Chemo-enzymatic synthesis of the active enantiomer of the anorressant 2- benzylmorpholine

Article abstract of DOI:10.1016/S0957-4166(98)00432-7

Baker's yeast reduction of (Z)-α-bromocinnamaldehyde 1 in the presence of absorbing resins allows the easy preparation of the corresponding saturated bromo-alcohol 2 in high yields and enantiomeric excess. The absolute configuration is assigned through conversion into the (R)-phenyl oxirane 3 and 1-phenyl-2-propanol 4 of (S) absolute configuration. The (R)- epoxide is transformed into 6, the pharmacologically active enantiomer of the appetite suppressant 2-benzylmorpholine, to which the (R) configuration is assigned.

Full text of DOI:10.1016/S0957-4166(98)00432-7

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 4021–4026
Chemo-enzymatic synthesis of the active enantiomer of the
anorressant 2-benzylmorpholine
Paola D’Arrigo, Maria Lattanzio, Giuseppe Pedrocchi Fantoni and Stefano Servi ∗
CNR Centro di Studio sulle Sostanze Organiche Naturali and Dipartimento Chimica Politecnico Milano, Via Mancinelli, 7,
20131 Milano, Italy
Received 21 September 1998; accepted 26 October 1998
Abstract
Baker’s yeast reduction of (Z)-α-bromocinnamaldehyde 1 in the presence of absorbing resins allows the easy
preparation of the corresponding saturated bromo-alcohol 2 in high yields and enantiomeric excess. The absolute
configuration is assigned through conversion into the (R)-phenyl oxirane 3 and 1-phenyl-2-propanol 4 of (S)
absolute configuration. The (R)-epoxide is transformed into 6, the pharmacologically active enantiomer of the
appetite suppressant 2-benzylmorpholine, to which the (R) configuration is assigned. © 1998 Elsevier Science Ltd.
All rights reserved.
1. Introduction
The ability of microorganisms to reduce stereoselectively triply substituted C_C double bonds of α,β-
unsaturated carbonyl compounds is a unique property of some yeasts and bacteria and has been exploited
in the preparation of enantiomerically enriched compounds.1–17 Recently homogeneous hydrogenation
catalysts have been successfully used for the same purpose.18 Because of the mild and environmentally
friendly reaction conditions, whole cell biocatalysts are used in stereoselective reduction even on an
industrial scale.19,20 We have previously shown, however, that compounds of type 1 undergo reduction
with unsatisfactory yields and selectivity, giving rise to the formation of several reduction products
with incomplete enantioselectivity.21 Recently we have applied absorbing resins in what we define as
extractive biocatalysis and have shown that by controlling the substrate concentration of hydropho-
bic substrates, not only is high product recovery achieved, but improved enantio-chemoselectivity is
obtained.22,27 In this article we report on the baker’s yeast reduction of 1, resulting in the preparation
of 2 of (S) absolute configuration and its transformation into benzylmorpholine 6, a potent appetite
suppressant drug. Compound 6 has previously been prepared in a racemic form and the two enantiomers
∗
Corresponding author. E-mail: servi@dept.chem.polimi.it
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00432-7

