β-isocupreidine-catalyzed Baylis-Hillman reaction of chiral N-Boc-α-amino aldehydes

Article abstract of DOI:10.1021/ol0622561

β-Isocupreidine (β-ICD)-catalyzed Baylis-Hillman reaction of chiral N-Boc-α-amino aldehydes and 1,1,1,3,3,3-hexafluoroisopropyl acrylate (HFIPA) takes place without racemization and exhibits the match-mismatch relationship between the substrate and the catalyst. In the case of acyclic amino aldehydes, α,-substrates show excellent syn selectivity and high reactivity in contrast to L-substrates. On the other hand, in the case of cyclic amino aldehydes, D-substrates rather than L-substrates show excellent anti selectivity and high reactivity.

Full text of DOI:10.1021/ol0622561

ORGANIC
LETTERS
2
006
Vol. 8, No. 23
357-5360
â
-Isocupreidine-Catalyzed
Baylis Hillman Reaction of Chiral
N-Boc- -Amino Aldehydes
5
−
r
Ayako Nakano, Keisuke Takahashi, Jun Ishihara, and Susumi Hatakeyama*
Graduate School of Biomedical Sciences, Nagasaki UniVersity,
Nagasaki-852-8521, Japan
susumi@nagasaki-u.ac.jp
Received September 11, 2006
ABSTRACT
â
-Isocupreidine (
takes place without racemization and exhibits the match
amino aldehydes, -substrates show excellent syn selectivity and high reactivity in contrast to
of cyclic amino aldehydes,
â
-ICD)-catalyzed Baylis
−
Hillman reaction of chiral N-Boc-
mismatch relationship between the substrate and the catalyst. In the case of acyclic
-substrates. On the other hand, in the case
-substrates show excellent anti selectivity and high reactivity.
r-amino aldehydes and 1,1,1,3,3,3-hexafluoroisopropyl acrylate (HFIPA)
−
L
D
D-substrates rather than L
1
4
The Baylis-Hillman reaction of chiral R-amino aldehydes
has attracted considerable interest due to the synthetic utility
of the products having a multifunctional R-methylene-â-
hydroxy-γ-amino acid structure. However, the major problems
associated with this reaction are the racemization of the sub-
During the course of our synthesis of epopromycin B, we
demonstrated that â-isocupreidine (â-ICD)-catalyzed Baylis-
Hillman reaction of N-Fmoc-L-leucinal with 1,1,1,3,3,3-
5
hexafluoroisopropyl acrylate (HFIPA) took place without
racemization in DMF at -55 °C to produce the correspond-
ing adduct in excellent syn selectivity and in good yield.
Interestingly, N-Fmoc-D-leucinal was found to exhibit poor
reactivity and low diastereoselectivity, suggesting the match-
mismatch relationship between the substrate and the catalyst.
However, one serious drawback of this reaction is that 1
equiv of â-ICD is required to promote the reaction at a
reasonable rate. Recently, we found that the azeotropically
2
strate and the difficulty in securing high diastereoselectivity.
Recently, Coelho et al. successfully developed the racemiza-
tion-free DABCO-mediated reaction of chiral N-Boc-R-amino
aldehydes with methyl acrylate under the influence of
3
ultrasound; however, the diastereoselectivity was moderate.
(
1) For reviews, see: (a) Methot, J. L.; Roush, W. R. AdV. Synth. Catal.
2
004, 346, 1035-1050. (b) Basavaiah, D.; Rao, A. J.; Satyanarayana, T.
Chem. ReV. 2003, 103, 811-892. (c) Iwabuchi, Y.; Hatakeyama, S. J. Synth.
Org. Chem. Jpn. 2002, 60, 4-16. (d) Langer, P. Angew. Chem., Int. Ed.
(3) Coelho, F.; Diaz, G.; Abella, C. A. M.; Almeida, W. P. Synlett 2006,
435-439.
(4) Iwabuchi, Y.; Sugihara, T.; Esumi, T.; Hatakeyama, S. Tetrahedron
Lett. 2001, 42, 7867-7871.
(5) (a) Iwabuchi, Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama, S. J.
Am. Chem. Soc. 1999, 121, 10219-10220. (b) Iwabuchi, Y.; Furukawa,
M.; Esumi, T.; Hatakeyama, S. Chem. Commun. 2001, 2030-3031. (c)
Kawahara, S.; Nakano, A.; Esumi, T.; Iwabuchi, Y.; Hatakeyama, S. Org.
Lett. 2003, 5, 3103-3105. (d) Nakano, A.; Kawahara, S.; Akamatsu, S.;
Morokuma, K.; Nakatani, M.; Iwabuchi, Y.; Takahashi, K.; Ishihara, J.;
Hatakeyama, S. Tetrahedron 2006, 62, 381-389.
2
000, 39, 3049-3052. (e) Ciganek, E. Org. React. 1997, 51, 201-350. (f)
Basavaiah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron 1996, 52, 8001-8062.
Drewes, S. E.; Roos, G. H. P. Tetrahedron 1988, 44, 4653-4670.
(
2) (a) Ameer, F.; Drewes, S. E.; Houston-McMillan, M. S.; Kaye, P. T.
S. Afr. J. Chem. 1986, 39, 57-62. (b) Roos, G.; Manickum, T. Synth.
Commun. 1991, 21, 2269-2274. (c) Drewes, S. E.; Khan, A. A.; Rowland,
K. Synth. Commun. 1993, 23, 183-188. (d) Manickum, T.; Roos, G. H. P.
S. Afr. J. Chem. 1994, 47, 1-16. (e) Nayak, S. P.; Thijs, L.; Zwanenburg
Tetrahedron Lett. 1999, 40, 981-9984. (f) Almeida, B. W. P.; Coelho, F.
Tetrahedron Lett. 2003, 44, 937-940.
1
0.1021/ol0622561 CCC: $33.50
© 2006 American Chemical Society
Published on Web 10/20/2006

