Microbial side-chain degradation of ergosterol and its 3-substituted derivatives: A new route for obtaining of deltanoids

Article abstract of DOI:10.1016/j.steroids.2010.04.001

The strain of Mycobacterium sp. VKM Ac-1815D was found to convert ergosterol and its 3-acetate mainly to androst-4-ene-3,17-dione (AD) thus demonstrating ability to reduce 7(8)-double bond and hydrolyze sterol ester in addition to oxidation of 3β-hydroxy group,δ5-δ4 isomerization and side-chain degradation. Ergosterol bioconversion in the presence of isoflavones and ions of some bivalent metals known inhibitors of 3-hydroxysteroid dehydrogenase, did not alter products composition. Protection of ergosterol 3β-hydroxyl with methoxymethyl group allowed the formation of bioconversion products retaining the δ 5,7-configuration. The major product was identified by mass-spectrometry and proton NMR as 3- methoxymethoxy-androsta-5,7-diene-17-one (MA). TheMAproducing activity was found to be inducible with sterols, cholestenone or lithocholic acid, but not with dehydroepiandrosterone, AD, androsta-1,4- ene-3,17-dione or organic acids. Under the optimized conditions, the yield ofMAreached 5 g/l from 10 g/l O-methoxymethyl-ergosterol (approx. 60% molar conversion) for 120 h. The results might be applied at the production of novel vitamin D derivatives.

Full text of DOI:10.1016/j.steroids.2010.04.001

Steroids 75 (2010) 653–658
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Microbial side-chain degradation of ergosterol and its 3-substituted derivatives:
A new route for obtaining of deltanoids
Dmitry V. Dovbnya∗, Olga V. Egorova, Marina V. Donova
G.K. Skryabin Institute of Biochemistry & Physiology of Microorganisms Russian Academy of Sciences, Pushchino, Moscow Region, 142290, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
The strain of Mycobacterium sp. VKM Ac-1815D was found to convert ergosterol and its 3-acetate mainly
to androst-4-ene-3,17-dione (AD) thus demonstrating ability to reduce 7(8)-double bond and hydrolyze
sterol ester in addition to oxidation of 3␤-hydroxy group, ꢀ5-ꢀ4 isomerization and side-chain degrada-
tion. Ergosterol bioconversion in the presence of isoflavones and ions of some bivalent metals – known
inhibitors of 3␤-hydroxysteroid dehydrogenase, did not alter products composition. Protection of ergos-
terol 3␤-hydroxyl with methoxymethyl group allowed the formation of bioconversion products retaining
the ꢀ5,7-configuration. The major product was identified by mass-spectrometry and proton NMR as 3-
methoxymethoxy-androsta-5,7-diene-17-one (MA). The MA producing activity was found to be inducible
with sterols, cholestenone or lithocholic acid, but not with dehydroepiandrosterone, AD, androsta-1,4-
ene-3,17-dione or organic acids. Under the optimized conditions, the yield of MA reached 5 g/l from 10 g/l
O-methoxymethyl-ergosterol (approx. 60% molar conversion) for 120 h. The results might be applied at
the production of novel vitamin D derivatives.
Received 27 February 2010
Received in revised form 31 March 2010
Accepted 1 April 2010
Available online 10 April 2010
Keywords:
Deltanoids
Ergosterol
Mycobacterium
Bioconversion
Sterol side-chain degradation
© 2010 Elsevier Inc. All rights reserved.
1. Introduction
dehydroepiandrosterone (DHEA) that are converted to the corre-
sponding 5,7-dienes with further additional modifications [7,8], or
The biological activities of vitamin D metabolites determine
wide potent therapeutic applications of vitamin D hormone for the
treatment of hyperproliferative disorders (e.g. cancer and psoria-
sis), immune dysfunction (autoimmune diseases), and endocrine
disorders (e.g. hyperparathyroidism), as well as osteoporosis
[1,2]. Unfortunately, the pharmacologically relevant doses of the
naturally occurring, biologically active form of vitamin D (i.e.
1␣,25-dihydroxylated ergo- or cholecalciferol derivatives) can pro-
duce hypercalcemia, or phosphataemia. In order to separate the
therapeutically important properties from the undesirable toxic
activities, the synthetic analogs of the vitamin D are being creating
[3].
