Efficient cyclization of squalene epoxide to lanosterol with immobilized cells of baker's yeast

Article abstract of DOI:10.1016/S0040-4020(02)00755-X

The cyclization of squalene epoxide to lanosterol with baker's yeast (Saccharomyces cerevisiae) can conveniently be carried out in aqueous solution with glass cored immobilisates of cells in calcium alginate. This enables the manifold use of the microorganism to obtain lanosterol in a single biocatalytic step using the immobilisates repeatedly.

Full text of DOI:10.1016/S0040-4020(02)00755-X

Tetrahedron 58 (2002) 7291–7293
Efficient cyclization of squalene epoxide to lanosterol with
immobilized cells of baker’s yeast
*
Olaf Rotthaus and Martin Demuth
¨ ¨
Max-Planck-Institut fur Strahlenchemie, P.O. Box 101 365, D-45413 Mulheim an der Ruhr, Germany
Received 4 April 2002; revised 14 June 2002; accepted 28 June 2002
Abstract—The cyclization of squalene epoxide to lanosterol with baker’s yeast (Saccharomyces cerevisiae ) can conveniently be carried out
in aqueous solution with glass cored immobilisates of cells in calcium alginate. This enables the manifold use of the microorganism to obtain
lanosterol in a single biocatalytic step using the immobilisates repeatedly. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
In recent years baker’s yeast (Saccharomyces cerevisiae )
gained increasing importance as a biocatalyst in asymmetric
synthesis due to good yields and high stereoselectivity
shown in many examples.1–6 Yeast reactions have also
successfully been applied in natural product synthesis.7–10
Although most of the known applications are carried out
with whole cell systems, examples are known where the
microorganisms have been immobilized prior to their
use.7,11–13
Whole cells of baker’s yeast deliver upon ultrasonic
pretreatment in aqueous suspension 2 with 42% yield
starting with racemic 1. With respect to the highly
hydrophobic substrate the addition of a detergent (here:
Triton X 100) is necessary.17 However, the detergent affects
the cell membranes and leads to an inactivation of the cells
within a single conversion; a re-use of the cells is therefore
not possible.18 Herein we report on investigations to achieve
re-usable immobilisates of yeast cells by different tech-
niques protecting the microorganisms from decomposition.
Among the numerous yeast enzymes lanosterol synthase is
doubtless the most remarkable one because of its ability to
build up four rings and to introduce seven asymmetric
centers selectively during the conversion of squalene
epoxide (1) to lanosterol (2, see Scheme 1; for mechanistic
aspects of this and related transformations, see Corey
et al.14–16).
A common method to entrap cells constitutes the immobil-
ization in polyacrylic amide that has been applied for
different organisms and also for the yeast-mediated
transformation of sekoketone to sekoketol.7 In our case
this procedure could not be used successfully as the number
of cells that could be immobilized in the polymer was far too
low.
Scheme 1.
Keywords: baker’s yeast; squalene epoxide; immobilization; calcium alginate; lanosterol synthase.
*
Corresponding author. Tel.: þ49-208-306-3671/3680/3684; fax: þ49-208-306-3951; e-mail: demuthm@mpi-muelheim.mpg.de
http://www.mpi-muelheim.mpg.de/mpistr_home.html
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00755-X

