Catalytic, asymmetric synthesis of 1,4-benzoxazinones: A remarkably enantioselective route to α-amino acid derivatives from o-benzoquinone imides

Article abstract of DOI:10.1002/anie.200602801

(Chemical Equation Presented) Cycloaddition of o-benzoquinone imides with chiral ketene enolates derived from cinchona alkaloid catalysts is the basis of a catalytic asymmetric synthesis of 1,4-benzoxazinones and 1,4-benzoxazines. The resulting cycloadduc

Full text of DOI:10.1002/anie.200602801

Communications
Amino Acids
methodology, a variety of substituents and N-acyl groups
can be incorporated into the products.
DOI: 10.1002/anie.200602801
We also highlight the synthesis of several biologically
significant a-amino acid derivatives (e.g., a-fluoro,[6] b,g-
alkynyl,[7] and b,g-alkenyl species) whose optically pure
synthesis has not been solved by asymmetric catalysis and
that are very difficult to prepare by using other methods.[8] For
example, a-fluoro a-amino acid derivatives are of great value
in the preparation of peptidomimetics and transition-state
analogues of peptide-containing therapeutic agents.[9] Sim-
ilarly, b,g-alkynyl and b,g-alkenyl a-amino acid derivatives
display analogous activity, postulated to be caused by
conformational constraint.[7]
Catalytic, Asymmetric Synthesis of 1,4-
Benzoxazinones: A Remarkably Enantioselective
Route to a-Amino AcidDerivatives from
o-Benzoquinone Imides**
Jamison Wolfer, Tefsit Bekele, Ciby J. Abraham,
Cajetan Dogo-Isonagie, and Thomas Lectka*
Molecules that serve as versatile branch points for the
synthesis of pharmaceutically or biologically active products,
besides being of interest in their own right, are especially
valuable targets for asymmetric catalysis. The 1,4-benzoxazi-
none[1] and 1,4-benzoxazine[2] systems are intriguing because
they are present in clinically significant pharmaceuticals and
other biologically active molecules. On the basis of previous
success in the preparation of a-oxygenated carboxylic acid
derivatives from benzodioxinones,[3] we speculated that chiral
1,4-benzoxazinone intermediates could also serve as flexible
precursors for the efficient synthesis of highly enantiomeri-
cally enriched a-amino acids and related derivatives.[4]
Herein, we present the first catalytic, asymmetric synthesis
of 1,4-benzoxazinones that relies on the highly enantioselec-
tive [4+2]cycloaddition of o-benzoquinone imides with chiral
ketene enolates (derived from acid chlorides and cinchona
alkaloid[5] catalysts; Scheme 1). These cycloadducts can be
We have had an interest in the catalytic, enantioselective
reactions of a-imino esters, chemistry which provides a
variety of useful products (b-lactams and a- or b-amino
acids) with high enantioselectivity upon alkylation at the
carbon atom.[10] Although we noted that o-benzoquinone
imides share structural similarity to a-imino esters, they
prefer to alkylate at the nitrogen atom instead, thus providing
products in which the aromaticity is restored.[11] An approach
employing chiral ketene enolates provides three points of
modification: the quinone core, the N substituent, and the
acid chloride. We noted that electron-withdrawing N-acyl
groups should increase the reactivity of the quinone unit
towards cycloaddition, besides serving as protecting groups
that can subsequently be removed. Initially, we examined
several N-acylated quinone imides[12] and chose experimental
conditions that worked well for the asymmetric cycloaddition
of ketene enolates and o-quinones.[3] We found that by
employing 4-methylvaleryl chloride 1a (R1 = iBu), imide 2a
(R2 = p-NO2PhCO, R3 = H, R4 = R5 = Cl), 10 mol% benzoyl-
quinidine (3a; BQd), and Hünigꢀs base in THF at À788C, we
formed cycloadduct 4a in 65% yield with over 99% ee
(Scheme 2).[13] Several other 1,4-benzoxazinones were syn-
thesized from different quinone imides and acid chlorides to
provide products in good yield with uniformly excellent
enantiomeric excess.[14] One reaction of special interest would
be the conversion of chiral 1,4-benzoxazinones into 1,4-
benzoxazines, skeleta that are present in biologically relevant
molecules, such as levofloxacin.[15] For example, benzoxazi-
[16]
Scheme 1. Synthesis of 1,4-benzoxazinones.
