A biocatalytic route to P-chirogenic compounds by lipase-catalyzed desymmetrization of a prochiral phosphine-borane

Article abstract of DOI:10.1021/ol0519893

(Chemical Equation Presented) Available methods for synthesis of P-chirogenic compounds are limited. We set out to find biocatalytical means to introduce asymmetry in a phosphine-borane. After screening different lipases, Candida antarctica lipase B was f

Full text of DOI:10.1021/ol0519893

ORGANIC
LETTERS
2005
Vol. 7, No. 22
4991-4994
A Biocatalytic Route to P-Chirogenic
Compounds by Lipase-Catalyzed
Desymmetrization of a Prochiral
Phosphine−Borane
Daniel Wiktelius,*,† Magnus J. Johansson,*,‡ Kristina Luthman,† and Nina Kann‡
Medicinal Chemistry, Department of Chemistry, Go¨teborg UniVersity,
SE-41296 Go¨teborg, Sweden, and Organic Chemistry, Department of Chemical and
Biological Engineering, Chalmers UniVersity of Technology,
SE-41296 Go¨teborg, Sweden
daniel.wiktelius@chem.gu.se; magjo@chembio.chalmers.se
Received August 17, 2005
ABSTRACT
Available methods for synthesis of P-chirogenic compounds are limited. We set out to find biocatalytical means to introduce asymmetry in
a phosphine borane. After screening different lipases, Candida antarctica lipase B was found to give excellent results in the desymmetrization
of prochiral phosphine boranes. Both enantiomers can be obtained in up to >98% optical purity via acetylation or hydrolysis in processes
−
−
that allow recycling of the substrates.
Asymmetric catalysis is a rapidly growing area of research,
much due to the demands from the pharmaceutical industry
to be able to prepare chiral material in high optical purity
using atom efficient methods. Thus, ligand design and the
development of new asymmetric reactions are becoming
increasingly important. The challenging task of preparing
P-chirogenic phosphines is a bottleneck in the search for new
ligands in transition metal catalysis, since the number of
methods whereby one can synthesize such compounds is still
limited. The preparation and use of P-chirogenic phosphines
has been reviewed.1
most interest; Evans’ method for the desymmetrization of
prochiral phosphine-boranes using (-)-sparteine/s-BuLi3 and
the (-)-ephedrine chiral auxiliary approach by Juge´ et al.4
However, they are both stoichiometric in either chiral base
or auxiliary and have limitations in substrate scope and in
giving high enough optical purity of the desired P-chirogenic
phosphine (the optical purity can, however, often be im-
proved by recrystallization). The preparation of both enan-
tiomers of a P-chirogenic phosphine is rarely reported.2d,5
Herein, we wish to report a biocatalytic approach as a
highly interesting alternative for the preparation of P-
In recent years, focus has been directed toward ligand
design utilizing already established methods to prepare
P-chirogenic compounds.2 Two procedures have gained the
(2) (a) Imamoto, T.; Watanabe, J.; Wada, Y.; Masuda, H.; Yamada, H.;
Tsuruta, H.; Matsukawa, S.; Yamaguchi, K. J. Am. Chem. Soc. 1998, 120,
1635. (b) Yamanoi, Y.; Imamoto, T. J. Org. Chem. 1999, 64, 2988. (c)
Tang, W.; Zhang, X. Angew. Chem., Int. Ed. 2002, 41, 1612. (d) Hoge, G.
J. Am. Chem. Soc. 2003, 125, 10219.
(3) Muci, A. R.; Campos, K. R.; Evans, D. A. J. Am. Chem. Soc. 1995,
117, 9075.
(4) Juge´, S.; Stephan, M.; Laffitte, J. A.; Geneˆt, J. P. Tetrahedron Lett.
1990, 31, 6357.
(5) (a) Johansson, M. J.; Schwartz, L. O.; Amedjkouh, M.; Kann, N. C.
Eur. J. Org. Chem. 2004, 1894. (b) Johansson, M. J.; Schwartz, L. O.;
Amedjkouh, M.; Kann, N. Tetrahedron: Asymmetry 2004, 15, 3531. (c)
Liu, D.; Zhang, X. Eur. J. Org. Chem. 2005, 646.
† Go¨teborg University.
‡ Chalmers University of Technology.
(1) (a) Pietrusiewicz, K. M.; Zabłocka, M. Chem. ReV. 1994, 94, 1375.
(b) Brunel, J. M.; Faure, B.; Maffei, M. Coord. Chem. ReV. 1998, 180,
665. (c) Quin, L. D. A Guide to Organophosphorous Chemistry; John Wiley
& Sons: New York, 2000. (d) Cre´py, K. V. L.; Imamoto, T. Top. Curr.
Chem. 2003, 229, 1. (e) Valentine, D. H.; Hillhouse, J. H. Synthesis 2003,
2437. (f) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029. (g) Johansson,
M. J.; Kann, N. C. Mini-ReV. Org. Chem. 2004, 1, 233.
10.1021/ol0519893 CCC: $30.25
© 2005 American Chemical Society
Published on Web 10/07/2005