7292
O. Rotthaus, M. Demuth / Tetrahedron 58 (2002) 7291–7293
In a second attempt to encapsulate baker’s yeast cells in
polyamide microcapsules, a procedure by Green et al.17 for
the reduction of 2-phenyl-1,2-propanedione to the corre-
sponding 2-hydroxy compound, was adopted and modified
in this work. Due to the fact that the microcapsules contain
the cells in an aqueous environment, but are applied in
organic solvents—in the present case in octane and
decane—the achievement of this method could have been
to carry out the reaction directly in a lipophilic medium with
no need of a detergent at all. But incapsulated yeast cells
showed no cyclization activity which is most likely related
to the well known fragility of the enzyme lanosterol
synthase that seems to be inactivated even by traces of
organic solvents present in the capsule cores. However, the
permeability of the capsule shells for the substrates was
experimentally proved.
Table 1. Cyclization of squalene epoxide with immobilized cells of baker’s
yeast
Number of conversionsa
Cells in calcium
alginate
Cells in calcium
alginate on glass
beads
Yield (%)
cfu/gb
Yield (%)
cfu/gb
1
2
3
4
5
12
11
8
7
4
1.4£108
4.9£107
3.4£107
9.3£107
8.2£107
28
24
19
11
7
1.6£108
5.1£107
3.2£107
9.1£107
7.9£107
Added yields
42
89
a
Each with a fresh solution of squalene epoxide.
b
cfu/g, cell forming units per gram of immobilisate (excluded the weight
of the glass beads) counted on agar plates. Fresh cells: 9.8£108 cfu/g.
Immobilisates have been stored in nutrition medium over night between
the third and fourth conversion.
Thirdly, calcium alginate was used as a polymer for the
immobilization. In this case, highly stable gel beads were
obtained that were capable of carrying out the cyclization
and could be used in five consecutive conversions. Even
thereafter active cells could still be detected in the
suspension as shown in Table 1 (first part). However, the
overall yield of 42% was still low and we assumed that
the substrate did not reach the center of the beads by
diffusion and therefore, only cells just beneath the surface
were reactive at all. To overcome this problem a second type
of alginate immobilisates was prepared. This type consisted
of a glass core being coated with a thin layer of alginate
containing the cells to make them more easily accessible for
the substrate. With these immobilisates the total yield of
lanosterol formation could be more than twice up to 89%
upon a fivefold use of the glass-cored beads, however, with
decreasing yields within the experimental series (see Table
1, second part). Notably, for each run of this and the
previously described experimental set a fresh solution of
squalene epoxide was employed.
In addition to these investigations we tested if the high
chemoselectivity of lanosterol synthase could also be
applied to cyclizations of smaller polyene epoxides. There-
fore, conversions of the epoxides 3–10 (see Scheme 2) were
investigated using again immobilized cells of baker’s yeast
under conditions that were successful for the cyclization of
squalene epoxide. However, substrates 3, 4, 6 and 8 were
converted to vicinal diols instead upon hydrolysis of the
epoxides in 6–23% yields (spectroscopic data of these
products are available from Ref. 18 or upon request from the
corresponding author). Cyclic products could not be
detected in these cases and notably, the epoxides 5, 7, 9
and 10 were unreactive at all. According to these results we
have to assume that either the selectivity of the lanosterol
synthase is too high to accept non-native epoxides or
another enzyme—an epoxide hydrolase—reacts faster than
the cyclase with these shorter chain substrates.
With these experiments we have shown that a manifold use
of yeast cells for the cyclization of squalene epoxide to
lanosterol is possible in spite of the necessity of Triton X
100 as a detergent to dissolve the polyene substrate if the
cells are protected by immobilization. With calcium
alginate coated glass beads higher overall yields could be
obtained than with a single use of free cells. This is a crucial
step in the development of a multi-usable biocatalytic
system for one-step steroid synthesis. With respect to the
still limited repeated use of the beads and the relatively large
amount of cells needed for the immobilisates these
investigations will still have to be continued. A further
increase of the immobilisate surface seems to be a
promising goal in view of the results that have been
obtained with glass-cored versus solid beads.
3. Experimental
3.1. General
All reactions were carried out with racemic 2,3-epoxy
squalene. The cyclization product, lanosterol, was identified
by thin layer chromatography on Merck 60 F254 precoated
silica gel plates and comparison with commercial available
material. Yeast cells were purchased from Sigma (baker’s
Yeast, Type I), sodium alginate (BioChemika) from Fluka.
Water was always used tridistilled with a Millipore Q
system. Reactions with alginate immobilisates were carried
out in tris-(hydroxymethyl)-aminomethane (TRIS) buffered
solution at a pH of 7.4.
Scheme 2.