none 4l reacts smoothly with BH3·SMe2
to provide the
corresponding benzoxazine 7l in 68% yield and with full
preservation of the enantiomeric excess.
functionalized in situ to provide 1,4-benzoxazines and a-
amino acid derivatives in good-to-excellent yields and with
virtual enantiopurity, only rivaled by that of enzymatic amino
acid synthesis. As a testament to the flexibility of this
[*] J. Wolfer, T. Bekele, C. J. Abraham, C. Dogo-Isonagie, Prof. T. Lectka
Department of Chemistry
Johns Hopkins University
3400 North Charles Street, Baltimore, MD 21218 (USA)
Fax: (+1)410-516-7044
E-mail: lectka@jhu.edu
[**] T.L. thanks the NIH (GM064559); the Sloan, Guggenheim, and
Dreyfus Foundations; and Merck & Co. for support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Scheme 2. Chiral 1,4-benzoxazinone and1,4-benzoxazine products.
Fmoc=9-fluorenylmethoxycarbonyl, Bn=benzyl.
7398
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 7398 –7400

Angewandte
Chemie
We discovered that cycloadduct 4a undergoes rapid ring-
opening methanolysis to afford ester 5a, thus indicating that
in situ transformation to a-amino acid derivatives by various
nucleophiles would occur. Thus, we sought to convert the
cycloadducts directly into the a-amino acids in one pot
(Scheme 3).[17]
Most notably, we accomplished the synthesis of three
biologically significant derivatives—b,g-alkynyl acid 5g, b,g-
alkenyl acid 5h, and a-fluoro acid 5p (Table 1, entries 7, 8,
and 16, respectively)—all in good yield and with excellent
enantiomeric excess. To our knowledge, a-fluoro a-amino
acid derivatives have not heretofore been addressed by
asymmetric catalysis.[6]
Finally, oxidation by using ceric ammonium nitrate
(CAN)[19] removes the aryl group in good yield under mild
conditions [Eq. (1)]; for example, see Table 2, entries 1
Scheme 3. One-pot amino acidsynthesis.
We chose o-benzoquinone imides derived from halogen-
ated o-aminophenols as templates for a-amino acid synthesis
for a variety of reasons—the starting materials are inexpen-
sive and the electron-withdrawing groups in the 3- and 4-
positions block undesired reactivity at the quinone ring and
enhance the overall reactivity of the system.[18] Following the
cycloaddition (ca. 5 h at À788C), MeOH was added to the
reaction mixture, which was then warmed to room temper-
ature to produce products 5 in high yield. Having chosen R2 =
p-NO2PhCO as the group that performed the best overall, we
then screened a variety of R1 substituents and cores, (Table 1,
entries 1—13 and 16). In each case, the reaction occurred in
good yield and with excellent enantiomeric excess.
Table 2: Deprotection using CAN.
Entry R2
R1
Product
ee
[%]
Yield[a]
[%]
1
2
3
4
5
6
Fmoc
Bn
p-NO2PhCO Bn
p-NO2PhCO Bn
p-NO2PhCO Ph
p-NO2PhCO PhOCH2
6o >99 71
6d >99 71
6d >99 64
6e >99 58
6 f >99 72
6h >99 74
=
p-NO2PhCO PhCH CH
[a] Reactions run with 5 (0.55 mmol) andCAN (1.65 mmol) in water/
MeCN (1:3) at 08C. Yieldafter column chromatography.
Table 1: Synthesis of a-amino acidderivatives.
