Search and discovery of actinobacteria capable of transforming deoxycholic and cholic acids

Article abstract of DOI:10.1016/j.molcatb.2016.12.010

The capability of 54 selected actinobacteria strains of different phyla to convert deoxycholic (DCA) and cholic (CA) acids under aerobic conditions was studied. Except for the two species, the strains did not grow on DCA (1 g/l) as a sole carbon source, but some of them effectively converted DCA performing 7β- and 9α- hydroxylation, 3α- and 12α-dehydrogenation, partial cleavage of the isoprenoic side chain and Δ⁴-dehydrogenation. Ursocholic acid, 9α-hydroxy-3,12-dioxo-23,24-bisnorchol-4-ene-22-oic acid, 3-keto-DCA and other end metabolites had been firstly identified in the actinobacteria strains. The total yield of 12α-hydroxy-3-oxo-chol-4-ene-24-oic and 3,12-dioxochol-4-ene-24-oic acids from DCA with Rhodococcus erythropolis VKM Ac-1152 reached 95%. Almost 80% DCA were converted to 9α-hydroxy-3,12-dioxo-23,24-bisnorchol-4-en-22-oic acid by Rhodococcus sp. MTS-77. Unlike DCA, cholic acid (CA) was confirmed to be a growth substrate for majority of the examined strains, but only three Rhodococcus strains exhibited 7α- and/or 12α-HSDH activities thus forming 7-keto-DCA and 12-keto-chenodeoxycholic acid as major products from CA. Steroid metabolites were identified by TLC, GC, MS, ¹H- and ¹³C NMR analyses. The results may contribute to the knowledge of biocatalytic potential of diverse soil-dwelling actinobacteria towards bile acids, and could be applied at the development of novel bioprocesses for production of the valuable cholanic acids.

Full text of DOI:10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Research paper
Search and discovery of actinobacteria capable of transforming
deoxycholic and cholic acids
N.O. Deshcherevskayaa, T.G. Lobastovaa, V.V. Kollerova,∗, A.V. Kazantsevb, M.V. Donovaa
a G.K. Skryabin Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Prospekt Nauki, 5, 142290, Pushchino, Moscow
region, Russia
b Moscow State University, GSP-1, Leninskiye Gori, 1, Chemical Department, Moscow, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
The capability of 54 selected actinobacteria strains of different phyla to convert deoxycholic (DCA) and
cholic (CA) acids under aerobic conditions was studied. Except for the two species, the strains did not
grow on DCA (1 g/l) as a sole carbon source, but some of them effectively converted DCA perform-
ing 7␤- and 9␣- hydroxylation, 3␣- and 12␣-dehydrogenation, partial cleavage of the isoprenoic side
chain and ꢀ4-dehydrogenation. Ursocholic acid, 9␣-hydroxy-3,12-dioxo-23,24-bisnorchol-4-ene-22-oic
acid, 3-keto-DCA and other end metabolites had been firstly identified in the actinobacteria strains.
The total yield of 12␣-hydroxy-3-oxo-chol-4-ene-24-oic and 3,12-dioxochol-4-ene-24-oic acids from
DCA with Rhodococcus erythropolis VKM Ac-1152 reached 95%. Almost 80% DCA were converted to 9␣-
hydroxy-3,12-dioxo-23,24-bisnorchol-4-en-22-oic acid by Rhodococcus sp. MTS-77. Unlike DCA, cholic
acid (CA) was confirmed to be a growth substrate for majority of the examined strains, but only three
Rhodococcus strains exhibited 7␣- and/or 12␣-HSDH activities thus forming 7-keto-DCA and 12-keto-
chenodeoxycholic acid as major products from CA. Steroid metabolites were identified by TLC, GC, MS,
1H- and 13C NMR analyses.
Received 1 September 2016
Received in revised form
21 November 2016
Accepted 23 December 2016
Available online xxx
Keywords:
Deoxycholic acid
Cholic acid
Actinobacteria
Transformation
Ursocholic acid
The results may contribute to the knowledge of biocatalytic potential of diverse soil-dwelling acti-
nobacteria towards bile acids, and could be applied at the development of novel bioprocesses for
production of the valuable cholanic acids.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
steroid substrates: 3-oxo- and 3␤-hydroxy steroids of pregnane
and androstane series, as well as sterols (phytosterol, cholesterol,
Actinobacteria are known to possess diverse steroid-
transforming activities and are capable of performing different
modifications of steroid molecule, such as hydroxylation, intro-
duction of double bonds, oxidation of carbonyl groups, reduction
of steroid alcohols, de-esterification, partial, or full oxidation of
the aliphatic side chains at C-17, and others [1]. Actinobacteria
may transform broad spectrum of both natural and synthetic
ergosterol) and their derivatives [2–4].
Bile acids (BAs) differ from other steroids by saturated
cyclopenthanoperhydrophenantrene nucleus showing cis-A/B ring
juncture and C5-acyl side chain at C-17, being hydroxylated C24
3␣-hydroxy-5␤-H-cholanoic acids. Due to their amphipathic fea-
tures, BAs are the powerful surface-active compounds that are
essential in the digestion and re-sorption of fats, fatty acids and
lipid-soluble vitamins in the digestive tracts of vertebrates. BAs
also participate in cholesterol balancing, apoptosis, colonic salvage,
and serve other important physiological functions [5–7]. BAs may
enter the environment in big amounts by extraction, and accumu-
late as multi-tons wastes during cattle and swine production [8].
Industrial wastes of ursodeoxycholic acid (UDCA) production from
bile may also contain deoxycholic acid (DCA), which is a hazardous
environmental pollutant [9].
Abbreviations: BA, Bile acid; DCA, deoxycholic acid (3␣,12␣-dihydroxy-
5␤-cholan-24-oic acid); CA, cholic acid (3␣,7␣,12␣-trihydroxy-5␤-cholan-24-oic
acid); LCA, lithocholic acid (3␣-hydroxy-5␤-cholan-24-oic acid); UCA, ursocholic
acid (3␣,7␤,12␣-trihydroxy-5␤-cholan-24-oic acid); CDCA, chenodeoxycholic acid
(3␣,7␣-dihydroxy-5␤-cholan-24-oic acid); 3-keto-DCA, (12␣-hydroxy-3-oxo-5␤-
cholan-24-oic acid); 3␣-HSDH, 3␣-hydroxysteroid dehydrogenase; 3␤-HSDH,
3␤-hydroxysteroid dehydrogenase; 12␣-HSDH, 12␣-hydroxysteroid dehydroge-
nase.
The reports on microbial transformation of BAs mainly relate
to commensal and pathogenic microorganisms in the human
gastrointestinal tract [10]. Intestinal and soil anaerobic bacteria
∗
Corresponding author.
E-mail addresses: slava-kollerov@yandex.ru, svkollerov@rambler.ru
(V.V. Kollerov).
http://dx.doi.org/10.1016/j.molcatb.2016.12.010
1381-1177/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
2
are capable of deconjugation, desulfation, oxidation and epimer-
ization of hydroxyl groups at C-3, C-7 and C-12, as well as
7-dehydroxylation, and esterification of BAs [11,12]. In spite of very
few bacteria can synthesize BAs [13], many aerobic bacterial species
are able to grow on BAs as sole carbon and energy sources.
biocatalysts that can be exploited further for the production of
potentially physiologically active cholanic acids.
