Effecient kinetic resolution of racemic amino aldehydes by oxidation with N-iodosuccinimide

Article abstract of DOI:10.1002/anie.200804188

(Chemical Equation Presented) Selective recognition: The first efficient method for the kinetic resolution of racemic amino aldehydes (see scheme, PG=protecting group) is based on a copper(II)/(R,R)-Ph-BOX complex. The coordinated amino aldehydes were tra

Full text of DOI:10.1002/anie.200804188

Communications
DOI: 10.1002/anie.200804188
Asymmetric Catalysis
Effecient Kinetic Resolution of Racemic Amino Aldehydes by
Oxidation with N-Iodosuccinimide**
Daishirou Minato, Yoko Nagasue, Yosuke Demizu, and Osamu Onomura*
Amino acids are very useful as synthetic building blocks for
various biologically active compounds.[1] Recently, several
pseudopeptides containing natural or non-natural amino
acids have been developed because they have pharmacolog-
ically important characteristics.[2] Although natural amino
acids are prepared by biochemical techniques such as
Scheme 1. Oxidative kinetic resolution of racemic aminoaldehydes.
fermentation, there is scant information on the preparation
of non-natural amino acids by using this approach.[3] Among
the asymmetric catalytic methods for the synthesis of natural
and non-natural amino acids,[4–6] the kinetic resolution of
amino acid derivatives is frequently used.[7] However, there
are few examples applicable to the synthesis of various
optically active amino acids, including cyclic amino acids, and
to the best of our knowledge, there is no chemical oxidation
method for their preparation. We report herein the first
efficient kinetic resolution of racemic amino aldehydes by
oxidation.
PG=protecting group.
First, we applied the previous reaction conditions for
asymmetric oxidation of 1,2-diols using N-bromosuccinimide
(NBS) in the presence of K2CO3 for the oxidative kinetic
[9a]
resolution of rac-N-benzoyl-2-piperidinecarboaldehyde (rac-
1a; Table 1, entry 1). (R)-2a[10] was obtained with a high
Table 1: Oxidative kinetic resolution of racemic N-benzoyl-2-piperidine-
carboaldehyde (1a).[a]
Recently, we accomplished the oxidative kinetic resolu-
tion of 1,2-diols, which was based on their recognition by a
copper(II)/(R,R)-Ph-BOX complex (see Scheme 2 for struc-
ture),[8] to afford optically active a-ketoalcohols.[9a] Moreover,
we have reported the asymmetric electrochemical oxidation
of N-protected 1,2-amino aldehydes to afford optically active
amino acid methyl esters in low yield, but with good
enantioselectivity.[9b] In line with our previous work, we
investigated the reaction conditions for oxidative kinetic
resolution of racemic amino aldehydes to improve the yields
and enantioselectivities of the optically active amino acids. To
our delight, we found a simple method for a highly efficient
kinetic resolution of racemic N-protected amino aldehydes.
The use of a chiral copper catalyzed oxidation procedue with
N-iodosuccinimide (NIS) afforded optically active amino acid
methyl esters, including cyclic and acyclic compounds, with
high enantioselectivity (Scheme 1). Additionally, instead of
recovering the starting material, the corresponding optically
active aminoaldehyde dimethyl acetals were preferentially
obtained.
Entry
1
Base
(R)-2a
(S)-1a
(S)-3a
s[c]
K2CO3
12% yield
94% ee
39% yield
85% ee
51% yield
4% ee[b]
–
–
2
none
46% yield
50% ee
20
[a] A mixture of 1a (0.5 mmol), Cu(OTf)2 (0.05 mmol), (R,R)-Ph-BOX
(0.05 mmol), and NBS (0.25 mmol) in MeOH (2 mL) in the presence or
absence of K2CO3 (0.25 mmol) was stirred at RT for 12 h. [b] Determined
after its reduction to the corresponding amino alcohol. [c] s=stereose-
lectivity factor for kinetic resolution.[12] Bz=benzoyl, Tf=triflate.
enantiomeric excess, but the yield was low; the enantiomeric
excess of recovered 1a was also very low. On the other hand,
the absence of a base drastically changed the reaction
(Table 1, entry 2) such that the yield of (R)-2a was signifi-
cantly increased and the optically active aminoaldehyde
dimethyl acetal (S)-(3a)[11] was obtained in an acceptable
yield with good enantioselectivity (s = 20).
