Selective lithiation of 2-methyloxazoles. Applications to pivotal bond constructions in the phorboxazole nucleus

Article abstract of DOI:10.1021/ol990027r

(equation presented) R = Alkyl, Vinyl, Aryl 6:7 >95:5 The lithiation of 2-methyloxazoles with alkyllithium and hindered lithium amide bases generally results in the competitive formation of a mixture of 5-lithio-and 2-(lithiomethyl)oxazole isomers. Herein a synthetically useful lithiation method which allows for the selective formation of 2-(lithiomethyl)oxazole is described. Diethylamine has been found to be a kinetically competent proton source that will mediate the equilibration of the kinetically formed 5-lithiooxazole to its more stable 2-(lithiomethyl)oxazole counterpart. Application of this metalation strategy with lithium diethylamide to two important bond constructions relevant to a projected phorboxazole synthesis is presented.

Full text of DOI:10.1021/ol990027r

ORGANIC
LETTERS
1999
Vol. 1, No. 1
87-90
Selective Lithiation of 2-Methyloxazoles.
Applications to Pivotal Bond
Constructions in the Phorboxazole
Nucleus
David A. Evans,* Victor J. Cee, Thomas E. Smith, and Keith J. Santiago
Department of Chemistry and Chemical Biology, HarVard UniVersity,
Cambridge, Massachusetts 02138
eVans@chemistry.harVard.edu
Received April 9, 1999
ABSTRACT
The lithiation of 2-methyloxazoles with alkyllithium and hindered lithium amide bases generally results in the competitive formation of a
mixture of 5-lithio- and 2-(lithiomethyl)oxazole isomers. Herein a synthetically useful lithiation method which allows for the selective formation
of 2-(lithiomethyl)oxazole is described. Diethylamine has been found to be a kinetically competent proton source that will mediate the equilibration
of the kinetically formed 5-lithiooxazole to its more stable 2-(lithiomethyl)oxazole counterpart. Application of this metalation strategy with
lithium diethylamide to two important bond constructions relevant to a projected phorboxazole synthesis is presented.
The discovery of marine natural products containing 2,4-
disubstituted oxazoles such as hennoxazole A,1 theonezolide
A,2 the virginiamycins,3 the ulapaulides,4 and the phorbox-
azoles5 has renewed interest in the chemistry of the oxazole
nucleus. Our interest in the synthesis of the phorboxazoles
(Figure 1) has led us to consider methods for the construction
(1) Ichiba, T.; Yoshida, W. Y.; Scheuer, P. J.; Higa, T.; Gravalos, D. G.
J. Am. Chem. Soc. 1991, 113, 3173-3174.
(2) Kobayashi, J.; Kondo, K.; Ishibashi, M.; Walchli, M. R.; Nakamura,
T. J. Am. Chem. Soc. 1993, 115, 6661-6665.
(3) (a) Paris, J. M.; Barriere, J. C.; Smith, C.; Bost, P. E. In Recent
Progress in the Chemical Synthesis of Antibiotics; Lukacs, G., Ohno, M.,
Eds.; Springer-Verlag: Berlin, 1990; pp 183-248. (b) Cocito, C. Microbiol.
ReV. 1979, 43, 145-198 and references therein.
(4) Roesener, J. A.; Scheuer, P. J. J. Am. Chem. Soc. 1986, 108, 846-
847.
(5) Isolation and structure: (a) Searle, P. A.; Molinski, T. F. J. Am. Chem.
Soc. 1995, 117, 8126-8131. Synthetic studies: (b) Williams, D. R.; Clark,
M. P. Tetrahedron Lett. 1999, 40, 2291-2294. (c) Williams, D. R.; Clark,
M. P.; Berliner, M. A. Tetrahedron Lett. 1999, 40, 2287-2290. (d) Wolbers,
P.; Hoffmann, H. M. R. Tetrahedron 1999, 55, 1905-1914. (e) Paterson,
I.; Arnott, E. A. Tetrahedron Lett. 1998, 39, 7185-7188. (f) Pattenden,
G.; Plowright, A. T.; Tornos, J. A.; Ye, T. Tetrahedron Lett. 1998, 39,
6099-6102. (g) Ye, T.; Pattenden, G. Tetrahedron Lett. 1998, 39, 319-
322. (h) Ahmed, F.; Forsyth, C. J. Tetrahedron Lett. 1998, 39, 183-186.
(i) Cink, R. D.; Forsyth, C. J. J. Org. Chem. 1997, 62, 5672-5673. (j)
Lee, C. S.; Forsyth, C. J. Tetrahedron Lett. 1996, 37, 6449-6452. Total
synthesis of phorboxazole A: (k) Forsyth, C. J.; Ahmed, F.; Cink, R. D.;
Lee, C. S. J. Am. Chem. Soc. 1998, 120, 5597-5598.
Figure 1. Phorboxazoles A and B.
of the C19-C20 and C32-C33 bonds. An attractive strategy
for the formation of such bonds, outlined in Scheme 1, is
the selective generation of a 2-(lithiomethyl)oxazole6 (2) and
its reaction with an electrophile (eq 1). This approach has
(6) The carbanion of 2 may be delocalized into the CdN π system.
10.1021/ol990027r CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/17/1999