2. Experimental
Cholate degradation has been best studied in Pseudomonas
sp. strain Chol1 [14] and Rhodococcus jostii RH11 [15]. The well-
known catabolic route includes steroid core degradation via the
9,10-secosteroid pathway releasing intermediates with a 3-keto-
2.1. Materials
Deoxycholic
acid
(3␣,12␣-dihydroxy-5␤-cholanic
acid,
DCA), lithocholic acid (3␣-hydroxy-5␤-cholanic acid, LCA)
and ursodeoxycholic (3␣,7␤-dihydroxy-5␤-cholanic acid, UDCA)
acids were obtained from ACROS Organics (USA). Cholic acid
(3␣,7␣,12␣-trihydroxy-5␤-cholan-24-oic acid, CA); ursocholic
ꢀ
1,4-diene structure of the steroid skeleton [16]. As shown by
genomic analysis, the 9,10-secosteroid pathway for cholate degra-
dation was conserved in members of Rhodococcus genus, while
patchy distributed among Proteobacteria [17].
acid
(3␣,7␤,12␣-trihydroxy-5␤-cholan-24-oic
acid,
UCA);
The existence of distinct cholate degradation pathway has been
demonstrated for actinobacteria of Dietzia sp. strain Chol2 which
degraded cholate via the intermediates with a 3-keto-ꢀ4,6-diene-
7-deoxy structure of the steroid skeleton such as 3,12-dioxo-4,6-
choldienoic acid [18].
chenodeoxycholic
acid (3␣,7␣-dihydroxy-5␤-cholan-24-oic
acid, CDCA); 3-keto-deoxycholic acid (12␣-hydroxy-3-oxo-
5␤-cholan-24-oic acid, 3-keto-DCA), 7-keto-deoxycholic acid
(12␣-hydroxy-7-oxo-5␤-cholan-24-oic acid, 7-keto-DCA) and
12-keto-chenodeoxycholic acid (3␣,7␣-dihydroxy-12-oxo-5␤-
cholan-24-oic acid, 12-keto-CDCA) were kind gifts from Prodotti
Chimici e Alimentari S.p.A. (Basaluzzo, Italy). Randomly methy-
lated ␤-cyclodextrin (RAMEB) was purchased from Wacker Chemie
(Germany); yeast extract - from Difco (USA), cornsteep solids -
from Sigma-Aldrich (USA). Methoxyamine hydrochloride, N,O-
bis(trimethylsilyl)trifluoroacetamide, N-trimethylsilylimidazole,
trimethylchlorosilane and pyridine were purchased from SERVA
(Germany). All other reagents were of the best purity grade from
commercial suppliers (Russia).
Biotransformation of cholic acid (CA) and deoxycholic acid
(DCA) may be a method of choice for production of valuable phar-
maceuticals and their precursors [19], as well as for the obtaining of
the compounds which are essential for cosmetics, material science
and environmental cleansing [20]. The products of dehydroxyla-
tion, oxidation of hydroxyl groups, side chain degradation, as well
as regio- and stereospecific hydroxylation are of potential practical
importance. For instance, microbial 7␤-hydroxylation of lithocholic
acid (LCA) results in ursodeoxycholic acid (UDCA) which is widely
used in medicine for treatment of cholestasis, sclerosing cholan-
gitis, hepatitis and cirrhosis [21–23]. Recently, we have reported
effective method for UDCA production that provides the yield of
almost 90% [24]. Chemoenzymatic routes from CA to UDCA were
recently reviewed [25].
Some proteobacteria were shown to convert CA. For example,
the strains of Xanthomonas, Pseudomonas and Acinetobacter genera
transformed CA to a mixture of 3␣,12␣-dihydroxy-7-oxo-cholanic,
3␣-hydroxy-7,12-dioxo-cholanic and 3,7,12-triketo-cholanic acids
[26]. Under anoxic conditions Pseudomonas sp. NCIB 10590 mod-
ified CA mainly to 12␤-hydroxy-androsta-4,6-diene-3,17-dione
while 12␣-hydroxy-3-oxochola-4,6-dien-24-oic acid was detected
in minor amounts [27].
2.2. Microorganisms
The strains were obtained from All-Russian Collection of
Microorganisms (VKM IBPM RAS) and working collection of Lab-
oratory of Microbiological Transformation of Organic Compounds
(MTOC) at G.K. Skryabin Institute of Biochemistry and Physiology
of Microorganisms, Russian Academy of Sciences (IBPM RAS).
2.3. Growth of strains on DCA, or CA as sole carbon sources
1 mL of cell suspension containing 1 × 106 CFU were sprayed
onto agar plates with mineral medium composed of (g/L): KH2PO4 –
1.0; (NH4)2SO4 −3.0; MgSO4 − 0.2; FeSO4 − 0.01; ZnSO4 – 0.002; pH
7.0 supplemented with either DCA (1 g/L), or CA (1 g/L). The plates
were incubated for 5–7 days at 30 ◦C. For each strain, the controls
without BAs were used to quantify any background growth.
Certain
proteobacteria,
mainly,
pseudomonads,
and
anaerobic bacteria of Clostridium, Eubacterium and Bifi-
dobacterium genera were reported to oxidize hydroxyl
group at C-3 [16]. Few Pseudomonas sp. mutant strains
were selected capable of accumulating 12␣- and 12␤-
hydroxyandrosta-1,4-diene-3,17-dione
stereoisomers,
and
2.4. Bioconversion
3␣,12␤-dihydroxy-9(10)-secoandrosta-1,3,5(10)-triene-9,17-
dione as major products from DCA [28]. Rhodococcus ruber
transformed CA and DCA to 3␣,7␣,12␣-trihydroxy-9-oxo-
The strains were grown in a nutrient medium (50 mL) contain-
ing (g/L): glucose − 7, yeast extract − 4.5, corn steep solids − 5,
CaCO3 – 0.05, dissolved in distilled water (рН 7.0). For DCA, or
CA bioconversion, 5% (v/v) of the seed culture was inoculated in
the transformation medium (medium A, or B) to a final volume of
100 mL. Medium A was composed of (g/L): glycerol − 5; KН2РO4 – 1,
K2НРO4 − 4, (NH4)2SO4 – 3, carbamide − 0.25, MgSO4 – 0.2, FeSO4
– 0.01, ZnSO4 – 0.002; рН 7.0. Medium B of the same composition
as medium A additionally contained RAMEB (6.7 g/L).
DCA, or CA (100 mg) was added as a hot methanol solution to
a final concentration of 1 g/L. Final solvent concentration did not
exceed 2% (v/v). Bioconversions were carried out aerobically in
Erlenmeyer flasks (750 mL) on a rotary shaker (200 rpm) at 29 ◦C
for 48–120 h.
9,10-seco-23,24-bisnorchola-1,3,5(10)-trien-22-oic
acid
and
3,12␣-dihydroxy-9-oxo-9,10-seco-23,24-bisnorchola-1,3,5(10)-
trien-22-oic acid, correspondingly [29].
In spite of the progress in the field of steroid microbial con-
version which was observed at the past decade, the discovery of
the suitable strains capable of performing desired reactions is still
a challenge, and along with bioinformatics approach, traditional
experimental screenings are necessary.
In our previous work, the strains of actinobacteria capable of
regio- and stereospecific hydroxylation of LCA at positions 7␣, 7␤
and 12␣ and those with 3␣-hydroxysteroid dehydrogenase (3-
HSDH) activity were revealed on the base of a broad screening
[24].
In this study, we investigated the ability of selected actinobac-
teria of different taxonomy belongings to transform DCA and CA,
identified key metabolites formed and revealed the most active
2.5. Isolation of metabolites
Steroids were isolated from the culture medium by extrac-
tion with ethyl acetate (EtOAc). After the rotary evaporation
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
3
1H- and 13C NMR spectra were recorded on BrukerAvance 400
with working frequency 400 and 100.6 MHz in accordance with
producer protocols.
of the solvent, crude residues (25–30 mg) were applied to a
preparative chromatography glass plates with 2.0 mm silica gel
with fluorescent indicator UV254 (Pre-coated TLC-plates SIL G-200
UV254, Germany). Individual compounds were eluted with EtOAc
and evaporated to dryness. Chromatographic purity of the com-
pounds was controlled by TLC using Systems A, or B (see below).
Further identification of BAs derivatives was performed by GC
analysis using the authentic standards. The individual 3-keto-4-
ene-metabolites were identified by mass spectrometry (MS) and
nuclear magnetic resonance (NMR) techniques.