Next, we sought to improve the reaction conditions by
varying the amount and type of cationic halogen species
(Table 2). Increasing the amount of NBS from 0.5 equivalents
to 0.75 equivalents had no effect on the yield or the selectivity
(Table 2, entry 1), and the use of other bromo cationic species
(NBPI, DBDMH, Br2) led to a lower s value than that
obtained with NBS (Table 2, entries 2–4). N-Chlorosuccini-
mide (NCS) did not oxidize 1a to afford methyl ester 2a, but
transformed it into acetal 3a in racemic form (Table 2,
entry 5). In contrast, the use of NIS led to a higher s value
than that obtained with NBS (Table 2, entries 6–8). The
[*] D. Minato, Y. Nagasue, Y. Demizu, Prof. Dr. O. Onomura
Graduate School of Biomedical Sciences, Nagasaki University
1-14, Bunkyo-machi, Nagasaki 852-8521 (Japan)
Fax: (+81)95-819-2476
E-mail: onomura@nagasaki-u.ac.jp
[**] We thank the JSPS Research Fellowships for Young Scientists, the
Sumitomo Foundation, a Grant-in-Aid for Young Scientists (B;
19790017) from the Ministry of Education, Science, Sports and
Culture (Japan), and a Grant-in-Aid for Scientific Research (C;
19550109) from Japan Society for the Promotion of Science.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804188.
9458
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9458 –9461

Angewandte
Chemie
Table 2: Effect of the oxidant on the oxidative kinetic resolution of 1a.[a]
when there was no activation by the copper catalyst, oxidation
of 1a by NIS hardly proceeded in the absence of the
CuII/(R,R)-Ph-BOX complex. We believe that the difference
in reactivity of the oxidants and the accelerating effect of the
molecular recognition led to the high selectivity in the cases of
oxidation with NIS.
Next, we screened various N-protecting groups on the
2-piperidinecarboaldehydes (Table 4).[13] Subtrates with
acetyl groups and the alkoxycarbonyl groups (such as
Entry
Oxidant (equiv)
Yield [%] (ee [%])
(S)-1a
s
(R)-2a
(S)-3a
1
2
3
4
NBS (0.75)
NBPI[d] (0.5)
DBDMH[e] (0.5)
Br2 (0.5)
NCS (0.5)
NIS (0.5)
NIS (0.75)
NIS (1.0)
NIS (0.75)
I2 (0.5)
43 (79)
28 (84)
60 (53)
43 (45)
–
22 (97)
38 (97)
43 (91)
16 (99)
trace
20
17
–
–
–
–
–
–
63
–
31 (77)
51 (39)
31 (86)
19 (17)
48 (0)
65 (29)
60 (51)
51 (85)
–
20
17
7
8
5[b]
6[b]
7[b]
8[b]
9[b,c]
10[b]
11[b]
87
109
57
Table 4: Effect of N-protecting groups on the oxidative kinetic resolution
of 2-piperidinecarboaldehydes 1b–e.[a]
92 (0)
20 (30)
PhI(OAc)2 (0.5)
–
42
[a] A mixture of 1a (0.5 mmol), Cu(OTf)2 (0.05 mmol), (R,R)-Ph-BOX
(0.05 mmol), and oxidant (0.25, 0.375, or 0.50 mmol) in MeOH (2 mL)
was stirred at RT for 12 h. [b] Reaction time of 24 h. [c] At 08C.
[d] N-Bromo phthalimide. [e] 1,3-Dibromo-5,5-dimethylhydantoin.
Entry
PG
Yield [%] (ee [%])
(S)-1b–e
s
(R)-2b–e
(S)-3b–e
1
2
3
4
1b
Ac
39 (92)
26 (95)
37 (95)
–
–
–
–
81
47 (52)
63 (25)
63 (29)
19 (0)
40
50
52
1c
CO2Me
Cbz
Ts
1d[b]
1e
conversion was additioanlly improved as the amount of the
NIS used was increased, and the ee value of acetal 3a
improved. The use of 0.75 equivalents of NIS led to the
highest s value of 109 (Table 2, entry 7), whereas 1.0 equiv-
alent of NIS slightly decreased the enantioselectivity of 2a
(Table 2, entry 8). Although at 08C 2a was obtained in
99% ee, 3a was not detected (Table 2, entry 9). Other iodo
cationic species (I2, PhI(OAc)2) were not effective for the
oxidation of 1a (Table 2, entries 10 and 11).
