An efficient enantioselective synthesis of (R,R)-formoterol, a potent bronchodilator, using lipases

Article abstract of DOI:10.1016/S0957-4166(00)00238-X

The potent β₂-adrenergic receptor agonist formoterol (R,R)-1 has been obtained in enantiomerically pure form by a convenient chemoenzymatic approach by coupling of epoxide (R)-6 with the unprotected primary amine (R)-9. Both chiral precursors have been prepared by enantiodifferentiation processes involving Pseudomonas cepacia (lipase PS) and Candida antarctica lipase (CALB), respectively. For the resolution of amine 9, we have found that utilization of triethylamine as non-reactive base enhances the reaction rate and the enantioselectivity of the process. The key coupling reaction of (R)-6 and (R)-9 has been conducted through derivatization of the amine with the labile trimethylsilyl group, which liberates the amino group of the resulting amino alcohol (R,R)-11 upon column chromatography purification. In this way, the overall approach is shorter than others previously described. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00238-X

Tetrahedron: Asymmetry 11 (2000) 2705±2717
An ecient enantioselective synthesis of (R,R)-formoterol, a
potent bronchodilator, using lipases
Francisco Campos, M. Pilar Bosch and Angel Guerrero*
Department of Biological Organic Chemistry, IIQAB (CSIC), Jordi Girona 18-26, 08034 Barcelona, Spain
Received 5 May 2000; accepted 15 June 2000
Abstract
The potent b2-adrenergic receptor agonist formoterol (R,R)-1 has been obtained in enantiomerically pure
form by a convenient chemoenzymatic approach by coupling of epoxide (R)-6 with the unprotected
primary amine (R)-9. Both chiral precursors have been prepared by enantiodi€erentiation processes
involving Pseudomonas cepacia (lipase PS) and Candida antarctica lipase (CALB), respectively. For the
resolution of amine 9, we have found that utilization of triethylamine as non-reactive base enhances the
reaction rate and the enantioselectivity of the process. The key coupling reaction of (R)-6 and (R)-9 has
been conducted through derivatization of the amine with the labile trimethylsilyl group, which liberates the
amino group of the resulting amino alcohol (R,R)-11 upon column chromatography puri®cation. In this
way, the overall approach is shorter than others previously described. # 2000 Elsevier Science Ltd. All
rights reserved.
1. Introduction
It is well known that b2-adrenergic receptor agonists are ecient bronchodilators in the
therapy of asthma and chronic bronchitis. A variety of b2-adrenoceptor agonists are currently
available, although some of them are not selective and produce important side e€ects, such as
isoproterenol (A),1 or display a short e€ect (seldom over 4±6 h), such as salbutamol (B) and
terbutaline (C).2 Formoterol (R,R)-1 (Fig. 1) is a new, long-lasting, b2-adrenoceptor agonist
which o€ers high selectivity for b2-adrenoceptors, a fast onset of action and an excellent safety
and tolerance pro®le.3 These compounds and other receptor agonists have in common a
substituted arylethanolamine functionality, an important feature in the structure of b3-adrenoceptor
agents which are being assayed in clinical trials for the treatment of obesity.4,5 Formoterol 1 is
commercialized in the racemic form, but the di€erent stereoisomers, corresponding to the two
stereogenic centers of the molecule, have di€erent pharmacological potencies, the order being
R,R>R,S>>S,R>S,S.6 In spite of its interesting properties, very few syntheses of the most
* Corresponding author . Fax: 34-93-2045904; e-mail: agpqob@iiqab.csic.es
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00238-X

2706
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
Figure 1.
active R,R enantiomer of formoterol have been described employing a resolution process of one
pair of enantiomers,6 a HPLC-semipreparative separation of diastereomers,7 with low overall
yields in both cases, and an asymmetric borane reduction using a chiral oxazaborolidine as
reducing agent.8
Enzyme-catalyzed reactions have great potential for the synthesis of enantiomerically pure
products through kinetic resolution of the racemic precursors or by asymmetrization of prochiral
substrates.9 Continuing our e€orts directed to the synthesis of biologically active compounds
through lipases,10 we report herein an enantioselective synthesis of formoterol (R,R)-1 in which
the required chiral synthons (R)-6 and (R)-9 are generated from convenient enantiomeric
resolutions through lipases.
2. Results and discussion
Epoxide (R)-6, the ®rst chiral synthon needed, had been previously prepared through
enantioselective borane reduction of the corresponding ketone catalyzed by chiral cis-1-amino-2-
indanol.8 This process required expensive starting reagents, so a more simple and lower-cost
procedure was envisioned. Bromination of methyl ketone 211 was carried out by bubbling a
stream of nitrogen onto the reaction mixture to remove HBr and thus minimizing formation of
the dibrominated product. The reaction furnished the corresponding mono-bromoketone 3,1
which was subjected to a conventional borane reduction to obtain bromoalcohol (RS)-4 in 71.3%
overall yield. A number of reports have been found in the literature to obtain optically active
halohydrins. They comprise enantioselective oxazaborolidine,12 borane13 or microbial
reduction14 of a-haloketones, enantioselective hydrolysis of a-halo acetates,15 and resolution of
halohydrins by acetylation.16 In our case, bromoalcohol (RS)-4 was subjected to an enzyme-
directed enantiodi€erentiation process by treatment with some commercially available lipases.
Although porcine pancreatic lipase (PPL, Sigma) has been reported to be useful for resolution of
benzylic alcohols,17 in our case the enzyme did not recognize compound 4 as a good substrate for
the esteri®cation and no acetate was detected after 22 h reaction (Table 1). However, Pseudomonas
cepacia (lipase PS, Amano), either as commercially received or immobilized on Celite, eciently
resolved the racemic compound to provide alcohol (R)-4 and acetate (S)-5 in good yield and ee.
The resolution was carried out on a thermostated bath at 37ꢀC in tert-butyl methyl ether as
solvent and vinyl acetate as acylating agent. The lipase:alcohol ratio was 2:1 in weight when the
crude enzyme was used, and 0.2:1 for the immobilized enzyme. The amount of vinyl acetate was