DCA (standard sample):1H NMR (DMSO-d6) ␦: 4.44 (br.s., 1H,
COOH), 4.17 (br.s., 1H, 12-OH), 3.77 (br.s., 1Н, 12␤-H), 2.24-0.95
(m, 28H), 0.90 (d, J = 6.5 Hz, 3H, 21-CH3), 0.83 (s, 3H, 19-CH3), 0.58
(s, 3H, 18-CH3).13C-NMR (DMSO-d6) ␦: 174.9 (24-C), 71.0 (C-12),
70.0 (C-3), 47.4, 46.2, 46.0, 41.6, 36.3, 35.7, 35.2, 35.0, 33.8, 32.9,
30.8, 30.7, 30.2, 28.6, 27.2, 27.0, 26.1, 23.5, 23.1 (C-19), 16.9 (C-21),
12.4 (C-18).
3. Results and discussion
2.6. Analyses
3.1. DCA conversion
2.6.1. Thin layer chromatography (TLC)
Samples of cultivation broth (1 mL) were taken every 24 h and
extracted with 2–5 mL of EtOAc. The extracts were applied to TLC
plates (ALUGRAM SIL G/UV254, Germany). The TLC plates were
developed in a mixture of chloroform − acetone − glacial acetic
acid 50:50:0.5 (v/v) (System A), or benzene-acetone 30:10 (v/v)
(System B). Staining of TLC plates was carried out using MnCl2 [30],
or the Komarowski reagent [31] and heating to 105 ◦C. The steroids
with 3-oxo-4-ene moiety were visualized under UV light (254 nm)
using hemiscope Desaga HP-UVIS (France). The authentic standards
of DCA, CA, ursocholic acid (UCA), 3-keto-DCA, 7-keto-DCA and
12-keto-chenodeoxycholic acid (12-keto-CDCA) have been used
during TLC analysis.
Actinobacteria of different phyla (totally, 5 strains) which exhib-
ited steroid-transforming activity during our previous works [1,24]
were screened for the ability to transform DCA. All the strains were
cultivated both in medium A and B during the screening. In some
cases the accumulation of metabolites was observed only when
the strains cultured in the medium B. Probably, ␤-cyclodextrin
derivative (RAMEB) enhanced bioconversion of DCA and CA whose
aqueous solubility is rather low (0.24 and 0.28 g/L, correspond-
ingly).
Thirty one strains of 12 different genera belonging mainly
to Corynebacterineae, as well as to Propionibacterineae, Micro-
coccineae, Pseudonocardineae and Streptosporangineae suborders
expressed transforming activity toward DCA with maximum rate
for the representatives of Nocardiaceae (Table 1). Other 23 strains
of Agromyces, Catellatospora, Clavibacter, Dactylosporangium, Gly-
comyces and some other genera did not transform DCA, − the
substrate remained unconverted at least for 5 days of incuba-
tion. Complete removal of DCA without any detectable steroid
metabolites accumulation was observed exclusively for rhodococci
strains: R. corynebacteroides VKM Ac-870, R. rhodochrous VKM Ac-
1227, Rhodococcus sp. S-67 and Rhodococcus sp. X-5 (Table 1).
Fourteen strains of Amycolatopsis, Curtobacterium, Dactylospo-
rangium, Mycobacterium, Lentzea, Nocardia, Nonomuraea, Nocar-
diopsis, Rhodococcus, Saccharopolyspora and Saccharothrix genera
showed low DCA-transforming activity, − the metabolites were
observed in small amounts (the yield under conditions used did
not exceed 3–5%), and their structures were not defined. These
metabolites were indicated in Table 1 as “other products” with
non-determined structure (“n.d.”).
2.6.2. Gas chromatography (GC)
Samples (1 mL) of the cultivation broths were extracted trice
with EtOAc (5–10 mL), the solvent was evaporated under vacuum
and the residue was redissolved in methanol (1–1.5 mL).
BAs were analyzed in accordance with [24] with some mod-
ifications. Per-trimethylsilyl derivatives of BAs were obtained as
follows: an aliquot 50 ␮L of methoxyamine hydrochloride solution
in dry distilled pyridine (12 ␮g/␮L) was added to the dry residue
which was obtained after the evaporation of methanol solution;
the mixture was heated for 30 min at 70 ◦C and evaporated to
dryness under vacuum. Silanization was carried out by treatment
with 100 ␮L of N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA),
N-trimethylsilylimidazole and trimethylchlorosilane (100:30:2,
v/v/v) for 1 h at 90 ◦C.
Analyses were performed on HP 5890 chromatograph (“Hewlett
Packard”, USA) equipped with fused silica capillary column
(20 m × 0.25 mm × 0.25 ␮m) with OV-1 immobilized stationary
phase; carrier gas (helium) flow was equal to 1.4 mL/min, flame
ionization detector and HP 3396A integrator were used. Usual
split injection method (1:20) was applied, injector and detector
temperatures were equal to 290 and 325 ◦C, respectively. Col-
umn temperature was programmed from 250 ◦C up to 315 ◦C
(100 ◦C/min) and then held at this temperature for 3 min.
Calibration curves were prepared using standard solutions
of DCA, CA, UCA, 3-keto-DCA, 7-keto-DCA and 12-keto-CDCA.
Retention times: DCA, 5.98 min; CA, 6.16 min; UCA, 6.29 min;
3-keto-DCA, 6.25 min; 7-keto-DCA, two major peaks: 6.31 and
6.93 min; 12-keto-CDCA, two major peaks: 6.04 and 6.96 min.
Noteworthy, several strains which had been earlier shown to
transform LCA (e.g. Nocardia nova VKM Ac-1971, Pseudonocardia
autotrophica VKM Ac-1067, Saccharopolyspora hirsuta subsp. hirsuta
VKM Ac-666, Saccharothrix longispora VKM Ac-1265 and Strepto-
myces hygroscopicus subsp. hygroscopicus VKM Ac-831) [24], did
not exhibit any activity, or showed very low activity towards DCA
(e.g. Saccharopolyspora rectivirgula VKM Ac-810 and Nonomuraea
carminata VKM Ac-1780).
3.1.1. 3˛-Hydroxysteroid dehydrogenase activity
Nine strains belonging to Dietziaceae, Mycobacteriaceae, Nocar-
diaceae, Nocardioidaceae, Pseudonocardiaceae and Streptosporangi-
naceae families converted DCA to 12␣-hydroxy-3-oxo-5␤-cholan-
24-oic acid (3-keto-DCA) with maximum rate for the strains of
Mycobacterium, Nocardiodes and Rhodococcus genera (Table 1). The
structure of 3-keto-DCA was confirmed by TLC and GC analyses by
comparison with an authentic standard. The formation of 3-keto-
DCA indicated the presence of 3␣-hydroxysteroid dehydrogenase
(3␣-HSDH) activity in the selected actinobacteria.
2.6.3. Mass-spectrometry (MS), 1H- and 13C NMR spectroscopy
MS spectra were recorded on a Bruker Maxis Impact spec-
trometer. MS data for 3-keto-4-ene-metabolites of DCA were
as follows: 12␣-hydroxy-3-oxochol-4-ene-24-oic acid: m/z calcd
for [C24H36O4(-1)-]:387.2616; found: 387.2625; 3,12-dioxochol-4-
ene-24-oic acid: m/z calcd for [C24H34O4(-1)-]: 385.2616; found:
385.2441; 9␣-hydroxy-3,12-dioxo-23,24-bisnorchol-4-en-22-oic
acid: m/z calcd for [C22H30O5(-1)-]:373.2390; found: 373.2104.
Oxidation of the hydroxyl function at C-3 is generally con-
sidered as a starting reaction of steroid core degradation by
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

Table 1
DCA and CA bioconversions by actinobacteria (Actinomycetales order)a.
Suborder
Family
Genus
Strain
Medium
Transformation products of
Growth on DCA/CA as a
sole organic substrate
g
b
DCA conversion
CA conversion
DCA
CA
3␣-hydroxy-5␤-H
C
-steroids
Others
3-keto-4-en-steroids
24
UCA
Dietziaceae
Dietzia
maris VKM Ac-593
neoaurum VKM
B
B
–
–
+3-keto- DCA
+n.d.
+n.d
+ n.d.
