1967, 6, 423. Therefore, it is somewhat
surprising that such high conversions of δ-hydroxy-R,â-acetylenic esters
to 4-alkoxy-R,â-unsaturated lactones under base catalysis (eq 7) have
been reported unless there is a mechanistic pathway available for the
isomerization of the preferentially formed Z-isomer (such as (Z)-29) to
the E-isomer which is required for lactonization. Obviously, in our case
the base-catalyzed Z to E isomerization is not observed. Therefore, one
must conclude that the oxygen substituent at C5 is in some way
inhibiting this transformation since no other reported examples include
an alkoxy group at the C5 position.
(32) The isomerization of (Z)-29 to (E)-29 was surveyed using a
variety of acid catalysts and solvents, including catalytic CSA or HCl
in CHCl3, THF, or MeOH. All of these conditions were inferior to those
described, resulting either in little isomerization or much more
extensive decomposition of the labile functionality (methyl vinyl ether
and MOM ether) in the starting material and/or product. Obviously,
there is a fine line between the acidic reaction conditions that lead to
decomposition and those that result in isomerization.
(33) These calculations were carried out using the recently published
Merck Molecular Force Field. We would like to thank Dr. Christopher
Bayly for assistance in carrying out these calculations. (a) Halgren,
T. J . Comput. Chem. 1996, 17, 490. (b) Halgren, T. J . Comput. Chem.
1996, 17, 520. (c) Halgren, T. J . Comput. Chem. 1996, 17, 553. (d)
Halgren, T.; Nachbar, R. B. J . Comput. Chem. 1996, 17, 587. (e)
Halgren, T. J . Comput. Chem. 1996, 17, 616.
(34) The thermodynamic stability of related systems (â-amino-R,â-
unsaturated esters) has also been described.30

Total Synthesis of (()-Dihydrokawain-5-ol
J . Org. Chem., Vol. 61, No. 26, 1996 9109
at 0 °C was added NaN(TMS)2 (11.6 mL of a 1 M solution in
THF, 2.5 equiv) dropwise. After 20 min, water was added and
the mixture was extracted with ether. The organics were
washed with water and brine and dried. After concentrating,
the residue was dissolved in THF (30 mL) and cooled to 0 °C.
A solution of TBAF (7 mL in 1 M in THF, 1.5 equiv) was added
and after 30 min, water was added. The mixture was
extracted with ether, and the organics were washed with brine,
dried, and concentrated. Chromatography (1:1 to 1:2 hexane/
ether) provided the alkyne diol 10a (780 mg, 89%) as a
colorless oil: IR 3360, 2115 cm-1; 1H NMR (300 MHz, CDCl3)
δ 1.76-1.89 (m, 1H), 1.97-2.08 (m, 1H), 2.47 (d, 1H, J ) 2.2
Hz), 2.72 (ddd, 1H, J ) 7.0, 9.6, 13.8 Hz), 2.88 (ddd, 1H, J )
5.2, 9.9, 13.8 Hz), 3.35 (br s, 1H), 3.64-3.72 (m, 2H), 4.20 (br
d, 1H, J ) 6.5 Hz), 7.16-7.32 (m, 5H); 13C NMR δ 31.6, 33.9,
65.9, 74.0, 74.7, 82.2, 125.8, 128.3, 128.4, 141.5. Anal. Calcd
for C12H14O2: C, 75.76; H, 7.42. Found: C, 75.60; H, 7.24.
syn -3,4-Dih yd r oxy-6-m eth yl-1-h ep tyn e (10b). To a solu-
tion of diol 7b (800 mg, 2.96 mmol) in DMF (10 mL) at 0 °C
was added NaN(TMS)2 (10.4 mL of a 1M solution in THF, 3.5
equiv) dropwise. After 20 min, water was added and the
mixture was extracted with ether. The organics were washed
with water and brine, dried, and concentrated. Chromatog-
raphy (1:1 hexane/ether) provided the alkyne diol 10b (225
syn -2-Iod o-4-(m e t h oxym e t h oxy)-6-p h e n yl-1-h e xe n -
3-ol (8a ). A solution of diol 7a 10 (250 mg, 0.79 mmol), CH-
(OMe)3 (175 µL, 2 equiv), and catalytic CSA (5 mg) in CH2Cl2
(12 mL) was stirred at rt for 45 min. The solution was cooled
to -78 °C, and neat DIBAL-H (1.4 mL, 10 equiv) was added.