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2707
10 equiv. with respect to the substrate and the reaction was stopped when ca. 50% conversion
was achieved (Table 1). The mixture was ®ltered and the non-reactive alcohol (R)-4 separated
from the acetate (S)-5 by column chromatography. (R)-4 was obtained in 48% yield and 96% ee
as determined by integration of the CF3 signals of the 19F NMR spectrum of the corresponding
Mosher ester with (S)-(+)-MTPA chloride (S,R diastereomer at ꢀ ^71.70; R,R diastereomer at ꢀ
^71.92 ppm). Acetate (S)-5 was hydrolyzed to the expected alcohol (S)-4 but this spontaneously
cyclized to epoxide (S)-6 under the basic conditions used (K2CO3/MeOH) in 84% overall yield. The
enantiomeric purity of the latter compound (77% ee) was calculated in base to its speci®c rotation8
and the enantiomeric ratio of the enzymatic process (E=65) determined from the ee of the residual
substrate and the extent of conversion.18 Celite-immobilized lipase PS also resulted in an excellent
resolving agent with the additional advantage of requiring lower amounts of lipase, ca. one order of
magnitude, than that used for the non-immobilized enzyme (E=111) (Table 1). Cyclization of
unreacted alcohol (R)-4, under identical conditions to those used for hydrolysis/cyclization of (S)-5,
a€orded the required intermediate epoxide (R)-6 in 89% yield and 94% ee (Fig. 2).
Table 1
Resolution of (RS)-4 through lipases
ꢀ
ꢀ
.
Figure 2. (a) Br2, CHCl3, N2 bubbling, 25 C, 82%; (b) BH3 THF (1.2 equiv.), THF, ^20 C, 87%; (c) lipase, t-BuOMe,
vinyl acetate, 37ꢀC; (d) anh. K2CO3, MeOH, 25ꢀC, 2.5 h, 89% for (R)-6, 84% for (S)-6
Synthesis of racemic amine (RS)-9, a well-known amphetamine derivative,19 is outlined in
Fig. 3. Nitrostyrene 8 was eciently prepared by Henry reaction of anisaldehyde with nitroethane
in the presence of ammonium acetate20 at re¯ux for a short time (35 min, 81% yield). Reduction

2708
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
of nitroderivative 8 with lithium aluminum hydride in ether directly provided amine (RS)-9 in
78% yield. This yield resulted as being superior to others involving the intermediate formation of
a ketoxime19 or the catalytic reduction with Pd/C at high pressure.21 Other procedures, such as
utilization of RED-AL [sodium bis-(2-methoxyethoxy)aluminum dihydride] in benzene,20 in our
hands gave inferior results (only 54% isolated yield), due to the rather dicult separation of the
salts formed during the work-up of the reaction.
Figure 3. (a) EtNO2, NH4AcO (1.25 equiv.), re¯ux 35 min, 81%; (b) LiAlH4, Et2O, re¯ux 2 h, 78%; (c) RED-AL,
benzene, re¯ux 2.5 h, 54%; (d) CALB, AcOEt, Et3N cat., 30ꢀC; (e) KOH 3 M, re¯ux, 3 h, 68%
Unlike the enzymatic resolution of alcohols, which has become a very useful strategy to obtain
both enantiomers simultaneously,9 selective enzymatic acylation of amines is less appealing since
two drawbacks arise. First, amines are much more nucleophilic than alcohols and may react in a
conventional way with the acyl donor (normally esters) used in the enzymatic process, and
second, after acylation of the amine the deacylation process very often requires harsh conditions
to liberate the free amine. This precludes utilization of enzymatic deacylation of amides and also
restricts further reaction of the chiral amide after the enzymatic acylation. Therefore, and to
circumvent these drawbacks, several approaches for the enzymatic resolution of amines have been
developed.22 In our case we decided to initially test those enzymes (subtilisin Carlsberg, subtilisin
BPN0, etc.) which preferentially recognize the S-enantiomer leaving the required R-enantiomer
unreacted. However, subtilisin Carlsberg protease failed to esterify racemic (RS)-9 with
tri¯uoroethyl butyrate in anh. 3-methyl-3-pentanol22a after 12 days of incubation at 30ꢀC (only
7% of conversion). Better results were obtained with subtilisin BPN0 and diallyl carbonate in
phosphate bu€er 0.1M (pH 8.0)22b for 15 h at room temperature (50% conversion), but the
enantioselectivity was poor (22% ee for the unreactive (R)-9 by 19F NMR spectra of the
diastereomeric Mosher amide) (Table 2). We then turned our attention to the utilization of
lipases, which, in contrast to proteases, show preference to acylate the R-enantiomer. Incubation
of our substrate with Candida antarctica lipase B (CALB) in the presence of ethyl butyrate as
transesteri®cation reagent yielded the desired butylated amide, which resulted in being extremely
resistent to hydrolysis even under acid (no reaction after 24 h re¯ux in the presence of tri-
¯uoroacetic acid or conc. HCl) or enzymatic conditions (no reaction with CALB in phosphate
bu€er pH=7.5). Use of ethyl acetate as acyl donor and solvent was more encouraging, since the
obtained amide (R)-10 could be successfully hydrolyzed by re¯uxing overnight with a 3 M KOH
soln. The resolution was performed at 30ꢀC for 15 h to give unreactive (S)-9 (47% yield, 68% ee)
and amide (R)-10, which was hydrolyzed as described to furnish (R)-9 in 32% overall yield and
84% ee. Determination of the enantiomeric excess was based on the 19F NMR spectra of the

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2709
diastereomeric Mosher amides (R,R diastereomer at ꢀ ^69.21; S,R diastereomer at ꢀ ^69.51 ppm).
In order to enhance the enantioselectivity of the resolution we followed two di€erent approaches:
sequential kinetic resolution, and use of a non-reactive organic base. Utilization of an
appropriate sequential kinetic resolution is an useful tool to increase the enantiomeric purity of a
speci®c enantiomer.10a,23 In our case, a batch of racemic amine (RS)-9 was treated with CALB in
ethyl acetate for 18 h. After this time a 48% conversion was achieved, and the resulting amide
was puri®ed and hydrolyzed to the corresponding amine (R)-9 (26% overall yield, 74% ee). This
latter amine was subjected to a new enzymatic resolution for 19 h at 69% conversion, and the
new amide hydrolyzed to provide chiral amine (R)-9 in 11% overall yield from the racemic
material with a good enantioselectivity (96% ee) (Table 2).
Table 2
Resolution of amine (RS)-9 through lipases
The ability of bases to increase the enantioselectivity in enzymatic resolutions has been
previously documented. Thus, in transesteri®cation reactions of alcohols the base was added to
neutralize any acetic acid liberated by hydrolysis of the acylating agent, particularly vinyl
acetate,24 or the carboxylic acid formed when acid anhydrides are used as acyl donors.25 Rakels et
al.,26 on the other hand, utilized non-reactive amines in water-saturated organic solvents to
enhance the ee and yield of enzyme-mediated ester hydrolysis, while Turner et al. used
triethylamine in CALB-promoted alcoholysis of (þ)-2-phenyl-4-benzyloxal-5(4H)-one27a and to
catalyze resolution of other 4-substituted oxazolones with Mucor miehei.27b With these precedents
in mind, we decided to test triethylamine as catalyst to enhance the enantioselectivity of the
resolution. To our satisfaction, the presence of 0.15 equiv. of this amine with regard to the
substrate provided, after only 4 h reaction, a 42% conversion of racemic amine (RS)-9 into the
corresponding amide (R)-10. Hydrolysis of the latter under the same conditions cited above led to
the required amine (R)-9 in 21% overall yield and 94% ee (E=66) (Table 2). To our knowledge,
this is the ®rst time that a non-reactive amine is described to catalyze the reaction rate and
enantioselectivity of an enzyme-mediated resolution of amines.

