Kinetic resolution of terminal epoxides via highly regioselective and enantioselective ring opening with TMSN3. An efficient, catalytic route to 1,2-amino alcohols

Full text of DOI:10.1021/ja961708+

7420
J. Am. Chem. Soc. 1996, 118, 7420-7421
Scheme 1
Kinetic Resolution of Terminal Epoxides via Highly
Regioselective and Enantioselective Ring Opening
with TMSN3. An Efficient, Catalytic Route to
1,2-Amino Alcohols
Jay F. Larrow, Scott E. Schaus, and Eric N. Jacobsen*
Department of Chemistry and Chemical Biology
HarVard UniVersity, Cambridge, Massachusetts 02138
ReceiVed May 20, 1996
Scheme 2
The value of enantiopure 1,2-amino alcohols lies in their
utility both as intermediates for the synthesis of a wide range
of biologically important compounds1 and as precursors to
effective and versatile ligands for asymmetric catalysis.2 Exist-
ing synthetic routes to amino alcohols rely heavily on the chiral
pool,2b and particularly on the reduction of R-amino acids. While
this approach provides ready access to many 2-amino-1-ols, the
regioisomerically transposed 1-amino-2-ols cannot be prepared
by this method. Asymmetric routes to the latter, synthetically-
valuable compounds include the nitroaldol reaction,3 hydrocya-
nation of aldehydes,4 asymmetric dihydroxylation,5 and the ring
opening of enantiopure epoxides by amines or amine equiva-
lents.6 The latter approach is direct and particularly appealing,
especially in light of the advances in the development of
enantioselective catalytic methods for the synthesis of epoxides.7
Still, there are several classes of epoxides that are not easily
accessed by asymmetric catalysis, with terminal alkyl-substituted
epoxides certainly heading the list in terms of importance.
Given the accessibility of a wide range of racemic terminal
epoxides at low cost, a kinetic resolution strategy for the
synthesis of optically pure 1-amino-2-alkanols becomes viable.
In this communication, we report the application of the recently
developed (salen)Cr-catalyzed epoxide ring-opening reaction8
to the efficient synthesis of 1-azido-2-trimethylsiloxyalkanes
from racemic epoxides (Scheme 1). The viability of this strategy
is illustrated in practical syntheses of (S)-propranolol, a widely-
used anti-hypertensive agent,9 and (R)-9-[2-(phosphonomethox-
y)propyl]adenine (PMPA), a compound recently demonstrated
to display prophylactic activity against SIV infection.10
The reaction of neat (()-propylene oxide with 0.5 equiv of
TMSN3 in the presence of (salen)CrN3 complex (R,R)-1 (1 mol
%) resulted in clean conversion to a mixture of epoxide and
ring-opened product after 18 h at 0 °C. Removal of the highly
volatile unreacted epoxide by rotary evaporation, followed by
distillation of the residue at 24 °C/<1 mm Hg, afforded 1-azido-
2-trimethylsiloxypropane in essentially quantitative yield based
on theory and in 97% ee (Table 1, entry 1).11 Careful analysis
of the product by GC revealed that the regioisomeric product
was formed to an almost negligible extent (<1%).
This method was found to be applicable to a series of other
terminal epoxides (Table 1). Unbranched alkyl-substituted
epoxides proved to be the best substrates with regard to
enantioselectivity, with krel values exceeding 100 (entries 2, 3).12
The kinetic resolution of epichlorohydrin led to highly enriched
1-azido-3-chloro-2-trimethylsiloxypropane (entry 4), a reaction
that is particularly noteworthy given the availability of the
racemic substrate and the functional versatility of the optically
active product. Epoxides bearing branched alkyl substituents
displayed reduced reactivity but no loss of enantioselectivity
in the ring opening (entry 8). As has been demonstrated
previously in the context of meso epoxides,8 excellent functional
group tolerance was exhibited in these (salen)Cr-catalyzed
reactions (entries 9-11). The kinetic resolution of 3,3-
diethoxypropene oxide (entry 10) provided the ring-opened
product in 96% isolated yield and 89% ee, despite the presence
of a potentially epimerizable stereogenic center in the product.