Natural and synthetic analogs of vitamins D2 and D3 termed
deltanoids are of raising interest for their relatively high non-
calciotropic activities that include inhibition of cell proliferation,
promotion of cell differentiation and apoptosis, modulation of
immune cell function and antimicrobial action. More than 1500
analogs have already been synthesized and biologically evaluated,
but very few of them are actively used in medicine [4,5]. Positive
therapeutic effects of artificial deltanoids related to the structural
modifications in vitamin D rings and side-chain [3,6]. The syn-
theses are based either on linear strategy starting from sterols or
converged strategy where desired structures are obtained by means
of combination of specially designed building blocks [9–11]. Spe-
cial conditions of photolysis of deltanoid molecules were developed
for effective opening of steroid core and rearrangement of double
bonds within the secosteroid products [8,12].
Recently, we have developed the method for production of
3␤-hydroxy-androsta-5,7-diene-17-one – key intermediate for the
synthesis of novel vitamin D analogs with modified side-chain [13].
The method includes microbial 7␣-hydroxylation of commercially
available DHEA followed by chemical introduction of the 7(8)-
double bond.
In this study, the strain of Mycobacterium sp. VKM Ac-
1815D capable of effective ␤-sitosterol and cholesterol side-chain
degradation was used for the conversion of ergosterol and its
3-substituted derivatives and a route for single stage microbial
production of ꢀ5,7-steroids of androstane family was suggested.
Ergosterol is a major membrane sterol in fungi and phy-
toplankton that potentially could be used as a natural source
for obtaining of deltanoids by microbial means. Microbial con-
version of ergosterol is poorly investigated. In few publications
known, degradation of ergosterol side-chain was accompanied
with steroid nucleus modification thus forming androst-4-ene-
3,17-dione (AD) as
a major product [14,15]. The obtaining
of 3-substituted 20-hydroxymethylpregna-5,7-diene-3␤-ol from
3␤-methoxymethoxy-ergosta-5,7,22-triene by bioconversion with
mycobacterial strains was patented [16]. Bioconversion of ergos-
∗
Corresponding author. Tel.: +7 4967730738; fax: +7 4959563370/4967733835.
E-mail address: anagoge@rambler.ru (D.V. Dovbnya).
0039-128X/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2010.04.001

654
D.V. Dovbnya et al. / Steroids 75 (2010) 653–658
terol or its derivatives to a product with fully degraded side-chain
HPLC: The analysis was carried out on a Series 1200 system (Agi-
lent, USA) with Symmetry C18, 5 ␮m, 250 mm × 4.6 mm column
(Waters, USA) using linear gradient of acetonitrile 50–100% start-
ing from 15 min and ending at 25 min after injection (Method 1) or
isocratic mobile phase composed of acetonitrile:2-propanol:water
55:45:5 (Method 2) at 50 ◦C, flow rate 1 ml/min and UV absorbance
detection at 282, 240 and 200 nm. Non-identified ꢀ5,7-steroids
were preliminary quantified using ME as a standard.
and preserved ꢀ5,7-configuration was hitherto unreported.
2. Experimental
2.1. Chemicals and reagents
Sterols: ergosterol (ergosta-5,7,22-trien-3␤-ol; cholesta-
5,7,22-trien-24␤-methyl-3␤-ol), 98% purity, and cholestenone
(cholest-4-en-3-one) were obtained from Acros Organics
(Belgium); O-methoxymethyl-ergosterol (3␤-methoxymethoxy-
ergosta-5,7,22-triene, ME) was obtained from KNC Laboratories
(Japan) and purified to 95% by two re-crystallizations from
ethanol; ergosterol 3-acetate and stigmasterol were from Fluka
(Switzerland); ␤-sitosterol, 79.2% purity from S&D Chemicals
(England); cholesterol from Serva (Germany). Steroids: androst-
4-ene-3,17-dione (AD), androst-1,4-diene-3,17-dione (ADD),
dehydroepiandrosterone (DHEA) and others were obtained from
Steraloids (USA). ␣-Naphtoflavone was purchased from Sigma
(USA), formononetin and biochanin A from Fluka (Switzerland).
Yeast extract and agar were from Difco (USA), randomly methy-
lated ␤-cyclodextrin (MCD) from Wacker Chemie (Germany).
Other materials and solvents were of research or analytical grade
and purchased from Panreac (Spain).