chirogenic compounds. This route comprises the first reported
desymmetrization of a prochiral phosphine-borane using
lipases, yielding both enantiomers of the corresponding
P-chirogenic phosphine-borane independently. Earlier work
with lipases to obtain enantioenriched P-chirogenic products
include kinetic resolution of hydroxymethyl(dialkyl)phos-
phine-boranes6 and desymmetrization of prochiral phosphine
oxides.7
Scheme 2. Stereoselective Desymmetrization of 2 or 4 by
Lipasea-Catalyzed Acetylation or Hydrolysisb
The prochiral starting material bis(hydroxymethyl)phen-
ylphosphine-borane (2, Scheme 1) was synthesized using
Scheme 1. Synthesis of
Bis(hydroxymethyl)phenylphosphine-Borane (2)
a Lipases used along with ee’s are given in Table 1 and text.
b See ref 12 for reaction conditions.
phosphine-boranes6 indicated that it may have relevance also
for this type of compounds.
Five lipases10 were screened for their ability to desym-
metrize diol 2 by acetylation using vinyl acetate as acyl donor
and solvent (Scheme 2, Table 1), and the conversion was
monitored by HPLC.
a one-pot solvent-free condensation between paraformalde-
hyde and phenylphosphine (1) followed by dilution with THF
and protection of the phosphorus atom by BH3‚SMe2.
Compound 2 was obtained in 92% yield after recrystallization
in toluene.
The prochiral diol 2 is similar in structure to its “all
carbon” analogues, the 1,3-propanediols, which are known
substrates for lipase catalyzed desymmetrization.8 Although
recent spectroscopic and computational studies have con-
tributed significantly to the understanding of lipase enantio-
selectivity,9 empirical rules are still widely used to predict
the enantiotope that reacts fastest. Application of the spacial
size preferences of BCL10 for primary alcohols suggested
by Weissfloch and Kazlauskas11 on the desymmetrization
of 2 into acetyloxymethyl(hydroxymethyl)phenylphosphine-
borane (3, Figure 1, Scheme 2) suggested the resulting
Table 1. Lipase Screening for Enantioselective
Desymmetrization of 2 by Acetylationa
lipaseb reaction time prod distrib (%)c 2:3:4 ee of (R)-3d (%)
ANL
BCL
PPL
PFL
CALB
CALB
7 d
63:36:1
75:23:2
65:34:1
41:53:6
50:49:1
0:48:52
70
21
21
62
60
92
96 h
19 d
28 h
140 min
6 h
a See ref 12 for reaction conditions. b For explanation of lipase abbrevia-
tions, see ref 10. c Determined by HPLC of the filtered reaction mixture.
d Determined by chiral HPLC of the isolated product.
The screening procedure began with ANL, BCL, PFL, and
PPL,10,12a but disappointingly the conversion was slow, and
the formation of bis(acetyloxymethyl)phenylphosphine-
borane (4) limited the yield of monoacetate 3. The optical
purity of the isolated product 3 was assessed by chiral HPLC
(Table 1). The stereoselectivity of BCL and PPL was low
Figure 1. Preferred orientation of a primary alcohol in the active
site of BCL,10 with the relative directions of the large (L), medium
(M), and small (S) substituents.11 An alignment of prochiral diol 2
according to this model indicates that the pro-(R) group is the
favored site of reaction.
(10) Lipases are abbreviated according to species of origin: ANL )
Aspergillus niger lipase (Amano lipase A); BCL ) Burkholderia cepacia
lipase (formerly known as Pseudomonas cepacia, Amano lipase PS);
CALB ) Candida antarctica lipase B (Novozym 435); PFL ) Pseudomo-
nas fluorescens lipase (Amano lipase AK); PPL ) Pig pancreatic lipase
(Fluka).
stereochemistry to be R. The rule has good, but not perfect
accuracy,11 and although the interactions of diol 2 with the
enzyme may differ significantly from that of carbon ana-
logues, data on kinetic resolution of hydroxymethyl(dialkyl)-
(11) Weissfloch, A. N. E.; Kazlauskas, R. J. J. Org. Chem. 1995, 60,
6959.
(6) Shioji, K.; Kurauchi, Y.; Okuma, K. Bull. Chem. Soc. Jpn. 2003,
76, 833.
(7) Kiełbasin´ski, P.; Z˙ urawin´ski, R.; Albrycht, M.; Mikołajczyk, M.
Tetrahedron: Asymmetry 2003, 14, 3379.
(8) For a recent review, see: Garc´ıa-Urdiales, E.; Alfonso, I.; Gotor, V.
Chem. ReV. 2005, 105, 313.
(9) For a recent review, see: Ema, T. Curr. Org. Chem. 2004, 8, 1009.
(12) (a) Typical procedure for the lipase catalysed desymmetrization by
acetylation: 2 (100 mg, 0.54 mmol) and lipase (100 mg) were stirred at rt
in vinyl acetate (2.5 mL). (b) Typical procedure for the lipase catalyzed
desymmetrization by hydrolysis: 4 (75 mg, 0.28 mmol) and CALB (75
mg) were shaken vigourusly in a mixture of 0.1 M phosphate buffer pH 7
(2 mL), diisopropyl ether (1 mL) and THF (0.5 mL).
4992
Org. Lett., Vol. 7, No. 22, 2005

