Enzymatic resolution of diethyl 3-hydroxycycloalkenyl phosphonates

Article abstract of DOI:10.1016/j.tetasy.2003.12.027

Enantiomerically pure cyclic (S)-diethyl 3-hydroxy-1-alkenyl phosphonates 1-3 were obtained by enzymatic resolutions. The corresponding (R)-acetates were also obtained with high enantiomeric excess, except in the case of the five-membered ring compound. T

Full text of DOI:10.1016/j.tetasy.2003.12.027

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 827–830
Enzymatic resolution of diethyl 3-hydroxycycloalkenyl
phosphonates
Mireille Attolini,a Gilles Iacazio,b Gilbert Peiffer,a Yolande Charmassona
and Michel Maffeia,*
aLaboratoire des Organo-Phosphorꢀes (U.M.R. 6009 du CNRS), B.P. 552, Facultꢀe de Saint Jꢀer^ome, Avenue Escadrille
Normandie-Niꢀemen, 13397 Marseille Cedex 20, France
bLaboratoire de Bioinorganique Structurale (U.M.R. 6517 du CNRS), B.P. 432, Facultꢀe de Saint Jꢀer^ome, Avenue Escadrille
Normandie-Niꢀemen, 13397 Marseille Cedex 20, France
Received 26 November 2003; accepted 10 December 2003
Abstract—Enantiomerically pure cyclic (S)-diethyl 3-hydroxy-1-alkenyl phosphonates 1–3 were obtained by enzymatic resolutions.
The corresponding (R)-acetates were also obtained with high enantiomeric excess, except in the case of the five-membered ring
compound. The most efficient enzyme (lipase Amano AK or Amano PS) in each case depends on the structure of the substrate (five
to seven-membered ring compounds).
ꢀ 2004 Elsevier Ltd. All rights reserved.
OH
1. Introduction
OH
OAc
Enzyme, 30 C
˚
+
Functionalized vinyl phosphonates are useful building
blocks,1 particularly for the synthesis of biologically
active compounds. We have shown2 that dialkyl
3-acetoxy-1-alkenyl phosphonates can be used to prepare
phosphono amino acids, which are known to be active
against epilepsy and ParkinsonÕs disease.3 They can also
be used to prepare the corresponding allyl alcohols,
which have in turn been used as starting materials for
the synthesis of antiviral nucleosides.4;5 Since the con-
figuration is determinant for the biological activity of
these chiral compounds, an enantioselective synthesis of
dialkyl 3-acetoxy (or 3-hydroxy)-1-alkenyl phospho-
nates is of interest.
( )n
( )n
( )n
vinyl acetate
P(OEt)2
O
P(OEt)2
O
P(OEt)2
O
1 n = 2
2 n = 1
3 n = 3
4 n = 2
5 n = 1
6 n = 3
Scheme 1.
membered ring homologue did not lead to the expected
diethyl 3-hydroxy-1-cyclopentenyl phosphonate.
For these reasons, we turned our attention towards the
enzymatic resolution of these hydroxy phosphonates in
order to obtain enantiomerically pure compounds. This
methodology has been exploited for the enantioselective
esterification of hydroxy phosphonates7–10 and we
recently described the enzymatic resolution of diethyl
3-hydroxy butenyl phosphonate.11 Herein, we report the
enzymatic resolution of diethyl 3-hydroxy cycloalkenyl
phosphonates 1–3 (Scheme 1).
In this area, we have already reported an enantioselec-
tive synthesis of cyclic dialkyl 3-hydroxy-1-alkenyl
phosphonates via bakerÕs yeast asymmetric reduction of
the corresponding cyclic 3-phosphono-1-enones.6 This
methodology was found to be effective only for the six-
membered ring compounds, whereas the seven-mem-
bered ring homologue (i.e., diethyl 3-hydroxy-1-cyclo-
heptenyl phosphonate) was obtained only with up to
34% enantiomeric excess (ee). Furthermore, the five-
2. Results and discussion
First, different enzymes for the resolution of 1 were
assayed, and the results are summarized in Table 1.