dried â-ICD displayed remarkable catalytic ability in asym-
5d
metric Baylis-Hillman reaction of aldehydes with HFIPA.
Table 1. â-ICD-Catalyzed Reaction of 2a-j with HFIPAa
This finding prompted us to examine this procedure for the
reaction of various chiral R-amino aldehydes to solve the
above-mentioned drawback. Furthermore, it is thought to be
of great significance to probe the influence of the chirality
of the substrate in diastereoselectivity from the mechanistic
point of view.
To start our study, N-Boc-R-amino aldehydes 2a-h were
6
prepared in high enantiomeric purity (>95% ee) by LiAlH
4
reduction of the Weinreb amides derived from N-Boc-R-
amino acids 1a-h according to the procedure developed by
7
Taylor et al. (Scheme 1). N-Boc-L- and -D-Garner aldehydes
Scheme 1. Preparation of Chiral N-Boc-R-Amino Aldehydes
b
c
c,d,e
yield (%) (syn/anti) [ee %]
time
(h)
net selectivity
5f 6g (de %)h
entry
1
2
3
4
a (L)
46 77 (100:0) 4 (75:25)
>98, -] [>98, >98]
46 83 (100:0) 3 (100:0)
>98, -] [>98, -]
17 63 (100:0) 18 (21:79)
>98, -] [>98, nd]
96 10 (100:0) 20 (0:100) 20 -
>98, -] [-, >98]
96 11 (94:6) 1 (13:87)
>98, -] [>98, 95]
b (D) 96 45 (0:100) 19 (69:31)
-, >98] [>98, >98]
d (D) 96 10 (5:95) 18 (76:24)
-, nd] [>98, nd]
96 37 (0:100) 15 (50:50)
[-, nd] [96, nd]
h (D) 96 73 (0:100) 8 (0:100)
-, >98] [-, >98]
73 67 (0:100) 9 (0:100)
-, >98] [-, >98]
-
-
-
-
-
-
syn (97)
syn (100)
syn (66)
anti (60)
anti (32)
anti (60)
syn (2)
[
2
3
4
5
6
7
8
9
c (L)
e (L)
g (L)
i (L)
[
[
[
30 -
[
-
-
-
-
-
2
[
-
8
6
2
i,j were obtained in >93% ee by DIBAH reduction of the
[
8
methyl esters of 1i,j.
f (D)
anti (75)
anti (100)
anti (100)
The Baylis-Hillman reactions of 10 N-Boc-R-amino
aldehydes 2a-j were then examined using 1.3 equiv of
HFIPA and 0.1 equiv of â-ICD in DMF at -55 °C. The
syn/anti ratio and optical purity of esters 3a-j and diox-
anones 4a-j were determined by HPLC analysis using a
chiral column of the corresponding methyl esters 7a-j
-
-
[
10
j (D)
[
a
Reactions were carried out at -55 °C in DMF (0.2 M) using 2 (1 equiv),
HFIPA (1.3 equiv), and â-ICD (0.1 equiv). b Isolated yield. Determined
c
(
Scheme 2). The stereochemistry of 7a-j was confirmed by
by HPLC analysis of the corresponding methyl ester 7 using a chiral column.
d
The enantiomeric purity of the syn and anti isomers. e nd: not determined
f
due to poor separation of the enantiomers by HPLC. The syn/anti ratio
g
was regarded to be the same as that of 4. The syn/anti ratio was regarded
to be the same as that of 3. h Calculated on the basis of the yields and the
syn/anti ratio of 3-6.
Scheme 2. Preparation of 7a-j and 8a-f
and N-Boc-L-valinal matched with the enantio preference (R-
selectivity) of â-ICD affording the excellent syn selectivity
and high chemical yields (entries 1 and 2), whereas the
corresponding D-amino aldehydes mismatched with â-ICD
showing moderate anti selectivity or the lack of diastereo-
selectivity and poor reactivity (entries 6 and 7). N-Boc-L-
alaninal also exhibited much higher reactivity than N-Boc-
D-alaninal, although moderate diastereoselectivity was observed
in both cases (entries 3 and 8). Interestingly, the cyclic
substrates showed the opposite match-mismatch relationship
to that observed for the acyclic substrates (entries 4, 5, 9,
and 10). Namely, D-substrates rather than L-substrates
matched well with â-ICD leading to the excellent anti
selectivity and good yields. It was also observed that in the
the NOESY spectra of oxazolidines 8a-f and the comparison
of the spectral data with those reported for 7a,b,e-j.
The results are summarized in Table 1, which reveals the
marked match-mismatch situation. Thus, N-Boc-L-leucinal
9
5358
Org. Lett., Vol. 8, No. 23, 2006