(21% ee), whereas PFL and ANL gave 3 in moderate ee, 62
and 70%, respectively. With these discouraging results at
hand we tried CALB,10 and to our delight the rate of
conversion was excellent in this case. The reaction was
stopped at 50% conversion and the appearance of 4, but the
ee of the product was a rather disappointing 60%. Consider-
ing the good catalytic efficiency of CALB, the possibility
of kinetic amplification of the optical purity of (R)-3 by
tandem resolution if the second acetylation reaction pro-
ceeded with stereoselectivity was raised. Rewardingly, re-
symmetrization to 4 was fastest for the minor (S)-enantiomer
of 3 formed in the desymmetrization step. The self-correcting
process afforded (R)-3 in 92% ee at a product distribution
of 0:48:52 of 2:3:4.
Table 2. Optical Purity of (S)-3 during Desymmetrization of
Diacetate 4 by Hydrolysis Catalyzed by CALBa
reaction time (h)
prod distribb (%) 2:3:4
ee of (S)-3c (%)
8
31
70
97
4:22:74
26:50:24
37:54:9
48:49:3
19
28
69
90
a For reaction conditions, see ref 12. b Determined by HPLC of the
filtered reaction mixture. c Determined by chiral HPLC of the reaction
mixture after filtration and extraction.
The hydrolysis of 4 was slower and proceeded with lower
stereoselectivity in the first step than the acetylation reaction,
which may reflect the smaller size difference between the
acetyloxymethyl and the phenyl substituent compared to the
hydroxymethyl. The resymmetrization into diol 2 improved
the low stereoselectivity in the first step; however, an ee of
90% was only obtainable at high conversion when ap-
proximately half of the monoacetate 3 was converted into
2. At this stage, the reaction was impractically slow for
further improvement of the optical purity, and (S)-3 was
isolated in 89% ee and 31% yield along with 36% of 2.
To verify the stereochemistry of monoacetate 3 and
explore some possible transformations of this compound, we
wanted to transform enantiomerically enriched 3 into the
known P-chirogenic compound hydroxymethyl(methyl)-
phenylphosphine-borane (6, Scheme 3), of which the
This result prompted us to investigate the dependence of
the optical purity of 3 on the extent of conversion into 4.
An experiment was set up in which the conversion and ee
were monitored at intervals during the reaction. The results
are summarized in Figure 2.
Scheme 3. Confirmation of Stereochemistry of (-)-3 by
Chemical Correlation: Synthesis of (S)-6
Figure 2. Optical purity of (R)-3 (O) and product distribution
(monoacetate 3 (9), diacetate 4 (2)) monitored by HPLC during
desymmetrization of diol 2 by CALB-catalyzed acetylation. See
ref 12 for reaction conditions.
These data indicate that the second acetylation proceeded
with high enantiopreference, and that at most 27% resym-
metrization into diacetate 4 is required to improve the ee of
monoacetate (R)-3 from 60 to >98%. However, after running
the reaction at even higher conversion (Table 1), the ee of
the isolated product was 92%, thus we suspect some
racemization during purification, probably via acyl migration.
With this in mind, we desymmetrized 2 into (R)-3 in 92%
ee and 59% isolated yield along with 32% of 4 to be used
for subsequent hydrolysis experiments.
An advantage of lipase catalyzed desymmetrization of
prochiral diols is that it can give access to both enantiomers
via stereoselective hydrolysis of the corresponding diacetate.8
The diacetate 4 was conveniently obtained as a byproduct
of the desymmetrization of diol 2 and could be used to
synthesize monoacetate (S)-3 by the reverse reaction cata-
lyzed by CALB (Scheme 2). As in the acetylation experi-
ment, the conversion and optical purity of 3 was monitored
during the reaction (Table 2).
levorotary (R)-enantiomer is avaliable via desymmetrization
of dimethylphenylphosphine-borane by enantioselective
deprotonation with (-)-sparteine/s-BuLi and quenching with
triethyl phosphite and oxygen.13 The effort began with the
presumed (R)-enantiomer of 3 (92% ee) obtained by stereo-
selective acetylation by CALB of diol 2 (Scheme 2).
Transformation into (S)-6 requires deoxygenation and cleav-
age of the acetate of (R)-3. We realized that if the hydroxy
function of 3 is exchanged for a halide, simultaneous
reduction of the ester and halide can be accomplished by
treatment with lithium triethylborohydride. To this end, (R)-3
was reacted with tetrabromomethane and triphenylphosphine
(13) Nagata, K.; Matsukawa, S.; Imamoto, T. J. Org. Chem. 2000, 65,
4185.
Org. Lett., Vol. 7, No. 22, 2005
4993