O. Rotthaus, M. Demuth / Tetrahedron 58 (2002) 7291–7293
7293
3.2. Immobilization of yeast cells in calcium alginate
immobilisates were taken and their weight was determined
precisely. They were then dissolved in a 1% sodium
ethylendiamine tetraacetate (NaEDTA) solution and diluted
to exactly 10 mL. 1 mL of this solution was a second time
diluted to 10 mL and this procedure was repeated two more
times to yield a 1024 fold diluted sample. 100 mL of this
sample were distributed on an agar plate that was stored for
48 h at 308C. Then the cell forming units were counted
microscopically and the cfu/g values were calculated.
2.0 g sodium alginate were dissolved in 60 mL of water and
combined with a suspension of 16 g dried yeast cells in
another 60 mL of water. Using a syringe, this suspension
was now dropwise given into 800 mL of 2% aqueous
solution of calcium chloride. During this procedure the
droplets stiffened immediately when they had contact with
the calcium chloride solution and formed rigid beads with a
diameter of 2–3 mm in average. The mixture was stirred for
further 30 min to harden the immobilisates, then they were
collected with a sieve and washed three times with each
800 mL of water.
References
3.3. Immobilization of yeast cells in calcium alginate on
glass beads
1. Servi, S. Synthesis 1990, 1–24.
¨
2. Csuk, R.; Glanzer, B. Chem. Rev. 1991, 91, 49–97.
3. Heidlas, J.; Tressl, R. Eur. J. Biochem. 1990, 188, 165–174.
4. Ward, O. P.; Young, C. S. Enzyme Microb. Technol. 1990, 12,
482–493.
300 g of glass beads with a diameter of 1.5–2 mm were
added to the solution of 16 g yeast cells and 2.0 g sodium
alginate in 60 mL of water and mixed thoroughly. During
this procedure the glass beads were covered with a thin layer
of the highly viscid alginate solution. The coated glass beads
were slowly poured into 800 mL of an aqueous solution
containing 2% of calcium chloride and left in this
environment for 30 min. Then the glass-cored beads were
collected and washed like the solid alginate beads described
in the previous procedure.
5. North, M. Tetrahedron Lett. 1996, 37, 1699–1702.
6. Rotthaus, O.; Kru¨ger, D.; Demuth, M.; Schaffner, K.
Tetrahedron 1997, 53, 935–938.
7. Koshcheyenko, K. A.; Bychkovy, G. G.; Gulevskaya, S. A.;
Lusta, K. A.; Gulaya, V. E.; Anachenko, S. N. Iz. Akad. Nauk.
SSSR. Ser. Biol. 1978, 6, 862–871.
8. Naoshima, Y.; Nakamura, A.; Munakata, Y.; Kamezawa, M.;
Tachibana, H. Bull. Chem. Soc. Jpn 1990, 63, 1263–1265.
9. Naoshima, Y.; Hasegawa, H.; Nishiyama, T.; Nakamura, A.
Bull. Chem. Soc. Jpn 1989, 62, 608–610.
3.4. Cyclization of squalene epoxide with alginate beads
(solid or with glass core)
10. Sugai, T.; Katoh, O.; Ohta, H. Tetrahedron 1995, 44,
11987–11998.
The immobilisates that have been prepared by one of the
above described procedures were given into a solution of
80 mg squalene epoxide and 1.0 g Triton X 100 in 80 mL of
Tris buffer and stirred for 20 h at a temperature of 308C.
Then the beads were separated from the solution, washed
three times with 40 mL portions of water and the collected
aqueous solutions were dried by lyophilization. The
resulting dry powder was extracted with three portions
(each 60 mL) of ether, evaporated and the reaction product
was isolated by column chromatography on 200 g silica gel
(grain size: 0.200–0.063 mm, Merck).
11. Zaborsky, O. Immobilized Enzymes. CRC: Cleveland, OH,
1977; pp 83–163.
12. Chibata, I.; Tosa, T.; Sato, T.; Mori, T. Immobilized Enzymes:
Research and Development. Wiley: New York, 1978.
13. Klyosov, A. A. Report UNIDO/IPTC, 93, V-89-61316; Order
No. PB90-210360.
14. Wendt, K. U.; Schulz, G. E.; Corey, E. J.; Liu, D. R. Angew.
Chem. 2000, 112, 2930–2952.
15. Corey, E. J.; Matsuda, S. P. T. J. Am. Chem. Soc. 1991, 113,
8172–8174.
16. Corey, E. J.; Matsuda, S. P. T.; Bartel, B. Proc. Natl Acad. Sci.
USA 1994, 91, 2211–2215.
The recovered alginate beads (solid or with glass core) were
used for five consecutive runs, each with a fresh solution of
squalene epoxide.
17. Green, K. D.; Gill, I. S.; Khan, J. A.; Vulfson, E. N.
Biotechnol. Bioengng 1996, 49, 535–543.
18. Rotthaus, O. Ph.D. Thesis, Max-Planck-Institut fu¨r
Strahlenchemie/University of Essen, ISSN 0932-5131, 1999.
3.5. Determination of the cell forming units per gram of
immobilisates (cfu/g) on agar plates
After every conversion samples of approximately 1 g of the