Asymmetric synthesis of chiral 1,3-diaminopropanols: Bisoxazolidine- catalyzed C-C bond formation with α-keto amides

Article abstract of DOI:10.1002/anie.201105778

Three high-yielding steps lead to the formation of chiral 1,3-diaminopropanols from aliphatic and aromatic α-keto amides. In this approach, a nitroaldol reaction, which is catalyzed by Cu(SO₂CF ₃)₂ and the bisoxazolidine ligand L1, is followed by two mild reduction reactions (see scheme). Laborious protection and deprotection steps can be avoided by using this method.

Full text of DOI:10.1002/anie.201105778

DOI: 10.1002/anie.201105778
Asymmetric Catalysis
Asymmetric Synthesis of Chiral 1,3-Diaminopropanols:
À
Bisoxazolidine-Catalyzed C C Bond Formation with a-Keto Amides**
Hanhui Xu and Christian Wolf*
The ever-increasing demand for enantiopure drugs and
agrochemicals, and the use of chiral building blocks in
polymers, liquid crystals, and other materials has generated
a strong stimulus for the development of asymmetric methods
that utilize previously unexplored starting materials to gen-
erate efficient access to new and emerging target com-
pounds.[1] New drug candidates that bear a chiral 1,3-
diaminopropanol moiety or a corresponding oxazolidinone
derivative have recently been introduced for the treatment of
tuberculosis, Alzheimerꢀs disease, and nosocomial infections
caused by bacteria that are resistant to common antibiotics.[2]
The formation of these challenging structures relies on
multistep syntheses from chiral epoxy alcohols or esters and
typically involves laborious protection/deprotection proto-
cols. We envisioned that the enantioselective synthesis of N-
substituted 1,3-diaminopropanols could be accomplished in
three steps by nitroaldol reaction of a-keto amides, which are
unexplored starting materials in asymmetric catalysis, and
subsequent reduction of the nitro and amide groups
(Scheme 1). The first catalytic asymmetric nitroaldol reaction
been elusive.[6] We expected that careful reduction of the nitro
group in a-hydroxy b-nitro propanamides would avoid
problems with the facile retro-aldol reaction[7] and thus lead
to a practical route to a series of chiral 1,3-diaminopropa-
nols.[8]
In recent years, chiral 1,3-oxazolidines have found
increasing use as ligands and auxiliaries in asymmetric
synthesis, which may be attributed to the intriguing ring
topology and the possibility of modular synthesis from amino
alcohols.[9] We have previously introduced bisoxazolidine L1
and showed several applications of this C2-symmetric N,O-
diketal in asymmetric catalysis (Scheme 2).[10] Despite the
ease of preparation of L1, which can be obtained in a single
step from inexpensive cis-1-amino-2-indanol and 1,2-cyclo-
Scheme 1. Retrosynthetic analysis of N-monosubstituted 1,3-diamino-
propanols.
was developed in 1992 by Shibasaki and coworkers, who used
a BINOL-derived rare-earth-metal complex.[3] This reaction
has since received considerable attention and its value for the
synthesis of important chiral building blocks and complex
target compounds has been demonstrated by many research
groups.[4] Relatively few examples of asymmetric nitroaldol
reactions with ketones[5] are known compared to the wealth of
reactions with aldehydes. The use of a-keto amides has not
been reported to date, in fact, catalytic asymmetric intermo-
Scheme 2. Structures of bisoxazolidines L1–L7.
hexanedione, the synthesis of other diketone-derived bis-
oxazolidines is not straightforward and requires careful
selection of starting materials and reaction conditions.[11]
Accordingly, we chose to prepare a series of new ligands
L2–L7 derived from (1R,2S)-aminoindanol analogues and
several diketones to vary the rigidity of the N,O-diketal
backbone and to explore the catalytic performance of
fluxional bisoxazolidines in the nitroaldol reaction with keto
amides.