which furnished (S)-acetyloxymethyl(bromomethyl)phen-
ylphosphine-borane (5) in 44% yield. The reduction of (S)-5
to the target compound 6 with LiBHEt3 proceeded smoothly
and afforded (S)-6 in 82% yield with minimal racemization
detected (90% ee, [R]D +8.2) (Scheme 3).
confers minimal loss of material. We chose to focus
thoroughly on the phenyl derivative of bis(hydroxymethyl)-
phoshine-borane in this study; however, other compounds
of this type are of course candidates for enzymatic desym-
metrization, which will be the aim of further research.
The chemical correlation (Scheme 3) confirmed that the
absolute configuration of monoacetate (-)-3 (Scheme 2) is
R as anticipated. (R)-6 was synthesized according to the
previously reported method13 in 80% ee and was found to
be identical to (S)-6 by NMR, but with the opposite sign of
optical rotation ([R]D -7.1) and we could further confirm
the stereochemical correlation by chiral HPLC. Thus, the
current enzymatic method gives access to the opposite
enantiomer of 6 from that obtained by enantioselective
deprotonation with (-)-sparteine/s-BuLi. Some interesting
transformations of 6 have been studied by Imamato et al.,
such as synthesis of the BisP*-type ligands14 by dimerization,
oxidation, and decarboxylation or direct decarbonylation into
phosphine-borane precursors of secondary P-chirogenic
phosphines.13 In a related study on enzymatic desymmetri-
zation of the phosphine oxide analogues of 2 and 4, Kieł-
basin´ski and co-workers obtained monoacetate products in
fair yields and up to 79% optical purity and a chemical
correlation with 6 in nine steps was reported which proceeded
with substantial racemization.7 Reported attempts on kinetic
resolution of hydroxymethyl(dialkyl)phosphine-boranes6
gave analogues of 6 in varying yields and optical purities;
however, the enantioselectivity of the resolutions were
unpractically low.
The new compounds 3, 5, and (S)-6 (Scheme 3) are highly
interesting for further exploration of their potential as
precursors in syntheses of other optically active derivatives.
The main use of P-chirogenic phosphine-boranes is as ligands
within the field of transition metal catalyzed asymmetric
reactions.1 We find it intriguing that compounds presented
in this study (or derivatives thereof) may give the opportunity
to utilize the inherent chirality of a lipase in such reactions
by relaying the asymmetry of the enzyme onto the reactants
via the phosphine-borane.
In summary, we have shown that enantiomerically en-
riched phosphine-boranes are accessible by lipase catalyzed
desymmetrization of prochiral diol or diacetate precursors.
The developed methods allow the preparation of both
enantiomers of monoacetate analogues in excellent ee with
defined absolute configurations. The obtained products are
highly interesting for further elaboration into useful com-
pounds in ligand development. The method is atom efficient
since the enzymatic reactions allow recycling of material.
Acknowledgment. Financial support was obtained from
the Knut and Alice Wallenberg Foundation, the Swedish
Research Council, and the Swedish Foundation for Strategic
Research.
We believe that the current results may provide an entry
to new methodology in the field of asymmetric synthesis of
P-chirogenic phosphine-boranes. High optical purity is
attained at the expense of yield with the current method, but
since 2, 3, and 4 can be interconverted (Scheme 2), this
Supporting Information Available: Experimental pro-
cedures and spectroscopic data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(14) Cre´py, K. V. L.; Imamoto, T. Tetrahedron Lett. 2002, 43, 7735.
OL0519893
4994
Org. Lett., Vol. 7, No. 22, 2005