Deoxycholic acid transformations catalyzed by selected filamentous fungi

Deoxycholic acid transformations catalyzed by selected filamentous fungi

Accepted Manuscript Deoxycholic acid transformations catalyzed by selected filamentous fungi V.V. Kollerov, T.G. Lobastova, D. Monti, N.O. Deshcherevs...

864KB Sizes 0 Downloads 146 Views

Accepted Manuscript Deoxycholic acid transformations catalyzed by selected filamentous fungi V.V. Kollerov, T.G. Lobastova, D. Monti, N.O. Deshcherevskaya, E.E. Ferrandi, G. Fronza, S. Riva, M.V. Donova PII: DOI: Reference:

S0039-128X(15)00313-X http://dx.doi.org/10.1016/j.steroids.2015.12.015 STE 7888

To appear in:

Steroids

Received Date: Revised Date: Accepted Date:

24 July 2015 8 December 2015 19 December 2015

Please cite this article as: Kollerov, V.V., Lobastova, T.G., Monti, D., Deshcherevskaya, N.O., Ferrandi, E.E., Fronza, G., Riva, S., Donova, M.V., Deoxycholic acid transformations catalyzed by selected filamentous fungi, Steroids (2015), doi: http://dx.doi.org/10.1016/j.steroids.2015.12.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Deoxycholic acid transformations catalyzed by selected filamentous fungi V.V. Kollerov,a T.G. Lobastova,a D. Monti,b,** N.O. Deshcherevskaya,a E. E. Ferrandi,b G. Fronza,c S. Riva,b and 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

Istituto di Chimica del Riconoscimento Molecolare - C.N.R.,Via Mario Bianco 9, 20131 Milano,

Italy c

Istituto di Chimica del Riconoscimento Molecolare - C.N.R., UOS-Milano Politecnico, Via

Mancinelli 7, 20131 Milano, Italy.

*

Corresponding author. Tel.: +7-4967-318579; fax: +7-495-9563370.

E-mail address: [email protected] Full address: Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Prospekt Nauki 5, 142290 Pushchino, Moscow region, Russia. **

Corresponding author. Tel.: +39 02 285 000 25; fax: +39 02 289 012 39.

E-mail address: [email protected] Full address: Istituto di Chimica del Riconoscimento Molecolare – C.N.R., Via Mario Bianco 9, 20131 Milano, Italy 1

ABSTRACT More than 100 filamentous fungi strains, mostly ascomycetes and zygomycetes from different phyla, were screened for the ability to convert deoxycholic acid (DCA) to valuable bile acid derivatives. Along with 11 molds which fully degraded DCA, several strains were revealed capable of producing cholic acid, ursocholic acid, 12-keto-lithocholic acid (12-keto-LCA), 3keto-DCA, 15β-hydroxy-DCA and 15β-hydroxy-12-oxo-LCA as major products from DCA. The last metabolite was found to be a new compound. The ability to catalyze the introduction of a hydroxyl group at the 7(α/β)-positions of the DCA molecule was shown for 32 strains with the highest 7β-hydroxylase activity level for Fusarium merismoides VKM F-2310. Curvularia lunata VKM F-644 exhibited 12α-hydroxysteroid dehydrogenase activity and formed 12-ketoLCA from DCA. Acremonium rutilum VKM F-2853 and Neurospora crassa VKM F-875 produced 15β-hydroxy-DCA and 15β-hydroxy-12-oxo-LCA, respectively, as major products from DCA, as confirmed by MS and NMR analyses. For most of the positive strains, the described DCA-transforming activity was unreported to date. The presented results expand the knowledge on bile acid metabolism by filamentous fungi, and might be suitable for preparativescale exploitation aimed at the production of marketed bile acids.