The mixture was stirred at this temperature for 1 h and then
placed in an ice bath. After 10 min, 1 N HCl was added and
the mixture was extracted with EtOAc. The organics were
washed with 1 N HCl (2×) and brine, dried, and concentrated.
The crude reaction mixture was inspected by 1H NMR spec-
troscopy (Table 1, entry 1). Chromatography (10:1 hexane/
ether) provided the MOM ether 8a (250 mg, 88%) as a colorless
1
oil: IR 3420, 1610 cm-1; H NMR (300 MHz, CDCl3) δ 1.69-
1.92 (m, 2H), 2.65 (ddd, 1H, J ) 6.6, 9.8, 13.8 Hz), 2.78 (ddd,
1H, J ) 5.6, 9.9, 13.8 Hz), 3.43 (s, 3H), 3.55-3.67 (m, 2H),
3.82 (d, 1H, J ) 4.0 Hz), 4.62 (d, 1H, J ) 6.7 Hz), 4.76 (d, 1H,
J ) 6.7 Hz), 5.95 (d, 1H, J ) 1.6 Hz), 6.44 (dd, 1H, J ) 1.0,
1.6 Hz), 7.15-7.31 (m, 5H); 13C NMR δ 31.4, 33.4, 56.0, 79.4,
82.7, 97.9, 113.1, 126.0, 127.8, 128.37, 128.42, 141.4. Anal.
Calcd for C14H19O3I: C, 46.42; H, 5.29. Found: C, 46.46; H,
5.39.
syn -4-(Meth oxym eth oxy)-6-p h en yl-1-h exyn -3-ol (11a ).
Following procedure A using diol 10a (200 mg, 1.05 mmol),
the MOM ether 11a (217 mg, 88%) was obtained as a colorless
1
mg, 53%) as a volatile, colorless oil: IR 3350, 2120 cm-1; H
1
oil: IR 3410, 3290, 2110 cm-1; H NMR (300 MHz, CDCl3) δ
NMR (300 MHz, acetone-d6) δ 0.91 (d, 3H, J ) 6.6 Hz), 0.93
(d, 3H, J ) 6.6 Hz), 1.35-1.50 (m, 2H), 1.87 (m, 1H), 2.84 (d,
1H, J ) 2.2 Hz), 3.60 (m, 1H), 3.74 (br d, 1H, J ) 4.6 Hz),
4.07 (m, 1H), 4.43 (br d, 1H, J ) 5.5 Hz); 13C NMR δ 21.6,
23.6, 24.4, 41.3, 66.3, 73.0, 74.4, 82.5. Anal. Calcd for
C8H14O2: C, 67.57; H, 9.92. Found: C, 67.81; H, 10.04.
syn -3,4-Dih yd r oxy-4-cycloh exyl-1-bu tyn e (10c). Fol-
lowing the procedure described for the conversion of 7b to 10b,
diol 7c (750 mg, 2.53 mmol) was converted to alkyne diol 10c
(216 mg, 51%), a white solid: mp 74.8-75.4 °C; IR (KBr) 3350,
3280, 2120 cm-1; 1H NMR (400 MHz, CDCl3) δ 1.05-1.31 (m,
5H), 1.60-1.80 (m, 6H), 2.47 (d, 1H, J ) 2.2 Hz), 2.70 (br s,
1H), 3.06 (br s, 1H), 3.39 (br s, 1H), 4.34 (br s, 1H); 13C NMR
δ 25.9, 26.15, 26.24, 27.3, 29.7, 39.1, 63.5, 74.3, 78.7, 82.9. Anal.
Calcd for C10H16O2: C, 71.39; H, 9.59. Found: C, 71.68; H,
9.46.
syn -3,4-Dih yd r oxy-5,5-d im eth yl-1-h exyn e (10d ). A so-
lution of diol 7d (1.0 g, 3.7 mmol), 2-methoxypropene (1.3 mL,
3.5 equiv), and catalytic CSA in DMF (3 mL) was stirred for
30 min. The solution was cooled to 0 °C, and NaN(TMS)2 (4.4
mL of a 1 M solution in THF, 1.2 equiv) was added dropwise.
