Substituted oxocanes by intramolecular allylboration reactions. Entry to an efficient synthesis of (+)-laurencin

Article abstract of DOI:10.1021/ja970914u

Δ⁵-Oxocenes, such as cis-3-hydroxy-2-vinyl-Δ⁵-oxocenes, can be generated by an intramolecular aldehyde-allylboration reaction. This allowed a rapid and stereoselective access to the trisubstituted oxocene 24 which had been transforme

Full text of DOI:10.1021/ja970914u

J. Am. Chem. Soc. 1997, 119, 7499-7504
7499
Substituted Oxocanes by Intramolecular Allylboration
Reactions. Entry to an Efficient Synthesis of (+)-Laurencin
Jochen Kru1ger and Reinhard W. Hoffmann*
Contribution from the Fachbereich Chemie, Philipps-UniVersita¨t Marburg,
Hans-Meerwein-Strasse, D-35032 Marburg, Germany
ReceiVed March 21, 1997X
Abstract: ∆5-Oxocenes, such as cis-3-hydroxy-2-vinyl-∆5-oxocenes, can be generated by an intramolecular aldehyde-
allylboration reaction. This allowed a rapid and stereoselective access to the trisubstituted oxocene 24 which had
been transformed by the Holmes group into (+)-laurencin. The key feature of this effective intramolecular allylboration
reaction is that both the aldehyde and the allylboronate functionality have been generated in a masked form in situ.
They are liberated by an aqueous workup which initiates the ring closure reaction.
Introduction
Scheme 1
Interest in saturated and unsaturated six- to nine-membered
oxygen heterocycles has been stimulated by the discovery of
the polyether toxins,1 such as brevetoxin,2 ciguatoxin,3 the
halichondrins,4 maitotoxin,5 or yessotoxin.6 In his classical
account of the first brevetoxin synthesis7 Nicolaou devoted a
whole section to the “oxocene-problem”, which he addressed
as an ill-famous, hard-to-overcome obstacle. Nicolaou himself
developed some ingenious routes to ring-fused oxocenes, yet
this left the challenge to uncover general inroutes to this class
of oxygen heterocycles.
Scheme 2
Driven by the interest in the polyether toxins, most progress
has been made in the synthesis of fused tetrahydropyran rings.8
Successful routes include ring closure of γ-hydroxy epoxides,9
intramolecular Michael addition of a hydroxy group to an R,â-
unsaturated ester,10 and ring closing metathesis reactions11 as
well as intramolecular allylmetalation reactions of aldehydes,
cf. Scheme 1.
by the action of boron trifluoride etherate to give mainly the
trans-disubstituted tetrahydropyran 3.12-15 This reaction has
been implemented into the synthesis of multifused heterocyclic
systems.16 In contrast, the report17 that 1 (M ) Bu3Sn) is
cyclized to the cis-disubstituted tetrahydropyran 2 by the action
of trifluoromethanesulfonic acid has not found further applica-
tion in synthesis. Our interest in intramolecular allylboration
reactions as a route to heterocyclic compounds18 led us to the
allylboronate 1 (M ) B(OR)2), which cyclized exclusively to
the cis-disubstituted tetrahydropyran 2.19 It was therefore
attractive to extend these studies to the synthesis of the
ill-famous oxocane and oxocene ring systems. Laurencin (4),20
a long known natural oxocene-derivative, was chosen as a target,
cf. Scheme 2.
Studies by the groups of Y. Yamamoto and J. D. Martin have
shown that the trialkyltin or trimethylsilyl derivatives 1 cyclize
X Abstract published in AdVance ACS Abstracts, July 15, 1997.
(1) Yasumoto, T.; Murata, M. Chem. ReV. 1993, 93, 1897.
(2) Lin, Y.-Y.; Risk, M.; Ray, S. M.; Van Engen, D.; Clardy, J.; Golik,
J.; James, J. C.; Nakanishi, K. J. Am. Chem. Soc. 1981, 103, 6773.
(3) Murata, M.; Legrand, A. M.; Ishibashi, Y.; Yasumoto, T. J. Am.
Chem. Soc. 1989, 111, 8929.
(4) (a) Uemura, D.; Takahashi, K.; Yamamoto, T.; Katayama, C.; Tanaka,
J.; Okumura, Y.; Hirata, Y. J. Am. Chem. Soc. 1985, 107, 4796. (b) Hirata,
Y.; Uemura, D. Pure Appl. Chem. 1986, 58, 707.
(5) (a) Sasaki, M.; Matsumori, N.; Maruyama, T.; Nonamura, T.; Murata,
M.; Tachibana, K.; Yasumoto, T. Angew. Chem. 1996, 108, 1782; Angew.
Chem., Int. Ed. Engl. 1996, 35, 1675. (b) Nonomura, T.; Sasaki, M.;
Matsumori, N.; Murata, M.; Tachibana, K.; Yasumoto, T. Angew. Chem.
1996, 108, 1786; Angew. Chem., Int. Ed. Engl. 1996, 35, 1672.
(6) (a) Murata, M.; Kumagai, M.; Lee, J. S.; Yasumoto, T. Tetrahedron
Lett. 1987, 28, 5869. (b) Takahashi, H.; Kusumi, T.; Kan, Y.; Satake, M.;
Yasumoto, T. Tetrahedron Lett. 1996, 37, 7087.
(7) Nicolaou, K. C. Angew. Chem. 1996, 108, 645; Angew. Chem., Int.
Ed. Engl. 1996, 35, 588.
(8) Alvarez, E.; Candenas, M.-L.; Pe´rez, R.; Ravelo, J. L.; Martin, J. D.
Chem. ReV. 1995, 95, 1953.
(9) (a) Nicolaou, K. C.; Prasad, C. V. C.; Somers, P. K.; Hwang, C.-K.
J. Am. Chem. Soc. 1989, 111, 5330. (b) Suzuki, T.; Sato, O.; Hirama, M.
Tetrahedron Lett. 1990, 31, 4747. (c) Mukai, C.; Ikeda, Y.; Sugimoto, Y.;
Hanaoka, M. Tetrahedron Lett. 1994, 35, 2179.