mismatched cases HFIPA reacted further with ester 3 or
dioxanone 4 to give 5 or 6 because of the longer reaction
time (entries 4-6 and 8). It is evident from the ee values
given that no racemization occurred during the reaction.
Table 2 shows that the appreciable racemization took place
above 0 °C but did not below -30 °C. It is also observed that
competition leading to the observed mismatched situation.
The match-mismatch relationship observed for the cyclic
amino aldehydes is consistent with the interpretation based
on the Felkin-Anh model of 15 and 16 (Scheme 3).1
c,d,8
Scheme 3. Plausible Reaction Mechanism
Table 2. â-ICD-Catalyzed Reaction of 2a with HFIPA at
Various Temperaturesa
b
c
ee (%)c
yield (%) (syn/anti)
temp time
entry (°C) (h)
net selectivity
(de %)d
3a
4a
syn anti
1
2
3
4
-55 46 77 (100:0) 4 (75:25) >98 >98
-30 46 80 (100:0) 6 (79:21) >98 95
syn (97)
syn (97)
syn (74)
syn (52)
0
96 52 (85:15) 16 (94:6)
97 70
86 66
rt 96 26 (76:24) 3 (75:25)
a
Reactions were carried out at the indicated temperature in DMF (0.2
b
M) using 2a (1 equiv), HFIPA (1.3 equiv), and â-ICD (0.1 equiv). Isolated
yield. Determined by HPLC analysis of methyl ester 7a using a chiral
column. Calculated on the basis of the yields and the syn/anti ratio of 3a
c
d
and 4a.
the total yield of the Baylis-Hillman products and the diaster-
eoselectivity (net de) markedly decreased above 0 °C.
At such temperatures, HFIPA was found to largely dimerize
to produce di(1,1,1,3,3,3-hexafluoroisopropyl) 2-methylene-
glutarate.
The stereochemical outcomes observed for the acyclic
amino aldehydes can be interpreted by assuming the cyclic
Cram model of 13 and 142c,d,10 governed by hydrogen bond-
ing. If an aldehyde approaches from the sterically less hin-
dered si face of E-enolate 911 in anti-periplaner fashion, in
the case of L-substrates, 13 is much more favorable than 14.
Consequently, the reaction proceeds predominantly via 13 to
produce zwitterionic intermediate 10, which undergoes the
1
2,13
E1cB process
via 12 (X ) R, Y ) H) to give the syn
adduct. It is assumed that, in the case of N-Boc-L-alaninal,
the energy difference between 13 and 14 would not be large
enough to lead to the high syn selectivity due to the sterically
less demanding methyl group. Thus, in addition to the major
reaction via 13 giving 10, the reaction via 14 largely occurs
to give zwitterionic intermediate 11. However, 11 suffers
from severe steric interactions between the substituent of the
aldehyde and the branched HFIPA ester in the E1cB step
(see 12 (X ) H, Y ) R)) and thus undergoes reaction with
a second aldehyde molecule rather than elimination to form
the dioxanone. On the contrary, in the case of D-amino
aldehydes, approach via either 13 or 14 is disfavored by the
steric interactions so that these two reaction courses are under
Scheme 4 exemplifies the synthetic utility of the above-
(6) Determined by HPLC analysis of the corresponding benzoate,
prepared by NaBH4 reduction followed by benzoylation.
mentioned methodology. Treatment of 3a,c,e,h with iodot-
Org. Lett., Vol. 8, No. 23, 2006
5359

