Stereoselective bioreduction for the resolution of racemic mixtures of bicyclo[3.3.1]nonane-2,6-dione using vegetables

Article abstract of DOI:10.1016/j.molcatb.2013.01.022

Screening of various vegetables as biocatalysts for the stereoselective bioreduction and resolution of racemic bicyclo[3.3.1]nonane-2,6-dione was performed. Vegetables of the family Apiaceae, i.e. the roots of parsnip (Pastinaca sativa), celery (Apium gra

Full text of DOI:10.1016/j.molcatb.2013.01.022

Journal of Molecular Catalysis B: Enzymatic 90 (2013) 66–69
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Stereoselective bioreduction for the resolution of racemic mixtures of
bicyclo[3.3.1]nonane-2,6-dione using vegetables
Albinas Zilinskasa,b, Jolanta Sereikaitea,∗
a Department of Chemistry and Bioengineering, Vilnius Gediminas Technical University, Vilnius, Lithuania
b Department of Organic Chemistry, Vilnius University, Vilnius, Lithuania
a r t i c l e i n f o
a b s t r a c t
Article history:
Screening of various vegetables as biocatalysts for the stereoselective bioreduction and resolution of
racemic bicyclo[3.3.1]nonane-2,6-dione was performed. Vegetables of the family Apiaceae, i.e. the roots
of parsnip (Pastinaca sativa), celery (Apium graveolens), parsley (Petroselinum crispum) and carrot (Daucus
carota) were found to be the best. The methodology described here allows the efficient enantioseparation
of (+)-enantiomer of bicyclo[3.3.1]nonane-2,6-dione. The reaction time, temperature and the amount of
celery roots were optimized and (+)-enantiomer was obtained with the optical purity of 90–98% and the
yield of more than 90%.
Received 2 November 2012
Received in revised form 15 January 2013
Accepted 24 January 2013
Available online 1 February 2013
Keywords:
Racemic bicyclo[3.3.1]nonane-2,6-dione
Vegetables
© 2013 Elsevier B.V. All rights reserved.
Enantioselective reduction
1. Introduction
because many enzymes exhibit a high degree of enantioselectivity
[11–15].
In 2013 we will mark the 100th anniversary of the well known
publication in what Meerwein and Schürmann described the syn-
thesis of so-called Meerwein’s ester and of bicyclo[3.3.1]nonane-
2,6-dione which was obtained from the first one [1]. During
the last century a lot of various interesting researches have
been carried out with bicyclo[3.3.1]nonane and their derivatives
in the field of synthetic organic chemistry and stereochem-
istry [2]. The bicyclo[3.3.1]nonane framework is a common motif
amongst natural products, it is also a useful precursor for the
synthesis of new compounds displaying a broad spectrum of
biological activities [3,4]. For example, the transformation of bicy-
clo[3.3.1]nonane system into bicyclo[5.3.1]undecane ring system
is a substantial step in taxoids synthesis [5]. Undoubtedly, the use
of bicyclo[3.3.1]nonane-2,6-dione for the synthesis of chiral com-
pounds and for the research of their chiroptical properties is a topic
of special importance because a lot of natural compounds with
bicyclo[3.3.1]nonane’s moiety are optically active [6–8]. Thus, the
separation of enantiomers from racemic bicyclo[3.3.1]nonane-2,6-
dione is still actual. Various procedures such as chromatographic
means [9], fractional crystallization of diastereomeric deriva-
tives of bicyclo[3.3.1]nonane-2,6-dione [10] are used for the
enantiomeric enrichment. Enzymatic methods are also applied
In general, for biotransformations isolated enzymes, microor-
ganisms or cells cultures of higher eukaryotes can be used. Plants
are the potential source of various enzymes, and plant cell cultures
as well as plant-derived enzymes are involved in the synthesis
of important chemical compounds. For that purpose, intact plant
materials also can be used because two first mentioned systems
have some limitations as high cost, nonrenewable reagents or
difficulties associated with plant cells cultivation [16,17]. Among
vegetables of the family Apiaceae, a root of carrot is the most known
chemical reagent. To the best of our knowledge, it was the first
plant material used as biocatalyst by Baldassare et al. Racemic 2-
methylcyclohexanone and 2-hydroxycyclohexanone were reduced
with fresh carrot root and chiral alcohols were obtained [18]. Enan-
tioselective hydrolysis of racemic acetates [19], the preparation of
chiral organochalcogeno-␣-methylbenzyl alcohols [20], enantiose-
lective reduction of aliphatic and aromatic ketones, cyclic ketones,
azidoketones and ␤-ketoesters [21,22], reduction of cyclic 3-oxo-
amines [23], chiral resolution of p- or m-substituted racemic aryl
methyl carbinols [24] using carrot root as biocatalyst are described.