2710
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
As cited above, the b-amino alcohol function is a common structural unit of a variety of
bioactive natural products, in particular in the pharmaceutical ®eld, and therefore a variety of
methods have been developed for their synthesis.28 The classical procedure, reaction of an
epoxide with an excess of amine in a protic solvent, presents a number of limitations, such as the
requirement of an excess of amine, elevated temperatures and poor yields when non-nucleophilic
(e.g. aromatic amines) or bulky amines are used.29 When a primary amine is used, another
drawback arises: the lack of regioselectivity. Thus, while nucleophilic attack at the less substituted
carbon to give the desired b-amino alcohols usually predominates, signi®cant amounts of the
a products as well as bis-alkylation adducts are also obtained.30 To overcome these drawbacks,
Overman protected primary amines as dialkylaluminum amides31 and Weigel as N-(trimethyl-
silyl)amines.30 More recently, these authors almost suppressed bis-alkylation products by using 1
equiv. of bis-(trimethylsilyl)acetamide (BSA) and an amine:epoxide ratio of 1:1.2.32 In our case, a
mixture of amine (R)-9 and BSA (1.1 equiv.) in anh. DMSO was stirred for 30 min under
nitrogen. Then epoxide (R)-6 (1 equiv.) in DMSO was added and the mixture stirred for 87 h at
75±80ꢀC. After work-up, the crude was puri®ed by column chromatography on neutral alumina,
which promoted concomitant hydrolysis of the silylated intermediate (Fig. 4) to provide amino
alcohol 11 in 80% overall yield from (R)-9. Compound 11 was a mixture of only two (R,R and
R,S), out of the four possible diastereomers, in 75:25 ratio by integration of the 1H and 13C NMR
signals of the benzylic protons next to the hydroxyl group. (R,R)-11 presented a doublet of
doublets at ꢀ 4.53 ppm (J=9.0, J0=3.6 Hz) while the same signal in (R,S)-11 appeared at ꢀ 4.61
ppm. In the 13C NMR the aliphatic carbons at ꢀ 70.3, 54.4, 54.0, 42.8 and 20.3 ppm appeared
duplicated accounting for the presence of two diastereomers.
Figure 4. (a) BSA, anh. DMSO, 25ꢀC, 30 min; (b) (R)-6, DMSO, 80ꢀC, 87 h; (c) neutral Al2O3 III, 80% overall from
(R)-9; (d) Fe, 1 M HCl, MeOH re¯ux 45 min, 67%; (e) anh. HCOOH, py, 60ꢀC, 6.5 h, 69%; (f) H2, Pd/C, EtOH, 3 kg/
cm2, rt overnight, 73%
Although the present approach is not highly diastereoselective, it is convenient since it does not
need to protect/deprotect amine (R)-9 as described8,33 and provides good yields of the new amino
alcohol (R,R)-11 in one step. At this stage this compound could not be separated from the R,S
diastereomer, so we decided to proceed forward and free the active material from the undesired
diastereomer in the next step. Reduction of the nitro group of (R,R)-11 was easily achieved by
treatment with iron turnings (4 equiv.) and 1 M HCl in MeOH. Careful column chromatographic
puri®cation on neutral alumina (activ. IV) a€orded diastereomerically pure amine (R,R)-12 in

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2711
67% yield. Selective N-formylation of the primary amino group (R,R)-12 was accomplished by
reaction of the amine with an excess of anh. formic acid in pyridine at 60ꢀC for 6.5 h to provide
(R,R)-13 in 69% yield after puri®cation on column chromatography. The progress of the reaction
was followed by HPLC using a Spherisorb ODS 25 mm column. It should be noted that under the
more classical conditions, such as treatment with a mixture of formic acid and acetic anhydride
(3:2) in CHCl3,33 a mixture of the corresponding mono N- and O-formyl compounds along with
the bis-formylated derivative was obtained. Formyl amine (R,R)-13 was obtained as a 70:30
mixture of the syn and anti conformers as deduced from its spectroscopic features. Thus, in the
13C NMR spectrum signals at ꢀ 158.7 and 161.2 ppm were assigned to the carbonyl groups of the
syn and anti conformers, respectively. In the 1H NMR spectrum the NH signals, which
disappeared upon treatment with D2O, appeared as broad singlets at ꢀ 8.75 and 7.82 ppm and
were attributed to the anti and syn conformers, respectively. Particularly noteworthy is the highly
deshielding e€ect promoted by the carbonyl on the vicinal aromatic proton of the syn conformer
(ꢀ 8.39) while the corresponding proton of the anti conformer resonates at ꢀ 7.22 ppm. Finally,
deprotection of the benzyl group of (R,R)-13 with Pd/C in EtOH at moderate pressure a€orded
(R,R)-1 as a yellow amorphous solid in 73% isolated yield, after puri®cation on a reverse-phase
Isolute SPE C18 column. The (l)-(+)-tartrate salt of (R,R)-1 showed an enantiomeric purity
ꢁ99% based on its speci®c rotation ‰ꢁŠ2D0=^29.4 (H2O, c 0.61) in comparison to the reported
value.34
3. Conclusion
In summary, a convenient new synthesis of enantiomerically pure formoterol (R,R)-1 was
achieved through the enzyme-directed preparation of the chiral precursors epoxide (R)-6 and
amine (R)-9. For this amine we found that use of triethylamine e€ectively catalyzes the reaction
rate and enantioselectivity of the process. This e€ect has been noticed in other processes but, to
our knowledge, not for resolution of primary amines. The synthetic approach is shorter than
others previously reported since in situ derivatization of amine (R)-9 with the labile trimethylsilyl
group avoids the generally assumed protection/deprotection of the primary amine. Further
work to study the scope of the use of non-reactive amines as catalyst in the enzymatic
resolution of amines is in progress in our laboratory and the results will be reported in due
course.
4. Experimental
Melting points were determined on a Ko‚er apparatus and are uncorrected. Elemental
analyses were carried out on Carlo Erba models 1106, 1107 and 1500. IR spectra were recorded
on a Bomem MB-120 instrument with Fourier transform. [1H] and [13C] NMR spectra were
obtained in CDCl3 solutions on a Varian XL-200 and Unity 300 spectrometers, operating at 200
and 300 MHz for [1H] and 50 and 75 MHz for [13C]. The values are expressed in ꢀ scale relative to
internal Me4Si. [19F] NMR spectra were recorded on a Varian Unity 300 instrument operating at
282 MHz and the values are reported in ꢀ scale relative to trichloro¯uoromethane as internal
standard. GC analyses were performed on Carlo Erba model 4130, equipped with a FID detector,
using a BPX 35 (SGE) 25mÂ0.25 mm ID fused silica capillary column and hydrogen as carrier