(1) (a) Shioiri, Y.; Hamada, Y. Heterocycles 1988, 27, 1035. (b) Barlow,
C. B.; Bukhari, S. T.; Guthrie, R. D.; Prior, A. M. Asymmetry in
Carbohydrates; Dekker: New York, 1979; pp 81-99.
(2) (a) Pfaltz, A. In AdVances in Catalytic Processes; Doyle, M. P., Ed.;
JAI Press: Greenwich, CT, 1995; pp 61-94. (b) Blaser, H.-U. Chem. ReV.
1992, 92, 935. (c) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. ReV. 1994, 94, 2483. (d) Denmark, S. E.; Nakijima, N.; Nicaise, O.
J.-C.; Faucher, A.-M.; Edwards, J. P. J. Org. Chem. 1995, 60, 4884.
(3) (a) Sasai, H.; Suzuki, T.; Arai, S.; Arai, T.; Shibasaki, M. J. Am.
Chem. Soc. 1992, 114, 4418. (b) Sasai, H.; Suzuki, T.; Itoh, N.; Shibasaki,
M. Tetrahedron Lett. 1993, 34, 851. (c) Sasai, H.; Itoh, N.; Suzuki, T.;
Shibasaki, M. Tetrahedron Lett. 1993, 34, 855. (d) Sasai, H.; Suzuki, T.;
Itoh, N.; Arai, S.; Shibasaki, M. Tetrahedron Lett. 1993, 34, 2657. (e)
Shibasaki, M.; Sasai, H. J. Synth. Org. Chem., Jpn. 1993, 51, 972. (f) Sasai,
H.; Suzuki, T.; Itoh, N.; Tanaka, K.; Date, T.; Okamura, K.; Shibasaki, M.
J. Am. Chem. Soc. 1993, 115, 10372.
(9) (a) Wang, Z.-M.; Zhang, X.-L.; Sharpless, K. B. Tetrahedron Lett.
1993, 34, 2267. (b) Also see refs 6c and 3c.
(4) See (a) Effenberger, F. Angew. Chem., Int. Ed. Engl. 1994, 33, 1555.
(b) Nitta, H.; Yu, D.; Kudo, M.; Mori, A.; Inoue, S. J. Am. Chem. Soc.
1992, 114, 7969, and references therein.
(5) Chang, H.-T.; Sharpless, K. B. Tetrahedron Lett. 1996, 37, 3219. A
breakthrough in the direct, asymmetric catalytic synthesis of amino alcohols
from alkenes was also reported recently by the Sharpless group: Li, G.;
Chang, H.-T.; Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1996, 35,
451.
(6) (a) Askin, D.; Eng, K. K.; Rossen, K.; Purick, R. M.; Wells, K. M.;
Volante, R. P.; Reider, P. J. Tetrahedron Lett. 1994, 35, 673. (b) Hanson,
R. M. Chem. ReV. 1991, 91, 437. (c) Klunder, J. M.; Ko, S. Y., Sharpless,
K. B. J. Org. Chem. 1986, 51, 3710.
(10) (a) Tsai, C.-C.; Follis, K. E.; Sabo, A.; Beck, T. W.; Grant, R. F.;
Bischofberger, N.; Benveniste, R. E.; Black, R. Science 1995, 270, 1197.
(11) An oven dried 10 mL flask equipped with a stir bar was charged
with 32 mg (0.05 mmol) of (R,R) -1. The flask was sealed, purged with
N2, and cooled to 0 °C in an ice bath, followed by the sequential addition
of (()-propylene oxide (350 µL, 5.0 mmol) and TMSN3 (330 µL, 2.5
mmol). The reaction was allowed to stir at 0-2 °C for 18 h, upon which
time the remaining epoxide was removed by rotary evaporation, and the
desired product was vacuum distilled (24 °C/<1 mm Hg) into a cooled
collection flask to yield 0.425 g (2.45 mmol, 98%) of a colorless oil. Chiral
GC analysis (Cyclodex-B) indicated that the 1-azido-2-trimethylsiloxypro-
pane was obtained in 97% ee.