The use of celery root for the enantioselective hydrolysis of aryl
ethyl acetates and the reduction of aryl methyl ketones and bromo-
and methoxy-acetophenone derivatives are also described [19,25].
Only one reference was found about the use of roots of parsnip and
parsley in the field of synthetic organic chemistry. Whole tissue
of celery and parsley was used as biocatalyst for the hydrolysis of
acetates and benzoates of 5-hydroxy-4-methyl-3-heptanone [26].
The aim of our study was to screen various vegeta-
bles for the enantioseparation of bicyclo[3.3.1]nonane-2,6-dione.
∗
Corresponding author at: Department of Chemistry and Bioengineering, Faculty
of Fundamental Sciences, Vilnius Gediminas Technical University, Sauletekio al. 11,
LT-10223 Vilnius-40, Lithuania. Tel.: +370 5 2744972; fax: +370 5 2744844.
E-mail address: jolanta.sereikaite@vgtu.lt (J. Sereikaite).
1381-1177/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.01.022

A. Zilinskas, J. Sereikaite / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 66–69
67
H
5
OH
1
O
O
O
5
5
5
1R,5R,6S
O
Plant enzymes
1
1
1
+
H
5
OH
O
O
O
1S,5S
1S,5S
1R,5R
1
H
1R,2S,5R,6S
OH
Scheme 1. Reduction of racemic bicyclo[3.3.1]nonane-2,6-dione by vegetables enzymes.
Stereoselective reduction of racemic diketone by plant enzymes
was performed. The remaining (+)-enantiomer was extracted from
the reaction mixture by organic solvent whereas (−)-enantiomer
had undergone an enzymatic reduction, and the reaction product,
i.e. 6-hydroxybicyclo[3.3.1]nonane-2-one was found.
31.41 (1 C, CH2), 37.09 (2 C, CH2), 43.54 (2 C, CH), 212.64 (2 C,
O).
C
2.2. Reduction of bicyclo[3.3.1]nonane-2,6-dione
The amount of 50 mg of racemic (1) was added to the suspen-
sion of vegetables (8 g) in 30 mL of distilled water. The mixture was
stirring for 24–72 h at 25 ◦C, 30 ◦C or 35 ◦C temperature. Finally, the
suspension was filtered off, and slices of vegetables were washed
with distilled water. The aqueous solution was then extracted with
chloroform (3 × 20 mL), and the organic phase was dried with anhy-
drous Na2SO4, and then evaporated in a vacuum. Additionally, the
reaction product was purified by column chromatography on silica
gel as mentioned above.
2. Experimental
2.1. General
Racemic bicyclo[3.3.1]nonane-2,6-dione (1) was synthesized
from Meerwein’s ester according to the procedure previously
described in [15]. All vegetables were purchased in the local
market. To increase the contact of substrate with a biocatalyst,
vegetables were freshly cut into small thin pieces (approximately
1 cm long slice). Column chromatography was performed on a
silica gel 60 (230–400 mesh, Merck) using chloroform/acetone
(9/1, v/v) as an eluent. TLC was performed on silica gel cov-
ered aluminum plates (60F254, Merck). Compounds were detected
by spraying with 0.5% KMnO4 solution in water. The optical
activity of the final product in chloroform was recorded in a
10 cm path length cuvette using polarimeter Polamat A (Carl
Zeiss/Jena). The optical purity was calculated as followed: optical
2.3. UPLC analysis of reaction aqueous solution and of chloroform
extract
UPLC analysis was performed on the Acquity UPLC system
(Waters) equipped with the Acquity UPLC photodiode array detec-
tor using Acquity UPLC HSS T3 (100 mm × 2.1 mm I.D., 1.8 ␮m)
column at 30 ◦C. A volume of 2 ␮L was injected using a partial
loop injection mode and “needle overfill” as the injection tech-
nique. The separation was achieved within 4 min at a flow rate of
0.25 mL/min. A solvent mixture of water/acetonitrile (20/80, v/v)
containing 15 mM formic acid was used as the mobile phase.
purity = (˛observed/˛pure enantiomer) × 100%. For pure (+)-1, [˛]25
=
546
+205◦ was used [6]. Diketone: white crystals, m.p. 134–135 ◦C.