[a] A mixture of 1b–e (0.5 mmol), Cu(OTf)2 (0.05 mmol), (R,R)-Ph-BOX
(0.05 mmol), and NIS (0.375 mmol) in MeOH (2 mL) was stirred at RT
for 24 h. [b] Yield was determined by 1H NMR analysis. Cbz=benzyloxy-
carbonyl, Ts=4-toluenesulfonyl.
CO2Me and Cbz), which are generally used as protecting
groups, led to desirable selectivities (Table 4, entries 1–3).
Since these values are comparable to those substrates having
a benzoyl group, these results might enhance the value of this
oxidative kinetic resolution. In contrast, the reaction of
N-tosylated amino aldehyde 1e did not afford the corre-
sponding methyl ester product 2e.
To confirm the accelerating effect of the recognition of 1a
by the CuII/(R,R)-Ph-BOX complex on the oxidation with
halogen cationic species (Table 3), we performed the reaction
Table 3: Accelerating effect based on recognition of 1a.[a]
The oxidation using NIS was applied to oxidative kinetic
resolution of the various acyclic amino aldehydes 4a–f
(Table 5).[14,15] The kinetic resolution of 4a–e proceeded
smoothly and excellent s values, which are attractive for
industrial applications, were attained. Particularly, the reac-
tion of amino aldehydes possessing linear alkyl groups gave
excellent selectivity (Table 5, entries 1–3 and 5). In the case of
Entry
Oxidant (equiv)
Condition[b]
Yield [%]
2a
1a
3a
1[c]
2[c]
3[d]
4[d]
NBS (0.5)
NBS (0.5)
NIS (0.75)
NIS (0.75)
A
B
A
B
13
22
trace
2
14
12
42
trace
58
62
45
81
[a] A mixture of 1a (0.5 mmol), Cu(OTf)2 (0 or 0.05 mmol), (R,R)-Ph-
BOX (0 or 0.05 mmol), and oxidant (0.25-0.375 mmol) in MeOH (2 mL)
was stirred at RT. [b] Condition A: In the absence of Cu(OTf)2 and (R,R)-
Ph-BOX. Condition B: In the absence of (R,R)-Ph-BOX. [c] Reaction time
of 12 h. [d] Reaction time of 24 h.
Table 5: Oxidative kinetic resolution of acyclic aminoaldehydes 4a–f.[a]
in the absence of both Cu(OTf)2 and (R,R)-Ph-BOX (Con-
dition A) and in the absence of (R,R)-Ph-BOX (Condi-
tion B). Similar tendencies were observed for both NBS and
NIS reactions. Condition A gave a much lower yield of 2a
(Table 3, entries 1 and 3), whereas Condition B (Table 3,
entries 2 and 4) led to a slight improvement in the yield
compared to that obtained with Condition A. However, the
reaction in the presence of Cu(OTf)2 and (R,R)-Ph-BOX
afforded 2a in a much higher yield, suggesting that 1a is
recognized by the CuII/(R,R)-Ph-BOX complex and thus
activated. Although oxidation with NBS proceeded even
Entry
R1
R2
Equiv of NIS
Yield [%] (ee [%])
(R)-5a–f (S)-6a–f
s
1
2
3
4
5
6
4a Me
4b Et
4c nPr
4d iPr
4e nBu
H
H
H
H
H
0.75
0.75
0.75
0.75
0.75
40 (97)
43 (99)
42 (94)
43 (82)
41 (96)
38 (55)
54 (65)
57 (64)
52 (69)
56 (50)
51 (71)
58 (25)
129
368
67
17
104
4
4 f
iPr
Me 2.0
[a] A mixture of 4a-f (0.5 mmol), Cu(OTf)2 (0.05 mmol), (R,R)-Ph-BOX
(0.05 mmol), and NIS (0.375 or 1.0 mmol) in MeOH (2 mL) was stirred
at RT for 24 h.