at -78 °C, and the resulting anions were quenched with
methyl triflate to provide the 2-ethyloxazole 6 and 2,5-
dimethyloxazole 7. While n-butyllithium and lithium diiso-
propylamide (LDA) either are selective for formation of the
5-methylated product 7 (substrate 1a) or are relatively
nonselective (substrates 1b-d), lithium diethylamide displays
remarkable selectiVity for the formation of the desired
product 6 in all cases. This effect is not limited to oxazoles.
In the case of thiazole 1e, n-butyllithium and LDA are
selective for the 5-methylated product 7e, consistent with
the findings of Meyers and Knaus,12 while lithium diethyl-
amide is highly selective for the formation of 6e.
Scheme 1
II. Synthetic Utility. With selective lithiation now pos-
sible, we investigated its utility for the elaboration of 2-meth-
yloxazoles into naturally occurring oxazole motifs. The (E)-
2-vinyloxazole of the phorboxazoles inspired the model study
depicted in Scheme 2, aimed at the generation of the C19-
proven difficult due to competitive formation of the 5-lithio-
oxazole 3, as illustrated by the work of Hamana and
Sugasawa (eq 2).7,8 Only a handful of successful examples
exist, including the alkylation reported by Whitney and
Rickborn (eq 3).9 However, the factors responsible for this
selectivity and its divergence from the Hamana result are
poorly understood.10
Scheme 2
I. Selective Alkylation. We felt that the lack of a general
method for the selective elaboration of 2-methyloxazoles
warranted further investigation. Our studies began with a
survey of the regioselectivity afforded by different bases in
an alkylation reaction (Table 1).11 The oxazoles were lithiated
C20 bond. Previous approaches to similar bonds have
employed activated 2-(phosphonomethyl)oxazoles;13 how-
ever, the use of an unfunctionalized 2-methyloxazole has
obvious advantages. To test this strategy, oxazole 1c was
treated with lithium diethylamide followed by hydrocinna-
maldehyde to give 8 as a single regioisomer in 73% yield.
The alcohol was then dehydrated with the Martin sulfurane14
to give 9 in quantitative yield and 95:5 E/Z selectivity. This
approach provides a useful alternative for the construction
of (E)-2-vinyloxazoles.
Table 1. Survey of Methods for the Selective Alkylation of Oxazoles
and Thiazolesa
The construction of the masked â-ketooxazole phorbox-
azole subunit is also amenable to this strategy. We envisioned
(7) (a) Hamana, H.; Sugasawa, T. Chem. Lett. 1983, 333-336. For related
examples see: (b) Williams, D. R.; Brooks, D. A.; Meyer, K. G.; Pagel,
M. Tetrahedron Lett. 1998, 39, 8023-8026. (c) Meyers, A. I.; Lawson, J.
P. Tetrahedron Lett. 1981, 22, 3163-3166. For a review of oxazole
metalation see: (d) Iddon, B. Heterocycles 1994, 37, 1321-1346.
(8) For approaches developed in response to this difficulty see refs 7a,c
and: (a) Gangloff, A. R.; Akermark, B.; Helquist, P. J. Org. Chem. 1992,
57, 4797-4799. (b) Wood, R. D.; Ganem, B. Tetrahedron Lett. 1983, 24,
4391-4392. (c) Nagao, Y.; Yamada, S.; Fujita, E. Tetrahedron Lett. 1983,
24, 2287-2290.
(9) (a) Whitney, S. E.; Rickborn, B. J. Org. Chem. 1991, 56, 3058-
3063. See also: (b) Entwistle, D. A.; Jordan, S. I.; Montgomery, J.;
Pattenden, G. Synthesis 1998, 603-612. (c) Garey, D.; Ramirez, M.;
Gonzales, S.; Wertsching, A.; Tith, S.; Keefe, K.; Pena, M. R. J. Org. Chem.
1996, 61, 4853-4856. (d) Bowie, J. H.; Donaghue, P. F.; Rodda, H. J.;
Cooks, R. G.; Williams, D. H. Org. Mass Spectrom. 1968, 1, 13-29.
(10) Dilithiated 3-methylthiophene-2-carboxylic acid has been reported
to give a different distribution of regioisomers with alkyl bromides than
with other electrophiles (acetone, methyl iodide, and deuterium oxide):
Gould, N. P.; Lee, T.-J. J. Org. Chem. 1980, 45, 4530-4532.
a All reactions were carried out in THF and proceeded to g90%
conversion unless otherwise noted. b Product ratios were determined by 1H
NMR unless otherwise noted. c One equivalent of n-BuLi was used, unless
otherwise noted. d Product ratios were determined by GC. e Two equivalents
of n-BuLi and 2.4 equiv of the lithium amides were used. f Multiple products
were observed. g Reaction proceeded to 55% conversion. h The thiazole and
base were warmed to -50 °C for 30 min and then cooled to -78 °C prior
to the addition of methyl triflate. i Reaction proceeded to 76% conversion.
88
Org. Lett., Vol. 1, No. 1, 1999