4022
P. D’Arrigo et al. / Tetrahedron: Asymmetry 9 (1998) 4021–4026
resolved with (−)-dibenzoyltartaric acid.23 Appetite suppressant tests on dogs showed that the biological
activity completely resides in the enantiomer showing positive rotation in plane polarised light. The
absolute configuration of the active enantiomer was not assigned.23
2. Results and discussion
Baker’s yeast reduction of 1 was reported to give the bromo-alcohol 2 of undefined enantiomeric
excess, to which the (S) absolute configuration was assigned by analogy with similar products.24 The
study of the yeast reduction of 1 in different operating conditions shows that the enantiomeric purity of
the saturated bromo-alcohol obtained strongly depends on the initial substrate concentration. With an
initial substrate concentration of 5 g/l the enantiomeric excess was only 62.7%, but could be raised to
91.8% when the concentration was kept as low as 0.1 g/l. A behaviour of this type can be explained in
terms of competition of the same substrate between similar enzymes with different Km.25,26 Dilutions
of this order cannot be considered for preparatively useful biotransformations. Absorbing resins are able
to effectively control the substrate activity in the medium. High substrate concentration on the resin still
results in a low substrate activity in the water phase and in the yeast environment, resulting in a direct
control of the enantiomeric excess of the product: with a resin:substrate ratio of 1 and an initial substrate
concentration of 5 g/l, compound 2 was obtained in nearly quantitative yield and 98.6% e.e.27
Fig. 1 summarises the enantiomeric excess of the product 2 obtained in different conditions, as
a function of the initial substrate concentration. The control of the enantiomeric excess by means
of absorbing resins added is evident. Already at a resin:substrate ratio of 0.1, the increase in ee of
2 is dramatic (95.6%), although a slight improvement is still observed by increasing the ratio. The
improved performance makes it possible to work at practically interesting concentrations (5 g/l). A higher
resin:substrate ratio improves the efficiency of product extraction. In this case, solvent extraction can be
minimised. The enantiomeric excess of 2 was evaluated from chiral GC and comparison with the racemic
bromohydrines obtained from allylbenzene and N-bromosuccinimide† in water.23
Treatment of 2 with base gave the epoxide 3 which was then converted into 1-phenyl-2-propanol
4 of (S) absolute configuration, as established by comparison with physical data from the literature.28
Compound 3 of high enantiomeric excess could easily be prepared on a 10 g scale. The same epoxide
†
NBS treatment of allylbenzene gives a mixture of isomeric bromoalcohols in racemic form: they are separated into four well
resolved signals in GC on chiral stationary phase (see experimental).

P. D’Arrigo et al. / Tetrahedron: Asymmetry 9 (1998) 4021–4026
4023
Figure 1. Enantiomeric excess of 2 as a function of initial concentration of 1: with no resin added (A); with addition of
XAD1180, resin:substrate ratio (g:g) (B) 0.1; (C) 0.5; (D) 1; (E) 2; (F) 5
can be prepared in about 25% yield in a tedious three step procedure from the corresponding amino-
acid, involving nitrous acid treatment, LAH reduction and base cyclisation.29,30 Treatment of 3 with
ethanolamine sulphate gave 5 as the major product in a mixture with the nitrogen substituted compound,
from which it could be purified by column chromatography after base catalysed ring closure to 6.23
Compound 6, showing a positive rotation sign, was identified as the biologically active enantiomer of
the appetite suppressant drug. Since the hydrochloride of compound 6 showed a specific rotation value
sensibly higher than reported in the literature for the resolved enantiomer, the MTPA derivatives of 6 with
(R)-MTPACl were prepared and analysed on an achiral HPLC column. Experiments with the racemic
derivatives showed that the limit of detectability of a second enantiomer was lower then 1% and thus an
enantiomeric excess of approximately 99% could be assigned to 6.
3. Experimental
3.1. Analytical
GC chiral analyses were performed on a DANI 8610 with a FID detector, fitted with a glass capillary
column, Megadex DACTBSβ-cdx (Mega, Legnano Italy), 25 m×0.25 mm i.d., film thickness 0.25 µm.
HPLC analyses were recorded on a Merck Hitachi L6000 equipped with L4000 UV detector and D2500
20
1
integrator. [α]D were recorded with a Propol automatic digital polarimeter. H NMR were recorded
on a Varian EMX 250 MHz with TMS as internal standard. All spectra are recorded in CDCl3 unless
otherwise indicated.
3.2. Absorption of the substrates onto the resin
The crude commercial resin XAD 1180 was washed successively with deionised water and acetone
(3 ml for 1 ml of resin). The substrate was dissolved in acetone and the resin, once dried, added to the
solution (i.e. 10 g of substrate, 100 ml of acetone and 10 g of dry resin). The mixture was shaken for
10 min, avoiding the use of the magnetic bar which can damage the resin, and then the acetone was
evaporated at reduced pressure. The solid so obtained was poured directly into the fermentation vessel.