C9H12O2 (152.19): calcd. C 71.03, H 7.95; found C 70.95, H
7.98. IR (CCl4): ꢀ = 1712 cm−1 (C O). 1H NMR (400 MHz, CDCl3):
ı(ppm) = 2.01–2.14 (m, 4 H, CH2), 2.20 (m, 2 H, CH2), 2.39 (quin-
tet, J = 8 Hz, 2 H, CH2), 2.54–2.62 (m, 2 H, CH2), 2.73 (m, 2 H,
CHCO). 13C NMR (100 MHz, CDCl3): ı(ppm) = 26.65 (2 C, CH2),
2.4. Mass spectrometry analysis
The mass spectrometry analysis was carried out on
a
quadrupole, time-of-flight mass spectrometer (micrOTOF-Q II,
Table 1
Results of the enantioseparation of racemic bicyclo[3.3.1]nonane-2,6-dione by some vegetables.a,b,c
25
No
Vegetables
[˛]
(◦)
Overall yield of diketone (%)
Optical purity (%)
59
546
1
2
3
4
5
6
7
8
9
Roots of Daucus carota
Roots of Apium graveolens
Stems of Allium porrum
Leaves of Brassica oleracea
Cores of Brassica oleracea
Rootstocks of Zingiber officinale
Roots of Raphanus sativus var. niger
Roots of Beta vulgaris atrorubra
Roots of Petroselinum crispum
Roots of Pastinaca sativa
+(38 18)
+(82 44)
+(25 7)
+(9 2)
+(27 9)
+(46 26)
+(4 3)
+(23 3)
+(52 2)
+(91 24)
50 16
43 18
53 10
51 12
51 19
34 15
60 20
71 12
46 12
4
70 11
56
52
57
61
51
56
63
72
2
1
2
6
1
1
1
6
10
36
9
a
The experimental data are presented as mean values standard deviations of 3–6 parallel experiments.
Reaction was carried out at 30 ◦C for 72 h.
Entries 1, 2, 9, 10; entry 3; entries 4, 5, 7; entry 6 and entry 8 belong to the families Apiaceae, Amaryllidaceae, Brassicaceae, Zingiberaceae and Amarathaceae, respectively.
b
c

68
A. Zilinskas, J. Sereikaite / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 66–69
3. Results and discussion
Previously, racemic mixtures of bicyclo[3.3.1]nonane-2,6-dione
were separated using a horse liver alcohol dehydrogenase cat-
alyzed reduction coupled with the regeneration of the coenzyme
NAD by dithionite or ethanol [13]. For the stereospecific reduc-
tion and the resolution of the enantiomers of that compound,
baker’s yeast were also applied [11,15]. Thus, having in mind pre-
vious papers we could expect that biocatalytic enantioselective
reduction would take place (Scheme 1). In this sense, preliminary
screening revealed that vegetables of the family Apiaceae, i.e. the
roots of parsnip (Pastinaca sativa), celery (Apium graveolens), pars-
ley (Petroselinum crispum) and carrot (Daucus carota) showed the
best results in the enantioseparation of racemic (1) (Table 1). The
final product, i.e. diketone extracted with chloroform was enriched
by (+)-enantiomer of (1). The results obtained with rootstocks of
ginger (Zingiber officinale), which belongs to another plant family
of Zingiberaceae, were similar as for carrots. Roots of celery were
chosen for more detailed investigation and optimization of reac-
tion conditions (Fig. 1). Additional experiments showed that the
highest optical purity of (+)-(1) can be reached at 25 ◦C after 48 h
(Fig. 1A). Moreover, the enantioseparation of racemic mixture was
performed under those conditions using four-fold higher amount
of (1), i.e. 200 mg of (1) was added to the suspension of vegetables
(20 g) in 140 mL of water. (+)-Enantiomer was obtained in the yield
of 50–54% and with the excellent optical purity of 90–98%. The 1.5-
fold increase of celery amount, i.e. 30 g of vegetables for 200 mg of
racemic diketone had no effect on (+)-1 purity, but lower amount of
celery (8 g for 200 mg of diketone) was not enough, and the optical
purity of (+)-1 was only 59%. Under reaction conditions mentioned
above (200 mg of diketone, 20 g of vegetables, 25 ◦C, 48 h) we also
obtained (+)-enantiomer in the yield of 40–45% and with optical
purity of 92–96% using roots of parsnip (P. sativa). To increase the
yield of (+)-1, we changed the procedure of diketone extraction pre-
viously mentioned in Section 2.2. When the suspension had been
filtered off the slices of vegetables (roots of celery) were ground, and
water was poured over the mash. After centrifugation both water
fractions were joined and extracted with chloroform as mentioned
in Section 2.2. This procedure increased the yield of pure (+)-1 up
to more than 90%.