(10) (a) Cooper, A. J.; Salomon, R. G. Tetrahedron Lett. 1990, 31, 3813.
(b) Aicher, T. D.; Buszek, K. R.; Fang, F. G.; Forsyth, C. J.; Jung, S. H.;
Kishi, Y.; Scola, P. M. Tetrahedron Lett. 1992, 33, 1549. (c) Mart´ın, V.
S.; Palazo´n, J. M. Tetrahedron Lett. 1992, 33, 2399. (d) Gung, B. W.;
Francis, M. B. J. Org. Chem. 1993, 58, 6177.
The ethyl side chain in laurencin (4) can be derived from a
vinyl group; therefore, the cis-hydroxy compound 5 appears as
(12) Yamada, J.-i.; Asano, T.; Kadota, I.; Yamamoto, Y. J. Org. Chem.
1990, 55, 6066.
(13) Suzuki, T.; Sato, O.; Hirama, M.; Yamamoto, Y.; Murata, M.;
Yasumoto, T.; Harada, N. Tetrahedron Lett. 1991, 32, 4505.
(14) (a) Kadota, I.; Gevorgyan, V.; Yamada, J.-i.; Yamamoto, Y. Synlett
1991, 823. (b) Yamamoto, Y.; Yamada, J.-i.; Kadota, I. Tetrahedron Lett.
1991, 32, 7069.
(15) Ravelo, J. L.; Regueiro, A.; Martin, J. D. Tetrahedron Lett. 1992,
33, 3389.
(16) Alvarez, E.; D´ıaz, M. T.; Pe´rez, R.; Ravelo, J. L.; Regueiro, A.;
Vera, J. A.; Zurita, D.; Mart´ın, J. D. J. Org. Chem. 1994, 59, 2848.
(17) Gevorgyan, V.; Kadota, I.; Yamamoto, Y. Tetrahedron Lett. 1993,
34, 1313.
(18) Hoffmann, R. W.; Hense, A. Liebigs Ann. Chem. 1996, 1283.
(19) (a) Hoffmann, R. W.; Mu¨nster, I. Tetrahedron Lett. 1995, 36, 1431.
(b) Hoffmann, R. W.; Mu¨nster, I. Liebigs Ann. Chem. 1997, 1143.
(20) (a) Irie, T.; Suzuki, M.; Masamune, T. Tetrahedron Lett. 1965, 1091.
(b) Irie, T.; Suzuki, M.; Masamune, T. Tetrahedron 1968, 24, 4193. (c)
Cameron, A. F.; Cheung, K. K.; Ferguson, G.; Robertson, J. M. J. Chem.
Soc., Chem. Commun. 1965, 638. (d) Cameron, A. F.; Cheung, K. K.;
Ferguson, G.; Robertson, J. M. J. Chem. Soc. B 1969, 559.
(11) (a) Clark, J. S.; Kettle, J. G. Tetrahedron Lett. 1997, 38, 123. (b)
Clark, J. S.; Kettle, J. G. Tetrahedron Lett. 1997, 38, 127.
S0002-7863(97)00914-1 CCC: $14.00 © 1997 American Chemical Society

7500 J. Am. Chem. Soc., Vol. 119, No. 32, 1997
Kru¨ger and Hoffmann
Scheme 3
Scheme 5
Scheme 6
Scheme 4
aldehyde and the allylboronate, be liberated which then should
enter into a ring closure reaction.
a reasonable precursor because the substitution of a hydroxyl
group under inversion of configuration had already been realized
in previous syntheses of laurencin.21 The ∆5-oxocene 5 should
be the product of an intramolecular allylboration of 6. The point
to be clarified, though, is whether the resident stereogenic center
in 6 would induce ring closure to 5 with the desired relative
configuration.
Thus, allyloxyhexanal (12) was treated with lithium N-
methoxy-N-methyl amide to generate 13. Subsequent addition
of sec-butyllithium should metalate the allyl ether moiety.25 The
resulting organolithium compound was trapped by addition of
an trialkylborate. Finally, hydrolysis liberated both the allyl-
boronate and the aldehyde functions in 10. To our dismay, none
of the desired cyclization product 11 could be obtained from
this reaction sequence. We suspect that ring closure of 10 was
not able to compete with polymerization.
However, the reaction sequence delineated in Scheme 4 had
worked fine to generate the allylboronate 1 (M ) B(OR)2)19
which cyclized readily to 2. We were therefore confident that
the model system 15 with a Z-double bond should have a higher
predisposition for cyclization, cf. ref 16.
Nevertheless, projecting the aldehyde 15 as a cyclization
precursor entailed a further problem: The double bond in 15 is
â,γ- to the aldehyde function. We were therefore afraid that
treatment of the ∆3-analog of 12 with a lithium amide base
would lead to enolization rather than to masking of the aldehyde
function by addition. Thus, we decided to generate the aldehyde
function as well directly in a masked form. This we hoped to
attain by a DIBAH reduction of the Weinreb amide 20, cf.
Scheme 6.
Allylmetalation Reactions Leading to Oxocanes
The formation of an oxocane ring by an intramolecular
aldehyde/allylstannation reaction had been studied before. Both
the Yamamoto12 and the Martin16 group found that the simple
oxocane 8 can be obtained by an intramolecular allylstannation
reaction, albeit in low yield; cf. Scheme 3. They found that
cyclization yields could be raised to between 30 and 65%, when
substituents or a Z-double bond in the precursor molecule
induced a conformation favorable for cyclization. While these
routes led to the trans-disubstituted oxocane 8, we report here
our efforts to reach cis-disubstituted oxocanes 11 by cyclization
of the allylboronate 10.
A problem germane to all intramolecular allylmetalation
reactions is the necessity to generate an aldehyde function in
the presence of the acid-labile allylmetal moiety or vice versa.