4024
P. D’Arrigo et al. / Tetrahedron: Asymmetry 9 (1998) 4021–4026
3.3. (S)-2-Bromo-3-phenyl-propan-1-ol 2
In a 5 l beaker containing 2 l of tap water at 30°C were added, in one portion, 100 g of D-glucose
and 500 g of fresh baker’s yeast (Distillerie Italiane, Eridania group). The mixture was pre-fermented
for 30 min. 10 g (47 mmol) of 1 adsorbed onto 10 g of XAD1180 resin were added in one portion and
the mixture was stirred at 25°C for 48 h. The reaction was filtered on a glass sintered funnel (porosity
0) and the collected resin was washed three times with tap water. The resin and the water phase were
extracted with ethyl acetate. The collected organic phases gave crude 2 (9.7 g). The alcohol was purified
by distillation to give pure 2 (9.1 g, 43, mmol, 91% yield). An analytical sample was obtained by bulb
1
to bulb distillation, 110°C at 0.1 mmHg. H NMR δ (CDCl3) 2.41 (1H, OH, s), 3.20 (2H, CH2, m),
20
3.77 (2H, CH2, m), 4.31 (1H, CH, m), 7.27 (5H, Ph, m). [α]D −22.6 (c 5, CHCl3). Anal. calcd for
C9H11BrO: C, 50.26; H, 5.15; Br, 37,15. Found: C, 50.25; H, 5.12; Br, 37.19. Temperature program for
the GLC chiral analysis on the MEGA column: 40°C 1 min, 20°C/min, 125°C 2 min, 1°C/min, 210°C;
injector 250°C, detector 250°C. Retention times: (R)-2 31.50, (S)-2 32.32. Racemic 2 was synthesised
as reported in the literature,23 by treating allyl benzene with NBS in water. The two peaks assigned to
the enantiomeric couple 2 were accompanied by a second pair of peaks attributed to the isomer of α-Br
addition. The e.e. determined by GLC was 98.6%. In a parallel experiment without the use of resin, the
product recovery was similar but the e.e. was only 68%.
3.4. (R)-2-Benzyl-oxirane 3
Literature procedure23 was followed for the preparation of 3. 8 g (37 mmol) of 2 were suspended in 15
ml of water and 2 g (50 mmol) of sodium hydroxide pellets were added to the suspension. The mixture
was heated at 65°C for 45 min, cooled to 25°C and diluted with water (30 ml). The water was extracted
with diethyl ether so as to obtain after distillation (95°C, 15 mmHg) pure 3, 3.6 g (27 mmol, 73% yield),
[α]D20 +25.9 (c 5, benzene), [α]D20 +19.1 (neat), lit.30 +19.2 (neat). 1H NMR δ (CDCl3) 2.55 (1H, CH2,
q), 2.79 (1H, CH2, m), 2.87 (2H, CH2, m), 3.15 (1H, CH, m) and 7.28 (5H, Ph, m). The e.e. determined
by GLC was 98.8%. Temperature program for the GLC chiral analysis on the MEGA column: 40°C 1
min, 20°C/min, 110°C 2 min, 1°C/min, 210°C; injector 250°C, detector 250°C. Retention times: (R)-3
12.01, (S)-3 12.43 (racemic 3 was prepared in the same way from the mixture of racemic 2 obtained as
described earlier).
3.5. (S)-1-Phenyl-propan-2-ol 4
3 g (22 mmol) of 3, dissolved in 15 ml of dry diethyl ether, were added dropwise to a suspension
of LiAlH4 (0.23 g, 6 mmol) in 50 ml of anhydrous diethyl ether. The mixture was stirred at reflux for
2 h, after which time the reaction was quenched with ice and the organic phase separated; the aqueous
layer was extracted twice with diethyl ether and the combined organic phases were dried over anhydrous
Na2SO4. Evaporation of the solvent under reduced pressure gave 3 g of crude material which was purified
by bulb to bulb distillation (10 mm Hg, 100°C) so as to obtain 2.6 g (19 mmol, 86% yield) of pure 4,
20
[α]D +40.7 (c 5, benzene, lit.28 +41). Anal. calcd for C9H12O: C, 79.37; H, 8.88. Found: C, 79.35;
1
H, 8.74. H NMR δ (CDCl3) 1.22 (3H, CH3, d), 1.71 (1H, OH, s), 2.72 (2H, CH2, m), 3.99 (1H, CH,
m), 7.25 (5H, Ph, m). The e.e. determined by GLC was 98.4%. Temperature program for the GLC chiral
analysis on the MEGA column: 40°C 1 min, 20°C/min, 75°C 2 min, 1°C/min, 113°C, 10°C/min, 210°C;
injector 250°C, detector 250°C. Retention times: (S)-4 35.50, (R)-4 36.52 [racemic 4 was of commercial
origin (Fluka)].