As can be seen in Table 1, entry 7 (Raphanus sativus var. niger) has
no enzymatic activity for (1) biotransformation or an enzyme does
not exhibit enantioselectivity (Table 1). The same conclusion could
be drawn in regard to Table 1 entries 3 (Allium porrum), 4 (Brassica
oleracea) and 5 (B. oleracea). About 50% of diketone amount was
lost, but the optical purity increased only slightly. The prolongation
of reaction time (Fig. 1A) or the increase of reaction temperature
(Fig. 1B and C) did not increase the optical purity. It is possible
that a reverse reaction takes place or there are two enzymes of
different enantioselectivity, i.e. one for (+)-1 and another for (−)-1
enantiomer, and the rates of their catalyzed reaction are different.
Thus, the product of slower reaction decreases the optical purity
Fig. 1. Dependence of overall diketone yield (ꢀ) and optical purity (ꢁ) on the reac-
tion duration at 25 ◦C (A), 30 ◦C (B) and 35 ◦C (C). Reaction mixture contained 50 mg
of (1) and 8 g of celery in 30 mL of water.
Bruker Daltonik GmbH). The instrument was modified for medium
masses (50–1500 amu). The following instrumental parameters
were used: transfer RF1, 200 Vpp; RF2, 300 Vpp; hexapole RF,
300 Vpp; ion energy, 5.0 eV; low mass, 100 Vpp; collision energy,
8.0 eV; collision RF, 200 Vpp; transfer time, 80.0 ␮s; pre pulse stor-
age, 5.0 ␮s. The ion source was operated in the positive mode on
the following set parameters: capillary voltage, 4.5 kV; nebulizer,
1.2 bar; flow rate, 7 mL/min; dry temperature, 180 ◦C. All mass spec-
tra were calibrated externally using the low concentration tuning
mix (ESI-L, Agilent Technologies) and were processed with Data-
Analysis 4.0 software (Bruker Daltonik GmbH).
Fig. 2. Analysis of reaction aqueous solution using Acquity UPLC HSS T3
(100 mm × 2.1 mm I.D., 1.8 ␮m) column.

A. Zilinskas, J. Sereikaite / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 66–69
69
4. Conclusions
Vegetables of the family Apiaceae are promising biocatalysts
for the stereoselective bioreduction and enantioseparation of
racemic bicyclo[3.3.1]nonane-2,6-dione. This methodology allows
to recover the (+)-enantiomer with a very good enantiomeric
purity. This biotechnological process is ecofriendly and cheap, the
isolation of the final product is easy and vegetables as biocata-
lysts are easy available. The method for the enantioseparation of
bicyclo[3.3.1]nonane-2,6-dione described in this work has some
advantages as compared with other enzymatic methods published
previously. First, it is more cheaper than the use of horse liver alco-
hol dehydrogenase and does not require additional experiments for
cofactor regeneration. Second, the use of vegetables is very simple
as compared with the enantioseparation of bicyclo[3.3.1]nonane-
2,6-dione by baker’s yeast.
Fig. 3. Analysis of chloroform extract using Acquity UPLC HSS T3 (100 mm × 2.1 mm
I.D., 1.8 ␮m) column.
References
[1] H. Meerwein, W. Schurmann, Justus, Liebigs Ann. Chem. 398 (1913)
196–242.
[2] E. Butkus, Synlett 12 (12) (2001) 1827–1835.
[3] P. Camps, R. El Achab, J. Morral, D. Munoz-Torrero, A. Badia, J.E. Banos,
N.M. Vivas, X. Barril, M. Orozco, F.J. Luque, J. Med. Chem. 43 (2000)
4657–4666.
Fig. 4. Mass spectrometry analysis of peak 1 from Figs. 2 and 3.
[4] K. Murayama, T. Tanabe, Y. Ishikawa, K. Nakamura, S. Nishiyama, Tetrahedron
Lett. 50 (2009) 3194–3196.