2.5. Isolation of bioconversion products
Bioconversion products were isolated by solid-phase extraction
from 10% or 20% (v/v) acetonitrile solutions using Lichrolute RP-18
columns (Merck, Germany), eluted with mixture of acetonitrile and
2-propanol (55:45), concentrated under vacuum and purified two
times by semi-preparative HPLC on Lichrocart-Lichrospher RP-18
10 ␮m, 250 mm × 10 mm column (Merck, Germany) at a flow rate
5 ml/min. Other conditions were as described above for Method 2.
2.6. Identification of products structure
UV-spectra were recorded on a Shimadzu UV-1700 spec-
trophotometer (Japan) in acetonitrile solutions. Mass-spectra were
obtained on a Finnigan SSQ-710 spectrometer (USA) at the energy
of ionization 70 eV using direct sample injection technique. 1H NMR
spectra were recorded on a Varian Unity +400 spectrometer (USA)
at 400 MHz using CDCl3 as a solvent; the signal of CHCl3 resting
protons (ı 7.24) was used as an internal standard.
2.2. Microorganism, cultivation, bioconversion of ME by resting
cells
A strain of Mycobacterium sp. VKM Ac-1815D was obtained from
All-Russian Collection of Microorganisms (VKM IBPM RAS) and cul-
tivated as reported earlier [17]. Preparation of resting cells and
bioconversion were carried out as per [18]. ME concentration was
2 g/l (1.78 mM).
3. Results
3.1. Ergosterol bioconversion by Mycobacterium sp. VKM
Ac-1815D
2.3. Bioconversion of ergosterol and its derivatives by growing
cells
The strain of Mycobacterium sp. VKM Ac-1815D was shown
to be capable of transforming phytosterol and cholesterol to AD
[17–19]. When incubating with ergosterol, the strain also formed
AD as a major product thus demonstrating the ability to reduce
7(8)-double bond in addition to modification of 3␤-ol-5-ene into
3-keto-4-ene functionality and cleavage of ergosterol side-chain.
The molar yield of AD from 12.06 mM (4.78 g/l) ergosterol reached
58.6% for 120 h; ADD and 20-hydroxymethylpregn-4-ene-3-one
(HMP) were formed as minor products; approximately 12% of the
Bioconversions by growing cultures were carried out in shake-
flasks at 30 ◦C, 200 rpm in 100 ml of the medium containing (g/l):
ergosterol, ergosterol acetate or ME – 2–10; glycerol – 5; (NH4)2SO4
– 3; MgSO4 – 0.2; FeSO4 – 0.01; ZnSO4 – 0.002, Tween 80 – 0.5
in 0.05 M potassium phosphate buffer (pH 7.0), where mentioned
supplemented with MCD. The portions of ME were loaded to the
medium in three different ways: before sterilization in autoclave
(110 ◦C, 30 min); sterilized separately with 0.05 g Tween 80 and
30 ml water (110 ◦C, 30 min) or added to 30 ml of the sterile aqueous
Tween 80, sonicated during 40 min at 70 ◦C on ultrasonic bath till
total homogenization and then combined with the rest of the sterile
medium. The target activity was induced during 6–20 h; inducers
were added as sterile aqueous suspensions, hot concentrated solu-
tions in dimethylformamide, or directly (organic acids) to the final
concentration 0.04 mM if not mentioned otherwise.
substrate remained unconverted (Fig. 1A). No steroids with ꢀ5,7
-
configuration were revealed. Conversion of ␤-sitosterol (12.06 mM,
5 g/l) was faster and more efficient; the substrate was totally con-
verted to 72 h with total molar yield of products with 3-keto-4-ene
configuration more than 96% (Fig. 1B) as compared with 80% in the
case of ergosterol.
␣-Naphtoflavone and formononetine did not cause any notice-
able effect on the ergosterol bioconversion. The presence of Fe2+
,
Mn2+ or Mg2+ as well as biochanin A only slightly influenced the
rate and yield of AD(D) formation from ergosterol. The addition of
ZnSO4 resulted in lower yield of AD that was accompanied by cor-
responding increase of ADD yield. CoCl2 in concentration of 5 mM
slowed down the bioconversion with the final AD yield of 35-37%
for 140 h. No steroidal metabolites were detected at the addition of
CuSO4 (data not shown).
In some experiments the medium was supplemented with ␣-
naphtoflavone, formononetin, or biochanin A (10–20 mM), as well
as 2–5 mM CoCl2, CuSO4, FeSO4, MgSO4 or MnSO4.