* Corresponding author. Tel.: +33-4-91-28-27-03; fax: +33-4-91-28-27-
38; e-mail: michel.maffei@univ.u-3mrs.fr
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2003.12.027

828
M. Attolini et al. / Tetrahedron: Asymmetry 15 (2004) 827–830
Table 1. Enzymatic resolution of 1
Entry Enzyme
Time
Conversiona Recovered 1
(%)
4
Ed
Yield (%)
Eeb (%)
Absolute
configuration
Yield (%)
Eeb (%) Absolute
configuration
1
2
3
4
5
Amano AY
PPLc
10 d
9 d
15
11
49
50
50
73
79
43
48
47
11
11
(S)
(S)
(S)
(S)
(S)
12
9
60
93
(R)
5
(R)
(R)
(R)
(R)
33
155
Amano PS
Amano AK
Amano AK
22 h
6 h
92
98
49
50
50
96
>99
>99
>200
>200
9 h
>99
a Conversion, determined by chiral HPLC.
b Determined by chiral HPLC after column chromatography.
c Porcine Pancreatic Lipase.
d Calculated following Ref. 12.
It appears that with lipase Amano AY (entry 1) or
Porcine Pancreatic Lipase (entry 2), the reaction was
slow and occurred with low enantioselectivity. Lipase
Amano PS (entry 3) gave good results, but the most
satisfactory enantioselectivities were obtained with
lipase Amano AK (entry 4), where enantiomerically
pure 4 and unreacted 1 were obtained, both with almost
quantitative yields. On a preparative scale (1 g of 1), we
obtained essentially the same results (entry 5). Those
two reactions occurred with an enantioselectivity fac-
tor12 E superior to 200.
and led to (S)-())-2 (35% yield) whose enantiomeric
purity was found to be 61% ee by chiral HPLC after
acetylation (see Experimental). Its specific rotation was
20
D
½aꢀ ¼ )24.0 (c 0.49, CH2Cl2).
4. Conclusion
Enantiomerically pure cyclic diethyl (S)-3-hydroxy-1-
alkenyl phosphonates 1–3 were obtained via enzymatic
resolution with lipases Amano AK or Amano PS. The
corresponding (R) acetates were also obtained with up
to 99% ee, except in the case of the five-membered ring
substrate. Thus, this method is far superior than the
bakerÕs yeast reduction of 3-phosphono-1-enones
described previously,6 whose chemical outcome is highly
dependent of the structure of the starting material.
The above results prompted us to look at five- and
seven-membered ring hydroxy phosphonates 2 and 3
using lipases Amano AK and Amano PS (Table 2).
Esterification of 2 was moderately enantioselective with
E ¼ 24 and 43, respectively (entries 1 and 2). The most
efficient lipase was Amano PS and a 500 mg scale reso-
lution with this lipase gave enantiomerically pure unre-
acted substrate (S)-2 and the acetate (R)-5 with 84% ee
(entry 3). With the substrate 3, the reaction kinetics were
slower but higher enantioselectivities were encountered.
Thus, lipase Amano AK was used on a preparative scale
(500 mg of 3) and led to enantiomerically pure (S)-(+)-3
and (R)-(+)-6 after 3 days (entry 6).
5. Experimental
5.1. General
All solvents were purified according to reported proce-
dures, and reagents were used as received. 1H NMR, 13
C
NMR and 31P NMR spectra were recorded in CDCl3,
on a Bruker Avance 300 spectrometer working at
300.00, 75.47 and 121.49 MHz, respectively (the usual
abbreviations are used: s: singulet, d: doublet, t: triplet,
q: quadruplet, qt: quintuplet, m: multiplet). The refer-
ences were d ¼ 7:26 ppm for residual chloroform (1H),
d ¼ 77 ppm (13C) and 85% H3PO4 as external standard
(31P). All chemical shifts are given in ppm. Optical
rotations were measured on a Perkin Elmer 341 polar-
imeter. Compounds 1–3 have been obtained by saponi-
fication of the corresponding acetates.18;19
3. Determination of absolute configurations
The absolute configurations for compounds 16 and 313
have already been determined by chemical correlation.