Keywords: deoxycholic acid, bile acid, bioconversion, filamentous fungi, 7β-hydroxylation, ursocholic acid

Abbreviations: BA, Bile acid; DCA, deoxycholic acid (3α,12α-dihydroxy-5β-cholan-24-oic acid); LCA, lithocholic acid (3α-hydroxy-5β-cholan-24-oic acid); UCA, ursocholic acid (3α, 7β,12α-trihydroxy-5β-cholan-24-oic acid); CA, cholic acid (3α,7α,12α-trihydroxy-5β-cholan24-oic acid); CDCA, chenodeoxycholic acid (3α,7α-dihydroxy-5β-cholan-24-oic acid); 12keto-LCA, 12-keto-lithocholic acid (3α-hydroxy-12-keto-5β-cholan-24-oic acid); 15β-hydroxyDCA,

3α,12α,15β-trihydroxy-5β-cholanoic

acid;

15β-hydroxy-12-oxo-LCA,

3α,15β-

dihydroxy-12-oxo-5β-cholanoic acid; 3α-HSDH, 3α-hydroxysteroid dehydrogenase; 12αHSDH, 12α-hydroxysteroid dehydrogenase.

2

1. Introduction

Bile acids (BAs) are steroid compounds with diverse important biological functions in the vertebrate digestion. They play a role in the solubilization and adsorption of fats, cholesterol, and lipid-soluble vitamins, and participate in the regulation of systemic endocrine functions [1]. Structurally, bile acids represent cholanic acid derivatives differing by the presence and orientation of hydroxyl groups attached to the steroid core. Primary bile acids, including cholic acid (CA; 3α,7α,12α-trihydroxy-5β-cholan-24-oic acid) and chenodeoxycholic acid (CDCA; 3α,7α-dihydroxy-5β-cholan-24-oic acid) are synthesized from cholesterol in the liver and secreted into the duodenum as the main component of bile (Fig. 1a) [2-4]. These bile acids can be subsequently transformed by the intestinal microflora in the colon into secondary bile acids, such as deoxycholic (DCA; 3α,12α-dihydroxy-5β-cholan-24-oic acid), lithocholic (LCA; 3αhydroxy-5β-cholanic acid), and ursodeoxycholic acids (UDCA, 3α,7β-dihydroxy-5β-cholan24-oic acid) (Fig.1b) [5]. The rigid steroid backbone with polar hydroxyl groups on one side and a hydrophobic convex face on another contributes to the amphipathic character of BAs. This feature in combination with the low market cost of these natural compounds ensures their broad application in different fields of medicine, supramolecular and material chemistry, and nanotechnology [6]. Being powerful biological detergents, BAs exhibit various biological activities such as antimicrobial [7], antiparasitic [8], anticancer [9], antifungal [10], and are also widely used in drug formulations, e.g., for facilitating transdermal penetration and providing drug stability [6,11]. Ursodeoxycholic acid (UDCA) is currently produced on an industrial scale as an effective drug useful for dissolving cholesterol gallstones and for the treatment of cholestatic diseases [12,13]. Cholic acid and CDCA, both extracted in good yields from bovine bile, are currently considered as the most appropriate starting materials for UDCA synthesis [14,15]. However, the industrial production of UDCA results in the accumulation of different waste products with DCA as one of the most abundant. Its accumulation in large amounts increases the ecological risks as well as the overall cost of the UDCA production process due to waste disposal [16-18].

3

Therefore, it would be economically advantageous to find either novel applications of DCA or straightforward methods to transform it into valuable derivatives. As far as direct exploitation of DCA concerns, synthetic DCA has recently found application in the non-surgical removal of fat in cosmetology [19]. In research practice, DCA is usually applied as important biomarker in the stimulation of colorectal cancer as well as a mild detergent for the isolation of membrane associated proteins [20, 21]. To date, the information on DCA conversion by microorganisms is scarce. Some bacterial strains have been reported to catalyze the oxidation of 3α-hydroxyl function to form 3-ketoDCA, or fully degrade BAs via the so-called 9(10)-seco pathway [5]. 12α-Hydroxyandrosta1,4-diene-3,17-dione and its 12β-epimer, as well as 3,12β-dihydroxy-9(10)-secoandrosta1,3,5(10)-triene-9,17-dione have been identified as major products from DCA by some Pseudomonas sp. mutant strains [22]. Filamentous fungi are known as powerful biocatalysts with a versatile enzymatic toolbox and have been shown to effectively carry out regio- and stereospecific hydroxylation of different type of steroids [23-25]. In our previous work, fungal strains capable to catalyze the 7βhydroxylation of LCA have been identified and exploited to achieve the formation of UDCA with high yield [26]. However, a restricted number of fungi has been described to transform DCA so far. Curvularia lunata NRRL 2380, Helicostylum piriforme ATTC 8992 and Pestalotia foedans ATCC 11817 have been reported as capable to catalyze the 1β-, 11β-, 15βhydroxylation and the oxidation of 12α-hydroxyl group of DCA with accumulation of the corresponding derivatives [27]. 12α-Hydroxysteroid dehydrogenase activity was also shown for the Penicillium sp. TTUR 422 strain, capable to catalyze the 7β-hydroxylation of DCA as well [28]. In this work, we investigated the ability of a large set of filamentous fungi from different phyla to convert DCA. Special attention was paid to the strains capable to selectively catalyze the hydroxylation and/or the oxidation of DCA to form useful starting molecules for UDCA production, e.g., ursocholic acid (UCA, 3α,7β,12α-trihydroxy-5β-cholan-24-oic acid), CA, or 12-keto-LCA.