In the case of the allyltin compound 1 this has been attained by
a mild oxidiation of a primary alcohol13 or a by oxidative
cleavage of a 1,2-diol.15 We envisioned an approach to 10, by
which the aldehyde function is temporarily masked during the
generation of the allylboronate function. This could be achieved
by an in situ protection of the type advocated by Comins.22 In
view of the high stability of the adducts23 of organolithium
compounds to Weinreb amides,24 we opted for a route to 10
described in Scheme 4. In this route the allylboronate moiety
is also generated in a masked form, as the ate complex 14. Only
upon aqueous workup will both reactive functionalities, the
Following an earlier approach by Corey26 the methoxycy-
clohexadiene 17 was used as a source of a Z-double bond in
the alcohol 18. The latter was allylated to give the ether 19.
Attempts to convert 19 directly into the Weinreb amide 2024,27
failed. We therefore took a more roundabout way to 20 via
saponification and formation of the acid chloride. The Weinreb
amide was then carried through in a one-pot reaction sequence
to the aldehyde 15: DIBAL reduction of the Weinreb amide to
generate the aluminum chelate 21, lithiation of the allyloxy
system with sec-butyllithium, borylation with a pinacol borate
ester and, finally, liberation of the reactive groups by aqueous
(21) (a) Masamune, T.; Matsue, H.; Murase, H. Bull. Chem. Soc. Jpn.
1979, 52, 127. (b) Masamune, T.; Murase, H.; Matsue, H.; Murai, A. Bull.
Chem. Soc. Jpn. 1979, 52, 135.
(25) (a) Evans, D. A.; Andrews, G. C.; Buckwalter, B. J. Am. Chem.
Soc. 1974, 96, 5560. (b) Still, W. C.; Macdonald, T. L. J. Am. Chem. Soc.
1974, 96, 5561.
(22) Comins, D. L. Synlett 1992, 615.
(23) (a) Evans, D. A.; Bender, S. L.; Morris, J. J. Am. Chem. Soc. 1988,
110, 2506. (b) Evans, D. A.; Polniaszek, R. P.; DeVries, K. M.; Guinn, D.
E.; Mathre, D. J. J. Am. Chem. Soc. 1991, 113, 7613.
(26) Corey, E. J.; Katzenellenbogen, L. A.; Gilman, N. W.; Roman, S.
A.; Erickson, B. W. J. Am. Chem. Soc. 1968, 90, 5618.
(27) Lewis, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982, 12,
989.
(24) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.

Substituted Oxocanes by Intramolecular Allylboration Reactions
J. Am. Chem. Soc., Vol. 119, No. 32, 1997 7501
Scheme 7
Scheme 8
pH 7 buffer solution. To our delight, this sequence resulted in
the formation of the oxocene 16 in 32% overall yield. While
this yield is still far from optimal, the result nevertheless shows
that the Z-double bond present in 15 raised the chances for the
intramolecular versus the intermolecular allylboration reaction.
The oxocene 16 was obtained as a single diastereomer. The
relative configuration of the two new stereogenic centers in 16
was not evident from the 1H-NMR spectra of the flexible
oxocene ring. We therefore converted the alcohol 16 into a
more rigid bicyclic derivative 23. This was effected by
hydroformylation of 16 to an anomeric mixture of the lactols
22, followed by oxidation to the lactone 23, cf. Scheme 7.
The two bridgehead protons in 23 showed a 2-3 Hz coupling.
Their disposition on the same side of the bicyclic framework
was also proven by a strong NOE contact between them.
The conversion of 20 into 16 set the stage for a direct access
to the oxocene ring system of laurencin. Even if the overall
yield of this transformation is only moderate, this will be
partially compensated by the high stereoselectivity and by the
low number of steps required to build the oxocene ring system.
Scheme 9
An Entry into an Efficient Laurencin Synthesis
Our route to ∆5-oxocenes seemed well suited to interface with
a recent synthesis of laurencin (4) carried out by the Holmes
group.28 In their synthesis the intermediate 24a has been
converted in nine operations into the enantiomerically pure (+)-
laurencin (4). We therefore wanted to reach the intermediate
24 by an intramolecular allylboration of the aldehyde 34. We
had, however, first to address the question, whether asymmetric
induction from the resident stereogenic center in 34 would lead
to the desired stereoisomer of 35. We therefore carried out
MM2-calculations on the transition states for the intramolecular
allylboration of 25, in which the methyl group serves as a model
for the side chain in laurencin. We used the force field
parameters developed by Gennari29 to model transition states
of the allylboronate aldehyde additions.
These calculations led us to expect an ca. 9:1 preference in
favor of 26, i.e., the relative configuration required for laurencin,
on the ring closure of 25. This was in keeping with the high
levels of asymmetric induction from resident stereocenters found
before30 in intramolecular allylboration reactions leading to six-
membered carbocycles.
tion to the alcohol followed by Dess-Martin oxidation (75%).
We intended to use a Wittig reaction with the known34
phosphorane 31 to arrive at the Z-double bond in 32. Using
the route described by Keinan35,36 the Wittig reaction furnished
the desired 32 in 94% yield with a Z/E ratio of 96:4. The
transformation of the OBO ester 32 into the Weinreb amide 33
was carried out as a one-pot sequence in 68% yield, involving
saponification to the carboxylic acid and DCC initiated forma-
tion of the amide.
At this point we were ready to launch the one-pot cyclization
to the oxocene 35. DIBAL reduction of the Weinreb amide
33, metalation with sec-butyllithium, borylation with the pinacol
borate ester and, finally, liberation of both the aldehyde and
the allylboronate function by aqueous pH 7 buffer solution
generated the reactive target 34 which cyclized in 38% yield
1
overall to the desired 35. NOE cross-peaks in the H-NMR
spectra between the hydrogens at C-2, C-3, and C-8 indicated
that all substituents at the oxocene ring were in a cis arrange-
ment. Since we observed no diastereomeric product, we
conclude that the cyclization proceeded under high asymmetric
induction from the resident stereogenic center present in 31.
At this point we had to address the selective hydrogenation
of 35 to 24b. High selectivities for hydrogenations of terminal
double bonds have been reported for ruthenium catalysts.37 We
had, however, no luck with these for the selective hydrogenation
of 35. After screening a variety of hydrogenation catalysts, we
The starting point of our synthesis was the alcohol 2831
derived from the ethyl ester 27 of unnatural malic acid.32
Attempts to allylate 28 under the usual basic conditions resulted
in partial migration of the silyl protecting group.33 Allylation
to 29 was, however, cleanly accomplished with allyl trichloro-
acetimidate and a catalytic amount of trifluoromethane-sulfonic
anhydride (91%). The ester group in 29 was now converted to
the aldehyde 30 by standard lithium aluminium hydride reduc-
(28) Holmes, A. B. Personal communication, 1996. Cf. the accompanying
paper.