After 45 min, water was added and the mixture was extracted
with ether. The organics were washed with water and brine,
dried, and concentrated. The residual oil was passed through
a plug of silica, eluting with 3% ether in hexane, and the
partially purified acetonide alkyne was redissolved in MeOH
(15 mL) and 6 N HCl (3 mL). After stirring at 65 °C for 90
min, the solution was cooled to rt and concentrated. The
residue was dissolved in ethyl acetate and washed successively
with saturated NaHCO3, water, and brine, dried, and concen-
trated. Chromatography (1:3 hexane/ether) provided the
alkyne diol 10d (220 mg, 42%) as a white solid: mp 59.8-
60.2 °C; IR (KBr) 3400, 3300, 2120 cm-1; 1H NMR (400 MHz,
CDCl3) δ 0.95 (s, 9H), 2.46 (d, 1H, J ) 2.2 Hz), 2.96 (d, 1H, J
) 5.3 Hz), 3.21 (d, 1H J ) 7.1 Hz), 3.35 (dd, 1H, J ) 3.7, 5.3
Hz), 4.40 (m, 1H); 13C NMR δ 26.4, 34.6, 61.8, 73.5, 80.7, 84.6.
Anal. Calcd for C8H14O2: C, 67.57; H, 9.92. Found: C, 67.11;
H, 9.61.
1.89-2.00 (m, 1H), 2.04-2.16 (m, 1H), 2.46 (d, 1H, J ) 2.2
Hz), 2.67 (ddd, 1H, J ) 6.6, 9.9, 13.8 Hz), 2.82 (ddd, 1H, J )
5.4, 10.2, 13.8 Hz), 3.44 (s, 3H), 3.44 (d, 1H, J ) 5.0 Hz), 3.59
(ddd, 1H, J ) 3.9, 6.0, 8.2 Hz), 4.37 (ddd, 1H, J ) 2.2, 5.0, 6.0
Hz), 4.68 (d, 1H, J ) 6.8 Hz), 4.83 (d, 1H, J ) 6.8 Hz), 7.15-
7.31 (m, 5H); 13C NMR δ 31.4, 32.7, 55.9, 64.8, 74.1, 82.2, 82.4,
97.5, 125.9, 128.29, 128.35, 141.5. Anal. Calcd for C14H18O3:
C, 71.77; H, 7.74. Found: C, 71.97; H, 7.88.
a n ti-2-Iod o-4-(m et h oxym et h oxy)-6-m et h yl-1-h ep t en -
3-ol (16). Following procedure A using diol 15 (100 mg, 0.37
mmol), the MOM ether 16 (109 mg, 94%) was obtained as a
colorless oil. IR 3450, 1615 cm-1; 1H NMR (300 MHz, acetone-
d6) δ 0.89 (d, 3H, J ) 6.6 Hz), 0.90 (d, 3H, J ) 6.7 Hz), 1.22
(ddd, 1H, J ) 2.7, 9.9, 14.3 Hz), 1.53 (ddd, 1H, J ) 4.0, 9.7,
14.3 Hz), 1.80 (m, 1H), 3.34 (s, 3H), 3.90 (ddd, 1H, J ) 2.7,
4.3, 9.6 Hz), 4.14 (tt, 1H, J ) 1.3, 4.3 Hz), 4.52 (d, 1H, J ) 4.3
Hz), 4.59 (d, 1H, J ) 6.9 Hz), 4.74 (d, 1H, J ) 6.9 Hz), 5.93 (t,
1H, J ) 1.3 Hz), 6.55 (t, 1H, J ) 1.3 Hz); 13C NMR δ 21.5,
23.8, 24.1, 37.4, 56.2, 77.9, 78.2, 96.6, 109.4, 126.4. Anal. Calcd
for C10H19O3I: C, 38.23; H, 6.10. Found: C, 38.57; H, 6.10.