that the C32-C33 bond could be formed by the union of a
2-methyloxazole and a lactone. Treatment of oxazole 10 with
lithium diethylamide followed by introduction of lactone 11
gave the C20-C38 region of the phorboxazoles as a single
regioisomer in 85% yield (Scheme 3).15
Table 2. Effect of Amines on the Selectivity of Oxazole Alkylationa
Scheme 3
a All reactions were carried out in with 1 equiv of n-BuLi in THF and
proceeded to g93% conversion. b Three equivalents of amine was used in
all reactions; this was necessary to facilitate a complete change in selectivity
for entries 7 and 8 in a reasonable amount of time. c All reactions were
quenched at -78 °C. d Product ratios were determined by 1H NMR. e A
small amount of decomposition was observed.
III. Origin of Regioselectivity. While the general utility
of lithium diethylamide for the selective lithiation of 2-me-
thyloxazoles was apparent, the origin of this selectivity was
unclear. A series of experiments led us to conclude that the
conjugate acid of the lithium amide was important in
determining the regioselectivity of lithiation. A convenient
method to investigate this hypothesis involved the lithiation
of 1d with n-butyllithium, followed by introduction of an
amine and finally reaction with methyl triflate. The results
of this study are shown in Table 2.
as the base (see Table 1). This effect is absent for the more
encumbered diisopropylamine and tetramethylpiperidine at
-78 °C (entries 3 and 4), but when the temperature is raised
to -50 °C for 10 min prior to the introduction of methyl
triflate, the product distribution begins to change (entries 5
and 6). Although diisopropylamine effects this change more
rapidly than tetramethylpiperidine, if the reaction mixtures
are stirred at -50 °C for 1 h prior to the introduction of
methyl triflate, both reactions proceed to give 6d in >95:5
selectivity. It is important to note that similar warming of a
reaction mixture in the absence of amine does not cause a
significant change in alkylation regioselectivity (entry 9).
While n-butyllithium alone generates a mixture of alky-
lated products (entry 1), the introduction of diethylamine at
-78 °C results in high selectivity for 6d. The selectivity is
identical with that seen when lithium diethylamide is used
Our interpretation of these findings is illustrated in Scheme
4. Treatment of the oxazole with n-butyllithium results in a
(11) An alkylation was chosen to simplify the product identification.
Reactions with other electrophiles occur with similar selectivity, provided
the reactions take place at low temperature. A representative procedure is
as follows. The oxazole 1d (39 mg, 0.126 mmol, 1 equiv) was dissolved in
anhydrous THF (0.76 mL) and the solution cooled with stirring to -78 °C
under an argon atmosphere. A solution of lithium diethylamide was prepared
by adding n-butyllithium (89 µL of a 1.97 M solution in hexanes, 0.176
mmol, 1.4 equiv) to a solution of diethylamine (20 µL, 0.189 mmol, 1.5
equiv) in THF (0.5 mL) at -78 °C. The lithium diethylamide solution was
warmed to 0 °C for 10 min, and then recooled to -78 °C and added to the
oxazole solution via cannula. The resulting yellow-orange reaction mixture
was stirred for 10 min. Methyl triflate (29 µL, 0.252 mmol, 2 equiv) was
then added, resulting in the immediate disappearance of color. The reaction
mixture was stirred for 20 min at -78 °C and was then partitioned between
saturated aqueous ammonium chloride and dichloromethane. The combined
organic phases were dried (MgSO4), filtered, and concentrated in vacuo to
afford a clear, colorless oil (41.5 mg) that was identified by 1H NMR to
contain 6d as the only methylated product and e3% of the starting material
1d.
Scheme 4
(12) Knaus, G.; Meyers, A. I. J. Org. Chem. 1974, 39, 1192-1195. This
study did not report the use of lithium diethylamide.
(13) See refs 5d,e and: (a) Celatka, C. A.; Liu, P.; Panek, J. S.
Tetrahedron Lett. 1997, 38, 5449-5452. (b) Pattenden, G. J. Heterocycl.
Chem. 1992, 29, 607-618.
(14) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327-
4329.
kinetic mixture of noninterconverting16 lithiated species 2
and 3, which, upon addition of methyl triflate, give rise to 6
(15) The details of the synthesis of fragments 10 and 11 will be reported
in due course.
(16) Meyers has found that, at low temperatures, lithiated thiazoles
prepared with n-butyllithium do not interconvert to a significant extent.12
Org. Lett., Vol. 1, No. 1, 1999
89