(7) (a) Jacobsen, E. N. In ComprehensiVe Organometallic Chemistry II;
Wilkinson, G., Stone, F. G. A., Abel, E. W., Hegedus, L. S., Eds.;
Pergamon: New York, 1995; Vol. 12, Chapter 11.1. (b) Johnson, R. A.;
Sharpless, K. B. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; VCH:
New York, 1993; Chapter 4.1. (c) Jacobsen, E. N. Ibid. Chapter 4.2. (d)
Katsuki, T. Coord. Chem. ReV. 1995, 140, 189.
(8) (a) Mart´ınez, L. E.; Leighton, J. L.; Carsten, D. H.; Jacobsen, E. N.
J. Am. Chem. Soc. 1995, 117, 5897. (b) Leighton, J. L.; Jacobsen, E. N. J.
Org. Chem. 1996, 61, 389.
(12) The values for krel were calculated using the equation krel ) ln[1 -
c(1 + ee)]/ln[1 - c(1 - ee)], where ee is the enantiomeric excess of the
product and c is the conversion. The conversion was set to equal the isolated
yield of the ring-opened product, so calculated values for krel actually
represent lower limits. For leading references on kinetic resolution, see:
(a) Eliel, E. L.; Wilen, S. H.; Mander, L. M. Stereochemistry of Organic
Compounds; Wiley-Interscience: New York, 1994; pp 395-415. (b) Kagan,
H. B.; Fiaud, J. C. In Topics in Stereochemistry; Allinger, N. L., Eliel, E.
L., Eds.; Interscience: New York, 1987; Vol. 14, p 249.
S0002-7863(96)01708-8 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 31, 1996 7421
Scheme 3a
aReagents: (a) 0.5 equiv TMSN3, 0.5 mol % (S,S)-1, 0 °C f rt; 98%, 97% ee; (b) CSA (cat.), MeOH, 92%; (c) 10% Pd/C, MeOH, H2, 91%;
(d) 5-amino-4,6-dichloropyrimidine, TEA, n-BuOH, ∆, 82%; (e) HC(OEt)3, conc. HCl, rt, 86%; (f) NH3, 65 °C, 91%; (g) NaH, DMF,
diethylphosphonomethoxytosylate, 0 °C f rt, 44%; (h) i. TMSBr, CH3CN, rt; ii. H2O, acetone, 4 °C, 92%.
Table 1. Kinetic Resolution of Monosubstituted Epoxides with
Azidotrimethylsilane Catalyzed by 1a
(Scheme 3). A desilylation/reduction sequence yielded the
synthetically important amino alcohol (R)-1-amino-2-propanol
in excellent yield. Further transformation of this compound to
(R)-PMPA was accomplished using known methods by conver-
sion of the amine to an adenine base,14 followed by alkylation
of the alcohol and standard deprotection of the phosphonate.15
In conclusion, the application of (salen)CrN3-catalyzed ep-
oxide ring-opening reaction to the kinetic resolution of racemic
terminal epoxides allows not only the recovery of epoxides with
high enantiomeric excess,8a but also the clean production of
1-azido-2-trimethylsiloxyalkanes in good yield and very high
enantiopurity. The synthetic utility of these amino alcohol
precursors, combined with the ready accessibility of the racemic
epoxides and of the catalyst,16 renders this an immediately useful
alternative to existing methods. In addition, the extremely high
enantioselectivity on such simple substrates as propylene oxide
with attendant broad substrate generality suggest very intriguing
mechanisms for chiral recognition by this catalyst system. This,
along with the expansion of this methodology to other classes
of nucleophiles and electrophiles, constitutes the subject of our
sustained efforts.