We used the same methodology for the determination of
the absolute configuration of hydroxy phosphonate 2
(Scheme 2). Thus, 3-bromo-2-cyclopentenone (easily
obtained from 1,3-cyclopentanedione14) was reduced to
the bromoalcohol 7 as already described,15 which was
submitted to enzymatic resolution by lipase Amano AK.
This led to a mixture of (S)-())-7 (20% yield) and (R)-8
(78% yield) whose enantiomeric purities could not be
determined at this stage. However, the absolute config-
uration of ())-7 was ascribed as (S) by comparison of its
specific rotation with the value given by Silverton
co-workers.15
Enantiomeric purities for 1, 2, 3, 4, 5 and 6 were mea-
sured by HPLC analysis on a Chiralcel OD-H column
with a UV detection (214 nm), eluent: hexane/isopro-
panol (90:10), flow rate: 0.5 mL/min. For compound 2,
Enantiomerically enriched (S)-())-7 was then used in a
This indicates that the enantiomeric purity of (S)-())-7 is at least 61%
ee.
palladium-catalyzed reaction16;17 with diethylphosphite

M. Attolini et al. / Tetrahedron: Asymmetry 15 (2004) 827–830
Table 2. Enzymatic resolution of 2 and 3
829
Entry Hydroxy
phosphonate
Enzyme
Time
Conversiona Recovered hydroxy phosphonate
(%)
Acetate
E
Yield (%) Eeb
(%)
Absolute
configuration
Yield (%) Eeb
(%)
Absolute
configuration
1
2
3
4
5
6
2
Amano AK
Amano PS
Amano PS
Amano AK
Amano PS
Amano AK
24 h
24 h
24 h
3 d
54
56
54
49
42
50
28
47
41
52
56
48
91c
>99c
>99c
96
(S)
(S)
(S)
(S)
(S)
(S)
63
52
48
45
42
50
76
78
(R)
(R)
(R)
(R)
(R)
(R)
24
43
57
2
2
3
3
3
84
98d
97d
>99d
>200
130
>200
7 d
3 d
70
>99
a Conversion, determined by chiral HPLC.
b Determined by chiral HPLC.
c Determined after acetylation.
d Determined after hydrolysis.
OH
OAc
OH
O
NaBH4, CeCl3
Lipase Amano AK
vinyl acetate
+
MeOH, -15 C
˚
30 C
˚
(89%)
Br
(S)-(-)-7
Br
Br
Br
(R)-8
( )-7
OH
OH
Diethyl phosphite, Pd(PPh3)4
e.e. = 61%
Et N, Toluene, 60 C
˚
3
(35%)
Br
(S)-(-)-7
P(OEt)2
O
(S)-(-)-2
Scheme 2.
20
D
ee were determined after acetylation since these hydroxy
phoshonates were not separated under these conditions.
For compound 6, ee were determined after saponifica-
tion since these acetoxyphosphonates were not sepa-
rated under these conditions. Retention times: (R)-1:
18.3 min; (S)-1: 16.8 min, (R)-3: 16.6 min; (S)-3:
22.2 min, (R)-4: 13.6 min; (S)-4: 11.3 min, (R)-5:
15.3 min; (S)-5: 18.4 min.
data: (S)-1: ½aꢀ ¼ )34.6 (c 0.5, CH2Cl2), (S)-3:
20
20
D
20
D
½aꢀ ¼ +9.9 (c 0.49, CH2Cl2), (R)-4: ½aꢀ ¼ +91.3 (c 0.51,
20
CH2Cl2), (R)-5: ½aꢀ ¼ +94.1 (c 0.49, CH2Cl2), (R)-6:
D
½aꢀ ¼ +15.5 (c 0.5, CH2Cl2).