2712
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
gas. Optical rotations were measured on Perkin±Elmer 141 and 341 polarimeters operating at 589
nm. Analytical-grade reagents were obtained from commercial suppliers and were used directly
without further puri®cation. Anhydrous tetrahydrofuran, benzene and diethyl ether were
prepared by drying with solid KOH followed by distillation from Na/benzophenone under N2,
n-hexane, dimethylsulfoxide and methylene chloride by distillation from CaH2, triethylamine and
pyridine by distillation from KOH, and formic acid by re¯ux and distillation from phthalic
anhydride.
4.1. Acylation of 1-(40-benzyloxy-30-nitro)phenyl-2-bromoethanol (RS)-4 with Pseudomonas
cepacia lipase. Preparation of (R)-(^)-4 and (S)-(+)-5
In a 10 ml Erlenmeyer ¯ask was placed a mixture of 101 mg (0.29 mmol) of (RS)-4 in 6.3 mL of
tert-butyl methyl ether, 202 mg of Pseudomonas cepacia lipase (lipase PS) and 0.247 g (2.87 mmol)
of vinyl acetate. The Erlenmeyer-¯ask was capped, placed in a thermostated bath at 37ꢀC and
shaken at 80 U/min. The reaction was monitored by TLC and when the transformation was 52%
(72 h reaction), the mixture was ®ltered o€ and the enzyme washed with diethyl ether. The solvent
was stripped o€ and the resulting crude puri®ed by column chromatography on silica gel eluting
20
with a hexane±CH2Cl2 mixture to furnish 56.8 mg (50%) of acetate (S)-(+)-5; 77% ee, ‰ꢁŠ
D
=+42.9 (CHCl3, c 2.84) and 48.4 mg (48%) of unreactive alcohol (R)-(^)-4, 96% ee, ‰ꢁŠ2D0=^33.6
(CHCl3, c 2.42). (R)-(^)-4: Rf: 0.09 (SiO2, hexane:CH2Cl2, 1:3). Elem. anal. calcd for
C15H14NO4Br: C, 51.16; H, 4.01; N, 3.98; Br, 22.69. Found: C, 51.27; H, 4.03; N, 3.96; Br, 22.64.
1
IR, ꢂ: 3531, 3066, 3033, 1622, 1575, 1531, 1353, 1280, 1265, 1018, 1000, 908, 736, 698 cm^1. H
NMR (300 MHz, CDCl3), ꢀ: 7.90 (d, J=2.1 Hz, 1H), 7.52 (dd, J=8.7 Hz, J0=2.4 Hz, 1H), 7.47±
7.26 (c, 5H), 7.12 (d, J=8.7 Hz, 1H), 5.24 (s, 2H), 4.92 (dt, J=8.7 Hz, J0=3.3 Hz, 1H), 3.62 (dd,
J=10.5 Hz, J0=3.6 Hz, 1H), 3.49 (dd, J=10.5 Hz, J0=8.4 Hz, 1H), 2.75 (d, J=3.6 Hz, 1H)
ppm.13C NMR (75 MHz, CDCl3), ꢀ: 151.6, 139.8, 135.2, 133.1, 131.5, 128.6, 128.2, 126.9, 123.3,
115.2, 72.1, 71.1, 39.3 ppm. (S)-(+)-(5): Rf: 0.09 (SiO2, hexane:CH2Cl2, 1:3). IR, ꢂ: 3066, 3033,
1
1747, 1622, 1533, 1454, 1371, 1355, 1286, 1263, 1236, 1022, 912, 819, 736 cm^1. H NMR (300
MHz, CDCl3), ꢀ: 7.88 (d, J=2.4 Hz, 1H), 7.50 (dd, J=8.6 Hz, J0=2.4 Hz, 1H), 7.47±7.31 (c, 5H),
7.12 (d, J=8.7 Hz, 1H), 5.93 (dd, J=7.20Hz, J0=5.7 Hz, 1H), 5.25 (s, 2H), 3.63 (dd, J=10.8 Hz,
J0=7.2 Hz, 1H), 3.56 (dd, J=10.8 Hz, J =5.4 Hz, 1H), 2.15 (s, 3H) ppm. 13C NMR (75 MHz,
CDCl3), ꢀ: 169.6, 152.1, 140.0, 135.2, 132.6, 130.4, 128.8, 128.4, 126.9, 124.0, 115.1, 73.2, 71.2,
33.5, 20.9 ppm.
4.2. Enantiomeric excess of alcohol (R)-(^)-4 and acetate (S)-(+)-5
(R)-(+)-a-Methoxy(tri¯uoromethyl)phenylacetic acid was converted into the acid chloride as
previously described.35 A sample of the alcohol (R)-(^)-4 (2.3 mg) was mixed with 134 ml of a
9.6Â10^2 M soln. of (S)-(+)-MTPA chloride in anh. CH2Cl2, 6 ml of Et3N and two crystals of
DMAP. The mixture was stirred for 2 h at room temperature under nitrogen, after which time no
starting material was detected on TLC. After evaporation of the solvent, direct 19F NMR (282
MHz, CDCl3) spectrum of the crude diastereomeric esters allowed calculation of the ee by
integration of the CF3 signals (S,R diastereomer, ꢀ: ^71.70 (s); R,R diastereomer, ꢀ: ^71.92 (s)).
The calculated enantiomeric purity of the alcohol (R)-(^)-4 was 96%. The ee of acetate (S)-(+)-5
was determined by hydrolysis and cyclization into the corresponding epoxide (S)-6 and
calculation of the enantiomeric excess of the latter (see below).