4

2.

Experimental

2.1. Materials Deoxycholic acid (DCA, 3α,12α-dihydroxy-5β-cholan-24-oic acid), ursocholic acid (UCA, 3α,7β,12α-trihydroxy-5β-cholan-24-oic acid), cholic acid (CA, 3α,7α,12α-trihydroxy-5βcholan-24-oic acid), and 12-keto-lithocholic acid (12-keto-LCA, 3α-hydroxy-12-keto-5βcholan-24-oic acid) were obtained from ACROS Organics (USA). Other bile acids and their derivatives were a kind gift from Prodotti Chimici e Alimentari S.p.A. (Basaluzzo, Italy). Yeast extract was purchased from Difco (USA), corn steep solids from Sigma-Aldrich (USA). All other reagents were of the best purity grade from commercial suppliers.

2.2. Microorganisms The strains were obtained from the All-Russian Collection of Microorganisms at the Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences (VKM IBPM RAS) and the working collection of the Laboratory of Microbial Transformation of Organic Compounds at the same Institute (Laboratory MTOC, IBPM RAS).

2.3. Screening of fungal strains If not otherwise mentioned, fungal strains were grown on the nutrient medium (50 mL) containing (g/L): starch - 45.0; corn extract – 10.0; yeast extract – 3.0; CaCO3 - 3.0, рН 7.0. Cultures were carried out aerobically on a rotary shaker (220 rpm) at 29°C for 48 h in Erlenmeyer flasks (750 mL). For DCA bioconversion, the same fresh medium was inoculated with a mycelium of first generation stage (10 % (v/v)). DCA was added as a hot methanol solution (final solvent concentration in transformation medium did not exceed 2%, v/v) to a final concentration of 1 g/L. Bioconversion was carried out on a rotary shaker (220 rpm) at 29°C for a maximum of 240 h and monitored daily by TLC as described below.

2.4. Isolation of metabolites Metabolites were isolated from the culture medium by ethyl acetate (EtOAc) extraction. After evaporation of the solvent under vacuum, crude residues were fractionated by silica gel column chromatography [29]. The individual compounds were analyzed by MS and NMR techniques.

5

2.5. Analyses

2.5.1. Thin layer chromatography (TLC) Samples of cultivation broth (1 mL) were taken every 24 h and extracted with 5 mL of EtOAc. The extracts were applied to Sorbfil UV 254 (Russia) TLC plates and developed in a mixture of CHCl3-acetone-CH3COOH 50:50:0.5 (System A), or CHCl3-MeOH-CH3COOH 70:30:0.5 (System B). Staining of TLC plates was carried out by using either MnCl2 [30], the Komarowski reagent [31] or the phosphomolybdic reagent, the latter prepared by dissolving of phosphomolybdic acid (50 g) in glacial acetic acid (950 mL) and concentrated sulfuric acid (50 mL), and heating to 105°C. 2.5.2. HPLC For selected reactions, the conversion of DCA into the 7α- and 7β-hydroxylated derivatives (CA and UCA, respectively) was monitored by HPLC analyses on a 5 µm C-18 reverse phase column (GraceSmart, 4.6 x 250 mm, Grace, Deerfield, USA) and using a Jasco 880-PU pump equipped with a Jasco 875-UV/Vis detector. The mobile phase consisted of 0.3% ammonium carbonate aqueous solution-acetonitrile (t = 0 min, 85:15 (v/v); t = 10 min, 75:25 (v/v); t = 20 min, 65:35 (v/v)) with a flow rate of 1.0 mL min-1. Peaks were detected at 220 nm. Calibration curves were prepared using standard solutions of DCA, UCA and CA. Retention times: UCA, 10.6 min; CA, 18.1 min; DCA, 19.8 min.