2.4. Steroid analyses
TLC: Steroids were extracted from samples of cultivation broth
with 5 volumes of ethyl acetate. The extracts were spotted on
Sorbfil PTLC-AF-A-UV TLC plates (Sorbpolymer, Russia), devel-
oped in benzene/acetone (3:1, v/v), benzene/acetone (30:2, v/v)
or hexane:diethyl ether:methanol (30:30:2, v/v/v). Steroids were
visualized under UV light (254 nm) then TLC plates were sprayed
with 4% phosphomolybdic acid hydrate solution in ethyl alcohol.
3.2. Bioconversion of 3-substituted ergosterol by Mycobacterium
sp. VKM Ac-1815D
Bioconversion of 3-esterified and 3-etherified ergosterol deriva-
tives was studied. As presented in Table 1, AD and ADD were
accumulated as major products from ergosterol 3-acetate yielding

D.V. Dovbnya et al. / Steroids 75 (2010) 653–658
655
Fig. 1. Time course of bioconversions of ergosterol and ␤-sitosterol by growing culture of Mycobacterium sp. VKM Ac-1815D.
totally 1.62 mM (35.5% of theor.) for 187 h. Free ergosterol (de-
esterified substrate) was detected in small amounts, while 45% of
ergosterol 3-acetate remained unconverted after 187 h.
Unlike the bioconversion of the esterified substrate, the
bioconversion of ergosterol with methoxymethyl protection of 3␤-
hydroxyl – ME resulted in very slow accumulation of products
with UV absorbance maximum near 282 nm. No formation of 3-
keto-4-ene-steroids (AD, ADD, or HMP) was observed. The target
ME-converting activity of the resting cells increased at the induc-
tion with ␤-sitosterol. The induced cells exhibited 3–4 times higher
activity and mainly produced one product as compared to non-
induced cells (data not shown). HPLC retention time values (Rt) for
the major product were 23.5 min using Method 1 and Rt 3.11 min
using Method 2, while Rt 12.1 min and Rt 8.2 min (Method 2) were
obtained for ME and ergosterol, respectively.
UV-spectrum of thepurifiedmajorproduct was characterizedby
the triple peak with absorbance maximums at 293, 282 and 271 nm
(Fig. 2) indicating the presence “ergosterol-like” conjugated ꢀ5,7
double bond system. The spectra obtained for ergosterol and ME
were mostly identical and good correlated with known UV-spectra
of ergosterol in ethanol [20].
Fig. 2. UV-spectrum of MA in acetonitrile in comparison with the spectra of ME and
AD.
Two big ions m/z 330 and m/z 268 presented in the mass-
spectrum of the major product. The former was attributed to a
3-methoxymetoxy-17-ketoandrostadiene molecular ion, the latter
was considered to be a fragment lacking 3-methoxymetoxy moiety.
Comparison of 1H NMR spectrum of the major product with
obtained spectra of standard compounds also evidenced presence
of intact ꢀ5,7 double bonds system in the ring B of the product as
the signals of 6-H and 7-H were observed in the same range of ı
at ı 5.18 ppm). Based on the results obtained the product with Rt
3.11 (Method 2) was identified as 3-methoxymethoxy-androsta-
5,7-diene-17-one (MA) (Fig. 3).
Along with MA, some minor products were accumulated dur-
ing Mycobacterium sp. VKM Ac-1815D incubation with ME. Based
on the data of MS and 1H NMR analyses, the product with
Rt 3.51 (Method 2) and m.w. 374 has been identified as 3-
methoxymethoxy-20-hydroxymethylpregna-5,7-diene (MHMPD)
(Fig. 3) and the product with Rt 4.74 (Method 1) and m.w. 456
was suggested as monohydroxy-ME with hydroxyl at one of the
primary carbons in the side-chain.