[5] G. Mislin, M. Miesch, Eur. J. Org. Chem. 9 (2001) 1753–1759.
[6] E. Butkus, A. Zilinskas, S. Stoncius, R. Rozenbergas, M. Urbanova, V. Setnicka, P.
Bour, K. Volka, Tetrahedron: Asymmetry 13 (2002) 633–638.
[7] S. Stoncius, E. Butkus, A. Zilinskas, K. Larsson, L. Ohrstrom, U. Berg, K. Warnmark,
J. Org. Chem. 69 (2004) 5196–5203.
[8] G. Bagdziunas, E. Orentas, U. Berg, A. Zilinskas, E. Butkus, Eur. J. Org. Chem. 25
(2007) 4251–4256.
[9] E. Butkus, U. Berg, J. Chem. Res. Synop. (1993) 116–117.
[10] H. Gerlach, Helv. Chim. Acta 61 (1978) 2773–2776.
[11] G. Hoffmann, R. Wiartalla, Tetrahedron Lett. 23 (1982) 3887–3888.
[12] M. Miyazawa, M. Nobata, S. Okamura, O. Muraoka, G. Tanabe, H. Kameoka, J.
Chem. Technol. Biotechnol. 71 (1998) 281–284.
[13] J. Malinauskiene, P. Kadziauskas, A. Malinauskas, J. Kulys, Monatsh. Chem. 130
(1999) 1513–1517.
[14] E. Butkus, U. Berg, J. Malinauskiene, J. Sandstrom, J. Org. Chem. 65 (2000)
1353–1358.
Fig. 5. Mass spectrometry analysis of peak 2 from Figs. 2 and 3.
[15] C.J. Wallentin, E. Orentas, E. Butkus, K. Warnmark, Synthesis
0864–0867.
[16] K. Ishihara, H. Hamada, T. Hirata, N. Nakajima, J. Mol. Catal. B: Enzym. 23 (2003)
145–170.
[17] G.A. Cordell, T.L.G. Lemos, F.J.Q. Monte, M.C. de Mattos, J. Nat. Prod. 70 (2007)
478–492.
[18] F. Baldassare, G. Bertoni, C. Chiappe, F. Marioni, J. Mol. Catal. B: Enzym. 11 (2000)
55–58.
[19] W.K. Maczka, A. Mironowicz, Tetrahedron: Asymmetry 13 (2002)
2299–2302.
[20] J.V. Comasseto, A.T. Omori, A.L.M. Porto, L.H. Andrade, Tetrahedron Lett. 45
(2004) 473–476.
[21] J.S. Yadav, P. Thirupathi Reddy, S. Nanda, A. Bhaskar Rao, Tetrahedron: Asym-
metry 24 (2002) 3381–3385.
[22] J.S. Yadav, S. Nanda, P. Thirupathi Reddy, A. Bhaskar Rao, J. Org. Chem. 67 (2002)
3900–3903.
5 (2009)
with the prolongation of reaction time. At higher temperature this
effect is observable in a shorter time, and at 35 ◦C the optical purity
of (1) is almost the same after 24 and 72 h.
For the identification of reaction products, UPLC separation
and identification of compounds by mass spectrometry analy-
sis were applied. Two peaks are seen on the chromatograms
of both reaction mixture in water (Fig. 2) and of chloro-
form extract (Fig. 3). Mass spectrometry analysis shows that
bicyclo[3.3.1]nonane-2,6-dione is eluted first and the reduction
product, i.e. 6-hydroxybicyclo[3.3.1]nonane-2-one is eluted as the
second peak (Figs. 4 and 5). It is noteworthy that under the reaction
conditions used in the present work bicyclo[3.3.1]nonane-2,6-diol
was not found. It should be mentioned that (+)-1 is the remaining
enantiomer in our reaction as well as in yeast-catalyzed reduction
[15]. On contrary, a horse liver alcohol dehydrogenase catalyzes the
reduction of (+)-1 and (−)-1 is the remaining enantiomer [13].
[23] R. Lacheretz, D.G. Pardo, J. Cossy, Org. Lett. 11 (2009) 1245–1248.
[24] H. Nagaoka, Biotechnol. Progr. 20 (2004) 128–133.
[25] W.K. Maczka, A. Mironowicz, Tetrahedron: Asymmetry 15 (2004)
1965–1967.
[26] B. Bohman, L.R. Cavonius, C.R. Unelius, Green Chem. 11 (2009) 1900–1905.