–
–
–
–
–
–
Corynebacterineae
c
Mycobacteriaceae
Mycobacterium
Ac-1815D
sp. VKM Ac-1817D
sp. VKM Ac-1060
coprophilus VKM Ac-571
A
B
A
–
–
–
++3-keto-DCA
+n.d.
+n.d.
–
–
–
–
–
–
–
–
+
–
+ n.d.
–
Nocardiaceae
Rhodococcus
d
corynebacteroides VKM
Ac-870
erythropolis VKM
Ac-1150
erythropolis VKM
Ac-1152
rhodochrous VKM
Ac-1227
B
B
B
B
B
–
–
–
–
–
−R
–
+ n.d.
–
–
–
–
–
+
+
–
+
+
–
+ n.d.
+ n.d.
e
–
+++
–
–
–
−R
–
rhodochrous VKM
Ac-1282
++3-keto- DCA
+ n.d.
ruber VKM Ac −1021
A
B
–
–
–
+n.d.
–
–
+ 7-keto- DCA
–
–
–
+
+
ruber VKM Ac-1167
f
sp. MTS-77
B
B
B
B
A
B
–
–
–
–
–
–
–
–
+n.d.
–
+++
–
–
–
–
–
–
–
+
–
+
+
+
+
+
+
sp. VKM Ac-857
sp. VKM Ac-1148
sp. VKM Ac-1153
sp. X-5
–
–
–
–
–
+ 7- keto-DCA
–
+ 7- keto-DCA
−R
sp. E-14C
+n.d.
+ 12- keto-CDCA
sp. S-26
sp. S-67
sp. SIMT-105
A
A
B
–
–
–
+3-keto-DCA
−R
–
–
–
–
–
–
–
–
–
+
+
–
++3-keto-DCA
Micrococcineae
Microbacteriaceae
Nocardioidaceae
Nocardiopsaceae
Curtobacterium
Nocardioides
Nocardiopsis
sp. D-7
B
A
A
–
–
+
+n.d.
++3-keto-DCA
+n.d.
–
–
–
–
–
–
–
–
–
–
+
–
simplex VKM Ac-2033D
dassonvillei subsp.
dassonvillei VKM Ac-836
sp. TB-1353
Propionibacterineae
B
A
–
+
+n.d.
–
–
–
–
–
–
+
+
coloradensis VKM
Ac-1732
+3-keto-DCA
Pseudonocardineae
Pseudonocardiaceae
Amycolatopsis
orientalis VKM Ac-866
alni VKM Ac-901
rectivirgula VKM Ac-810
waywayandensis VKM
Ac-1970
A
B
A
A
+
+
+
+
+n.d.
+3-keto-DCA
+n.d.
–
–
–
–
–
–
–
–
+
–
–
–
+
+
+
+
Pseudonocardia
Saccharopolyspora
Lentzea
+n.d.
Actinosynnematoideae
Streptosporanginaceae
Saccharothrix
Nonomuraea
longispora VKM Ac-907
A
A
+
+
+n.d.
+n.d.
–
–
–
–
–
–
+
–
roseoviolacea subsp.
carminata VKM Ac-1780
sp. MTOC-129
Streptosporangineae
A
+
+3-keto-DCA
–
–
–
–
a
No conversion of DCA and CA was observed with the following strains: Mycobacterium lacticum VKM Ac-1145, Nocardia nova VKM Ac-1971, Nocardia sp. VKM Ac-1059, Glycomyces rutgersensis VKM Ac-1248, Agromyces fucosus
VKM Ac-1346, Clavibacter michiganensis subsp. insidiosus VKM Ac-1402, Catellatospora sp. 122v, Dactylosporangiu matsuzakiense VKM Ac-1321, Nocardiopsis dassonvillei subsp. albirubida VKM Ac-1882, N. lucentensis VKM Ac-1962,
Pseudonocardia alni VKM Ac-914, P. autotrophica VKM Ac-1067, P. hydrocarbonoxydans VKM Ac-799, P. petroleophila VKM Ac-865, Saccharopolyspora erythraea VKM Ac-1189, Saccharopolyspora hirsuta subsp. hirsuta VKM Ac-666,
Saccharothrix coeruleofusca VKM Ac-855, S. espanaensis VKM Ac-1969, S. longispora VKM Ac-1265, S. clavuligerus 8v, Saccharothrix sp. VKM Ac-1757, Streptomyces hygroscopicus subsp. hygroscopicus VKM Ac-831.
b
Conversion level: «-» − no product detected; “+” − less than 15%; “++” − over 20%, “+++” − over 50%.
c
“n.d.”: the structure of products was not determined.
d
“R”: complete removal of steroid with no detectable metabolite accumulation.
The mixture of 12␣-Hydroxy-3-oxo-chol-4-ene-24-oic and 3,12-dioxo-chol-4-ene-24-oic acids.
9␣-hydroxy-3,12-dioxo-23,24-bisnor-chol-4-en-22-oic acid.
Growth on DCA/CA as sole organic substrate: “-” − no; “+” − yes.
e
f
g

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
5
bacteria. Cholesterol oxidase and 3␤-hydroxysteroid dehydroge-
nase (3␤-HSDH) which are responsible for the reaction towards
3␤-OH-5-ene-steroids may also carry out dehydrogenation of 3␣-
hydroxyl function of BAs [32]. This activity towards LCA had
been mentioned for cholesterol oxidase ChoD from Mycobacterium
neoaurum VKM Ac-1815D (re-classified from Mycobacterium sp.
VKM Ac-1815D) [33,34]. The presence of cholesterol oxidase and
3␤-HSDH activities in Rhodococcus strains was reported repeat-
edly, and in some cases the activity towards 3␣-hydroxy steroids
had been also shown [32,35]. In addition, the 3␣-HSDH activity
towards BAs was also previously reported for anaerobic bacteria
of Clostridium genus, such as C. perfringens [36], C. scindens [37]
and C. hiranonis [38], as well as for Peptostreptococcus productus
[39], Eggerthellalenta (Eubacterium lentum) [40] and Pseudomonas
testosteroni [41].
However, the accumulation of 3-keto-DCA from DCA was
hitherto unreported for the representatives Nocardioides, Mycobac-
terium, Nonomuraea, Pseudonocardia, Rhodococcus, and Amycolatop-
sis. None of the strains which accumulated 3-keto-DCA from DCA
could grow on DCA as a sole carbon source (Table 1).
3.1.2. 7ˇ-Hydroxylase activity
The strains of Amycolatopsis, Lentzea, Nocardiopsis, Nonomu-
raea, Pseudonocardia, Saccharothrix and Saccharopolyspora genera
generated UCA from DCA thus evidencing the presence of 7␤-
hydroxylase activity (Table 1). Occurrence of 7␤-hydroxylase
activity toward DCA for all 9 selected strains has been shown here
for the first time. As reported in our previous work [24], the strains
of Amycolatopsis, Lentzea, Pseudonocardia and Saccharopolyspora
genera possessed 7␤-hydroxylase activity towards LCA thus form-
ing UDCA as a major product.
Fig. 1. TLC chromatogram of crude extract metabolites formed from DCA by selected
actinobacteria strains (system A, visualization by UV light, 254 nm): 1, Rhodococcus
erythropolis VKM Ac-1152 (20 ␮g in spot); 2, Rhodococcus sp. MTS-77 (10 ␮g in spot);
a, 3,12-dioxochol-4-ene-24-oic acid; b, 12␣-hydroxy-3-oxochol-4-ene-24-oic acid;
c, 9␣-hydroxy-3,12-dioxo-23,24-bisnorchol-4-en-22-oic acid.
In this study, the strains of Amycolatopsis coloradensis VKM Ac-
1732, Nonomuraea sp. MTOC-129 and Pseudonocardia alni VKM
Ac-901 along 7␤-hydroxylation of DCA carried out oxidation of
3␣-hydroxyl group leading to UCA and 3-keto-DCA, respectively
(Table 1).
Maximum activity was observed for Rhodococcus erythropolis VKM
Ac-1152 and Rhodococcus sp. VKM MTS-77 (Fig. 1).