Angew. Chem. Int. Ed. 2008, 47, 9458 –9461
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9459

Communications
branched amino aldehyde 4d, the selectivity was within an
addition, optically active aminoaldehyde dimethyl acetals
that are easy to handle were obtained. Additional mechanistic
studies and synthetic applications are underway.
acceptable range (Table 5, entry 4) and the reaction of N,N-
disubstituted amino aldehyde 4 f required excess NIS because
of reduced reactivity (Table 5, entry 6), and gave moderate
selectivity.
A plausible reaction mechanism is shown in Scheme 2. We
believe that the high selectivity was obtained because of the
asymmetric oxidation of aminoaldehydes 1 and 4 associated
with the chiral copper catalyst was considerably faster than
that of the acid catalyzed acetalization of their noncoordi-
nated counterparts. In our previous study on oxidative kinetic
resolution,[9] bases performed an important role of neutraliz-
ing generated HX to accelerate the reaction. However, here
Received: August 25, 2008
Published online: October 30, 2008
Keywords: amino acids · amino aldehydes · copper ·
.
kinetic resolution · oxidation · synthetic methods
[1] a) G. M. Coppola, H. F. Schuster, Asymmetric Synthesis Con-
struction of Chiral Molecules Using Amino Acids, New York,
Wiley, 1987; b) F. J. Sardina, H. Rapo-
port, Chem. Rev. 1996, 96, 1825 – 1872.
[2] a) D. R. W. Hodgson, J. M. Sanderson,
Chem. Soc. Rev. 2004, 33, 422 – 430;
b) K. Izawa, T. Onishi, Chem. Rev. 2006,
106, 2811 – 2827.
[3] a) R. O. Duthaler, Tetrahedron 1994, 50,
1539 – 1650; b) R. C. Lloyd, M. C.
Lloyd, M. E. B. Smith, K. E. Holt, J. P.
Swift, P. A. Keene, S. J. C. Taylor, R.
McCague, Tetrahedron 2004, 60, 717 –
728; c) D. Arosio, A. Caligiuri, P. DꢀAr-
rigo, G. Pedrocchi-Fantoni, C. Rossi, C.
Saraceno, S. Servi, D. Tessaro, Adv.
Synth. Catal. 2007, 349, 1345 – 1348;
d) D. A. Schichl, S. Enthaler, W. Holla,
T. Riermeier, U. Kragl, M. Beller, Eur.
J. Org. Chem. 2008, 3506 – 3512.
[4] Hydrogenation: T. Ohkuma, M. Kita-
mura, R. Noyori in Catalytic Asymmet-
ric Synthesis, 2nd ed. (Ed.: I. Ojima),
Wiley-VCH, New York, 2000.
Scheme 2. Plausible reaction mechanism for the oxidative kinetic resolution of racemic N-
protected aminoaldehydes.
the oxidative kinetic resolution without any base improved
[5] Using chiral phase transfer catalysts: a) M. J. OꢀDonnel, W. D
Bennett, S. Wu, J. Am. Chem. Soc. 1989, 111, 2353 – 2355; b) E. J.
Corey, F. Xu, M. C. Noe, J. Am. Chem. Soc. 1997, 119, 12414 –
12415; c) T. Ooi, M. Kameda, K. Maruoka, J. Am. Chem. Soc.
1999, 121, 6519 – 6520.
[6] The Strecker reaction: a) H. Ishitani, S. Komiyama, S. Kobaya-
shi, Angew. Chem. 1998, 110, 3369 – 3372; Angew. Chem. Int. Ed.
1998, 37, 3186 – 3188; b) M. S. Sigman, E. N. Jacobsen, J. Am.
Chem. Soc. 1998, 120, 4901 – 4902; c) L. Yet, Angew. Chem. 2001,
113, 900 – 902; Angew. Chem. Int. Ed. 2001, 40, 875 – 877; d) S.
Masumoto, H. Usuda, M. Suzuki, M. Kanai, M. Shibasaki, J. Am.
Chem. Soc. 2003, 125, 5634 – 5635; e) H. Geoeger, Chem. Rev.
2003, 103, 2795 – 2827; f) T. Ooi, Y. Uematsu, Y. Maruoka, J. Am.