À
lecular C C bond formations with a-keto amides have so far
[*] H. Xu, Prof. Dr. C. Wolf
Department of Chemistry, Georgetown University
37th and O Streets, Washington, DC 20057 (USA)
E-mail: cw27@georgetown.edu
The synthesis of L2 started with the bromination of 1,2-
dihydronaphthalene (1) and mild hydrolysis of dibromide 2 to
give racemic trans-2-bromo-1-hydroxytetrahydronaphthalene
(3) in a high yield (Scheme 3).[12] The Ritter reaction of 3 then
produced racemic cis-1-amino-2-hydroxytetrahydronaphtha-
lene (4) in 79% yield. Resolution of the enantiomers of 4 by
crystallization with mandelic acid furnished (1R,2S)-4 in more
than 99.5% ee according to HPLC analysis of the tert-
butoxycarbonyl (tBoc) protected analogue (see the Support-
[**] Funding from the National Science Foundation (CHE-0848301) is
gratefully acknowledged.
Supporting information (containing general experimental proce-
dures, details on method development, and product character-
ization, including spectroscopic and chromatographic data) for this
article is available on the WWW under http://dx.doi.org/10.1002/
anie.201105778.
Angew. Chem. Int. Ed. 2011, 50, 12249 –12252
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12249

Communications
enantioselectivity but proved detrimen-
tal to the yield. This trend was also
observed when the reaction was per-
formed without solvent. Screening of
different solvents showed that the yields
and ee values of the nitroaldol reaction
vary significantly. Most importantly, we
found that (S)-14 is produced in 89%
yield and 86% ee when acetonitrile is
used as solvent.[14]
A thorough evaluation of the effect
of a wide range of base additives in both
THF and acetonitrile did not improve
yields and ee values (see Table 2 in the
Supporting Information). However, we
were surprised to find that the enantio-
selectivity of the nitroaldol reaction
with keto amide 13 is very sensitive to
the order of addition of reagents. Com-
pound 14 was obtained in essentially the
same yield but in only 9% ee when the
base was introduced to a solution of
ligand L1 and Cu(OTf)2 prior to the
addition of nitromethane and the sub-
strate. Further fine-tuning of the proce-
dure showed that (S)-14 can be pro-
duced in 93% yield and 90% ee when
nitromethane is added first and the
Scheme 3. Synthesis of ligands L2 and L3. THF=tetrahydrofuran, PTSA=p-toluenesulfonic acid.
ing Information for details). Finally, L2 was isolated in 76%
reaction mixture is stirred at room temperature for 30 min
before addition of Et3N and 13 at 08C. With this optimized
reaction protocol established, we evaluated the substrate
scope with a range of aromatic and aliphatic keto amides
yield after condensation of (1R,2S)-4 with cyclohexanedione.
We chose a similar route toward ligand L3. We first prepared
7-methyl-2,3-dihydro-1H-inden-1-one (7) in two steps by a
Diels–Alder reaction between cyclopentenone (5) and 1,3-
pentadiene to give 6 in 92% yield, and subsequent dehydro-
genation at high temperature.[13] Reduction and elimination
of alcohol 8 then provided 4-methyl-1H-indene (9), which was
employed in the previously established bromination, hydrol-
ysis, and Ritter reaction protocols to afford cis-7-methylami-
noindanol (12). The enantiomers of 12 were separated by
crystallization with phenylpropionic acid and we obtained the
(1R,2S)-enantiomer in more than 99% ee based on HPLC
analysis of its tBoc-protected derivative. The condensation
with cyclohexanedione finally provided L3 in 52% yield.
Bisoxazolidines L4–L7 were prepared accordingly from two
equivalents of (1R,2S)-aminoindanol and the corresponding
diketones in the presence of formic acid (see the Supporting
Information for details).
With ligands L1–L7 synthesized, we investigated the
nitroaldol reaction with a-keto amides. We first employed
10 mol% of copper(II) triflate and the bisoxazolidine ligands
in the reaction between N-phenyl 2-oxo-2-phenylacetamide
(13) and nitromethane in THF as solvent. We were pleased to
find that the desired N-phenyl 2-hydroxy-3-nitro-2-phenyl-
propanamide (14) was formed in high yield and 57–60% ee
when L1, L4, and L6 were used at 08C for 24 h (see Table 1 in
the Supporting Information). The other bisoxazolidine
ligands generated 14 in similar yields but low ee values,
whereas the sense of asymmetric induction was reversed when
ligand L3 was used. A decrease in temperature improved the
Scheme 4. Enantioselective synthesis of 2-substitued 2-hydroxy-3-nitro-
propanamides by using 10 mol% of Cu(OTf)2 and ligand L1. Tf=tri-
fluoromethanesulfonyl.