When incubating with DCA, R. erythropolis VKM Ac-1152 formed
two major products: 12␣-hydroxy-3-oxochol-4-ene-24-oic acid
(M 388) (Fig. 1,b): 1H NMR (DMSO-d6) ␦: 5.61 (s, 1H, 4-H), 4.17
(br. s, 1H, 12-OH), 3.83 (br. s., 1Н, 12␤-H), 2.42-0.95 (m, 23H), 1.11
(s, 3H, 19-CH3), 0.91 (d,J = 6.5 Hz, 3H, 21-CH3), 0.65 (s, 3H, 18-CH3).
13C-NMR (DMSO-d6) ␦: 198.0 (C-3), 174.9 (C-24), 171.2 (C-5), 123.0
(C-4), 70.7 (C-12) 46.7, 46.6, 46.0, 45.8, 37.7, 35.1, 35.0, 34.8, 33.5,
32.1, 31.7, 30.8, 30.7, 28.5, 27.0, 23.3, 16.9 (C-21), 16.7 (C-19), 12.3
It should be noted that none of 54 tested actinobacteria strains
was able to hydroxylate DCA at 7␣, whereas some of these
strains, namely, representatives of Catellatospora, Lentzea, Nocar-
dia, Nocardiopsis, Pseudonocardia, Saccharopolyspora, Saccharothrix
and Streptomyces genera were previously shown to possess 7␣-
hydroxylase activity toward LCA [24]. Probably the presence of
12␣-OH function in DCA molecule interferes hydroxyl group inser-
tion into 7␣-position, while does not influence stereoisomeric
7␤-hydroxylation. It is well known that position of micro-
bial hydroxylation depends on the structure of steroid. For
instance, during incubation of Gongronella butleri mycelium with
androst-4-ene-3,17-dione (AD), reactions of 14␣-, 6␣-, 7␣- and
6␤-hydroxylation were observed with accumulation of the corre-
sponding derivatives, while neither 6␣-, nor 7␣-hydroxylation was
evidenced with 9␣-OH-AD as a substrate [42].
(C-18) and 3,12-dioxochol-4-ene-24-oic acid (М 386) (Fig. 1,a): 1
H
NMR (DMSO-d6) ␦: 5.65 (s, 1H, 4-H), 2.71-0.90 (m, 23H), 1.22 (s, 3H,
19-CH3), 1.03(s, 3H, 18-CH3), 0.76 (d,J = 6.2 Hz, 3H, 21-CH3). 13C-
NMR (DMSO-d6) ␦: 212.8 (C-12), 197.4 (C-3), 174.7 (C-24), 169.3
(C-5), 123.7 (C-4), 56.7, 56.3, 55.5, 46.1, 38.6, 37.6, 35.0, 34.7, 34.4,
33.4, 31.7, 31.0, 30.5, 30.3, 26.9, 23.8, 18.6 (C-21), 16.3 (C-19), 11.2
(18-CH3).
At the initial step, the formation of 12␣-hydroxy-3-oxochol-
4-ene-24-oic acid (Rt 6.57 min) was observed thus evidencing
oxidation of 3␣-hydroxyl group and introduction of ꢀ4- double
bond into the A-ring of DCA. This compound was further oxidized
to 3,12-dioxochol-4-ene-24-oic acid (two major peaks: Rt 6.86 and
7.21 min) yielding over 75% from DCA. Totally, R. erythropolis VKM
Ac-1152 converted over 95% of DCA to the two 3-keto-4-ene deriva-
tives as confirmed by GC analyses (Table 2). Interestingly, the strain
did not grow on DCA as a sole carbon source (Table 1).
Oxidation of the 12␣-hydroxyl function is considered as an
important reaction in the chemoenzymatic route to a high-value
UDCA [21]. The presence of 12␣-HSDH activity towards DCA had
been earlier reported for proteobacteria of Pseudomonas and Acine-
tobacter genera [26], as well as for Eubacterium lentum strain [44].
Interestingly, whole cells of E. lentum expressed 12␣-HSDH activity
towards DCA, but not towards CA [45], while Acinetobacter cal-
3.1.3. DCA conversion to 3-keto-4-ene-derivatives
In most cases, oxidation of BAs hydroxyl group at C3 is followed
by desaturation of the A-ring leading to 3-keto-1,4-diene structures
being a step of a whole degradation pathway [16]. The presence
in the genome of the corresponding genes encoding dehydroge-
nases that are responsible for insertion of ꢀ4- and/or ꢀ1-double
bonds in the A-ring was repeatedly shown for mycolic acid rich
actinobacteria like rhodococci and mycobacteria [34,43].
In this study, the capability to generate the metabolites with
3-keto-4-ene structure from DCA was revealed only for seven rep-
resentatives Corynebacterineae suborder. Along with the strains
of Mycobacterium and Rhodococcus genera, the representatives of
Dietzia and Nocardia formed 3-keto-4-ene derivatives (Table 1).
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
6
Table 2
Identification of 3-keto-4-ene metabolites formed from DCA by selected strains of Rhodococcus.
Strain
Metabolite
M.w.
Yield, %
(molar)
Putative structure
Rhodococcus erythropolis VKM Ac-1152
12␣-Hydroxy-3-oxo-chol-4-
ene-24-oic acid
388
386
20–23
75–80
3,12-dioxochol-4-ene-24-oic
acid
Rhodococcus sp. MTS-77
9␣-hydroxy-3,12-dioxo-23,24-
bisnorchol-4-en-22-oic
acid
374
73–80
coaceticus acidovorans B49, A. calcoaceticus lwoffii strains BS11 and
B37 oxidized both 7␣- and 12␣ hydroxyl groups of CA to form
3␣-hydroxy-7,12-dioxo-5␤-cholanic acid [26]. 12␣-HSDH activity
was also mentioned for the fungus Penicillium sp. TTUR 422, which
nevertheless catalyzed 12␣-hydroxysteroid dehydrogenation only
after hydroxyl group insertion to the 7␤-position of DCA thus form-
ing 12-keto-UCA as a final product [46]. In this study, 12␣-HSDH
activity was exhibited in R. erythropolis VKM Ac-1152 only after
dehydrogenation of 3␣-hydroxyl group and unsaturation of the
A-ring by introduction of the ꢀ4-double bond.
It is noteworthy that no 3-keto-1,4-diene derivatives has been
observed among the products from DCA with actinobacteria tested,
while the known BAs degradation pathway by bacteria hypoth-
esized both ꢀ4- and ꢀ1-dehydrogenations leading to ADD as a
central intermediate [16].
Another major metabolite (Rt 6.14 and 6.50 min) was generated
from DCA by a strain of Rhodococcus sp. MTS-77 (Table 2). 1H NMR
(DMSO-d6) ␦: 5.66 (s, 1H, 4-H), 4.58(s, 1H, 9-OH), 2.54-1.22 (m,
18H), 1.33 (s, 3H, 19-CH3), 1.03 (d,J = 6.6 Hz,3H, 21-CH3), 0.99 (s, 3H,
18-CH3). 13C-NMR (DMSO-d6) ␦: 210.9 (C-12), 197.7 (C-3), 177.4 (C-
22), 169.3 (C-5), 125.4 (C-4), 80.8 (C-9), 55.8,50.0, 44.5, 44.4, 43.4,
42.5, 36.7, 33.6, 31.0, 27.7, 26.4, 24.2, 23.4, 19.4 (C-19), 17.7 (C-21),
10.9 (18-CH3).
Fig. 3. The structures of CA and its 7- and 12-oxo derivatives formed by selected
actinobacteria strains: 1, Rhodococcus ruber VKM Ac −1021, Rhodococcus sp. VKM
Ac-1153; 2, Rhodococcus sp. E-14C.
3,12-dioxo-23,24-bisnorchol-4-en-22-oic acid seems to be a part of
cascade reactions involved in the degradation pathway.
Thestructuresof DCAmetabolitesformedby selectedactinobac-
teria are summarized in Fig. 2.