D
5.2.1.2. Diethyl 3-hydroxy-1-cyclopentenyl phospho-
nate 2. Colourless oil (Rf : 0.19, ethyl acetate/methanol
90:10). IR (thin film, cmꢁ1): 3373, 2981, 2936, 1614,
1235, 1025. 1H NMR: 1.2 (t, 6H, 2CH3, 3JHH ¼ 7:0 Hz);
1.8 (m, 1H); 2.4 (m, 2H, CH2); 2.6 (m, 1H); 3.4 (s, 1H,
5.2. Syntheses
3
3
OH); 4.0 (qt, 4H, 2CH2, JHH ¼ 7:0 Hz, JPH ¼ 7:0 Hz);
5.1 (s, 1H); 6.5 (dd, 1H, 3JPH ¼ 11:1 Hz; 3JHH ¼ 1:9 Hz).
13C NMR: 16.4 (d, 3JPC ¼ 6:4 Hz); 31.7 (d,
5.2.1. General procedure for the enzyme assay in the
resolution of hydroxy phosphonates 1–3.
2JPC ¼ 12:9 Hz); 33.9 (d, JPC ¼ 10:6 Hz); 62.0 (d,
3
2JPC ¼ 5:7 Hz); 77.3; 135.0 (d, JPC ¼ 185:7 Hz); 148.5
1
20
D
(d, 2JPC ¼ 12:4 Hz). 31P NMR: 15.0. ½aꢀ ¼ )49.8 (c 0.47,
CH2Cl2), ee > 99%.
5.2.1.1. Enzymatic resolution of diethyl phosphonates.
The enzyme (100 mg) was mixed to a solution of
hydroxy phosphonate (100mg) and vinyl acetate (15 mL),
the suspension was shaken at 30 ꢁC and the reaction
was monitored by chiral HPLC. After removal of the
enzyme by filtration and solvent in vacuo, the crude was
subjected to flash chromatography (silica, ethyl acetate/
methanol, 90/10) to yield the (R)-acetate followed by
unreacted (S)-hydroxy phosphonate.
5.2.2. Determination of absolute configuration of (S)-())-
2.
5.2.2.1. 3-Bromo-1-cyclopentenol 7. A solution of
cerium trichloride heptahydrate (9.2 g, 24.6 mmol) in
methanol (60 mL) was cooled to )15 ꢁC with stirring
under a nitrogen atmosphere. A solution of bromo-
cyclopentenone (3.3 g, 20.5 mmol) in methanol (10 mL)
Spectral data for compounds 1 and 3,6 and 4, 5 and 618
were identical to those already reported. Additional

830
M. Attolini et al. / Tetrahedron: Asymmetry 15 (2004) 827–830
was added, followed by addition of sodium borohydride
(935 mg, 24.6 mmol) in portions, and the reaction was
monitored by TLC. The mixture was quenched by slow
addition of water (10 mL), and most of the methanol
was removed in vacuo. Diethyl ether (50 mL) and water
(30 mL) were added, the aqueous layer was extracted
with diethyl ether (4 · 20 mL), the combined organic
layers were washed with saturated NaCl, and dried with
MgSO4. Filtration, removal of solvent, and flash chro-
matography (silica, diethyl ether/pentane 2:1) yielded
2.97 g (89%) of 7 as a colourless oil. (Rf : 0.49). 1H NMR:
1.8 (m, 1H); 2.5 (m, 4H); 4.7 (m, 1H); 5.9 (m, 1H). 13C
NMR: 34.0; 38.4; 76.4; 127.4; 133.6.
Acknowledgements
ꢀ ꢀ
Financial support from CNRS as a ÔDelegationÕ to
M.M. is gratefully acknowledged.
References and notes
1. Minami, T.; Motoyoshiya, J. Synthesis 1992, 333.
2. Attolini, M.; Maffei, M.; Principato, B.; Peiffer, G. Synlett
1997, 384.
3. The NMDA Receptor; Watkins, J. C., Collingridge, G. L.,
Eds.; 2nd ed.; Oxford University Press: Oxford, 1994.