12250 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12249 –12252

(Scheme 4). The L1-catalyzed nitroaldol reaction produced
several aromatic analogues, including thiophene 20, with
excellent results. Importantly, the high yields and ee values
were not compromised when aliphatic substrates were
employed, and 2-hydroxy-3-nitroalkanamides 22–24 were
isolated in 87–90% yield and 86–88% ee.[15]
when borane is used in THF at reflux for one hour.[17] This
two-step reduction sequence was then used to form the
desired series of chiral 1,3-diaminopropan-2-ols 25–35 in
excellent overall yields (Table 1, entries 1–11).
In summary, we have developed the first example of
À
catalytic intermolecular enantioselective C C bond forma-
With a general approach to chiral (S)-2-alkyl- and 2-aryl-
2-hydroxy-3-nitropropanamides established, we evaluated
methods for the careful transformation of the nitro and the
amide group to the corresponding primary and secondary
amino functions. As expected, all attempts to reduce 14 under
slightly basic or acidic conditions gave 13 because of the facile
retro-nitroaldol reaction. However, this complication could
essentially be avoided by palladium-catalyzed hydrogenation
in methanol at 558C, and we isolated the desired primary
amine in 94% yield (Table 1, entry 1). We then tested several
reported protocols for the amide reduction, including the use
of metal hydrides, ruthenium-catalyzed hydrogenation at high
temperatures, and the use of triethoxysilane in the presence of
catalytic amounts of zinc acetate, all of which were unsuc-
cessful.[16] Finally, we discovered that 1-amino-2-phenyl-3-
(phenylamino)propan-2-ol (25) can be obtained in 88% yield
tions with a-keto amides. The bisoxazolidine-catalyzed nitro-
aldol reaction of these unexplored bifunctional building
blocks provides a practical approach to a wide range of
chiral 1,3-diaminopropanols that bear a central, tertiary
alcohol group. The screening of seven C2-symmetric bis-
oxazolidine ligands and several reaction parameters showed
that both aliphatic and aromatic keto amides are converted to
2-alkyl- and 2-aryl-2-hydroxy-3-nitropropanamides in up to
93% yield and 90% ee in the presence of 10 mol% of L1 and
copper(II) triflate in acetonitrile. Selective hydrogenation
and borane reduction of the nitro and amide groups were
found to generate chiral 1,3-diaminopropanols in high overall
yields and without the need for functional-group protection.
We expect that the results from this study will direct
increasing attention to the use of keto amides as readily
available, versatile starting materials for asymmetric syn-
thesis.
Received: August 16, 2011
Revised: October 17, 2011
Table 1: Synthesis of chiral 1,3-diaminopropanols.
Published online: October 31, 2011
Keywords: a-keto amides · asymmetric catalysis ·
.
À
bisoxazolidines · C C bond formation · nitroaldol reaction
Entry
R
1,3-Diaminopropanol
Yield [%][a]
Step 1 Step 2
[1] a) G.-Q. Lin, Y.-M. Li, A. S. C. Chan, Principles and Applica-
tions of Asymmetric Synthesis, Wiley-VCH, New York, 2001;
b) C. Wolf, Dynamic Stereochemistry of Chiral Compounds, RSC
Publishing, Cambridge, 2008.
[2] a) S. J. Brickner, M. R. Barbachyn, D. K. Hutchinson, P. R.
Manninen, J. Med. Chem. 2008, 51, 1981; b) F. Russo, F.
Wangsell, J. Savmarker, M. Jacobsson, M. Larhed, Tetrahedron
2009, 65, 10047; c) A. Kamal, R. V. C. R. Shetti, S. Azeeza, P.
Swapna, M. N. A. Khan, I. A. Khan, S. Sharma, S. T. Abdullah,
Eur. J. Med. Chem. 2011, 46, 893.
1
2
3
4
Ph
(+)-25
(+)-26
(+)-27
(+)-28
94
95
90
90
88
88
87
78
4-MeC6H4
3-MeC6H4
4-tBuC6H4
[3] a) J. Sasai, T. Suzuki, S. Arai, T. Arai, M. Shibasaki, J. Am. Chem.