Chem. Soc. 2006, 128, 2548 – 2549.
the yield of the methyl esters (2 and 5) and the HX generated
acted as a catalyst for the transformation of noncoordinated
amino aldehydes. Since this acetalization might proceed
gradually, the ee values of acetals 3 and 6 are lower than
those of the corresponding methyl esters 2 and 5. It seems to
be reasonable that the Cu(OTf)2 was not involved in this
redox process by accepting one electron from the hemiacetal
oxygen. This hypothesis is supported by an experimental
result shown in Scheme 3, in which a chiral Zn(OTf)2/(R,R)-
[7] a) J. Liang, J. C. Ruble, G. C. Fu, J. Org. Chem. 1998, 63, 3154 –
3155; b) G. T. Notte, T. Sammakia, J. Am. Chem. Soc. 2006, 128,
4230 – 4231; c) Y. Ishii, R. Fujimoto, M. Mikami, S. Murakami, Y.
Miki, Y. Furukawa, Org. Process Res. Dev. 2007, 11, 609 – 615;
d) K. Ishihara, Y. Kosugi, S. Umemura, A. Sakakura, Org. Lett.
2008, 10, 3191 – 3194.
Scheme 3. Oxidative kinetic resolution of rac-1a with a chiral Zn-
(OTf)2/(R,R)-Ph-BOX complex.
[8] Asymmetric reactions catalyzed with CuII/Ph-BOX reported by
us: a) Y. Matsumura, T. Maki, S. Murakami, O. Onomura, J. Am.
Chem. Soc. 2003, 125, 2052 – 2053; b) M. Mitsuda, T. Tanaka, T.
Tanaka, Y. Demizu, O. Onomura, Y. Matsumura, Tetrahedron
Lett. 2006, 47, 8073 – 8077; c) K. Matsumoto, M. Mitsuda, N.
Ushijima, Y. Demizu, O. Onomura, Y. Matsumura, Tetrahedron
Lett. 2006, 47, 8453 – 8456; d) Y. Matsumura, D. Minato, O.
Onomura, J. Organomet. Chem. 2007, 692, 654 – 663; e) Y.
Demizu, K. Matsumoto, O. Onomura, Y. Matsumura, Tetrahe-
dron Lett. 2007, 48, 7605 – 7609; f) O. Onomura, M. Mitsuda,
M. T. T. Nguyen, Y. Demizu, Tetrahedron Lett. 2007, 48, 9080 –
Ph-BOX complex, lacking redox properties, catalyzed the
oxidative kinetic resolution of 1a to afford (R)-2a and (S)-3a
with moderate selectivity.
In conclusion, we have presented the first efficient
method for the kinetic resolution of racemic N-protected
amino aldehydes, which is based on the recognition by a
copper(II) /(R,R)-Ph-BOX complex to afford optically active
amino acid methyl esters with high enantiomeric excess. In
9460 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9458 –9461

Angewandte
Chemie
9084; g) Y. Demizu, Y. Kubo, Y. Matsumura, O. Onomura,
Synlett 2008, 433 – 437.
[9] a) O. Onomura, H. Arimoto, Y. Matsumura, Y. Demizu,
Tetrahedron Lett. 2007, 48, 8668 – 8672; b) D. Minato, H.
Arimoto, Y. Nagasue, Y. Demizu, O. Onomura, Tetrahedron
2008, 64, 6675 – 6683.
[10] The absolute stereoconfiguration of (R)-2a was determined by
comparison with the specific rotation of an authentic sample. See
reference [9b].
[12] H. B. Kagan, J. C. Fiaud, Topics in Stereochemistry, Vol. 18 (Ed.:
E. L. Eliel), Wiley, New York, 1988, pp. 249 – 330.
[13] Absolute stereoconfigurations of (R)-2b–d and (S)-3b–d shown
in Table 4 were deduced on the basis of that of (R)-2a.
[14] The absolute stereoconfiguration of (R)-5a was determined by
comparison with the specific rotation of an authentic sample. See
reference [9b]
[15] Absolute stereoconfigurations of (R)-5b–f and (S)-6a–f shown
in Table 5 were deduced on the basis of that of (R)-5a.
[11] Absolute stereoconfigurations of (S)-3a shown in Table 1 and
Table 2 were deduced on the basis of that of (R)-2a.
Angew. Chem. Int. Ed. 2008, 47, 9458 –9461
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9461