Making a parallel with previous findings which evidenced
potentially different pathways for cholate degradation in some acti-
nobacteria which can coexist in natural habitats and may interact
via interspecies cross-feeding [18], the end metabolites from DCA
revealed in this study which were accumulated by selected acti-
nobacteria strains can be considered as intermediate substrates for
further degradation catalyzed by other microbial strains in savage
nature.
The signals in 13C-spectrum of this compound showed that it
is a C-22 steroid thus indicating the removal of the two carbon
fragment from isoprenoic side chain of DCA. The presence of addi-
tional signal (80.8) in 13C-spectrum evidenced hydroxyl group at
C-9 thus allowing to identify the compound as 9␣-hydroxy-3,12-
dioxo-23,24-bisnorchol-4-en-22-oic acid. In accordance to GC data,
the yield of the product was over 75% (Table 2).
Therefore, along with 9␣-hydroxylation Rhodococcus sp. MTS-
77 provides partial oxidation of the DCA side chain.
The formation of 9␣-hydroxy-3,12-dioxo-23,24-bisnorchol-4-
en-22-oic acid in small amounts had been earlier reported only for
Mycobacterium mucosum 1210 [47], but no 1H- and 13C NMR data
were presented. DCA conversion to 3-keto-4-ene derivatives with
partial degraded isoprenoic side chain had been previously shown
mostly for Pseudomonas strains [28].
3.2. CA conversion by actinobacteria strains
Structurally, CA differs from DCA by the presence of 7␣-hydroxyl
function (Fig. 3), but unlike DCA this bile acid seems to be less
convertible by selected actinobacteria. Only five of 54 examined
strains showed CA-transforming activity, and all of them belong to
Rhodococcus genus (Table 1). Other strains tested did not express
any activity toward CA under conditions used, − no detectable
metabolite accumulation was observed. Among five rhodococci
strains selected, low CA-transforming activity was detected for R.
corynebacteroides VKM Ac-870 and R. erythropolis VKM Ac-1152,
whereas two other strains, − R. ruber VKM Ac-1021 and Rhodococ-
cus sp. VKM Ac-1153 exhibited 7␣-HSDH activity and generated
7-keto-DCA as a major product from CA (Table 1, Fig. 3). During CA
incubation with Rhodococcus sp. E-14C 12-keto-chenodeoxycholic
acid (12-keto-CDCA) was detected in culture broth by GC analysis,
As recently shown for Pseudomonas sp. strain Chol1, cholate
side chain degradation proceeds via the stepwise removal
of an acetyl and
a propionyl residues to form 7␣,12␣-
dihydroxy-3-oxopregna-1,4-diene-20-carboxylate as an interme-
diate and 7␣,12␤-dihydroxyandrosta-1,4-diene-3,17-dione (12␤-
DHADD) as an end product. 12␤-DHADD is then subjected
to 9␣-hydroxylation leading to 3,7,12-trihydroxy-9,10-seco-
1,3,5(10)-triene-9,17-dione, which is further degraded via the
known pathway [14].
In spite of the inability of Rhodococcus sp. MTS-77 to grow on
DCA as a sole carbon and energy source, generation of 9␣-hydroxy-
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
7
Fig. 2. The structures of DCA and its derivatives formed by selected actinobacteria strains: 1, Amycolatopsis longispora VKM Ac-907, Amycolatopsis orientalis VKM Ac-866,
Lentzea waywayandensis VKM Ac-1970, Nocardiopsis dassonvillei subsp. dassonvillei VKM Ac-836, Nonomuraea roseoviolacea subsp. carminata VKM Ac-1780, Pseudonocardia alni
VKM Ac-901, Saccharopolyspora rectivirgula VKM Ac-810, Saccharothrix coloradensis VKM Ac-1732; Nonomuraeae sp. MTOC-129; 2, Dietzia maris VKM Ac-593, Mycobacterium
sp. VKM Ac-1817D, Nocardiodes simplex VKM Ac-2033D, Pseudonocardia alni VKM Ac-901, Amycolatopsis coloradensis VKM Ac-1732, Rhodococcus rhodochrous VKM Ac-1282,
Rhodococcus sp. S-26, Rhodococcus sp. SIMT-105, Nonomurea sp. MTOC-129; 3,4, Rhodococcus erythropolis VKM Ac-1152; 5, Rhodococcus sp. MTS-77.
thus evidencing the presence of 12␣-HSDH activity (Table 1, Fig. 4).
12-keto-CDCA may be of interest as a precursor of pharmacologi-
cally important CDCA.
which were earlier shown to degrade steroids, or other BAs were
unable to degrade DCA. For example, Nocardioides simplex VKM
Ac-2033D was shown to grow on cholesterol, or LCA as an only
carbon and energy source, and possesses the sets of the genes
involved in steroid/cholate catabolism [51], but it is incapable of
DCA degrading (Table 1). As it was earlier suggested at the study
of microbial degradation of DCA by Pseudomonas sp. NCIB 10590,
the presence of 12␣-hydroxyl group in BA molecule may prevent
enzymatic cleavage to form C-17 and C-20 intermediates, − the hin-
drance was removed when the 12␣-hydroxyl group was oxidized,
or epimerized [52]. Probably, the DCA degradation by N. simplex
VKM Ac-2033D and some other strains may be restricted because
of the absence of enzyme systems accounting for the oxidation,
or epimerization of the hydroxyl group at C12. Inability of DCA
assimilation was previously mentioned for the strains of Gordo-
nia, Tsukamurella, and Rhodococcus which were however capable
to grow on cholesterol and other steroids including cholate, as sole
carbon source. It has been suggested that many steroid-degrading
bacteria possess a complex functional unit, integrated by several
specific pathways (peripheral routes) required for the conversion
of deoxycholate (or other bile salts) to common intermediates (AD,
ADD, and the hydroxylated derivatives). Thus, inability of some
actinobacteria to degrade DCA might be related to the lack of such
specific peripheral pathways [50].
It should be noted that none of the CA-transforming rhodococci
strains with except for R. corynebacteroides VKM Ac-870, was capa-
ble to transform DCA (Table 1). Obviously the presence of 7␣-OH
group in CA molecule could induce 7␣-HSDH activity in R. ruber
VKM Ac-1021 and Rhodococcus sp. VKM Ac-1153 as well as 12␣-
HSDH activity in Rhodococcus sp. E-14C strain. The capability to
transform CA to 7-oxo- and 12-oxo derivatives had been pre-
viously mentioned for selected proteobacteria of Xanthomonas,
Pseudomonas and Acinetobacter genera [26], as well as for the strains
of Arthrobacter sp. [48] and Bacteroides intestinalis AM-1 isolated
from human feces [49], and had not been reported for any repre-
sentatives of Rhodococcus genus, or relative actinobacteria so far.
3.3. Growth on DCA/CA as sole organic substrate
Only two strains, − Rhodococcus sp. X-5 and Amycolatopsis ori-
entalis VKM Ac-866 were able to grow on DCA as a sole organic
substrate under conditions used, whereas the most part of the acti-
nobacteria strains examined were able to grow on CA (Table 1).
Interestingly, complete removal of DCA with no detectable steroid
metabolites was observed under conditions used forRhodococcus
corynebacteroides VKM Ac-870, R. rhodochrous VKM Ac-1227 and
Rhodococcus sp. S-67 while the strains did not grow on DCA as a
sole organic substrate (Table 1). This phenomenon requires further
investigation.
Aerobic CA degradation by actinobacteria of Arthrobac-
ter, Corynebacterium, Mycobacterium, Nocardia and Streptomyces
genera had been previously reported [16], while growth on deoxy-
cholate was reported bacteria of Pseudomonas genus [50].
Hypothetical BA degradation pathway was proposed based on
the compounds, genes and proteins identified in different steroid-
degrading bacteria, and suggested oxidation of the hydroxyl group
at C3, followed by desaturation of the A-ring and side chain
cleavage yielding a 1,4-androstadiene-3,17-dione (ADD) deriva-
tives as central intermediates [16]. However, the actinobacteria
4. Conclusions
In this work, we studied the metabolic capacity of soil-dwelling
actinobacteria (totally, 54 strains) which had earlier demonstrated
various enzymatic activities towards different type steroids, to
transform DCA and CA. The selected actinobacteria exhibited broad
biocatalytic potency towards DCA, and in a less degree − towards
CA.