cat. mol
%
yield
(%)b
c
entry
R
ee (%)
krel
1
2
3
4
5
6
7
8
9
CH3
CH2CH3
(CH2)3CH3
CH2Cl
CH2OTBDMS
CH2O(1-naphthyl)
CH2C6H5
c-C6H11
(CH2)2CHdCH2
CH(OEt)2
1.0
2.0
2.0
2.0
3.0
5.0
2.0
2.0
2.0
2.0
2.0
98
83
89
94
96
74
94
84h
94
96
80
97d
97f
97d
95e
96e
93g
93f
97f
98d
89e
92e
230
140
160
100
150
48
71
140
280
44
10
11
CH2CN
45
a Reactions were run without solvent for 18-50 h at 0-2 °C.
b Isolated yield of the azido silyl ether based on TMSN3. c See ref 12.
d Determined by chiral GC analysis of the azido silyl ether. e Determined
by chiral GC analysis of the azido alcohol. f Determined by chiral GC
analysis of the azido O-acetate derivative. g Determined by chiral HPLC
analysis of the azido silyl ether. h Isolated yield of the azido alcohol.
Acknowledgment. We thank Ernest J. Prisbe from Gilead Sciences
for his helpful suggestions. This work was supported by the National
Institutes of Health (GM-43214) and by a generous gift from Versicor.
We thank the Packard Foundation, the Camille and Henry Dreyfus
Teacher-Scholar program, and the Sloan Foundation for awards to
E.N.J. and Eli Lilly for a predoctoral fellowship (J.F.L.).
Tolerance for Lewis basic functionality such as cyano groups
(entry 11) is also noteworthy.
A wide range of synthetic applications of this kinetic
resolution procedure are readily envisaged. The synthesis of
propranolol outlined in Scheme 2 serves to illustrate the efficient
and straightforward elaboration of the azido silyl ether products
of the ring-opening reaction to a biologically important amino
alcohol derivative. Kinetic resolution of the racemic epoxide
derived from epichlorohydrin and 1-naphthol afforded the
corresponding azido silyl ether in 74% isolated yield and in
93% ee (entry 6). In a one-pot, two-step procedure, transforma-
tion to (S)-propranolol was accomplished by desilylation fol-
lowed by azide reduction and in situ reductive alkylation in the
presence of acetone.3c
Supporting Information Available: Chromatographic analyses of
all racemic and enantiomerically enriched azido silyl ethers or deriva-
tives, full characterization of all isolated products, and complete
experimental procedures for the syntheses of (R)-PMPA and (S)-
propranolol (18 pages). See any current masthead page for ordering
and Internet access instructions.
JA961708+
(14) For preparation of 9-[2-(hydroxypropyl)]adenine see: Schaeffer, H.
J.; Vince, R. J. Med. Chem. 1967, 10, 689.
(15) For similar phosphonomethoxytosylate alkylation reactions, see: Yu,
K-L; Bronson, J. J.; Yang, H.; Patick, A.; Alam, M.; Brankovan, V.; Datema,
R.; Hitchcock, M. J. M.; Martin, J. C. J. Med. Chem. 1992, 35, 2958.
(16) For catalyst preparation, see ref 8b. The catalysts used in this study
were prepared using AgBF4 instead of AgClO4 but were identical to previous
catalyst batches in all other respects. For ligand preparation, see: Larrow,
J. F.; Jacobsen, E. N.; Gao, Y.; Hong, Y.; Nie, X.; Zepp, C. M. J. Org.
Chem. 1994, 59, 1939. Both enantiomers of the ligand are also commercially
available (Aldrich).
The synthesis of (R)-PMPA13 was effected similarly in a
highly efficient manner via kinetic resolution of propylene oxide
(13) (R)-PMPA has recently been synthesized from the chiral pool
utilizing D-(-)-lactic acid as the enantiopure starting material: Holy, A.;
Masoj´ıdkova´, M. Collect. Czech. Chem. Commun. 1995, 60, 1196.