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2713
4.3. Acylation of 1-(40-benzyloxy-30-nitro)phenyl-2-bromoethanol (RS)-4 with Pseudomonas
cepacia lipase immobilized on Celite
When immobilized lipase PS was used, starting from 4.30 g (12.2 mmol) of alcohol (RS)-4, 0.87
g of immobilized lipase and 11.4 mL of vinyl acetate dissolved in 18 mL of tert-butyl methyl ether
were obtained, after 99 h reaction at 37ꢀC and puri®cation as above, 2.30 g (48%) of acetate (S)-
(+)-5, ‰ꢁŠ2D0=+47.0 (CHCl3, c 2.23) (ee=86%) and 1.96 g (46%) of unreactive alcohol (R)-(^)-4,
‰ꢁŠ2D0=^33.6 (CHCl3, c 2.81) (ee=96%).
4.4. (R)-4-Benzyloxy-3-nitrostyrene oxide (R)-(^)-6
A mixture of 1.81 g (5.14 mmol) of alcohol (R)-(^)-4 and 1.07 g (7.71 mmol) of anh. K2CO3 in
49 mL of MeOH was stirred for 2.5 h at room temperature, at which time there was no starting
material left by TLC. The mixture was diluted with water and thoroughly extracted with CH2Cl2.
The combined organic phases were washed with water, dried (MgSO4) and the solvent removed
under vacuum. The crude was puri®ed by column chromatography on neutral alumina (act. III)
eluting with hexane±CH2Cl2 mixtures to leave 1.24 g (89%) of epoxide (R)-(^)-6, ‰ꢁŠ2D0=^12.1
(CHCl3, c 1.05) (ee=94%). M.p.: 86.0±87.5ꢀC. Rf: 0.46 (SiO2, CH2Cl2). Elem. anal. calcd for
C15H13NO4: C, 66.41; H, 4.83; N, 5.16. Found: C, 66.37; H, 4.82; N, 5.20. IR, ꢂ: 3056, 3037, 1625,
1575, 1531, 1498, 1355, 1278, 862, 736, 696 cm^1. 1H NMR (300 MHz, CDCl3), ꢀ: 7.79 (d, J=2.2 Hz,
1H), 7.48±7.26 (c, 6H), 7.10 (d, J=8.6 Hz, 1H), 5.24 (s, 2H), 3.85 (dd, J=3.8 Hz, J0=2.6 Hz, 1H),
3.15 (dd, J=5.1 Hz, J0=4.2 Hz, 1H), 2.76 (dd, J=5.3 Hz, J0=2.4 Hz, 1H) ppm. 13C NMR (75 MHz,
CDCl3), ꢀ: 151.7, 140.5, 135.4, 130.9, 130.6, 128.7, 128.3, 127.0, 123.0, 115.4, 71.3, 51.1 ppm.
4.5. (S)-4-Benzyloxy-3-nitrostyrene oxide (S)-(+)-6
Following a similar procedure as described for the preparation of epoxide (R)-6, hydrolysis of
bromoacetate (S)-(+)-5 (56.8 mg, 0.15 mmol) with K2CO3 in MeOH produced the expected
bromoalcohol (S)-(+)-4, which concomitantly cyclized to give 33.1 mg (84%) of (S)-(+)-6,
‰ꢁŠ2D0=+9.9 (CHCl3, c 1.66) (ee=77% by comparison with the literature value).34
4.6. Resolution of amine (RS)-9 with Candida antarctica lipase catalyzed by triethylamine
In a 10 mL Erlenmeyer-¯ask were placed a solution of 111.2 mg (0.67 mmol) of (RS)-9 in 2.0
mL of anh. ethyl acetate, 27.8 mg of Candida antarctica lipase and 13 mL of triethylamine. The
Erlenmeyer-¯ask was capped, placed in a thermostated bath and shaken at 30ꢀC and 80 U/min.
The reaction was monitored by TLC and when the transformation was ca. 50% (4 h, 42%
conversion), the mixture was ®ltered o€ and the enzyme washed with ethyl acetate, CH2Cl2 and
MeOH. The solvent was stripped o€ and the resulting crude dissolved in 15 mL of CH2Cl2,
acidi®ed with 1 M HCl and repeatedly extracted with CH2Cl2. The organic phases were washed
with water, dried (MgSO4), ®ltered and freed from solvent. The residue was puri®ed by column
chromatography on silica gel eluting with hexane±ethyl acetate mixtures to give acyl amine (R)-10
(43.9 mg, 31%). M.p.: 88.0±91.0ꢀC. Rf: 0.42 (SiO2, AcOEt:MeOH, 10:1). IR, ꢂ: 3327, 1635, 1610,
1
1537, 1512, 1242, 1033, 813 cm^1. H NMR (300 MHz, CDCl3), ꢀ: 7.09 (d, J=8.6 Hz, 2H), 6.83
(d, J=8.6 Hz, 2H), 5.78 (bd, J=7.8 Hz, 1H), 4.19 (m, 1H), 3.78 (s, 3H), 2.78 (dd, J=13.5 Hz,
J0=5.7 Hz, 1H), 2.63 (dd, J=13.6 Hz, J0=7.2 Hz, 1H), 1.92 (s, 3H), 1.09 (d, J=6.8 Hz, 3H) ppm.