P. D’Arrigo et al. / Tetrahedron: Asymmetry 9 (1998) 4021–4026
4025
3.6. (+)-(R)-2-Benzylmorpholine 6
Ethanolamine-O-sulphate (17.5 g) was added to a mixture of 3 (4 g, 29.9 mmol), 16 M NaOH solution
(9.5 g) and MeOH (6 ml) and the mixture stirred for 2 h at 40°C. To this mixture, solid NaOH (7.5 g)
and toluene (26 ml) were added and the reaction heated at 65°C for 7 h. To the cooled mixture, 45 ml of
water and 15 ml of toluene were added and the organic phase separated. The toluene was extracted with 2
N HCl and the acid extract was made alkaline to pH 11 with NaOH 4 M and extracted with toluene. The
toluene was washed with brine, dried, and evaporated to an oil (4.5 g). Crude 6 was chromatographed on
silica gel eluting first with ethyl acetate so as to obtain the N-substituted morpholine and then with ethyl
20
acetate:EtOH:Et3N (4:1:0.1) to obtain pure 6 (3.5 g, 19.8 mmol, 66% yield), [α]D +1.32 (c 5, CHCl3)
1
and +28.99 (c 5, benzene), H NMR δ (CDCl3) 2.42 (1H, NH, s), 2.63 (2H, CH2, m), 2.84 (4H, 2CH2,
m), 3.64 (2H, CH2, m) 3.88 (1H, CH, m) and 7.25 (5H, Ph, m).
3.7. (+)-(R)-2-Benzylmorpholine 6 hydrochloride
Compound 6 (3.5 g) was dissolved in diethyl ether (50 ml), and ethanol saturated with HCl (10 ml)
was added. The precipitate was collected by filtration and crystallisation from ethanol–ether gave pure
6·HCl, [α]D +13.8 (c 5, 2 N HCl), (lit.23 +8.4), m.p. 153, 1H NMR δ (DMSO-d6) 3.40 (9H, m), 7.22
20
(5H, m) and 9.45 (1H, m). Anal. calcd for C11H16NOCl: C, 61.82; H, 7.55; N, 6.55. Found: C, 61.84; H,
7.51; N, 6.49.
3.8. MTPA derivatives of 6
The hydrochloride of 6 (50 mg) was suspended in dry pyridine (0.5 ml) and (R)-(+)-MTPA-Cl (60 mg)
was added in one portion. The mixture was stirred for 18 h at room temperature. Pyridine was evaporated
at reduced pressure and the crude residue used directly for the HPLC analysis. The sample was dissolved
in hexane:i-PrOH:ethyl acetate (8:1:1) so as to obtain a solution of 1 mg/ml of both (R)-6 and racemic 6.
100 µL of this solution were injected into the HPLC column (Lichrospher Si60 5 µm, 25×4 mm, eluent
n-hexane:i-PrOH, 95:5; flow 0.6 ml/min, retention times, 8.68 for the derivative of the (R) enantiomer and
9.96 for that of the (S) enantiomer) and no traces of the (S) enantiomer could be detected. The presence
of the second enantiomer became detectable when 2 µl of racemic mixture was added to 100 µl of the
solution containing the single enantiomer.
Acknowledgements
We thank Rohm and Haas France S.A. for a generous gift of XAD 1180 and Mrs. Rosanna Bernardi
for the chiral GC analysis. This work is partially financed by CNR (Target project on biotechnology).
References
1. Simon, H.; Bader, J.; Günther, H.; Neumann, S.; Thanos, J. Angew. Chem., Int. Ed. Eng. 1985, 24, 539.
2. Belan, A.; Bolte, J.; Fauve, A.; Gourcy, J. G.; Veschambre, H. J. Org. Chem. 1987, 54, 256.
3. Leunberger, H. G. W.; Boguth, W.; Widmer, E.; Zell, R. Helv. Chim. Acta 1976, 59, 1832.
4. Fuganti, C.; Ghiringhelli, D.; Grasselli, P. J. Chem. Soc., Chem. Commun. 1975, 846.
5. Gramatica, P.; Manitto, P.; Monti, D.; Speranza, G. Tetrahedron 1988, 44, 1299.
6. Högberg, H. E.; Hedenström, E.; Fägerhag, J; Servi, S. J. Org. Chem. 1992, 57, 2052.