Gen er a l P r oced u r e B: Mon osilyla tion of Diols 7. The
procedure for the reaction of diol 7a is representative. Ex-
perimental details and characterization data for the monosi-
lylation of diols 7c-e can be found in the supporting infor-
mation.
syn -4-(ter t-Bu tyld im eth ylsiloxy)-2-iod o-6-p h en ylh ex-1-
en -3-ol (23a ) a n d syn -3-(ter t-Bu tyld im eth ylsiloxy)-2-iod o-
6-p h en ylh ex-1-en -4-ol (24a ). To a solution of diol 7a (2.4 g,
7.5 mmol) and i-Pr2NEt (3.3 mL, 2.5 equiv) in DMF (20 mL)
at -45 °C was added TBDMSOTf (2.2 mL, 1.25 equiv)
dropwise. After 2 h, another 1 mL of TBDMSOTf (0.6 equiv)
was added, and the mixture was stirred for a further 30 min
when TLC indicated the absence of starting material. In some
of the other cases, additional TBDMSOTf was required to
completely consume the starting material. Saturated aq NH4-
Cl was added, and the resulting mixture was extracted with
ether. The organics were washed with 1 N HCl (2×) and brine,
dried, and concentrated. After inspection by 1H NMR spec-
troscopy, the residue was subjected to chromatography (7:3
to 1:1 hexane/toluene), and the alcohols 23a and 24a were
obtained as colorless oils. The first compound to be eluted was
syn -3,4-Dih yd r oxy-4-p h en yl-1-bu tyn e (10e). Following
the procedure described for the conversion of 7d to 10d , diol
7e (700 mg, 2.41 mmol) was converted to alkyne diol 10e (216
mg, 55%), a white solid: mp 60.0-60.4 °C; IR (KBr) 3300, 2120
the major silyl ether 23a (2.9 g, 89%): IR 3510, 1620, 835 cm-1
;
1
cm-1; H NMR (300 MHz, CDCl3) δ 2.38 (d, 1H, J ) 2.0 Hz),
1H NMR (300 MHz, CDCl3) δ 0.07 (s, 3H), 0.08 (s, 3H), 0.90
(s, 9H), 1.74-1.86 (m, 1H), 1.95-2.08 (m, 1H), 2.60-2.77 (m,
2H), 3.11 (d, 1H, J ) 8.1 Hz), 3.96 (dm, 1H, J ) 8.1 Hz), 4.06
(ddd, 1H, J ) 2.8, 4.5, 7.4 Hz), 5.95 (dd, 1H, J ) 1.1, 1.6 Hz),
6.49 (t, 1H, J ) 1.6 Hz), 7.17-7.33 (m, 5H); 13C NMR δ -4.34,
-4.26, 18.1, 25.9, 31.3, 36.4, 71.9, 78.3, 113.8, 125.97, 126.01,
128.39, 128.44, 141.6. Anal. Calcd for C18H29O2ISi: C, 50.00;
H, 6.75. Found: C, 50.22; H, 7.06. The more polar minor
regioisomer 24a (0.25 g, 8%): IR 3580, 1610, 835 cm-1; 1H NMR
(300 MHz, CDCl3) δ 0.12 (s, 6H), 0.95 (s, 9H), 1.61-1.79 (m,
3.69 (br s, 2H), 4.34 (dd, 1H, J ) 2.0, 7.3 Hz), 4.64 (d, 1H, J )
7.3 Hz), 7.24-7.36 (m, 5H); 13C NMR δ 66.9, 75.2, 77.0, 81.5,
127.1, 128.2, 128.3, 138.8. Anal. Calcd for C10H10O2: C, 74.06;
H, 6.21. Found: C, 73.67; H, 6.29.
Gen er a l P r oced u r e A: Mon op r otection (MOM) of
Diols 7 a n d 10. The procedure described below for the
protection of 7a is general. Experimental details and char-
acterization data for the monoprotection (MOM) of diols 7b-e
and 10b-e can be found in the supporting information.

9110 J . Org. Chem., Vol. 61, No. 26, 1996
Friesen and Vanderwal
2H), 2.35 (dd, 1H, J ) 0.9, 3.8 Hz), 2.69 (ddd, 1H, J ) 6.8, 9.8,
13.8 Hz), 2.90 (ddd, 1H, J ) 5.2, 10.0, 13.8 Hz), 3.48 (dd, 1H,
J ) 0.7, 6.2 Hz), 3.63 (m, 1H), 5.95 (d, 1H, J ) 1.5 Hz), 6.37
(dd, 1H, J ) 0.7, 1.5 Hz), 7.16-7.30 (m, 5H); 13C NMR δ -4.89,
-4.22, 18.1, 25.8, 31.9, 34.3, 73.0, 81.8, 113.6, 125.8, 127.5,
128.3, 128.5, 141.9. Anal. Calcd for C18H29O2ISi: C, 50.00;
H, 6.75. Found: C, 50.24; H, 6.91.