Bioconversion of ME by growing mycobacterial culture was
more efficient as compared with that by resting cells. Non-induced
growing culture converted 2.8 mM (1.22 g/l) of ME for 260 h to
0.74 mM MA and accumulated approximately equal overall con-
3
(∼5.3–5.6 ppm) and with the same SCC (dd, 1H, J6-H, 7-H = 5.6 Hz,
3
4J = 2.6 Hz and dt, 1H, J6-H, 7-H = 5.7 Hz, 4J = 2.6 Hz, 2.6 Hz) as for
ergosterol and ME. Protective group 3-OCH2OCH3 also remained
intact (identity of singlet signals of OCH3 at ı 3.37 ppm, 3H and
signals of AB-proton system O–CH2–O with center at ı 4.69 ppm
2
(2H, JAB = 6.8 Hz)). The absence of ergosterol side-chain was con-
firmed by the absence of signals of CHCH3 (d, 3H, at ı 1.02, 0.90,
0.82 and 0.8 ppm) and 22-H and 23-H signals of C22H = C23H (m, 2H
Table 1
Bioconversion of ergosterol 3-acetate by Mycobacterium sp. VKM Ac-1815D.
Bioconversion substrate (2 g/l)
Time (h)
Residual substrate
Molar yield (%)
AD
ADD
HMP
Other 3-keto-ꢀ4-products
Ergosterol
Other ꢀ5,7-products
Ergosterol 3-acetate
Ergosterol (control)
187
140
45
20
20
45
15.5
30
0.5
3
<1
<2
<0.5
–
Not detected
Not detected

656
D.V. Dovbnya et al. / Steroids 75 (2010) 653–658
Fig. 3. Products of ME bioconversion of by Mycobacterium sp. VKM Ac-1815D.
Fig. 5. Accumulation of MA at bioconversion of ME (2 g/l) by non-induced (empty
symbols) and ␤-sitosterol-induced (0.4 mmol/l, filled symbols) growing culture at
the use of MCD (circles) and at the use of MCD and non-autoclaved ultrasonically
micronized ME (diamonds). Triangles – control (glycerol-mineral medium auto-
claved with ME).
Fig. 4. Effect of serial induction with ␤-sitosterol on ME (5 g/l) bioconversion by the
growing culture. Arrows indicate additions of the inducer (0.04 mmol/l).
centration of other 5,7-diene products with partially degraded
side-chain; then bioconversion stopped. At the subsequent serial
induction by portions of ␤-sitosterol, the bioconversion activity
increased over initial level, and then stabilized for approximately
96 h (Fig. 4) while the inducer totally disappeared (converted to
AD) within 20–30 h.
optimal conditions found 10 g/l of ME was converted to 5 g/l of MA
during 120 h that is respectively equal to 60% molar yield. Approx-
imately 15% of ME remained non-converted.
The effect of different type inducers on ergosterol side-
chain cleavage was studied (Table 2). Full side-chain oxidizing
activity was induced by sterols, but not by ADD, AD or
DHEA, propionic or laurinic acids. The induction was accom-
panied by preferential accumulation of MA. Among sterols,
the inducing effect was decreased in the following order:
␤-sitosterol ≥ stigmasterol > cholesterol > ergosterol and preferred
that for cholestenone. Lithocholic acid was found to be a weak
inducer. Inducers were presented in the medium for at least 24 h
before their disappearance (full conversion).
4. Discussion
Ergosterol conversion by Mycobacterium spp. was described ear-
lier; the yield of AD from 0.3 g/l ergosterol did not exceed 30–35%,
with high portion of unconverted substrate [15]. As shown, the
ability of some actinobacteria for side-chain cleavage of different
sterols decreases in the following order: cholesterol, campes-
terol, ␤-sitosterol, stigmasterol, dehydrocholesterol and ergosterol
[21]. The presence of double bond at C-7 was suggested to slow
down side-chain oxidation of ergosta-7,22-diene-3␤-ol by growing
Nocardia erythropolis [14]. Besides, activity of cholesterol oxidase
from Rhodococcus equi, the enzyme which is responsible for oxi-
dation/isomerization of 3␤-ol-5-ene moiety to 3-keto-4-ene, was
almost twice less active towards ergosterol than towards choles-
terol [22].
As known isoflavones, as well as ions of some bivalent
metals can inhibit the activity of cholesterol oxidase and/or 3␤-
hydroxysteroid dehydrogenase/ꢀ5-ꢀ4-isomerase enzyme system
which is accounting for 4-en-3-ketone formation from steroid
5-en-3␤-alcohols [23,24]. In our experiments on the whole
mycobacterial cells, none of the potential inhibitors affected the
composition of ergosterol catabolites in the range of concentrations
used.