4026
P. D’Arrigo et al. / Tetrahedron: Asymmetry 9 (1998) 4021–4026
7. Gramatica, P.; Manitto, P.; Monti, D.; Speranza, G. Tetrahedron 1987, 43, 4481.
8. Sato, T.; Hanayama, K.; Fujisawa, T. Tetrahedron Lett. 1988, 29, 2197.
9. Ohta, H.; Kobayashi, A.; Ozaki, K. J. Org. Chem. 1989, 54, 1802.
10. Zueger, M. F.; Giovannini, F.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1983, 12, 1012.
11. Goodhue, C. T.; Schaffer, J. R. Biotechnol. Bioeng. 1971, 13, 203.
12. Utaka, M.; Konishi, S.; Okubo, T.; Tsuboi, S.; Takeda, A. Tetrahedron Lett. 1987, 28, 1447.
13. Leunberger, H. G. W. In Biocatalysts in Organic Synthesis; Tramper, J.; Van der Plas, H. C.; Linko, P., Eds; Elsevier:
Amsterdam, 1985; p. 99.
14. Koul, S.; Crout, D. H. G.; Errington, W.; Tax, J. J. Chem. Soc., Chem. Commun. 1995, 2969.
15. Fuganti, C.; Pedrocchi Fantoni, G.; Sarra, A.; Servi, S. Tetrahedron: Asymmetry 1994, 5, 1135.
16. Fronza, G.; Fogliato, G.; Fuganti, C.; Lanati, S.; Rallo, R.; Servi, S. Tetrahedron Lett. 1995, 36, 123.
17. Utaka, M.; Konishi, S.; Takeda, A. Tetrahedron Lett. 1986, 27, 4737.
18. Kobayashi, H. Recl. Trav. Chim. Pays-Bas 1996, 115, 201.
19. Buendia, J.; Crocq, V.; Masson, C.; Prat, D.; Vivat, M. Eur. Pat. Appl. 574,317, 1993.
20. Crocq, V.; Masson, C.; Winter, J.; Richard, C.; Lemaitre, G.; Lenay, J.; Vivat, M.; Buendia, J.; Prat, D. Organic Process
R&D 1997, 1, 2.
21. Fuganti, C.; Grasselli, P.; Servi, S.; Högberg, H. E. J. Chem. Soc., Perkins Trans. 1 1988, 1061.
22. D’Arrigo, P.; Pedrocchi Fantoni, G.; Servi, S.; Strini, A. Tetrahedron: Asymmetry 1997, 8, 2375–2379.
23. Brown, G. R.; Forster, G.; Foubister, A. J.; Stribling, D. J. Pharm. Pharmacol. 1990, 42, 797–799.
24. Fuganti, C.; Grasselli, P. Chem. Ind. (London) 1977, 983.
25. Chen, C. S.; Zhou, B. N.; Girdaukas, G.; Shieh, W. R.; VanMiddlesworth, F.; Gopalan, A. S.; Sih, C. J. Bioorg. Chem.
1984, 12, 98–117.
26. Nakamura, K., Kondo, S.-I.; Nakajima, N.; Ohno, A. Tetrahedron 1995, 51, 687.
27. D’Arrigo, P.; Fuganti, C.; Pedrocchi Fantoni, G.; Servi, S. Tetrahedron 1998, 54, 15017–15026.
28. Arens, C. L.; Hallgarten, P. A. J. Chem. Soc. 1956, 2987–2990.
29. Koppenhoefer, B.; Schurig, V. Org. Synth. 1987, 66, 151.
30. Bradshaw, J. S.; Huszthy, P.; McDaniel, C. W.; Zhu, C. Y.; Kent-Dalley, N.; Izatt, R. M.; Lifson, S. J. Org. Chem. 1990,
55, 3129–3137.