2.5.3. Mass-spectrometry (MS) and NMR analyses ESI-MS spectra were recorded on a FT-ICR Mass Spectrometer APEX II & Xmass software (Bruker Daltonics, USA). ESI-MS for UCA and CA: m/z calcd for [C24H39O5(-1)-]: 407.28030; found: 407.26450 and 407.28059, respectively. ESI-MS data for metabolites X1-X4: X1: m/z calcd for [C24H39O5(-1)-]: 407.28030; found: 407.27988; X2: m/z calcd for [C24H37O5(-1)-]: 405.26465; found: 405.26398; X3: m/z calcd for [C24H39O5(-1)-]:407.28030; found: 407.28059; X4: m/z calcd for [C24H39O5(-1)-]: 407.28030; found: 407.27966. 1

H- and

13

1

H-NMR, selected signals (solvent: DMSO-d6): δ 3.80 (1 H, t, J = 2.8 Hz, H-12β), 3.33 (2 H,

C-NMR spectra were recorded on a Bruker AC500 spectrometer (500 MHz). UCA,

m, H-3β and H-7α), 0.94 (3 H, d, J = 7 Hz, C-21 Me), 0.86 (3 H, s, C-19 Me), 0.63 (3 H, s, C18 Me). CA, 1H-NMR, selected signals (solvent: DMSO-d6): δ 3.80 (1 H, t, J = 2.8 Hz, H12β), 3.61 (1 H, q , J = 3.0, H-7β), 3.18 (1 H, m, H-3β), 0.94 (3 H, d, J = 7 Hz, C-21 Me), 0.86 (3 H, s, C-19 Me), 0.63 (3 H, s, C-18 Me). 6

1

H- and

13

C-NMR spectroscopic data for compound X1 (3α,12α,15β-trihydroxy-5β-cholanoic

acid) and X2 (3α,15β-dihydroxy-12-oxo-5β-cholanoic acid), obtained in [2 H4]-methanol, are reported in Table 3 and Table 4, respectively, and discussed in the following.

2.5.4. Melting points (mp) Mps of compounds X1 and X2 were measured on an automated melting point apparatus EZMelt (SRS, USA). The free acids were extracted with isopropanol and crystallized from EtOAc - hexane mixture (3:1, v/v).

3.

Results and discussion

3.1. DCA conversion by fungal strains Fungal strains, mostly from Ascomycota, Zygomycota and Mucorales phyla (totally, 118), were chosen for screening trials based on their previously shown activity towards other steroids and bile acids, such as LCA [25, 26, 29]. Among the 118 tested strains, 45 microorganisms, including some strains of Bipolaris, Curvularia, Aspergillus, Paecilomyces, Penicillium, Acremonium, Fusarium, Sepedonium, Trichoderma,

Trichothecium,

Scopulariopsis,

Myceliophtora,

Nigrospora,

Spicaria,

Gongronella, Mucor, Mortierella, Conidiobolus, and Sporotrichum genera, did not reveal any transforming activity toward DCA (steroid substrate was remained unconverted with no product detection in the transformation medium during the fungal growth (3-5 days)) (Table 1).

The remaining 73 fungal strains demonstrated to be capable to transform DCA to some extent. In most cases, the consumption of DCA was accompanied by the formation of different products, while for a limited number of strains the full degradation of DCA, i.e., without the formation of detectable products under the analytical conditions used, was observed. As far as ascomycete fungi concern, DCA conversion was observed with 53 of the 91 tested strains. Only in three cases - Aspergillus flavus VKM F-1024, A. sydowi VKM F-2268, and Fusarium viride VKM F-2735 - the full degradation of DCA occurred under the conditions used (Table 1).