(conversion of 5o into 6o), 2 (5k
Entry
R2
R1
Product[a]
ee [%]
Yield[%]
into 6d), 3 (5d into 6d), 4 (5e into
6e), 5 (5 f into 6 f), and 6 (5h into
6h).
We then screened other qui-
none imides in which the N-acyl
group was varied (Table 1,
entries 14 and 15). In particular,
we were interested in highlighting
1[b]
2[b]
3[b]
4[b]
5[b]
6[b]
7[b]
8[b]
9[c]
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
iBu
Et
Me
Bn
Ph
PhOCH2
CH3CH2C C
PhCH CH
5a
5b
5c
5d
5e
5 f
5g
5h
5i
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
73
62
69
63
66
72
59
61
71
90
ꢀ
=
PhOCH2
Et
a
signature protecting group
10[d]
5j
important in peptide synthesis,
such as 9-fluorenylmethyl carba-
mate (Fmoc).[20] For example,
when R1 = Bn and R2 = Fmoc, the
reaction occurs smoothly in THFat
À788 C to form product 5o in 62%
yield with greater than 99% ee.
The Fmoc group was then removed
by piperidine to afford 8 in high
yield (92%) followed by deprotec-
tion with CAN to give 9 in good
yield and without loss of optical
activity (Scheme 4).[20]
11[b]
12[b]
13[b]
p-NO2PhCO
p-NO2PhCO
p-NO2PhCO
Bn
Et
iBu
5k
5l
5m
>99
>99
>99
83
71
73
14[e]
15[e]
16[b]
Fmoc
Fmoc
p-NO2PhCO
Et
Bn
F
5n
5o
5p
>99
>99
>99
59
62
60
In conclusion, we have illus-
trated the first highly enantioselec-
tive synthesis of 1,4-benzoxazi-
[a] Nu=OMe for all entries, except entry 9 (Nu=NH2) andentry 10 (Nu =BnNH). [b] Reactions run
with catalyst (10 mol%), Hünig’s base (0.55 mmol), acid chloride (0.55 mmol), and quinone imide
(0.55 mmol) at À788C followed by addition of MeOH and overnight stirring. Yield for cycloaddition and
methanolysis. [c] Reaction quenchedwith NH 4OH to yieldamide 5i. [d] Reaction quenched with
benzylamine in THF to yield 5j. [e] Quinone imide formed in situ at À78 8C; yieldfor both steps.
nones
and
1,4-benzoxazines.
These products are readily con-
Angew. Chem. Int. Ed. 2006, 45, 7398 –7400
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7399

Communications
[9]V. P. Kukhar in Fluorine-Containing Amino Acids, Synthesis and
Properties (Ed.: V.A. Soloshonok), Wiley, Chichester, 1995.
[10]a) D. Ferraris, B. Young, T. Dudding, T. Lectka, J. Am. Chem.
Soc. 1998, 120, 4548; b) A. E. Taggi, A. M. Hafez, H. Wack, B.
Young, W. J. Drury III, T. Lectka, J. Am. Chem. Soc. 2000, 122,
7831.
[11]For a detailed study of o-quinone imide cycloadditions: K. C.
Nicolaou, K. Sugita, P. S. Baran, Y.-L. Zhong, J. Am. Chem. Soc.
2002, 124, 2221.
Scheme 4. Synthesis of N-deprotected a-amino esters.
[12]In some cases, the quinone imides were isolated, in other cases,
they were generated in situ by simple Pb(OAc)4 oxidation.
[13]The opposite S enantiomer products can be made in comparable
selectivity with benzoylquinine (BQ, 3b) as the catalyst.
[14]Absolute configurations were determined by correlation to
several known a-amino acid derivatives; the sense of induction
observed was consistent with that obtained in other cinchona
alkaloid catalyzed reactions (see Reference [10]).
verted into virtually optically pure a-amino acid esters. This
study compares very favorably with other chiral a-amino acid
syntheses because of the remarkably high enantioselectivities
obtained and as it provides access important classes of chiral
compounds that are otherwise difficult to synthesize.
[15]K. F. Croom, K. L. Goa, Drugs 2003, 63, 2769.