(29) Gennari, C.; Fioravanzo, E.; Bernardi, A.; Vulpetti, A. Tetrahedron
1994, 50, 8815.
(30) Hoffmann, R. W.; Sander, T. Liebigs Ann. Chem. 1993, 1185.
(31) Saito, S.; Ishikawa, T.; Kuroda, A.; Koga, K.; Moriwake, T.
Tetrahedron 1992, 48, 4067.
(34) Corey, E. J.; Shimoji, K. J. Am. Chem. Soc. 1983, 105, 1662.
(35) Keinan, E.; Sinha, S. C.; Singh, S. P. Tetrahedron 1991, 47, 4631.
(36) The conversion of the bromoethyl compound into the iodo ethyl
compound worked better, when using sodium iodide, sodium bicarbonate
in butanone as solvent.
(32) Gao, Y.; Zepp, C. M. Tetrahedron Lett. 1991, 32, 3155.
(33) Mulzer, J.; Scho¨llhorn, B. Angew. Chem. 1990, 102, 433; Angew.
Chem., Int. Ed. Engl. 1990, 29, 431.
(37) (a) Rylander, M.; Berkowitz, P. N. J. Org. Chem. 1959, 24, 708. (b)
Tsuji, J.; Suzuki, H. Chem. Lett. 1977, 1083. (c) Hallman, P. S.; Evans, D.;
Osborn, J. A.; Wilkinson, G. J. Chem. Soc., Chem. Commun. 1967, 305.

7502 J. Am. Chem. Soc., Vol. 119, No. 32, 1997
Kru¨ger and Hoffmann
with 5 mL each of tert-butyl methyl ether. The combined organic
phases were washed with brine, dried (Na2SO4), and concentrated. The
residue (214 mg) was dissolved in anhydrous CH2Cl2 (8 mL). To this
solution (COCl)2 (240 mg, 1.89 mmol) and a catalytic amount of DMF
was added under ice-cooling. The mixture was maintained at 0 °C for
30 min and at room temperature for 15 min. The solvent and excess
of (COCl)2 were removed in vacuo, and the residue was diluted with
CH2Cl2 (5 mL). N,O-Dimethylhydroxylamine hydrochloride (136 mg,
1.39 mmol) and pyridine (224 µL, 2.77 mmol) were added under ice-
cooling. The mixture was stirred for 40 min at room temperature.
Aqueous HCl (1 mL, 2 M solution) was added, and the aqueous layer
was extracted twice with 3 mL each of CH2Cl2. The combined organic
layers were washed with saturated aqueous NaHCO3 and brine, dried
(Na2SO4), and concentrated. The resulting residue was purified by flash
chromatography (1:1 petroleum ether/tert-butyl methyl ether) to yield
20 (183 mg, 68%) as a pale yellow liquid. 1H-NMR (500 MHz, CDCl3)
δ 2.37 (dq, J ) 5.7 and 0.96 Hz, 2H), 3.17 (s, 3H), 3.23 (d, J ) 5.9
Hz, 2H), 3.45 (t, J ) 7.0 Hz, 2H), 3.68 (s, 3H), 3.97 (dt, J ) 1.4 and
5.6 Hz, 2H), 5.16 (dq, J ) 1.7 and 10.5 Hz, 1H), 5.25 (dq, J ) 1.6 and
17.3 Hz, 1H), 5.60-5.67 (m, 2H), 5.90 (ddt, J ) 5.6, 10.3 and 17.3
Hz, 1H); 13C-NMR (75 MHz, CDCl3) δ 28.4, 31.1, 32.3 (broad), 61.3,
69.6, 71.9, 116.7, 123.7, 128.8, 134.9. The resonance for the carbonyl-
C-atom of the amide was missing. Anal. Calcd for C11H19NO3: C,
61.95; H, 8.98. Found: C, 61.86; H 9.16.
Scheme 10
found that simple palladium on carbon differentiated well
enough between the two double bonds in 35 to realize a selective
hydrogenation to 24b in 97% yield. Careful monitoring of the
progress of the hydrogenation reaction was, however, necessary.
Finally, the TBDMS group was cleaved from 24b, and the
resulting diol was selectively reprotected with TBDPS chloride
to furnish 24a which showed 1H- and 13C-NMR spectra identical
to those kindly provided by Dr. A. B. Holmes.
(2R*,3R*)-3-Hydroxy-2-vinyl-3,4,7,8-tetrahydro-2H-oxocin (16).
To a solution of the Weinreb amide 22 (183 mg, 0.85 mmol) and
anhydrous THF (10 mL) was added DIBAH (0.82 mL, 1.02 mmol,
1.24 M solution in petroleum ether) dropwise via a syringe at -78 °C.
The reaction was maintained at -78 °C for 90 min. A solution of
s-BuLi (1.3 M in 92:8 cyclohexane/hexane, 1.44 mL, 1.87 mmol,
purchased from Acros) was added dropwise via a syringe, and the
resulting bright yellow solution was stirred for 60 min at -78 °C. A
solution of 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (380
mg, 2.04 mmol) in THF (2 mL) was added via a syringe within 5 min.