2714
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
13C NMR (75 MHz, CDCl3), ꢀ: 169.3, 158.0, 130.2, 129.9, 113.6, 55.1, 46.1, 41.3, 23.3, 19.7 ppm.
The aqueous layer, containing unreactive amine (S)-9 as the hydrochloride, was made basic with
2 M NaOH, extracted with CH2Cl2, washed with water and dried (MgSO4). After ®ltration,
evaporation of the solvent left chiral amine (S)-9 as an oil (64.3 mg, 58%).
4.7. Hydrolysis of acetamide (R)-10. Amine (R)-9
A mixture of amide (R)-10 (43.9 mg) and 11 mL of a 3 M KOH aq. soln. was re¯uxed
overnight. Completion of reaction was assessed by TLC analysis. The mixture was then diluted
with water, extracted with CH2Cl2 and washed with water. After drying (MgSO4), the solvent was
removed to obtain 23.8 mg (68%) of amine (R)-9 (94% ee, see below). Rf: 0.08 (SiO2,
1
CH2Cl2:MeOH, 5:1). IR, ꢂ: 3361, 3284, 2958, 1612, 1583, 1514, 1247, 1035, 804 cm^1. H NMR
(300 MHz, CDCl3), ꢀ: 7.11 (d, J=8.6 Hz, 2H), 6.85 (d, J=8.6 Hz, 2H), 3.79 (s, 3H), 3.10 (m, 1H),
2.66 (dd, J=13.3 Hz, J0=5.3 Hz, 1H), 2.44 (dd, J=13.4 Hz, J0=8.0 Hz, 1H), 1.10 (d, J=6.4 Hz,
3H), 1.22 (b, 2H) ppm. 13C NMR (75 MHz, CDCl3), ꢀ: 158.1, 131.8, 130.1, 113.8, 55.2, 48.6, 45.8,
23.5 ppm.
4.8. Enantiomeric excess of amines (S)-9 and (R)-9
Following a similar procedure as for alcohol (R)-4, formation of the diastereomeric Mosher
amides of (S)-9 and (R)-9 allowed determination of the ee of the latter compounds by their 19F
NMR spectra (282 MHz, CDCl3) (R,R-diastereomer, ꢀ: ^69.21 (s). S,R-diastereomer, ꢀ: ^69.51
ppm). The enantiomeric purities of (R)-9 and (S)-9 were 94 and 56%, respectively.
4.9. Coupling reaction of amine (R)-9 with epoxide (R)-6. Preparation of (R,R)-11
A mixture of 0.75 g (4.5 mmol) of amine (R)-9, 22.5 mL of anh. DMSO and 640 mL of N,O-bis-
(trimethylsilyl)acetamide was stirred under nitrogen for 30 min at room temperature. Then 1.23 g
(4.5 mmol) of epoxide (R)-6 dissolved in 22.5 mL of anh. DMSO were added and the resulting
mixture heated at 80ꢀC for 87 h. The mixture was cooled, the solvent distilled o€ (0.1 mm) and
the resulting crude puri®ed by column chromatography in neutral alumina (activity III) eluting
with hexane±ethyl acetate mixtures. Compound 11 was obtained as an orange gum (1.35 g, 80%
overall yield from (R)-9) as a mixture of diastereomers R,R and R,S in 75:25 ratio. Rf: 0.40 (SiO2,
AcOEt:MeOH, 3:1). ‰ꢁŠ2D0=^39.8 (CHCl3, c 1.21). Elem. anal. calcd for C25H28N2O5: C, 68.79; H,
6.47; N, 6.42. Found: C, 68.95; H, 6.46; N, 6.58. IR, ꢂ: 3282, 3068, 2933, 2835, 1620, 1531, 1512,
1454, 1352, 1276, 1245, 1035, 910, 817, 732 cm^1. 1H NMR (300 MHz, CDCl3), ꢀ: 7.83 (d, J=2.1
Hz, 1H), 7.47±7.29 (c, 6H), 7.07 (d, J=8.4 Hz, 2H), 7.05 (d, J=8.7 Hz, 1H), 6.83 (d, J=8.7 Hz,
2H), 5.21 (s, 2H), 4.53 (dd, J=9.0 Hz, J0=3.6 Hz, 1H), 3.78 (s, 3H), 2.90±2.84 (c, 2H), 2.7 (b, 2H)
2.64±2.57 (c, 3H), 1.08 (d, J=6.3 Hz, 3H) ppm. 13C NMR (75 MHz, CDCl3), ꢀ: 158.1, 151.1,
140.0, 135.6, 135.4, 131.3, 130.9, 130.1, 128.7, 128.2, 127.0, 123.1, 115.1, 113.9, 71.2, 70.3, 55.2,
54.4, 54.0, 42.8, 20.3 ppm.
4.10. Amine (R,R)-12
To a solution of 406 mg (0.93 mmol) of (R,R)-11 in 3.5 mL of methanol were added 208 mg
(3.72 mmol) of iron turnings and 1.2 mL of 1 M HCl. The mixture was brought to re¯ux for