4. Yokomatsu, T.; Shimizu, T.; Yuasa, Y.; Shibuya, S.
Synlett 1995, 1280.
5. Lau, W. Y.; Zhang, L.; Wang, J.; Cheng, D.; Zhao, K.
Tetrahedron Lett. 1996, 37, 4297.
6. Attolini, M.; Bouguir, F.; Iacazio, G.; Peiffer, G.; Maffei,
M. Tetrahedron 2001, 57, 537.
5.2.2.2. Enzymatic resolution of 7. Lipase Amano AK
(250 mg) was added to a solution of 7 (500 mg,
3.07 mmol) in vinyl acetate (20 mL) and the suspension
was stirred at 30 ꢁC. After 3 h, the mixture was filtered,
the solvent removed in vacuo and the residue was sub-
jected to flash chromatography on silica gel with pen-
tane/diethyl ether (75/25). (R)-8 (491 mg) was isolated
7. Zhang, Y.; Xu, C.; Li, J.; Yuan, C. Tetrahedron: Asym-
metry 2003, 14, 63.
8. Zhang, Y.; Yuan, C.; Li, Z. Tetrahedron 2002, 58, 2973.
9. Kielbasinski, P.; Albrycht, M.; Luczak, J.; Mikolajczyk,
M. Tetrahedron: Asymmetry 2002, 13, 735.
10. Zurawinski, R.; Nakamura, K.; Drabowicz, J.; Kielbasin-
ski, P.; Mikolajczyk, M. Tetrahedron: Asymmetry 2001,
12, 3139.
11. Attolini, M.; Iacazio, G.; Peiffer, G.; Maffei, M. Tetra-
hedron Lett. 2002, 43, 8547.
12. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294.
13. Attolini, M.; Iacazio, G.; Peiffer, G.; Maffei, M. Tetra-
hedron: Asymmetry, 2003, 14, 3857.
14. Piers, E.; Grierson, J. R.; Lau, C. K.; Nagakura, I. Can. J.
Chem. 1982, 60, 210.
15. Ito, S.; Kasai, M.; Ziffer, H.; Silverton, J. V. Can. J. Chem.
1987, 65, 574.
16. Hirao, T.; Masunaga, T.; Ohshiro, Y.; Agawa, T. Tetra-
hedron Lett. 1980, 21, 3595.
(78% yield) followed by 102 mg of (S)-7 (20% yield),
20
D
which displayed ½aꢀ ¼ )38.4 (c 0.59, CH2Cl2).
5.2.2.3. (S)-())-Diethyl (3-hydroxy-1-cyclopentenyl)
phosphonate 2.
A solution of (S)-())-7 (117 mg,
0.7 mmol), triethylamine (200 lL, 1.45 mmol) and di-
ethyl phosphite (100 lL, 0.7 mmol) in anhydrous toluene
(2 mL) was degassed by bubbling with nitrogen, and
added to tetrakis triphenylphosphine palladium (42 mg,
0.036 mmol) in a round bottomed flask under nitrogen.
The mixture was gradually heated to 70 ꢁC and main-
tained at this temperature for 1 h, after which a white
precipitate appeared. After cooling to room tempera-
ture, ethyl acetate was added (20 mL), the mixture was
filtered and the solvents were removed in vacuo. Flash
chromatography with ethyl acetate/methanol (90:10)
afforded 53 mg (35%) of (S)-())-2 as a colourless oil. Its
17. Hirao, T.; Masunaga, T.; Yamada, N.; Ohshiro, Y.;
Agawa, T. Bull. Chem. Soc. Jpn. 1982, 55, 909.
18. Attolini, M.; Peiffer, G.; Maffei, M. Tetrahedron 2000, 56,
2693.
spectral data were identical to those described above
20
D
purity was 61% ee (after acetylation).
{with ½aꢀ ¼ )24 (c 0.49, CH2Cl2)}, and its enantiomeric
19. Oehler, E.; Zbiral, E. Chem. Ber. 1991, 124, 175.