Actinobacteria of Corynebacterineae suborder such as Mycobac-
terium, Rhodococcus, Nocardia are characterized by high content
of mycolic acids in the cell envelope. In this study, the rep-
resentatives of such actinobacteria showed maximum level of
DCA-transforming activity, while lower level, or inability to trans-
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
8
form DCA was observed for the representatives of other taxa.
Probably, the features of the mycolic acid rich actinobacteria pro-
vide their better survival in the presence of DCA which is known to
be one of the most toxic BAs.
The toxicity of DCA may be a reason for the inability of the most
part of the examined strains to grow on DCA, but some of them
implemented different reactions of DCA transformation, such as
7␤-hydroxylation, dehydrogenations of 3␣- and/or 12␣-hydroxyl
groups, insertion of ꢀ4–double bond. These reactions may be con-
sidered as DCA detoxification similar to those described recently
for lower eukaryotes [53].
[15] W.W. Mohn, M.H. Wilbrink, I. Casabon, G.R. Stewart, J. Liu, R. van der Geize,
L.D. Eltis, Gene cluster encoding cholate catabolism in Rhodococcus spp, J.
Bacteriol. 194 (2012) 6712–6719.
[16] B. Philipp, Bacterial degradation of bile salts, Appl. Microbiol. Biotechnol. 89
(2011) 903–915.
[17] L.H. Bergstrand, E. Cardenas, J. Holert, J.D. Van Hamme, W.W. Mohn,
Delineation of steroid-degrading microorganisms through comparative
genomic analysis, mBio 7 (2016), http://dx.doi.org/10.1128/mBio.00166-16.
ˇ
[18] J. Holert, Y. Onur, V. Suvekbala, Z. Kulicı´, H. Möller, B. Philipp, Evidence of
distinct pathways for bacterial degradation of the steroid compound cholate
suggests the potential for metabolic interactions by interspecies
cross-feeding, Environ. Microbiol. 16 (2014) 1424–1440.
[19] O. Bortolini, G. Fantin, M. Fogagnolo, L. Mari, Control of the enantioselectivity
by keto bile acid derivatives in the epoxidation of alkenes with oxone,
Tetrahedron Asymmetry 15 (2004) 3831–3833.
Unlike DCA, the examined actinobacteria with few excep-
tions assimilated CA as a sole carbon source with restricted
capability to accumulate metabolites as end products. Effective
CA conversion was evidenced for three strains of rhodococci.
7␣/12␣-Hydroxysteroid dehydrogenase activities were shown to
be induced in R. ruber VKM Ac-1021, Rhodococcus sp. VKM Ac-1153
and Rhodococcus sp. E-14C strains by the presence of 7␣-OH group
when incubating with CA and afforded 7-keto-DCA and 12-keto-
CDCA as transformation products.
Novel data on DCA and CA bioconversions, specifically regard-
ing to the formation of UCA and 12-keto-CDCA, may have the
preparative-scale exploitation in the production of pharmacologi-
cally important UDCA and CDCA.
[20] E. Virtanen, E. Kolehmainen, Use of bile acids in pharmacological and
supramolecular applications, Eur. J. Org. Chem. 16 (2004) 3385–3399.
[21] O.L. Bortolini, A. Medici, S. Poli, Biotransformations on steroid nucleus of bile
acids, Steroids 62 (1997) 565–577.
[22] T. Ikegami, Y. Matsuzaki, Ursodeoxycholic acid: mechanism of action and
novel clinical applications, Hepatol. Res. 38 (2008) 123–131.
[23] C.I. Williams, E.A. Shaffer, Gallstone disease: current therapeutic practice,
Curr. Treat. Options Gastroenterol. 11 (2008) 71–77.
[24] V.V. Kollerov, D. Monti, N.O. Deshcherevskaya, T.G. Lobastova, E.E. Ferrandi, A.
Larovere, S.A. Gulevskaya, S. Riva, M.V. Donova, Hydroxylation of lithocholic
acid by selected actinobacteria and filamentous fungi, Steroids 78 (2013)
370–378.
[25] T. Eggert, D. Bakonyi, W. Hummel, Enzymatic routes for the synthesis of
ursodeoxycholic acid, J. Biotechnol. 10 (2014) 11–21.
[26] G. Fantin, S. Ferranini, A. Medici, P. Pedrini, S. Poli, Regioselective microbial
oxidation of bile acid, Tetrahedron 54 (1998) 1937–1942.
[27] R.W. Owen, R.F. Bilton, The degradation of cholic acid by Pseudomonas sp.
N.C.I.B. 10590 under anaerobic conditions, Biochem. J. 216 (1983) 641–654.
[28] R.A. Leppik, D.J. Sinden, Pseudomonas mutant strains that accumulate
androstane and seco-androstane intermediates from bile acids, Biochem. J.
243 (1987) 15–21.
Acknowledgments
[29] S. Costa, P.P. Giovannini, G. Fantin, A. Medici, P. Pedrini, New
9,10-secosteroids from biotransformations of bile acids with Rhodococcus
ruber, Helv. Chim. Acta 96 (2013) 2124–2133.
[30] H. Jork, W. Funk, W. Fischer, H. Wimmer, Thin-Layer Chromatography
Reagents and Detection Methods, FRG, Weinheim, 1990.
[31] I.A. Macdonald, Detection of bile acids with Komarowskyı´reagent and group
specific dehydrogenases, J. Chromatogr. 136 (1977) 348–352.
[32] L.F. de las Heras, R. van der Geize, O. Drzyzga, J. Perera, J.M. Lorens, Molecular
characterization of three 3-ketosteroid-ꢀ1-dehydrogenase isoenzymes of
Rhodococcus ruber strain Chol-4, J. Steroid Biochem. Mol. Biol. 132 (2012)
271–281.
[33] T.V. Ivashina, V.M. Nikolayeva, D.V. Dovbnya, M.V. Donova, Cholesterol
oxidase ChoD is not a critical enzyme accounting for oxidation of sterols to
3-keto-4-ene steroids in fast-growing Mycobacterium sp. VKM Ac-1815D, J.
Steroid Biochem. Mol. Biol. 121 (2012) 47–53.
[34] E.Y. Bragin, V.Y. Shtratnikova, D.V. Dovbnya, M.I. Schelkunov, Y.A. Pekov, O.V.
Egorova, T.V. Ivashina, S.L. Sokolov, V.V. Ashapkin, M.V. Donova, Comparative
analysis of genes encoding key steroid core oxidation enzymes in
fast-growing Mycobacterium spp. strains, J. Steroid Biochem. Mol. Biol. 138
(2013) 41–53.
[35] P. Fernandes, A. Cruz, B. Angelova, H.M. Pinheiro, J.M.S. Cabral, Microbial
conversion of steroid compounds: recent developments, Enzyme Microb.
Technol. 32 (2003) 688–705.
The work was supported by a grant from the Russian Science
Foundation (RSCF Project no. 14-24-00169). The authors are grate-
ful to Dr. K.S. Laurinavichus (IBPM RAS) for the performance of MS
analysis of some bioconversion derivatives.
References
[1] M.V. Donova, Transformation of steroids by actinobacteria: a review, Appl.
Biochem. Microbiol. 12 (2007) 1–14.
[2] V.V. Fokina, G.V. Sukhodol’skaya, S.A. Gulevskaya, E.Yu. Gavrish, L.I.
Evtushenko, M.V. Donova, The 1(2)-dehydrogenation of steroid substrates by
Nocardioides simplex VKM Ac-2033D, Microbiology 72 (2003) 24–29.
[3] D.V. Dovbnya, O.V. Egorova, M.V. Donova, Microbial side-chain degradation of
ergosterol and its 3-substituted derivatives: a new route for obtaining of
deltanoids, Steroids 75 (2010) 653–658.
[4] V.A. Andryushina, N.V. Rodina, T.S. Stytsenko, L.D. Huy, A.V. Druzhinina, V.V.