3.3. Optimization of ME bioconversion to MA
Conditions of ME bioconversion were optimized; the most crit-
ical steps of the optimization are presented in Fig. 5. Addition
of ␤-sitosterol fine powder as the inducer in the concentration
0.4 mM before sterilization of the bioconversion medium shortened
the induction period and provided the long-lasting activity (up to
160 h). ME exhibited strong agglomeration during steam steriliza-
tion resulting in its low (20–60%) accessibility to the culture; as
shown the problem was mostly solved by micronization of con-
centrated ME suspension on hot ultrasonic bath in the presence
of surfactant. Modification of the aqueous bioconversion medium
with MCD resulted in higher yield and faster conversion. In the

D.V. Dovbnya et al. / Steroids 75 (2010) 653–658
657
Table 2
The effect of induction of sterol side-chain degradation activity with different compounds on the bioconversion of ME to MA by Mycobacterium sp. VKM Ac-1815D.
Inducer
Max rate of MA formationa (␮mol/l h)
Bioconversion selectivityb
Degree of destruction (conversion) of inducer
%
Over a period (h)
W/o induction
␤-Sitosterol
Stigmasterol
Cholesterol
Ergosterol
Cholestenone
Lithocholic acid
ADD
AD
DHEA
Propionic acid
Laurinic acid
7.5 0.27
58.7 3.72
57.8 3.1
46.4 2.5
37.5 4.24
32.2 1.8
19.2 1.51
6.8 0.35
6.5 0.38
6.0 0.4
0.99
38.4
34.8
37.5
22.0
17.6
3.3
0.99
0.89
0.92
1.05
0.75
–
96
–
24
N.d.c
N.d.
90
N.d.
N.d.
96
N.d.
72
72
72
24
N.d.
N.d.
N.d.
56
12
10
98
6.1 0.32
4.7 0.27
N.d.
N.d.
a
The maximal bioconversion rates were calculated using linear fragments of the product accumulation curves (36–48 h) on the basis of three independent experiments.
ME concentration was 2 g/l.
b
Determined as final molar ratio of the MA to other products formed.
N.d. value was not determined.
c
The data demonstrated that 3-acetyl substitution of ergosterol
slowed down the substrate conversion by mycobacteria, but did
not influence the composition of the catabolites: the substrate
de-esterification probably preceded the side-chain oxidation. The
ability of actinobacteria to hydrolyze 3-esters of sterols (i.e. ␤-
sitosterol, cholesterol) is known [25].
As known, sterol side-chain oxidation in mycobacteria is a mul-
tienzyme process which includes at least 14 reactions catalyzing
by at least three groups of inducible enzymes. The enzymes can be
induced by either sterols or some oligoisopren derivatives [26]. It is
often mentioned that in bacteria oxidation of 3␤-hydroxy moiety
and ꢀ5-ꢀ4 isomerization are initial reactions of sterol catabolism.
From the other hand, chemical structure of sterol ring A does not
influence significantly the side-chain degradation that is in accor-
dance with the present results.
ME itself evidently did not induce side-chain oxidizing activ-
ity; the initial slow bioconversion observed (Fig. 4) was probably
induced by trace impurities of non-derivatized sterols in ME. The
possible products of ME conversion have ꢀ5,7-configuration and
can be easily recognized from the products originated from natural
sterols (e.i. 3-keto-4-ene-steroids). Therefore, application of ME as
a bioconversion substrate was used as a tool for study of induc-
tion of sterol side-chain oxidation in Mycobacterium sp. The results
(Table 2) demonstrated that simultaneous presence of 3-hydroxyl
(preferably 3␤-), double bond at C-5 along with side-chain at C-17
in inducer molecule can be of critical importance for appearance
of the inducing effect. Induction with ␤-sitosterol or stigmasterol
having the side-chains branched at C-24 was the most effective
and increased the rate of MA formation by 7.7–7.8 times. Interest-
ingly, the inducing effect of ergosterol was lowest among the tested
sterols that can be attributed to the presence of the conjugated 5,7
double bonds in its molecule. The weak inducing effect may play
a certain role in relatively low rates of ergosterol bioconversion by
mycobacterial strains that were observed earlier and in the present
work.
Further to above-mentioned we can suggest that known high
reactivity and low thermal stability of ergosterol and deriva-
tives can also contribute to low rate of their bioconversions
due to the substrates partial decomposition and agglomeration
at the routinely applied steam-sterilization procedures. To make
ME accessible for the transforming culture the mild steriliza-
tion conditions and micronization were necessary. Alteration of
physico-chemical properties of biocatalysts or their environment
by means of molecular capsulation or application of cloud-point
systems have been recently investigated and became a powerful
tool for improving of steroid bioconversions [18,27,28].