2.01 (m, 1H), 2.55 (br s, 1H), 2.72 (ddd, 1H, J ) 7.1, 9.5, 13.8
Hz), 2.89 (ddd, 1H, J ) 5.1, 9.9, 13.8 Hz), 3.39 (s, 3H), 3.76 (s,
3H), 3.75 (m, 1H), 4.28 (d, 1H, J ) 6.7 Hz), 4.65 (d, 1H, J )
6.8 Hz), 4.89 (d, 1H, J ) 6.8 Hz), 7.16-7.31 (m, 5H); 13C NMR
δ 31.5, 34.1, 52.8, 56.0, 70.0, 72.2, 78.3, 83.4, 95.0, 125.9, 128.4,
128.5, 141.4, 153.3. Anal. Calcd for C16H20O5: C, 65.74; H,
6.90. Found: C, 65.57; H, 6.87.
syn -4-(t er t -B u t y ld im e t h y ls ilo x y )-2-io d o -3-(m e t h -
oxym eth oxy)-6-p h en ylh ex-1-en e (25). A solution of alcohol
23a (4.54 g, 10.5 mmol) and MOM-Cl (10.6 mL, 12 equiv) in
i-Pr2NEt (30 mL) was heated at 85 °C until the starting
material was consumed (approximately 3 h). The mixture was
cooled to rt, and water was added. The mixture was extracted
with ether and the organics were washed with 1 N HCl (2×)
and brine, dried, and concentrated. Chromatography (95:5
hexane/ether) of the residue provided 25 (4.46 g, 89%) as a
colorless oil: IR 1610, 830 cm-1; 1H NMR (300 MHz, acetone-
d6) δ 0.161 (s, 3H), 0.164 (s, 3H), 0.97 (s, 9H), 1.67-1.93 (m,
2H), 2.70-2.81 (m, 2H), 3.39 (s, 3H), 3.59 (d, 1H, J ) 7.2 Hz),
3.94 (dt, 1H, J ) 7.2, 5.0 Hz), 4.62 (d, 1H, J ) 6.5 Hz), 4.66 (d,
1H, J ) 6.5 Hz), 6.13 (d, 1H, J ) 1.5 Hz), 6.62 (dd, 1H, J )
0.8, 1.5 Hz), 7.14-7.31 (m, 5H); 13C NMR δ -4.6, -4.0, 18.3,
26.0, 30.6, 35.4, 56.2, 73.5, 83.3, 94.3, 111.1, 125.8, 128.3, 128.4,
129.5, 142.4. Anal. Calcd for C20H33O3ISi: C, 50.42; H, 6.98.
Found: C, 50.56; H, 6.94.
(Z)-Met h yl
syn -3-Met h oxy-5-h yd r oxy-4-(m et h oxy-
m eth oxy)-7-p h en ylh ep t-2-en oa te ((Z)-29) a n d P yr a n on e
30. To a solution of hydroxy ester 28 (130 mg, 0.44 mmol) in
MeOH (1.5 mL) was added 550 µL of a solution of NaOMe in
MeOH (0.2 M, 0.25 equiv). The mixture was stirred at rt for
18 h, and then EtOAc was added. The mixture was washed
with 1 N HCl and brine, dried, and concentrated. Chroma-
tography (5:2 to 3:2 hexane/EtOAc) first provided (Z)-29 (84
mg, 58%) as a colorless oil. IR 3480, 1715, 1640 cm-1; 1H NMR
(400 MHz, CDCl3) δ 1.75-1.82 (m, 2H), 2.42 (br s, 1H), 2.68
(dt, 1H, J ) 16.5, 8.2 Hz), 2.86 (app qu, 1H), 3.37 (s, 3H), 3.66
(s, 3H), 3.72 (m, 1H), 3.85 (d, 1H, J ) 5.7 Hz), 3.92 (s, 3H),
4.60 (d, 1H, J ) 6.8 Hz), 4.68 (d, 1H, J ) 6.8 Hz), 5.27 (s, 1H),
7.14-7.27 (m, 5H); 13C NMR δ 31.8, 34.6, 51.2, 56.2, 61.3, 71.7,
80.8, 95.4, 97.3, 125.8, 128.3, 128.4, 141.7, 165.2, 167.4. Anal.