The cooling bath was removed, and the reaction mixture was allowed
to reach room temperature. It was diluted with petroleum ether (10
mL), and pH 7 phosphate buffer (10 mL) was added under vigorous
stirring. The mixture was maintained at ambient temperature for 3 h
and then extracted three times with 10 mL each of ether. The combined
organic layer was washed with brine (10 mL), dried (Na2SO4), and
concentrated. The resulting residue was purified by flash chromatog-
raphy (1:1 petroleum ether/ether) to afford the ∆5-oxocene 16 (41.0
mg, 32%) as a single diastereomer. 1H-NMR (500 MHz, CDCl3) δ
1.94 (s, broad, 1H), 2.16 (m, 1H), 2.34 (m, 1H), 2.43-2.63 (m, 2H),
3.42 (m, 1H), 3.76 (s, broad, 1H), 3.96 (ddd, J ) 3.7, 6.0 and 12.0 Hz,
1H), 4.02 (m, 1H), 5.21 (dq, J ) 2.0 and 10.6 Hz, 1H), 5.34 (dq, J )
1.8 and 17.3 Hz, 1H), 5.76-5.91 (m, 3H); 13C-NMR (75 MHz, CDCl3)
δ 29.4, 32.7, 70.4, 74.4, 81.8, 115.6, 128.8, 130.6, 136.9. Anal. Calcd
for C9H14O2: C, 70.10; H 9.15. Found: C, 70.34; H 9.35.
If our route to the oxocene 24b (16% overall yield from 28)
is implemented into the Holmes synthesis of laurencin, this
amounts to a 18-step synthesis with an overall yield of 3%.
Comparison with the other approaches to laurencin21,38,39 makes
it clear that direct ring closure reactions, as the one developed
here and the one utilized by Overman,39 characterize the most
efficient ways to laurencin.
Apart from opening an efficient route to laurencin, the
heterocyclization reaction by intramolecular allylboration of an
aldehyde function presented here has the novel feature that both
the reactive aldehyde and the allylboronate moiety are initially
generated in a masked form in situ and are liberated simulta-
neously by spontaneous hydrolysis of the precursor functions.
Experimental Section
All temperatures quoted are not corrected: 1H-NMR, 13C-NMR:
Bruker AMX-500; Bruker AC-300; boiling range of petroleum ether:
40-60 °C; pH 7 buffer: 56.2 g NaH2PO4 × 2 H2O + 213.2 g Na2-
HPO4 × 2 H2O in 1.0 L of water; flash chromatography: silica gel Si
60 E. Merck AG, Darmstadt, 40-63 µm.
Methyl (3Z)-6-(2-Propenyloxy)-3-hexenoate (19). A solution of
methyl (3Z)-6-hydroxy-3-hexenoate26 18 (0.64 g, 4.4 mmol), Ag2O (1.0
g, 4.4 mmol), allyl bromide (0.87 mL, 10 mmol), and 20 mL anhydrous
diethyl ether was refluxed for 8 h under exclusion of light in an nitrogen
atmosphere. The reaction was maintained at room temperature over
night. The solvent and excess of allyl bromide were removed in vacuo.
The resulting residue was purified by flash chromatography (10:1
petroleum ether/tert-butyl methyl ether) to afford 19 (0.40 g, 49%) as
a colorless liquid. 1H-NMR (300 MHz, CDCl3) δ 2.31 (m, 2H), 3.10
(d, J ) 5.5 Hz, 2H), 3.43 (t, J ) 6.8 Hz, 2H), 3.66 (s, 3H), 3.95 (dt,
J ) 5.6 and 1.4 Hz, 2H), 5.14 (dq, J ) 10.3 and 1.5 Hz, 1H), 5.24 (dq,
J ) 17.1 and 1.7 Hz, 1H), 5.62 (m, 2H), 5.88 (ddt, J ) 10.5, 17.2 and
5.6 Hz, 1H); 13C-NMR (50 MHz, CDCl3) δ 28.1, 32.8, 51.8, 69.3, 71.8,
116.8, 122.7, 129.3, 134.7, 172.2. Anal. Calcd for C10H16O3: C, 65.19;
H, 8.75. Found: C, 65.00; H 8.51.
(1R*,8R*)-2,9-Dioxabicyclo[6.4.0]dodec-5-en-10-one (23). Triphen-
ylphosphine (0.10 g, 0.39 mmol, 100 equiv) was added to a solution
of Rh(CO)2acac (1.0 mg, 1.9 mol%) in anhydrous benzene (5 mL),
and the mixture was stirred for 10 min. Homoallylic alcohol 16 (33
mg, 0.21 mmol) was added, and the reaction mixture was transferred
into an autoclave through a cannula. An atmosphere of CO (10 bar)
and H2 (10 bar) was established, and the mixture was stirred at 70 °C
for 2 h. The reaction vessel was flushed with nitrogen, and the solvent
was removed in vacuo. The residue was purified by flash chromatog-
raphy (1:1 petroleum ether/tert-butyl methyl ether) to yield 28 mg of
the lactol 22 as an anomeric mixture. The product was dissolved in
anhydrous CH2Cl2 (5 mL) and pyridine (0.12 mL, 1.5 mmol), and
Dess-Martin periodinane (76 mg, 0.18 mmol) was added under ice-
cooling. The mixture was maintained at 0 °C for 30 min. After TLC
indicated complete oxidation, the mixture was concentrated, and the
residue was purified by flash chromatography (1:2 petroleum ether/
ether) to afford the desired lactone 23 (16 mg, 60%): 1H-NMR (300
MHz, CDCl3) δ 1.86-2.11 (m, 3H), 2.36-2.47 (m, 2H), 2.55-2.84
(m, 3H), 3.34 (dt, J ) 12.0 and 1.3 Hz, 1H), 3.76 (dt, J ) 3.3 and 3.6
Hz, 1H), 4.10 (dt, J ) 3.5 and 12.2 Hz, 1H), 4.36 (ddd, J ) 2.7, 4.9
and 11.5 Hz, 1H), 5.69 (m, 1H), 5.92 (m, 1H); 13C-NMR (75 MHz,
CDCl3) δ 25.1, 25.8, 29.7, 30.3, 72.2, 72.7, 83.0, 126.8, 132.0, 170.8;
N-Methoxy-N-methyl-(3Z)-6-(2-propenyloxy)-3-hexenoicamide (20).
To a solution of the ester 19 (234 mg, 1.27 mmol), DME (5 mL), and
water (3 mL) was added 2 mL of an aqueous LiOH solution (5 M) at
room temperature. It was stirred for 10 min and then carefully brought
to pH 2-3 with 2 M aqueous HCl. The reaction mixture was diluted
with ether (10 mL), and the aqueous layer was extracted three times
(38) (a) Tsushima, K.; Murai, A. Tetrahedron Lett. 1992, 33, 4345. (b)
Mujica, M.; Afonso, M. M.; Galindo, A.; Palenzuela, J. A. Synlett 1996,
983.