Yaderetz, N.E. Voishvillo, Conversion of soybean sterols into
3,17-diketosteroids using actinobacteria Mycobacterium neoaurum,
Pimelobacter simplex, and Rhodococcus erythropolis, Appl. Biochem. Microbiol.
47 (2011) 270–273.
[36] I.A. Macdonald, E.C. Meier, D.E. Mahony, G.A. Costain, 3␣, 7␣- and
12␣-hydroxysteroid dehydrogenase activities from Clostridium perfringens,
Biochim. Biophys. Acta 450 (1976) 466–476.
[37] D.H. Mallonee, M.A. Lijewski, P.B. Hylemon, Expression in Escherichia coli and
characterization of a bile acid-inducible 3␣-hydroxysteroid dehydrogenase
from Eubacterium sp. strain VPI 12708, Curr. Microbiol. 30 (1995) 259–263.
[38] J.E. Wells, P.B. Hylemon, Identification and characterization of a bile acid
7␣-dehydroxylating operon in Clostridium sp. strain TO931, a highly active
7␣-dehydroxylating strain isolated from human feces, Appl. Environ.
Microbiol. 66 (2000) 1107–1113.
[5] A.F. Hofmann, Bile acids: the good, the bad, and the ugly, News Physiol. Sci. 14
(1999) 24–29.
[6] S.P. Saini, J. Sonoda, L. Xu, D. Toma, H. Uppal, Y. Mu, A novel constitutive
androstane receptor-mediated and CYP3A-independent pathway of bile acid
detoxification, Mol. Pharmacol. 65 (2004) 292–300.
[7] T. Matsubara, K. Yoshinari, K. Aoyama, M. Sugawara, Y. Sekiya, K. Nagata, Y.
Yamazoe, Role of vitamin D receptor in the lithocholic acid-mediated CYP3A
induction in vitro and in vivo, Drug Metab. Dispos. 36 (2008) 2058–2063.
[8] S. Mahato, E. Mukherjee, S. Banerjee, Advances in microbial biotechnology of
bile acids, Biotechnol. Adv. 12 (1994) 357–361.
[39] R. Edenharder, A. Pfützner, R. Hammann, Characterization of NAD-dependent
3␣- and 3␤-hydroxysteroid dehydrogenase and of NADP-dependent
7␤-hydroxysteroid dehydrogenase from Peptostreptococcus productus,
Biochim. Biophys. Acta 1004 (1989) 230–238.
[9] Global Ursodeoxycholic Acid Market Research Report and Industry Analysis
2015–2020, QYRESEARCH, 2015130–180.
[10] M. Begley, G.M. Gahan, C.H. Hill, The interaction between bacteria and bile,
FEMS Microbiol. Rev. 29 (2005) 625–651.
[40] I.A. Macdonald, J.F. Jellet, D.E. Mahony, Bile salt 3␣- and 12␣-hydroxysteroid
dehydrogenases from Eubacterium lentum and related strains, Appl. Environ.
Microbiol. 37 (1979) 992–1000.
[41] B.A. Skalhegg, On the 3␣-hydroxysteroid dehydrogenase from Pseudomonas
testosteroni: purifications and properties, Eur. J. Biochem. 46 (1974) 117–125.
[42] V.V. Kollerov, A.A. Shutov, V.V. Fokina, G.V. Sukhodol’skaya, M.V. Donova,
Biotransformation of 3-keto-androstanes by the strain of Gongronella butleri
VKM F-1033, J. Mol. Catal. B: Enzym. 55 (2008) 61–68.
[43] I. Uhía, B. Galán, S.L. Kendall, N.G. Stoker, J.L. García, Cholesterol metabolism
in Mycobacterium smegmatis, Environ. Microbiol. Rep. 4 (2) (2012) 168–182.
[44] S. Hirano, N. Masuda, Epimerization of the 7-hydroxy group of bile acids by
the combination of two kinds of microorganisms with 7␣- and 7␤-
[11] S.F. Baron, P.B. Hylemon, Biotransformation of bile acids, cholesterol and
steroid hormones, in: R.I. Mackie, B.A. White (Eds.), Gastrointestinal
Microbiology, Vol. I, Gastrointestinal Ecosystems and Fermentations,
International Thomson Publ., New York, 1997, pp. 470–510.
[12] J.M. Ridlon, D.J. Kang, P.B. Hylemon, Bile salt biotransformations by human
intestinal bacteria, J. Lipid Res. 47 (2006) 241–259.
[13] I. Kim, S.H. Ahn, T. Inagaki, Differential regulation of bile acid homeostasis by
the farnesoid X receptor in liver and intestine, J. Lipid Res. 48 (2007)
2664–2672.
ˇ
[14] J. Holert, Z. Kulicı´, O. Yücel, V. Suvekbala, M.J. Suter, H.M. Möller, B. Philipp,
Degradation of the acyl side chain of the steroid compound cholate in
Pseudomonas sp. strain Chol1 proceeds via an aldehyde intermediate, J.
Bacteriol. 195 (2013) 585–595.
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010

G Model
MOLCAB-3498; No. of Pages9
ARTICLE IN PRESS
N.O. Deshcherevskaya et al. / Journal of Molecular Catalysis B: Enzymatic xxx (2016) xxx–xxx
9
hydroxysteroid dehydrogenase activity, respectively, J. Lipid Res. 22 (1981)
1060–1068.
[45] S. Hirano, N. Masuda, Transformation of bile acids by Eubacterium lentum,
Appl. Environ. Microbiol. 42 (1981) 912–915.
[50] E. Merino, A. Barrientos, J. Rodrigez, G. Naharro, J.M. Luengo, E.R. Olivera,
Isolation of cholesterol and deoxycholate-degrading bacteria from soil
samples: evidence of a common pathway, Appl. Microbiol. Biotechnol. 97
(2013) 891–904.
[46] A. Okamura, H. Matsui, Bile acid converting microorganism and process for
preparing bile acid, US Patent 5989855, (1999).
[51] V.Y. Shtratnikova, M.I. Schelkunov, V.V. Fokina, Y.A. Pekov, T.V. Ivashina, M.V.
Donova, Genome-wide bioinformatics analysis of steroid
[47] L.O. Severina, I.V. Torgov, G.K. Skrjabin, V.I. Zaretskii, N.S. Wulfson, I.V.
Papernaja, The transformation of desoxycholic acid by the culture
Mycobacterium mucosum 1210, Tetrahedron 25 (1969) 5617–5622.
[48] M. Jirsa, M. Novakova, V. Bradova, M. Mara, Bacterial oxidation of cholic acid
by Arthrobacter sp, Folia Microbiol. 40 (1995) 621–626.
metabolism-associated genes in Nocardioides simplex VKM Ac-2033D, Curr.
Genet. (February) (2016) 1–14.
[52] R.F. Bilton, A.N. Mason, M.E. Tenneson, Microbial degradation of deoxycholic
acid by Pseudomonas sp. NCIB 10590. Characterisation of products and a
postulated pathway, Tetrahedron 37 (1981) 2509–2513.
[49] S. Fukiya, M. Arata, H. Kawashima, D. Yoshida, M. Kaneko, K. Minamida, J.
Watanabe, Y. Ogura, K. Uchida, K. Itoh, M. Wada, S. Ito, A. Yokota, Conversion
of cholic acid and chenodeoxycholic acid into their 7-oxo derivatives by
Bacteroides intestinalis AM-1isolated from human feces, FEMS Microbiol. Lett.
293 (2009) 263–270.
[53] V.V. Kollerov, T.G. Lobastova, D. Monti, N.O. Deshcherevskaya, E.E. Ferrandi, G.
Fronza, S. Riva, M.V. Donova, Deoxycholic acid transformations catalyzed by
selected filamentous fungi, Steroids 107 (2016) 20–29.
Please cite this article in press as: N.O. Deshcherevskaya, et al., Search and discovery of actinobacteria capable of transforming deoxy-
cholic and cholic acids, J. Mol. Catal. B: Enzym. (2016), http://dx.doi.org/10.1016/j.molcatb.2016.12.010