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2715
45 min, after which time no starting material was detected by TLC. After cooling, the mixture
was diluted with 5 mL of MeOH and the precipitate ®ltered under vacuum. The ®ltrate was made
basic with 1 M NaOH and ®ltered again. The ®ltrate was concentrated, diluted with water and
extracted with CH2Cl2. The organic phase was washed with water and dried (MgSO4). After
®ltration and evaporation of the solvent, the residue was puri®ed by careful column chromatography
on neutral alumina (activity IV) eluting with hexane±ethyl acetate mixtures to provide amine
(R,R)-12 as a yellow gum (253 mg, 67%). Rf: 0.24 (SiO2, AcOEt:MeOH, 2:1). ‰ꢁŠ2D0=^28.2
(CHCl3, c 1.01). IR, ꢂ: 3468, 3369, 3197, 3031, 2931, 1614, 1512, 1247, 1220, 1035, 808, 754 cm^1.
1H NMR (300 MHz, CDCl3), ꢀ: 7.46±7.29 (c, 5H), 7.07 (d, J=8.7 Hz, 2H), 6.83 (d, J=8.7 Hz,
2H), 6.80 (d, J=8.4 Hz, 1H), 6.74 (d, J=2.1 Hz, 1H), 6.64 (dd, J=8.3 Hz, J0=2.0 Hz, 1H), 5.06
(s, 2H), 4.47 (dd, J=8.7 Hz, J0=3.9 Hz, 1H), 3.83 (b, 2H), 3.78 (s, 3H), 2.88 (dd, J=6.5 Hz,
J0=6.5 Hz, 1H), 2.83 (dd, J=11.9 Hz, J0=4.1 Hz, 1H), 2.68 (dd, J=11.9 Hz, J0=8.9 Hz, 1H),
2.65 (dd, J=13.4 Hz, J0=6.9 Hz, 1H), 2.55 (dd, J=13.5 Hz, J0=6.6 Hz, 1H), 1.05 (d, J=6.3 Hz,
3H) ppm. 13C NMR (75 MHz, CDCl3), ꢀ: 158.0, 145.8, 137.1, 136.4, 135.7, 131.3, 130.2, 128.5,
128.0, 127.5, 115.7, 113.7, 112.6, 111.8, 71.7, 70.5, 55.2, 54.4, 54.3, 42.8, 20.3 ppm. HRMS: calcd
mass for C25H30N2O3: 406.225643; Found: 406.224823.
4.11. Formylation of amine (R,R)-12
To a solution of 243 mg (0.60 mmol) of amine (R,R)-12 in 15 mL of anh. pyridine were added
3.6 mL of anh. formic acid and the mixture heated at 60ꢀC for 6.5 h under an inert atmosphere.
Progress of the reaction was followed by HPLC on a Spherisorb ODS column 25 mm 12.5Â0.46
cm, isocratic CH3CN:bu€er soln. 65:35. The bu€er soln. contained Et3N±AcOH (40 mM) for
pH=6.8 (Rt of (R,R)-12: 4.9 min, (R,R)-13: 3.9 min). The solvent was evaporated o€ and the
residue diluted with 1 M Na2CO3 soln. and extracted with CH2Cl2. The organic phases were
washed with water, dried (MgSO4), ®ltered and the solvent removed. The residue was puri®ed by
column chromatography on neutral alumina (activity IV), eluting with hexane±ethyl acetate
mixtures, to a€ord 180 mg (69%) of formyl amine (R,R)-13 as a mixture of syn and anti (70:30)
conformers. Rf: 0.24 (SiO2, AcOEt:MeOH, 2:1). ‰ꢁŠ2D0=^37.4 (CHCl3, c 1.68). IR, ꢂ: 3398, 3327,
1
3064, 3033, 2960, 2931, 1693, 1610, 1598, 1531, 1512, 1247, 1035, 788, 761 cm^1. H NMR (300
MHz, CDCl3), ꢀ: 8.75 (b, 1H anti conformer), 8.39 (d, J=1.8 Hz, 1H syn conformer), 8.33 (d,
J=2.1 Hz, 1H syn conformer), 7.82 (b, 1H syn conformer), 7.44±7.34 (c, 5H), 7.22 (d, J=1.8 Hz,
1H, anti conformer), 7.12 (d, J=2.1 Hz, 1H, anti conformer), 7.08 (dm, J=8.7 Hz, 3H), 6.93 (dd,
J=8,4 Hz, J0=1.5 Hz, 1H), 6.83 (dm, J=8.7 Hz, 2H), 5.08 (s, 2H), 4.53 (dd, J=9.0 Hz, J0=3.6
Hz, 1H syn), 4.50 (dd, J=9.0 Hz, J0=3.6 Hz, 1H anti), 3.78 (s, 3H), 2.91±2.80 (c, 2H), 2.72±2.52
(c, 3H), 1.08 (d, J=6.0 Hz, 1H anti), 1.05 (d, J=6.3 Hz, 1H syn). 13C NMR (75 MHz, CDCl3), ꢀ:
161.2, 158.7, 158.0, 146.9, 146.3, 136.1, 135.8, 135.7, 131.2, 131.0, 130.14, 130.10, 128.7, 128.5,
128.4, 127.9, 127.6, 126.7, 126.4, 122.3, 121.4, 118.2, 113.8, 112.4, 111.3, 71.4, 71.1, 70.94, 70.86,
55.2, 54.4, 54.3, 54.2, 42.8, 20.4, 20.3 ppm.
4.12. Formoterol (R,R)-1
A mixture of 89 mg (0.20 mmol) of (R,R)-13 dissolved in 3.4 mL of abs. ethanol and 76 mg
(0.072 mmol) of 10% Pd/C was hydrogenated at 3 kg/cm2 and room temperature overnight. The
mixture was ®ltered under vacuum and the catalyst thoroughly washed with ether, ethanol and
methylene chloride. Removal of the solvent left a residue which was chromatographed on a

2716
F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
reverse-phase column (1 g, Isolute SPE C18) using mixtures of MeOH±H2O as eluent. The solvent
was evaporated o€ and the aqueous solution lyophilized (^35ꢀC, 0.6 bar) overnight to obtain 51.5
mg (73%) of formoterol (R,R)-1 as amorphous solid. Rf: 0.27 (SiO2, AcOEt:MeOH, 1:1).
‰ꢁŠ2D0=^41.5 (CHCl3, c 0.53). IR, ꢂ: 3383, 2967, 2923, 1674, 1668, 1610, 1514, 1442, 1247, 1033,
815 cm^1. 1H NMR (300 MHz, CDCl3), ꢀ: 8.11 (b, 1H), 7.46 (b, 1H), 6.99 (d, J=8.4 Hz, 2H), 6.9±
6.7 (c, 4H), 4.46 (m, 1H), 4.34 (b, 3H interchangeable), 3.74 (s, 3H), 2.90±2.45 (c, 5H), 1.02 (d,
J=5.7 Hz, 3H) ppm. 13C NMR (75 MHz, CDCl3), ꢀ: 160.2, 158.3, 147.7, 133.4, 130.6, 130.2 (2C),
125.7, 123.7, 119.5, 117.8, 114.0 (2C), 71.3, 55.3, 54.7, 53.6, 42.0, 19.4 ppm. CI (positive, LC-MS)
(m/z, %) 435 (M+1, 100). The tartrate salt was prepared by dissolving 13.8 mg (0.04 mmol) of
(R,R)-1 and 6.0 mg (0.04 mmol) of (l)-(+)-tartaric acid in 150 mL of 85% aqueous isopropanol.
The solution was left standing overnight and the resulting crystalline solid (7.6 mg) puri®ed on a
reverse-phase column (1 g, Isolute SPE C18) using mixtures of MeOH±H2O as eluent. The solvent
was removed under vacuum and the aqueous solution lyophilized (^35ꢀC, 0.6 bar) overnight. The
(l)-(+)-tartrate salt of (R,R)-1 showed an ‰ꢁŠ2D0=^29.4 (H2O, c 0.61) (>99% ee based on the
reported value 34).
Acknowledgements
We gratefully acknowledge CICYT (AGF 97-1217-C0201), EC (FAIR CT96-0836) and
Comissionat per a Universitats i Recerca (1999 SGR00187) for ®nancial support. We also thank
Amano Pharmaceutical Co. and Novo Nordisk for supply of Pseudomonas cepacia and Candida
antarctica lipases.
References
1. Kaiser, C.; Colella, D. F.; Schwartz, M. S.; Garvey, E.; Wardell Jr., J. R. J. Med. Chem. 1974, 17, 49.
2. Anderson, G. P. Life Sciences 1993, 52, 2145.
3. Pawels, R. Formoterol, a new generation of beta-2-agonist; Hogrefe and Huber Pub.: Toronto, 1990.
4. Lundell, K.; Kanerva, L. T. Tetrahedron: Asymmetry 1995, 6, 2281.
5. Kordik, C. P.; Reitz, A. B. J. Med. Chem. 1999, 42, 186.
6. Murase, K.; Mase, T.; Ida, H.; Takahashi, K.; Murakami, M. Chem. Pharm. Bull. 1978, 26, 1123.
7. Trofast, J.; Osterberg, K.; Kallstrom, B.-L.; Waldeck, B. Chirality 1991, 3, 443.
8. Hett, R.; Fang, Q. K.; Gao, Y.; Hong, Y.; Butler, H. T.; Nie, X.; Wald, S. A. Tetrahedron Lett. 1997, 38, 1125.
9. (a) Chen, C. S.; Sih, C. J. Angew. Chem., Int. Ed. Engl. 1989, 28, 695. (b) Faber, K. Biotransformations in Organic
Chemistry; Springer-Verlag: Heidelberg, 1992. (c) Klibanov, A. M. Acc. Chem. Res. 1990, 23, 114. (d) Scho€ers,
E.; Golebiowski, A.; Johnson, C. R. Tetrahedron 1996, 52, 3769.
10. (a) Jimenez, O.; Bosch, M. P.; Guerrero, A. J. Org. Chem. 1997, 62, 3496. (b) Gil, N.; Bosch, M. P.; Guerrero, A.
Tetrahedron 1997, 53, 15115. (c) Petschen, I.; Malo, E. A.; Bosch, M. P.; Guerrero, A. Tetrahedron: Asymmetry
1996, 7, 2135. (d) Petschen, I.; Bosch, M. P.; Guerrero, A. Tetrahedron: Asymmetry 2000, 11, 1691.
11. Larsen, A. A.; Gould, W. A.; Roth, H. R.; Comer, W. T.; Uloth, R. H.; Dungan, K. W.; Lish, P. M. J. Med.
Chem. 1967, 10, 462.
12. Hull, K. G.; Visnick, M.; Tautz, W.; She€ron, A. Tetrahedron 1997, 53, 12405.
13. Hong, Y.; Gao, Y.; Nie, X.; Zepp, C. M. Tetrahedron Lett. 1994, 35, 6631.
14. Wei, Z. L.; Li, Z. Y.; Lin, G. Q. Tetrahedron 1998, 54, 13059.
15. Conde, S.; Fierros, M.; Rodrõguez-Franco, M. I.; Puig, C. Tetrahedron: Asymmetry 1998, 9, 2229.
16. (a) Igarishi, Y.; Otsutomo, S.; Harada, M.; Nakano, S.; Watanabe, S. Synthesis 1997, 549. (b) Rotticci, D.;
Orrenius, C.; Hult, K.; Norin, T. Tetrahedron: Asymmetry 1997, 8, 359. (c) Adam, W.; Blancafort, L.; Saha-