(39) Bratz, M.; Bullock, W. H.; Overman, L. E.; Takemoto, T. J. Am.
Chem. Soc. 1995, 117, 5958.

Substituted Oxocanes by Intramolecular Allylboration Reactions
J. Am. Chem. Soc., Vol. 119, No. 32, 1997 7503
HRMS (EI) m/z 182.0938 (182.0943 calcd for C10H14O3) m/z 183.0979
(183.0976 calcd for 13C1C9H14O3).
1H-NMR (300 MHz, CDCl3) δ 0.03 (s, 6H), 0.78 (s, 3H), 0.87 (s, 9H),
2.25 (m, 2H), 2.46 (m, 2H), 3.39 (m, 1H), 3.57 (m, 2H), 3.88 (s, 6H),
4.06 (m, 2H), 5.11 (qd, J ) 1.8 Hz, and 10.5 Hz, 1H), 5.23 (qd, J )
1.8 Hz and 17.3 Hz, 1H), 5.57 (m, 2H), 5.88 (ddt, J ) 10.3, 17.2 and
5.6 Hz, 1H); 13C-NMR (75 MHz, CDCl3) δ -5.4, -5.3, 14.5, 18.3,
25.9 (3 C), 29.5, 30.4, 35.0 65.3, 71.1, 72.6 (3 C), 79.7, 108.8, 116.2,
123.2, 128.1, 135.6. Anal. Calcd for C21H38O5Si: C, 63.28; H 9.61.
Found: C, 63.01; H 9.81.
Ethyl 4-[(tert-Butyldimethylsilyl)oxy]-(3R)-(2-propenyloxy)bu-
tanoate (29). Four drops of trifluoromethane sulfonic anhydride were
added to a solution of ethyl 4-[(tert-butyldimethylsilyl)oxy]-(3R)-2-
hydroxybutanoate (28)31 (3.22 g, 12.3 mmol) and allyl trichloroace-
timidate (4.05 g, 20.0 mmol) in petroleum ether (25 mL), and the
mixture was stirred for 3 days at room temperature. The precipitated
trichloroacetamide was filtered off. The filtrate was washed with
saturated aqueous NaHCO3 and brine, dried with Na2SO4, and
concentrated. The resulting residue was purified by flash chromatog-
raphy (10:1 petroleum ether/tert-butyl methyl ether) to yield 3.39 g
(91%) of the allylated alcohol 39 as a colorless liquid. [R]25D +17.1°
(c 0.52, CH2Cl2); 1H-NMR (300 MHz, CDCl3) δ 0.04 (s, 6H), 0.87 (s,
9H), 1.25 (t, J ) 7.2 Hz, 3H), 2.45 (dd, J ) 7.9 and 15.6 Hz, 1H),
2.58 (dd, J ) 4.8 and 15.6 Hz, 1H), 3.54 (dd, J ) 5.9 and 10.3 Hz,
1H), 3.69 (dd, J ) 5.3 and 10.4 Hz, 1H), 3.85 (m, 1H), 4.06 (m,
overlayed by the adjacent quartett, 2H), 4.13 (q, J ) 7.2 Hz, 2H), 5.13
(qd, J ) 1.4 and 10.3 Hz, 1H), 5.24 (qd, J ) 1.7 and17.2 Hz, 1H),
5.87 (ddt, J ) 10.3, 17.3 and 5.6 Hz, 1H); 13C-NMR (75 MHz, CDCl3)
δ -5.4 (2 C), 14.2, 18.3, 25.9 (3 C), 37.6, 60.4, 64.8, 71.5, 76.6, 116.7,
135.1, 171.7. Anal. Calcd for C15H30O4Si: C, 59.56; H 10.00.
Found: C, 59.40; H 10.22.
4-[(tert-Butyldimethylsilyl)oxy]-(3R)-(2-propenyloxy)butanal (30).
A solution of the ester 29 (0.30 g, 1.0 mmol) in THF (1 mL) was
added dropwise via a syringe at 0 °C to a suspension of lithium
aluminium hydride (29 mg, 0.75 mmol) in THF (5 mL). TLC indicated
the reaction to be complete after 10 min. Saturated aqueous K2CO3
solution was added dropwise to the reaction mixture under ice-cooling
until there was no more evolution of gas. The mixture was stirred at
room temperature for 15 min, dried (Na2SO4), filtered, and concentrated.
The residue was purified by flash chromatography (2:1 petroleum ether/
N-Methoxy-N-methyl-7-[(tert-butyldimethylsilyl)oxy]-(6R)-(2-pro-
penyloxy)-(3Z)-heptenoic-amide (33). To a solution of the OBO ester
32 (0.70 g, 1.76 mmol) in DME (5 mL) was added under ice-cooling
5 mL of an aqueous NaHSO4 solution which was adjusted to pH 3-4.
The cooling bath was removed, and the mixture was stirred until TLC
indcated complete comsumption of the starting material (approximately
3 h). The aqueous layer was extracted four times with 5 mL each of
ethyl acetate. The combined organic layer was washed with brine and
concentrated. The residue was dissolved in DME (10 mL) and cooled
to 0 °C followed by addition of 5 mL of an aqueous LiOH (2 M)
solution. The mixture was maintained at room temperature for 1 h
and the brought to pH 5-6 by dropwise addition of aqueous HCl (1
M). The aqueous layer was extracted four times with 10 mL each of
tert-butyl methyl ether. The combined organic layer was washed with
brine, dried with Na2SO4, and concentrated to yield 0.56 g (1.76 mmol)
of the crude acid. This was taken up in anhydrous CH2Cl2 (10 mL),
and N,O-dimethylhydroxylamine hydrochloride (0.17 g, 1.77 mmol),
N-methylpiperidine (0.18 g, 1.77 mmol), DMAP (22 mg, 0.18 mmol),
and DCC (0.37 g 1.77 mmol) were added. The mixture was maintained
at room temperature over night, filtrated over zeolite, and concentrated.