and 7, respectively. In the presence of an amine, a pathway
for the equilibration of the lithiated regioisomers is available,
allowing for the conversion of 3 to the thermodynamically
more stable 2.17 The rate of equilibration is dependent on
the steric encumbrance of the amine, with diethylamine
effecting the most rapid equilibration and tetramethylpip-
eridine the slowest.18 Therefore, the remarkable selectivity
observed in metalations with lithium diethylamide is appar-
ently due to an equilibration process mediated by the
diethylamine liberated during the reaction.19 However, at
comparable temperatures, other bases generate a nonequili-
brating kinetic mixture of lithiated oxazoles.
The remarkable ability of amines to mediate the equilibra-
tion of metalated species has been largely overlooked. In
the metalation of 2,4-lutidine, for example, Levine noted a
divergence in the selectivity afforded by amide and alkyl-
lithium bases which was attributed to kinetic effects.21
Consistent with Levine’s results, lithium diethylamide and
n-butyllithium give exclusively 16 and 17, respectively
(Table 3, entries 1 and 2). However, when diethylamine is
Table 3. Selective Lithiation of 2,4-Lutidine
The feasibility of the proposed amine-mediated proton
transfer was investigated in the crossover experiments shown
in Scheme 5. When the lithiated 2-tert-butyloxazole 14 was
Scheme 5
introduced to the anion generated with n-butyllithium, the
selectivity is reversed (entry 3). While a kinetically controlled
lithiation with n-butyllithium gives rise to 17, amide bases
deliver 16 by equilibration to the thermodynamically more
stable 4-methyl anion.
In conclusion, we have described a selective metalation
which provides a regioselective approach to the elaboration
of 2-methyloxazoles. This method appears to be general and
has been applied to the synthesis of 2,4-disubstituted oxazole
systems relevant to the phorboxazoles. In addition, we have
identified that the selectivity of the process relies on the
unique ability of diethylamine to effect the rapid equilibration
of lithiated regioisomers at low temperatures.
mixed with an equal amount of 1d and treated with methyl
triflate, 15 was generated as the only methylated product.20
The introduction of diethylamine (1.5 equiv) prior to
treatment with methyl triflate results in the formation of 6d
as the only methylated product, thus establishing the
importance of the amine in promoting the equilibration of
lithiated oxazoles.
Acknowledgment. Support has been provided by the
NIH. An NSF Predoctoral Fellowship to V.J.C. and an
American Cancer Society Postdoctoral Fellowship to T.E.S.
are gratefully acknowledged. The authors wish to thank Mr.
Duke Fitch for helpful discussions.
(17) It is conceivable that 2 could give rise to the product 7 by ring
1
alkylation and aromatization. Low-temperature H NMR has allowed the
observation of distinct intermediates consistent with 2 and 3 as well as
their complete equilibration to 2. Analysis of the products 6 and 7 formed
upon quenching these reactions suggests that ring alkylation of 2 is not
occurring. For more details, see the Supporting Information.
Supporting Information Available: Text giving experi-
mental procedures and characterization data for the com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(18) This conclusion is supported by the results of Fraser et al., who
have reported that the rate of proton exchange between amines and lithiated
amides decreases with increasing steric bulk: Fraser, R. R.; Baignee, A.;
Bresse, M.; Hata, K. Tetrahedron Lett. 1982, 23, 4195-4198.
(19) It cannot be ruled out that the kinetic selectivity of lithium
diethylamide favors the formation of 2.
OL990027R
(21) Levine, R.; Dimmig, D. A.; Kadunce, W. M. J. Org. Chem. 1974,
39, 3834-3837.
1
(20) The products were identified by H NMR.
90
Org. Lett., Vol. 1, No. 1, 1999