F. Campos et al. / Tetrahedron: Asymmetry 11 (2000) 2705±2717
2717
Moller, C. R. Tetrahedron: Asymmetry 1997, 8, 3189. (d) Skupin, R.; Cooper, T. G.; Frohlich, R.; Prigge, J.;
Haufe, G. Tetrahedron: Asymmetry 1997, 8, 2453.
17. Morgan, B.; Oehlschlager, A. C.; Stokes, T. M. J. Org. Chem. 1992, 57, 3231.
18. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294.
19. Biel, J. H.; Schwarz, E. G.; Sprengeler, E. P.; Leiser, H. A.; Friedman, H. L. J. Am. Chem. Soc. 1954, 76, 3149.
20. Butterick, J. R.; Unrau, A. M. J. Chem. Soc., Chem. Commun. 1974, 307.
21. Ohshita, T.; Ando, H. Jpn. J. Toxicol. Environ. Health 1992, 38, 571.
22. (a) Kitaguchi, H.; Fitzpatrick, P. A.; Huber, J. E.; Klibanov, A. M. J. Am. Chem. Soc. 1989, 111, 3094. (b) Orsat,
B.; Alper, P. B.; Moree, W.; Mak, C.-P.; Wong, C.-H. J. Am. Chem. Soc. 1996, 118, 712. (c) Iglesias, L. E.;
Sanchez, V. M.; Rebolledo, F.; Gotor, V. Tetrahedron: Asymmetry 1997, 8, 2675. (d) Takayama, S.; Lee, S. T.;
Hung, S.-C.; Wong, C.-H. Chem. Commun. 1999, 127.
23. Wang, Y.-F.; Chen, C.-S.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1984, 106, 3695.
24. Stead, P.; Marley, H.; Mahmoudian, M.; Webb, G.; Noble, D.; Ip, Y. T.; Piga, E.; Rossi, T.; Roberts, S.; Dawson,
M. J. Tetrahedron: Asymmetry 1996, 7, 2247.
25. Berger, B.; Rabiller, C. G.; Konigsberger, K.; Faber, K.; Griengl, H. Tetrahedron: Asymmetry 1990, 1, 541.
26. Rakels, J. L. L.; Straathof, A. J. J.; Heijnen, J. J. Tetrahedron: Asymmetry 1994, 5, 93.
27. (a) Parker, M.-C.; Brown, S. A.; Robertson, L.; Turner, N. J. Chem. Commun. 1998, 2247. (b) Turner, N. J.;
Winterman, J. R.; McCague, R.; Parratt, J. S.; Taylor, S. J. C. Tetrahedron Lett. 1995, 36, 1113.
28. (a) Kunieda, T.; Ishizuka, T. Stud. Nat. Prod. Chem. 1993, 12, 411. (b) Reetz, M. T.; Drewes, M. W.; Lennick, K.;
Schmitz, A.; Holdgrun, X. Tetrahedron: Asymmetry 1990, 1, 375. (c) Koppenhoefer, B.; Trettin, U.; Wachtler, A.
Synthesis 1994, 1141. (d) Deloux, L.; Srebnik, M. Chem. Rev. 1993, 93, 763.
29. (a) Moller, F. In Methoden der Organischen Chemie (Houben-Weyl); Muller, E., Ed.; Thieme-Verlag: Stuttgart,
1957; Vol. 11/1, pp. 311. (b) Freifelder, M.; Stone, G. R. J. Org. Chem. 1961, 26, 1477.
30. Atkins, R. K.; Frazier, J.; Moore, L. L.; Weigel, L. O. Tetrahedron Lett. 1986, 27, 2451.
31. Overman, L. E.; Flippin, L. A. Tetrahedron Lett. 1981, 22.
32. Shuker, A. J.; Siegel, M. G.; Matthews, D. P.; Weigel, L. O. Tetrahedron Lett. 1997, 38, 6149.
33. Murase, K.; Mase, T.; Ida, H.; Takahashi, K.; Murakami, M. Chem. Pharm. Bull. 1977, 25, 1368.
34. Hett, R.; Senanayake, C. H.; Wald, S. A. Tetrahedron Lett. 1998, 39, 1705.
35. Ward, D. E.; Rhee, C. K. Tetrahedron Lett. 1991, 32, 7165.