The resulting residue was purified by flash chromatography (1:1
petroleum ether/tert-butyl methyl ether) to afford 0.43 g (68%) of the
desired Weinreb amide 33 as a pale yellow liquid. [R]20 -0.57° (c
D
1
1.25, CH2Cl2); H-NMR (300 MHz, CDCl3) δ 0.04 (s, 6H), 0.87 (s,
9H), 2.29 (m, 2H), 3.16 (s, 3H), 3.26 (m, 2H), 3.43 (m, 1H), 3.54 (dd,
J ) 5.2 and 10.6 Hz, 1H), 3.62 (dd, J ) 5.7 and 10.6 Hz, 1H), 3.67 (s,
3H), 4.07 (m, 2H), 5.12 (qd, J ) 1.8 and 10.3 Hz, 1H), 5.24 (qd, Jq )
1.6 and 17.3 Hz, 1H), 5.67 (m, 2H), 5.88 (ddt, J ) 10.4, 17.2 and 5.6
Hz, 1H); 13C-NMR (75 MHz, CDCl3) δ -5.4 (2 C), 18.3, 25.9 (3 C),
29.8, 31.1 (broad), 32.3 (very broad), 61.2, 65.0, 71.2, 79.5, 116.4,
123.7, 128.5, 135.4. The resonance for the carbonyl-C-atom of the
amide was missing. Anal. Calcd for C18H35NO4Si: C, 60.46; H, 9.87.
Found: C, 60.44; H 9.81.
tert-butyl methyl ether) to yield the alcohol (0.21 g, 80%) as a colorless
1
oil. [R]20 +29.5° (c 0.66, CH2Cl2); H-NMR (300 MHz, CDCl3) δ
D
0.06 (s, 6H), 0.88 (s, 9H), 1.78 (m, 2H), 2.67 (t, J ) 5.1 Hz, 1H), 3.58
(m, 2H), 3.67-3.79 (m, 3H), 4.04 (ddt, J ) 1.4, 4.5, and 12.6 Hz,
1H), 4.18 (ddt, J ) 1.4, 5.5, and 12.6 Hz, 1H), 5.16 (qd, J ) 1.3 and
10.3 Hz, 1H), 5.23 (qd, J ) 1.7 and 17.2 Hz, 1H), 5.91 (ddt, J ) 10.4,
17.2, and 5.7 Hz, 1H); 13C-NMR (75 MHz, CDCl3) δ -5.4 (2 C), 18.3,
25.9 (3 C), 34.4, 60.4, 65.3, 71.2, 79.1, 117.0, 135.0. Anal. Calcd for
C13H28O3Si: C, 59.95; H 10.84. Found: C, 59.80; H 10.91.
(8R)-[(tert-Butyldimethylsilyl)oxy]methyl-(3R)-hydroxy-(2R)-vinyl-
3,4,7,8-tetrahydro-2H-oxocin (35). DIBAH (1.12 mL, 0.89 M in
petroleum ether) was added via a syringe over 15 min at -78 °C to a
solution of the Weinreb amide 33 (357 mg, 1.00 mmol) in THF (10
mL). The mixture was maintained at -78 °C for 90 min. Addition of
1.62 mL s-BuLi (2.10 mmol, 1.3 M in cyclohexane/hexane 92:8) via
syringe produced a yellow color. The mixture was stirred for 30 min.
A solution of 2-methoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (347
mg, 2.20 mmol) in THF (2 mL) was added, and the cooling bath was
removed. After warming to room temperature the mixture was poured
into a vigorous strirred bilayer system of ether (20 mL) and pH 7 buffer
(20 mL). It was stirred over night, and the aqueous layer was extracted
three times with 10 mL each of ether. The combined organic layer
was washed with brine, dried (Na2SO4), and concentrated. The residue
was purified by flash chromatography (5:1 petroleum ether/ether) to
yield 112 mg (38%) of the desired product as a single diastereomer.
[R]20D -38.4° (c 0.5, CH2Cl2); 1H-NMR (500 MHz, CDCl3) δ 0.07 (s,
6H), 0.89 (s, 9H), 1.77 (d, J ) 7.6 Hz, OH), 2.29-2.39 (m, 3H), 2.60 (m,
1H), 3.44 (m, 1H), 3.49 (m, 1H), 3.74 (dd, J ) 5.5 and 10.0 Hz, 1H),
3.77 (m, broad, 1H), 4.16 (m, sharp, 1H), 5.23 (dt, J ) 2.0 and 10.6
Hz, 1H), 5.46 (dt, J ) 2.0 and 17.2 Hz, 1H), 5.76-5.88 (m, 3H); 13C-
NMR (75 MHz, CDCl3) δ -5.3 (2 C), 18.3, 25.9 (3 C), 30.7, 32.8,
65.9, 74.5, 81.3, 81.5, 115.9, 129.3, 129.9, 136.7. Anal. Calcd for
C16H30O3Si: C, 64.38; H, 10.13. Found: C, 64.24; H 10.16.
Dess-Martin periodinane (0.59 g, 1.4 mmol) was added to a solution
of the alcohol (0.26 g, 1.0 mmol) in anhydrous CH2Cl2 (10 mL),
pyridine (0.79 mL, 10 mmol), and approximately 100 mg of powdered
molecular sieves at 0 °C. It was stirred at room temperature for 2 h,
filtered, and concentrated. The residue was purified by flash chroma-
tography (5:1 petroleum ether/tert-butyl methyl ether) to yield 0.24 g
(95%) of the desired aldehyde 30 as a colorless liquid. [R]20D +17.4°
(c 0.80, CH2Cl2); 1H-NMR (300 MHz, CDCl3) δ 0.04 (s, 6H), 0.87 (s,
9H), 2.61 (m, 2H), 3.57 (dd, J ) 6.1 and 10.4 Hz, 1H), 3.73 (dd, J )
4.9 and 10.4 Hz, 1H), 3.91 (q, J ) 5.9 Hz, 1H), 4.07 (m, 2H), 5.16
(qd, J ) 1.3 and 10.3 Hz, 1H), 5.24 (qd, J ) 1.6 and 17.2 Hz, 1H),
5.87 (ddt, J ) 10.4, 17.2 and 5.6 Hz, 1H), 9.80 (t, J ) 2.2 Hz, 1H);
13C-NMR (75 MHz, CDCl3) δ -5.3 (2 C), 18.2, 25.8 (3 C), 46.3, 64.4,
71.1, 75.0, 117.0, 134.8, 201.0. Anal. Calcd for C13H26O3Si: C, 60.42;
H 10.14. Found: C, 60.50; H 10.28.