Androsta-5,7-diene-3␤,17␤-diol, the corresponding 9(10)-
secosteroids and derivatives were patented as immune system
stimulating compounds effective against various infections [29].
The known key intermediate for their obtaining is 3␤-hydroxy-
androsta-5,7-dien-17-one that is in turn can be produced from
dehydroepiandrosterone by multi-stage chemical synthesis [11].
The presently obtained MA is analog of the key intermediate with
protected 3␤-hydroxyl. Such a protection is usual in deltanoid syn-
theses so the product can be used directly in further modification
or the protection can be easily removed.
Known methods for convergent synthesis of deltanoids are
applicable for obtaining of wide spectra of novel structures that
is necessary for laboratory and clinical research. At the same time
they are of little use for production because of restricted stereose-
lectivity and yield. Total synthesis of 1R-hydroxyvitamin D5 from
commercially available vitamin D2, requires not less than 40 steps
and the overall yield barely exceeds 3% [8,10]. Efficiency of meth-
ods for chemical introduction of ꢀ7-moiety into different steroids
with already protected 3␤-hydroxyl usually is within the range
1.2–50% [7,8]. Taking into account the reported here yield for MA
and 70–80% steroid recovery from bioconversion broth by conven-
tional approaches we can suggest efficiency of proposed method as
quite competitive.
Therefore the applied methoxymethyl protection of 3␤-
hydroxyl of ergosterol was found to be stable with mycobacterial
culture, provided ability for the side-chain degradation but pre-
vented reduction and isomerization of ꢀ5,7-moiety. The present
data demonstrated principal possibility for obtaining of deltanoids
and their nearest precursor compounds from ergosterol in a single
stage bioprocess as an alternative or good supplementation to the
conventional chemical approaches.
Acknowledgements
The work was supported by grants from the Russian Basic
Research Foundation N 07-08-00783-a and ISTC P-3200. We thank
Dr. Sergey M. Khomutov for ME purification, Dr. Konstantin F.
Turchin for proton NMR spectroscopy and Mr. Boris P. Baskunov
for recording mass spectra.
References
[1] Norman AW. Pleiotropic actions of 1␣,25-dihydroxyvitamin D3: an overview.
J Nutr 1995;125:1687S–9S.
[2] Brown SE. Vitamin D and fracture reduction: an evaluation of the existing
research. Altern Med Rev 2008;13:21–33.
[3] Agoston ES. Anticancer Agents. Med Chem 2006;6:53–71.

658
D.V. Dovbnya et al. / Steroids 75 (2010) 653–658
[4] Gulliford T, English J, Colston KW, Menday P, Moller S, Coombes RC. A phase I
[16] Weber A, Kennekke M, Neef G. Process for the production of 20-methyl-
5,7-pregnadiene-3␤,21-diol derivatives using mycobacterium. US Patent
5,429,934; 1995.
study of the vitamin D analogue EB 1089 in patients with advanced breast and
colorectal cancer. Br J Cancer 1998;78:6–13.
[5] Gombart AF, Luong QT, Koeffler HP. Vitamin D compounds: activity against
microbes and cancer. Anticancer Res 2006;26:2531–42.
[6] Steinmeyer A, Kirsch G, Neef G, Schwarz K, Thieroff-Ekerdt R, Wiesinger H,
et al. Vitamin D derivatives with carbo- or heterocyclic substituents at C-25, a
process for their production, intermediate products and their use for producing
medicaments. US Patent 6,600,058; 2003.
[17] Donova MV, Dovbnya DV, Kalinichenko AN, Vagabova LM, Arinbasarova AY,
Morozova ZV, et al. Method for production of androst-4-ene-3,17-dione. RF
Patent 20,779,258; 1997.
[18] Donova MV, Nikolayeva VM, Dovbnya DV, Gulevskaya SA, Suzina NE. Methyl-
␤-cyclodextrin alters growth, activity and cell envelope features of sterol
transforming mycobacteria. Microbiology 2007;153:1981–92.
[19] Egorova OV, Gulevskaya SA, Puntus IF, Filonov AE, Donova MV. Production of
androstenedione using mutants of Mycobacterium sp. J Chem Technol Biotech-
nol 2002;77:141–7.