[16]P. Verma, S. Singh, D. K. Dikshit, R. Suprabhat, Synthesis 1987,
1, 68.
[17]Reactions were quenched with a THF solution of the desired
nucleophile at room temperature.
[18]R. Adams, J. M. Stewart, J. Am. Chem. Soc. 1952, 74, 5876.
[19]For an example of dearylation by using CAN, see: S. Kobayashi,
J. Kobayashi, H. Ishiani, M. Ueno, Chem. Eur. J. 2002, 8, 4185.
[20]For a review, see: L.A. Carpino, Acc. Chem. Res. 1987, 20, 401.
Experimental Section
General procedure: A solution of the quinone imide (0.12 mmol) in
THF (2 mL) was added to a reaction flask containing an acid chloride
(0.12 mmol), Hünigꢀs base (0.12 mmol), and BQd (3a; 0.012 mmol) at
À788C. After stirring for 6 h, the reaction was concentrated in vacuo
and the crude residue was purified by column chromatography.
Additional procedures and characterization data are presented in the
Supporting Information.
Received: July 14, 2006
Published online: October 12, 2006
Keywords: amino acids · asymmetric catalysis · benzoxazines ·
.
benzoxazinones · cycloaddition
[1]Benzoxazinones are known as antitumor agents and protease
inhibitors: a) S. K. Sengupta, D. H. Trites, M. S. Madhavarao,
W. R. Beltz, J. Med. Chem. 1979, 22, 797; b) R. L. Jarvest, S. C.
Connor, J. G. Gorniak, L. J. Jennings, H. T. Serafinowska, A.
West, Bioorg. Med. Chem. Lett. 1997, 7, 1733; c) F. Bihel, J.-L.
Kraus, Org. Biomol. Chem. 2003, 1, 793.
[2]Benzoxazines are active as antioxidative neuroprotectants and
antibiotics: a) M. Largeron, B. Lockhart, B. Pfeiffer, M.-B.
Fleury, J. Med. Chem. 1999, 42, 5043; b) Y.-G. Zhou, P.-Y. Yang,
X.-W. Han, J. Org. Chem. 2005, 70, 1679; c) J. Ilas, P. S.
Anderluh, M. S. Delenc, D. Kikelj, Tetrahedron 2005, 61, 7325.
[3]T. Bekele, M. H. Shah, J. Wolfer, C. J. Abraham, A. Weatherwax,
T. Lectka, J. Am. Chem. Soc. 2006, 128, 1810.
[4]For recent examples of asymmetric a-amino acid synthesis, see:
a) T. Abellan, R. Chinchilla, N. Galindo, G. Guillena, C. Najera,
J. M. Sansano, Eur. J. Org. Chem. 2000, 2689; b) J. Kobayashi, M.
Nakamura, Y. Mori, Y. Yamashita, S. Kobayashi, J. Am. Chem.
Soc. 2004, 126, 9192; c) T. Belser, M. Stohr, A. Pfaltz, J. Am.
Chem. Soc. 2005, 127, 8720.
[5]K. Kacprzak, J. Gawronski, Synthesis 2001, 7, 961.
[6]a) D. P. Huber, K. Stanek, A. Togni, Tetrahedron: Asymmetry
2006, 17, 658; b) B. Mohar, J. Baudoux, J.-C. Plaquevent, D.
Cahard, Angew. Chem. 2001, 113, 4666; Angew. Chem. Int. Ed.
2001, 40, 4214.
[7]P. Meffre, F. Le Goffic, Amino Acids 1996, 11, 313.
[8]Presumably, these amino acids are not very amenable to
synthesis through asymmetric hydrogenation (that is, 5g and
5h by virtue of their unsaturation), through the chiral phase-
transfer synthesis (that is, 5g and 5h as no suitable SN2 substrates
exist), or the Strecker reaction (that is, the conditions for the
hydrolysis of the nitrile group can be harsh for 5g, 5h, and 5p).
7400 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 7398 –7400