1-[6-[(tert-Butyldimethylsilyl)oxy]-(5R)-(2-propenyloxy)-(2Z)-hex-
enyl]-4-methyl-2,6,7-trioxabicyclo[2.2.2]octane (32). NaHMDS (0.7
mL of a 2 M solution in THF) was added to a stirred suspension of
2-(4-methyl-2,6,7-trioxabicyclo[2.2.2]octyl)ethyltriphenylphos-
phonium iodide34,35,36 (0.82 g, 1.5 mmol) in anhydrous THF (15 mL)
via a syringe at room temperature. The reaction mixture was stirred
at 45 °C for 3 h and then cooled to -78 °C. A solution of aldehyde
30 (0.22 g, 0.87 mmol) in THF (2 mL) was added dropwise, and the
mixture was allowed to warm to room temperature over night. It was
diluted with ether (10 mL) and quenched with water (10 mL). The
aqueous layer was extracted three times with 10 mL each of tert-butyl
methyl ether. The combined organic layer was washed with brine,
dried (Na2SO4), and concentrated. The resulting residue was purified
by flash chromatography (5:1 petroleum ether/tert-butyl methyl ether
containing 0.5% triethylamine) to yield the alkene 32 (0.33 g, 94%) as
(8R)-[(tert-Butyldimethylsilyl)oxy]methyl-(3R)-hydroxy-(2R)-ethyl-
3,4,7,8-tetrahydro-2H-oxocin (24b). A solution of the diene 35 (43.0
mg, 144 µmol), anhydrous methanol (3 mL), and a catalytic amount
of Pd/C (purchased from Fluka) was stirred in an H2 atmosphere at
1
room temperature. The reaction was monitored by H-NMR spectro-
scopy. After 30 min all olefinic signals for the vinyl group had
dissappeared. The reaction mixture was filtered over zeolite and
a 96:4 mixture of Z/E diastereomers. [R]20 -1.6° (c 0.75, CH2Cl2);
D

7504 J. Am. Chem. Soc., Vol. 119, No. 32, 1997
Kru¨ger and Hoffmann
concentrated resulting in 42.0 mg (97%) of the desired oxocene 24b
Cl2. The combined organic layer was dried with Na2SO4 and
concentrated. The resulting residue was purified by flash chromatog-
1
as a colorless liquid. [R]20 -11.3° (c 0.39, CH2Cl2); H-NMR (300
D
MHz, CDCl3) δ 0.05 (2 sharp s, 2 × 3H), 0.89 (s, 9H), 0.94 (t, J ) 7.4
Hz, 3H), 1.35-1.49 (m, 1H), 1.56-1.74 (m, 1H), 1.81 (m, broad, 1H),
2.29 (m, 3H), 2.54 (dt, J ) 12.5 and 9.0 Hz, 1H), 3.46 (m, 3H), 3.69
(m, 2H), 5.69-5.89 (m, 2H); 13C-NMR (75 MHz, CDCl3) δ -5.4, -5.3,
10.5, 18.3, 25.9 (3 C), 30.3, 33.5, 65.6, 74.3, 81.4, 82.4, 129.0, 129.5.
Anal. Calcd for C16H32O3Si: C, 63.95; H, 10.73. Found: C, 63.78;
H 10.75.
(8R)-(tert-Butyldiphenylsilyloxy)methyl-(3R)-hydroxy-(2R)-ethyl-
3,4,7,8-tetrahydro-2H-oxocin (24a). To a solution of the TBS ether
25b (9.0 mg, 33 µmol) and THF (1 mL) was added TBAF (70 µL of
a 1 M solution in THF) under ice-cooling. The mixture was stirred at
room temperature for 1 h and concentrated, and the residue was purified
by flash chromatography using ether as the eluent to yield 6.1 mg of
the diol which was not further characterized. Under a nitrogen
atmosphere tert-butyldiphenylsilyl chloride (31.5 µL, 0.12 mmol) and
imidazole (8.2 mg, 0.12 mmol) were added under ice-cooling to a
solution of the diol (21 mg, 0.11 mmol) in CH2Cl2 (3 mL). The mixture
was maintained at 0 °C for 2 h and then quenched with water (2 mL).
The aqueous layer was extracted three times with 3 mL each of CH2-
raphy (3:1 petroleum ether/ether) to afford 34 mg (67%) of the TBDPS
1
ether 24a as a colorless oil: [R]20 +0.6° (c 1.7, CHCl3); H-NMR
D
(300 MHz, CDCl3) δ 0.93 (t, J ) 7.4 Hz, 3H), 1.13 (s, 9H), 1.46 (m,
1H), 1.69 (m, 1H), 1.80 (broad, 1H), 2.37 (m, 3H), 2.58 (m, 1H), 3.46
(m, 1H), 3.50-3.62 (m, 2H), 3.71 (m, 1H), 3.79 (m, 1H), 5.86-5.88
(m, 2H), 7.46 (m, 6H), 7.75 (m, 4H); 13C-NMR (75 MHz, CDCl3) δ
10.3, 19.0, 25.7, 26.7 (3 C), 30.2, 33.3, 66.2, 74.1, 80.9, 82.2, 127.4 (4
C), 128.9, 129.2, 129.4 (2 C), 133.5 (2 C), 135.4 (4 C). The
spectroscopic data for 24a are identical to those kindly provided by
Dr. A. B. Holmes (Cambridge, UK).
Acknowledgment. We would like to thank the Deutsche
Forschungsgemeinschaft (SFB 260) and the European Com-
munity, Grant ERBCHRXCT940620 as well as the Fonds der
Chemischen Industrie for support of this study. We thank
Professor A. B. Holmes, Cambridge, for the exchange of
information and spectral data.
JA970914U