Soc. 1992, 114, 4418; b) J. Sasai, T. Suzuki, N. Itoh, M. Shibasaki,
Tetrahedron Lett. 1993, 34, 851 – 854; c) H. Sasai, N. Zuzuki, K.
Itoh, T. Tanaka, K. Date, K. Okamura, M. Shibasaki, J. Am.
Chem. Soc. 1993, 115, 10372 – 10373; d) Y. Sohtome, Y. Kato, S.
Handa, N. Aoyama, K. Nagawa, S. Matsunaga, M. Shibasaki,
Org. Lett. 2008, 10, 2231 – 2234; e) T. Nitabaru, N. Kumagai, M.
Shibasaki, Tetrahedron Lett. 2008, 49, 272 – 276.
[4] a) B. M. Trost, V. S. C. Yeh, Angew. Chem. 2002, 114, 889;
Angew. Chem. Int. Ed. 2002, 41, 861; b) C. Christensen, K. Juhl,
R. G. Hazell, K. A. Jørgensen, J. Org. Chem. 2002, 67, 4875; c) T.
Ooi, K. Doda, K. Maruoka, J. Am. Chem. Soc. 2003, 125, 2054;
d) D. A. Evans, D. Seidel, M. Rueping, H. W. Lam, J. T. Shaw,
C. W. Downey, J. Am. Chem. Soc. 2003, 125, 12692; e) C.
Palomo, M. Oiarbide, A. Mielgo, Angew. Chem. 2004, 116, 5558;
Angew. Chem. Int. Ed. 2004, 43, 5442; f) C. Palomo, M. Oiarbide,
A. Laso, Angew. Chem. 2005, 117, 3949; Angew. Chem. Int. Ed.
2005, 44, 3881; g) M. Bandini, F. Piccinelli, S. Tommasi, A.
Umani-Ronchi, C. Ventrici, Chem. Commun. 2007, 616; h) B. M.
Choudary, K. V. S. Ranganath, U. Pal, M. L. Kantam, B.
Sreedhar, J. Am. Chem. Soc. 2005, 127, 13167; i) T. Arai, M.
5
4-Cl
(+)-29
93
86
6
7
8
4-Br
(+)-30
(+)-31
87
82
86
82
85
80
3-C4H3S
4-MeOC6H4 (+)-32
9
PhC2H4
(+)-33
(+)-34
84
89
85
83
10
iPr
11
tBu
(+)-35
89
85
[a] Yields of isolated products.
Angew. Chem. Int. Ed. 2011, 50, 12249 –12252
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12251

Communications
Watanabe, A. Fujiwara, N. Yokoyama, A. Yanagisawa, Angew.
[9] a) C. A. Caputo, N. D. Jones, Dalton Trans. 2007, 4627; b) C.
Wolf, H. Xu, Chem. Commun. 2011, 47, 3339.
Chem. 2006, 118, 6124; Angew. Chem. Int. Ed. 2006, 45, 5978;
j) H. Y. Kim, K. Oh, Org. Lett. 2009, 11, 5682; k) M. Steurer, C.
Bolm, J. Org. Chem. 2010, 75, 3301; l) K. Lang, J. Park, S. Hong,
J. Org. Chem. 2010, 75, 6424; m) D. Uraguchi, S. Nakamura, T.
Ooi, Angew. Chem. 2010, 122, 7724; Angew. Chem. Int. Ed. 2010,
49, 7562; n) V. J. Mayani, S. H. R. Abdi, R. I. Kureshy, N.-u. H.
Khan, A. Das, H. C. Bajaj, J. Org. Chem. 2010, 75, 6191; o) W.
Jin, X. Li, B. Wan, J. Org. Chem. 2011, 76, 484.
[10] a) C. Wolf, S. Liu, J. Am. Chem. Soc. 2006, 128, 10996; b) S. Liu,
C. Wolf, Org. Lett. 2007, 9, 2965; c) S. Liu, C. Wolf, Org. Lett.
2008, 10, 1831; d) K. Yearick Spangler, C. Wolf, Org. Lett. 2009,
11, 4724; e) H. Xu, C. Wolf, Chem. Commun. 2010, 46, 8026;
f) H. Xu, C. Wolf, Synlett 2010, 2765; g) C. Wolf, P. Zhang, Adv.