[7] Siddiqui AU, Wilson WK, Swaminathan S, Schroepfer Jr GJ. Efficient preparation
of steroidal 5,7-dienes of high purity. Chem Phys Lipids 1992;63:115–29.
[8] Moriarty RM, Albinescu D. Synthesis of 1␣-hydroxyvitamin D5 using
a
modified two wavelength photolysis for vitamin D formation. J Org Chem
2005;70:7624–8.
[20] Quail MA, Kelly SL. Isolation and investigation of sterols from yeast. In: Evans IH,
editor. Methods in molecular biology. Totowa: Humana Press; 1996. p. 123–31.
[21] Donova MV. Transformation of steroids by actinobacteria: a review. Appl
Biochem Microbiol 2007;43:1–14.
[22] Pollegioni L, Piubelli L, Molla G. Cholesterol oxidase: biotechnological applica-
tions. FEBS J 2009;276:6857–70.
[23] Keung WM. Dietary estrogenic isoflavones are potent inhibitors of ␤-
hydroxysteroid dehydrogenase of P. testosteronii. Biochem Biophys Res
Commun 1995;215:1137–44.
[9] Cornella I, Suárez RM, Mourin˜o A, Sestelo JP, Sarandeses LA. Stereoselective
convergent synthesis of 24-substituted metabolites and analogues of vitamin
D. J Steroid Biochem Mol Biol 2004;89-90:19–23.
[10] Groth U, Richter N, Kalogerakis A. Synthesis of (3S)-hydroxyandrosta-5,7-
diene-17-ones via intramolecular cobalt-mediated [2 + 2 + 2] cycloaddition.
Synlett 2006;6:905–8.
[11] Okabe M, Sun R-C, Scalone M, Jubilian CH, Hutchings SD. Synthesis of 25-
hydroxycholecalcifer-16-en-23-ynol:
Chem 1995;60:767–71.
[12] Zmijewski MA, Li W, Zjawiony JK, Sweatman TW, Chen J, Miller DD, et al. Syn-
thesis and photo-conversion of androsta- and pregna-5,7-dienes to vitamin
D3-like derivatives. Photochem Photobiol Sci 2008;7:1570–6.
[13] Lobastova TG, Khomutov SM, Vasiljeva LL, Lapitskaya MA, Pivnitsky
KK, Donova MV. Synthesis of 3␤-hydroxy-androsta-5,7-dien-17-one from
3␤-hydroxyandrost-5-en-17-one via microbial 7␣-hydroxylation. Steroids
2009;74:233–7.
[14] Klimashina MM, Viktorovskii IV, Mikhailova NP, V’iunov KA. Analy-
sis of the products of ergosta-7,22-dien-3␤-ol biotransformation by
a
potential antipsoriatic agent.
J
Org
[24] Elalami A, Kreit J, Filali-Malfout A, Boudrant J, Germain PRP. Characterization of
a secreted cholesterol oxidase from Rhodococcus sp. GKI (CIP 105335). World J
Microbiol Biotechnol 1999;15:579–85.
[25] Ivshina IB, GrishkoVV, NogovitsinaEM, Kukina TP, Tolstikov GA. Bioconversion
of ␤-sitosterol and its esters by actinobacteria of the genus Rhodococcus. Appl
Biochem Microbiol 2005;41:626–33.
[26] Szentirmai A. Microbial physiology of side-chain degradation of steroids. J Ind
Microbiol 1999;6:101–16.
[27] Dovbnya DV, Khomutov SM, Nikolaeva VM, Donova MV. CD-medium control
of microbial sterol sidechain cleavage. In: Labandeira JJT, Jato JLV, editors. Pro-
ceedings of 9th international symposium on cyclodextrins. New York: Kluwer
Academic Publishers; 1999. p. 395–8.
a
Nocardia erythropolis culture. Appl Biochem Microbiol 1988;24:647–
52.
[28] Wang Z, Zhao F, Chen D, Li D. Cloud point system as a tool to improve the
efficiency of biotransformation. Enzyme Microb Technol 2005;36:589–94.
[29] Gates S, Loria RM. Compositions for regulation of immune response. US Patent
5,776,921; 1998.
[15] Weber A, Kennecke M, Process for the production of 4-androstene-3,17-dione
and 1,4-androstadiene-3,17-dione from ergosterol with Mycobacterium. US
Patent 5,516,649; 1996.