Calcd for C17H24O6: C, 62.95; H, 7.46. Found: C, 63.07; H,
7.59. Further elution yielded the pyranone 30 (43 mg, 33%)
as a colorless oil: IR 1710, 1630 cm-1 1H NMR (300 MHz,
;
syn -4-(ter t-Bu tyld im eth ylsiloxy)-3-(m eth oxym eth oxy)-
6-p h en ylh ex-1-yn e (26). To a solution of vinyl iodide 25 (3.30
g, 6.93 mmol) in DMF (30 mL) at 0 °C was added NaN(TMS)2
(12.4 mL of a 1 M solution in THF, 1.8 equiv) slowly over 5
min. After 30 min, water was added and the mixture was
extracted with ether. The organics were washed with water
and brine, dried, and concentrated. Chromatography (96:4
hexane/ether) of the residue provided alkyne 26 (2.26 g, 94%)
acetone-d6) δ 1.99-2.11 (m, 1H), 2.16-2.29 (m, 1H), 2.77 (ddd,
1H, J ) 7.0, 9.4, 13.7 Hz), 2.90 (ddd, 1H, J ) 5.4, 9.7, 13.7
Hz), 3.34 (s, 3H), 3.82 (s, 3H), 4.08 (dd, 1H, J ) 1.2, 2.3 Hz),
4.36 (ddd, 1H, J ) 2.3, 4.9, 8.8 Hz), 4.66 (d, 1H, J ) 6.7 Hz),
4.81 (d, 1H, J ) 6.7 Hz), 5.20 (d, 1H, J ) 1.2 Hz), 7.16-7.37
(m, 5H); 13C NMR δ 30.9, 31.5, 56.2 (double intensity), 69.5,
77.7, 92.1, 96.0, 126.1, 128.49, 128.50, 140.7, 165.9, 171.5.
Anal. Calcd for C16H20O5: C, 65.74; H, 6.90. Found: C, 65.16;
H, 6.80.
as a colorless oil: IR 2110, 830 cm-1 1H NMR (400 MHz,
;
CDCl3) δ 0.119 (s, 3H), 0.122 (s, 3H), 0.95 (s, 9H), 1.94-2.09
(m, 2H), 2.43 (d, 1H, J ) 2.2 Hz), 2.68 (ddd, 1H, J ) 5.9, 10.9,
13.6 Hz), 2.80 (ddd, 1H, J ) 6.0, 11.0, 13.6 Hz), 3.40 (s, 3H),
3.85 (dt, 1H, J ) 4.1, 6.2 Hz), 4.31 (dd, 1H, J ) 2.2, 6.2 Hz),
4.62 (d, 1H, J ) 6.6 Hz), 4.93 (d, 1H, J ) 6.6 Hz), 7.16-7.30
(m, 5H); 13C NMR δ -4.7, -4.4, 18.1, 25.9, 31.3, 35.1, 55.7,
69.7, 73.4, 75.0, 80.3, 94.6, 125.7, 128.31, 128.34, 142.4. Anal.
Calcd for C20H32O3Si: C, 68.92; H, 9.25. Found: C, 68.86; H,
9.61.
P yr a n on e 30 fr om Isom er iza tion of (Z)-29. A solution
of (Z)-29 (114 mg, 0.35 mmol) in CDCl3 (6 mL, Isotech) was
stirred at rt for 2 h. The intermediate (E)-29 was not
purified: 1H NMR (300 MHz, CDCl3) δ 1.66-1.78 (m, 2H), 2.66
(ddd, 1H, J ) 7.4, 9.5, 13.6 Hz), 2.87 (ddd, 1H, J ) 5.8, 9.7,
13.6 Hz), 3.37 (s, 3H), 3.62 (s, 3H), 3.66 (s, 3H), 3.87 (m, 1H),
4.68 (s, 2H), 5.15 (s, 1H), 5.50 (d, 1H, J ) 6.8 Hz), 7.12-7.29
(m, 5H). The solution was concentrated and dissolved in
MeOH (1.5 mL), and 500 µL of a solution of NaOMe in MeOH
(0.2 M, 0.25 equiv) was added. The mixture was stirred at rt
for 18 h, and then EtOAc was added. The mixture was washed
with 1 N HCl and brine, dried, and concentrated. Chroma-
tography (2:1 hexane/EtOAc) provided pyranone 30 (82 mg,
80%) as a colorless oil.