Acknowledgment. We are grateful to Professor Fernando
Coelho (Instituto de Qu ´ı mica da Unicamp, Brazil) for
providing us with the spectral data of 7. This work was
supported in part by a Grant-in-Aid for Scientific Research
on Priority Areas (16073213), The Ministry of Education,
Culture, Sports, Science and Technology (MEXT), Japan.
Scheme 4. Synthesis of Functionalized Pyrrolidinones
Supporting Information Available: Experimental de-
tails, characterization data for 3a-j, 4a-j, 5g,i, 6b,f, 7a-j,
1
13
8
a-f, and 17a,c,e,h, and H and C NMR spectra of 7a-j
and 17a,c,e,h. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL0622561
rimethylsilane in situ generated from chlorotrimethylsilane
and NaI in acetonitrile at 0 °C promoted cleavage of the
Boc group and concomitant cyclization of the resulting amine
to give the highly functionalized pyrrolidinones 17a,c,e,h
almost quantitatively.
In conclusion, we have developed a highly diastereose-
lective racemization-free Baylis-Hillman reaction of chiral
N-Boc-R-amino aldehydes for the first time. The observed
matched-mismatched results allow us to propose a mech-
anism where both the diastereoselectivity of the aldol step
and the feasibility of the E1cB process are crucial. The
present work provides an effective method for the preparation
of highly enantiomerically pure R-methylene-â-hydroxy-γ-
amino acid derivatives.
(9) See reference 2c,d for 7e,f,i,j. 7a,b,g,h were determined by com-
1
13
parison with the H and C NMR data provided by Professor F. Coelho.
10) Dondoni, A.; Fantin, G.; Fagagnolo, M.; Pedrini, P. J. Org. Chem.
1990, 55, 1439-1446.
11) Because molecular mechanics calculations (MMFF, Macro Model
.5) reveal that the E-enolate is energetically more stable than the Z-enolate
(
(
8
-1
by over 6 kcal mol , the E-enolate is assumed to preferentially intervene
in the aldol reaction. Molecular mechanics calculations also suggest an
enolate structure where the CF3 group sticks out from the enolate plane as
depicted in Scheme 3. By such an enolate, it is understood why the branched
5d
structure of HFIPA is so influential on enantioselectivity as reported earlier.
12) This E1cB process is assumed to involve a six-membered proton
(
transfer from the phenolic OH to the alkoxide of zwitterionic intermediate
10 or 11 with concomitant deprotonation of the R-methine to give an enolate,
followed by elimination of â-ICD via 12 as proposed by Aggarwal et al.
Aggarwal, V. K.; Fulford, S. Y.; Lloyd-Jones, G. C. Angew. Chem. Int.
Ed. 2005, 44, 1706-1708.
(13) For recent reports on the reaction mechanism see ref 12 and: (a)
Santos, L. S.; Pavam, W. P.; Coelho, F.; Eberlin, M. N. Angew. Chem.,
Int. Ed. 2004, 43, 4330-4333. (b) Price, K. E.; Broadwater, S. J.; Jung, H.
M.; McQuade, D. T. Org. Lett. 2005, 7, 147-150. (c) Price, K. E.;
Broadwater, S. J.; Walker, B. J.; McQuade, D. T. J. Org. Chem. 2005, 70,
3980-3987. (d) Buskens, P.; Klankermayer, J.; Leitner, W. J. Am. Chem.
Soc. 2005, 127, 16762-16763.
(
7) Campbell, A. D.; Raynham, T. M.; Taylor, R. J. K. Synthesis 1998,
707-1709.
8) Garner, P.; Park, J. M. J. Org. Chem. 1987, 52, 2361-2364.
1
(
5360
Org. Lett., Vol. 8, No. 23, 2006