Synth. Catal. 2011, 353, 760. This ligand is commercially
available at Strem (cat. no. 07-0488), US Patent, 11/737,371,
2007.
[11] We found that in most cases double condensation to generate a
bisoxazolidine structure does not occur with a wide range of
amino alcohols and diketones. In contrast, only one carbonyl
group reacts even under harsh conditions to give a monoox-
azolidine, or other by-products are formed.
[5] a) C. Christensen, K. Juhl, K. A. Jørgensen, Chem. Commun.
2001, 2222; b) D.-M. Du, S.-F. Lu, T. Fang, J. Xu, J. Org. Chem.
2005, 70, 3712; c) H. Li, B. Wang, L. Deng, J. Am. Chem. Soc.
2006, 128, 732; d) F. Tur, J. M. Saa, Org. Lett. 2007, 9, 5079; e) B.
Qin, X. Xiao, X. Liu, J. Huang, Y. Wen, X. Feng, J. Org. Chem.
2007, 72, 9323; f) A. Bulut, A. Aslant, O. Dogan, J. Org. Chem.
2008, 73, 7373; g) M. Bandini, R. Sinisi, A. Umani-Ronchi,
Chem. Commun. 2008, 4360; h) G. Blay, V. Hernandez-Olmos,
J. R. Pedro, Chem. Eur. J. 2011, 17, 3768.
[12] a) C. M. Bellucci, A. Bergamini, P. G. Cozzi, A. Papa, E.
Tagliavini, A. Umani-Ronchi, Tetrahedron: Asymmetry 1997, 8,
895; b) S. Higashijima, H. Itoh, Y. Senda, S. Nakano, Tetrahe-
dron: Asymmetry 1997, 8, 3107.
[6] For an example of asymmetric catalysis with an a-formyl amide,
see: D. A. Evans, Y. Aye, J. Am. Chem. Soc. 2006, 128, 11034.
During the completion of this manuscript, the asymmetric
intramolecular arylation of a-keto amides was reported: L. Yin,
M. Kanai, M. Shibasaki, Angew. Chem. 2011, 123, 7762; Angew.
Chem. Int. Ed. 2011, 50, 7620. Racemic nitroaldol products of
keto amides have been obtained by using catalytic amounts of
diethyl- or triethylamine: a) W. R. Conn, H. G. Lindwall, J. Am.
Chem. Soc. 1936, 58, 1236; b) Y.-J. Wang, Z.-X. Shen, Y.-W.
Zhang, Chin. J. Org. Chem. 2006, 26, 1291.
[7] Nitroaldol products are known to easily decompose and
racemize under basic and acidic conditions or upon heating.
[8] For desymmetrization of 1,3-propanediols and 1,2,3-propane-
triols, see: a) B. Jung, S. H. Kang, Proc. Natl. Acad. Sci. USA
2007, 104, 1471; b) J. Y. Lee, Y. S. You, S. H. Kang, J. Am. Chem.
Soc. 2011, 133, 1772.
[13] a) F. Fringuelli, R. Girotti, F. Pizzo, L. Vaccaro, Org. Lett. 2006,
8, 2487; b) K. Tanaka, T. Kobayashi, H. Mori, S. Katsumura,
J. Org. Chem. 2004, 69, 5906.
[14] The absolute configuration of the nitroaldol products was
confirmed by NMR analysis with a chiral solvating agent (see
the Supporting Information).
[15] The nitroaldol reaction with N-propyl and N-benzyl 2-oxo-2-
phenylacetamide gave the product in 92–95% yield and 26–30%
ee.
[16] a) A. A. NfflÇez Magro, G. R. Eastham, D. J. Cole-Hamilton,
Chem. Commun. 2007, 3154; b) S. Das, D. Addis, S. Zhou, K.
Junge, M. Beller, J. Am. Chem. Soc. 2010, 132, 1770.
[17] Formation of the tBoc derivative of 25 and HPLC analysis on a
chiral stationary phase (Chiralpak AD) showed that the ee value
was slightly reduced from 90% to 86% after the two reduction
steps (see the Supporting Information).
12252 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12249 –12252