Met h yl syn -5-(ter t-Bu t yld im et h ylsiloxy)-4-(m et h oxy-
m eth oxy)-7-p h en ylh ep t-1-yn oa te (27). To a solution of
alkyne 26 (2.27 g, 6.51 mmol) in THF (45 mL) at -78 °C was
added MeLi (7 mL of a 1.4 M solution in THF, 1.5 equiv), and
the resulting mixture was stirred at this temperature for 10
min and then at 0 °C for 30 min. Methyl chloroformate (754
µL, 1.5 equiv) was added slowly and after 15 min, water was
added. The mixture was extracted with ether, and the
organics were washed with 1 N HCl and brine, dried, and
concentrated. Chromatography (95:5 hexane/ether) of the
residue provided ester 27 (1.77 g, 67%) as a colorless oil: IR
(()-Dih yd r ok a w a in -5-ol (1). A solution of pyranone 30
(115 mg, 0.39 mmol) in 75% aq HOAc (2 mL) and concd H2-
SO4 (1 drop) was stirred at 70 °C for 3 h. The mixture was
diluted with EtOAc, washed with brine, dried, and concen-
trated. Chromatography (3:2 hexane/EtOAc) provided (()-1
(88 mg, 90%) as a white solid: mp 121-122 °C; IR (KBr) 3485,
1
2230, 1720, 830 cm-1; H NMR (400 MHz, CDCl3) δ 0.098 (s,
1
1695, 1625, 1225 cm-1; H NMR (400 MHz, CDCl3) δ 1.98-
3H), 0.105 (s, 3H), 0.93 (s, 9H), 1.89-2.08 (m, 2H), 2.65 (ddd,
1H, J ) 5.6, 11.0, 13.6 Hz), 2.78 (ddd, 1H, J ) 5.5, 11.3, 13.6
Hz), 3.39 (s, 3H), 3.76 (s, 3H), 3.87 (dt, 1H, J ) 7.2, 5.0 Hz),
4.39 (d, 1H, J ) 6.0 Hz), 4.62 (d, 1H, J ) 6.7 Hz), 4.86 (d, 1H,
J ) 6.7 Hz), 7.16-7.30 (m, 5H); 13C NMR δ -4.7, -4.5, 18.1,
25.8, 31.4, 35.0, 52.6, 55.8, 69.6, 73.0, 78.1, 84.4, 95.1, 125.8,
128.3, 142.0, 153.5. Anal. Calcd for C22H34O5Si: C, 64.99; H,
8.43. Found: C, 65.08; H, 8.33.
2.07 (m, 1H), 2.25-2.35 (m, 1H), 2.74-2.91 (m, 2H), 3.74 (s,
3H), 3.92 (dd, 1H, J ) 0.5, 2.6 Hz), 4.21 (ddd, 1H, J ) 2.6, 5.2,
8.7 Hz), 5.15 (d, 1H, J ) 0.5 Hz), 7.16-7.29 (m, 5H); 13C NMR
δ 30.9, 31.0, 56.4, 66.2, 78.1, 91.4, 126.2, 128.5, 128.6, 140.7,
166.3, 172.4. Anal. Calcd for C14H16O4: C, 67.73; H, 6.50.
Found: C, 67.36; H, 6.20.
Meth yl syn -5-Hyd r oxy-4-(m eth oxym eth oxy)-7-p h en yl-
h ep t-2-yn oa te (28). To a solution of ester 27 (1.53 g, 3.76
mmol) in THF (40 mL) at 0 °C was added TBAF (5.6 mL of a
1 M solution in THF, 1.5 equiv). Saturated NH4Cl was added
after 1 h, and the resulting mixture was extracted with EtOAc.
The organics were washed with brine, dried, and concentrated.
Chromatography (1:1 hexane/ether) of the residue provided
hydroxy ester 28 (870 mg, 79%) as a colorless oil. IR 3450,
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data (1H and 13C NMR, IR,
analysis) for 7b-e, 8b-e, 11b-e, 13d , 23c-e, and 24c-e (6
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
1
2235, 1720 cm-1; H NMR (400 MHz, CDCl3) δ 1.84 (m, 1H),
J O961653U