7

The capability to transform DCA was shown also for 17 of the 20 tested strains of the Mucorales order and 3 of 5 strains of Mortierella genus (Table 1). However, full DCA degradation without accumulation of any steroid intermediates in detectable amounts was observed for 8 strains, - the representatives of Absidia, Cunninghamella, Rhizomucor, and Benjaminiella (Table 1). Interestingly, all tested Cunninghamella japonica strains degraded DCA, thus suggesting a possible type specificity. As mentioned above, BAs are generally considered to be natural substrates for proteobacteria and actinomycetes which can fully utilize them as carbon and energy sources via the so-called 9(10)-seco pathway [3]. To the best of our knowledge, the capability of filamentous fungi to catabolize DCA to its full degradation has not been previously described. As far as products formation concerns, 31 fungal strains of different genera either accumulated DCA metabolites in too small amounts (the yield did not exceed 5%), or showed an extremely low selectivity in the respect of the formed products as in the case of DCA transformation by Fusarium culmorum VKM F-1017, Doratomyces purpureofuscus VKM F-2519, and Mortierella isobellina VKM F-525. As, in both cases, it was not possible to define the structure of these products, their formation has been indicated in Table 1 as "other products" with no determined structure ("n.d."). However, by TLC comparison with pure bile acids standards, we succeeded in the identification of several strains capable of producing cholic acid (CA), ursocholic acid (UCA), 12-keto-lithocholic acid (12-keto-LCA), and 3-keto-DCA as major products from DCA. In selected reactions, the conversions of DCA into CA and/or UCA were quantitatively estimated by HPLC analysis and the identity of these products further confirmed by mass and NMR analysis (see Experimental for details). A mixture of the stereoisomers UCA and CA was detected among the metabolites when incubating Aspergillus sp. MTOC F-246, Fusarium sp. MTOC F-3051, Neurospora crassa, Absidia coerulea, Rhizopus stolonifer VKM F-401, and Mortierella isobellina with DCA (Table 1), thus indicating the possible concomitant production of 7α- and 7β-hydroxylating activities. The strains of Acremonium cerealis VKM F-1542, Fusarium culmorum VKM F2303, F. moniliforme VKM F-670, and F. proliferatum VKM F-136 transformed DCA to CA more selectively, however, the conversion did not exceed 15% under the conditions used. Interestingly, when tested toward LCA in our previous studies [26], none of the positive Aspergillus and Fusarium strains showed any 7α-hydroxylase activity.

8

Most of the strains capable of selectively producing UCA from DCA belong to ascomycetes (particularly to Penicillium, Talaromyces, Acremonium, Fusarium, Gibberella, Metarhizum, and Scopulariopsis genera). In fact, among zygomycetes this exclusive activity was observed only for two strains from Cunninghamella and Actinomucor genera and in both cases with less than 15% conversions. Some 7β-hydroxylating strains showed high activity levels toward DCA affording more than 20% conversion into UCA, specifically Aspergillus sp. MTOC F-246 (23%), Fusarium merismoides VKM F-2310 (35%), F. wolgoense VKM F-2763 (25%), Fusarium sp. MTOC F159 (28%), Fusarium sp. MTOC F-3051 (23%), Gibberella zeae VKM F-2598 (23%), Scopulariopsis brevicaulis VKM F-2411 (22%), and Rhizopus stolonifer VKM F-401 (28%). It should be noted that the strains previously shown to be effective in the 7β-hydroxylation of LCA, e.g., Gibberella zeae VKM F-2600 [26], showed very low hydroxylase activity towards DCA, or its complete absence, as for example was the case for Bipolaris australiensis VKM F3040, Cunninghamella japonica VKM F-1205, and Cunninghamella sp. MTOC F-6. Therefore, the presence of an additional 12α-hydroxyl group in the DCA structure (Fig.1) seems to hinder the 7β-hydroxylation reaction by previously shown LCA-converting strains. Maximum UCA productivity during the screening trials was thus observed for Fusarium merismoides VKM F-2310 that afforded about 35% conversion of DCA into UCA without any concomitant production of CA. This strain was thus highlighted as the most promising wholecell catalyst for the selective 7β-hydroxylation of DCA. For all the selected strains the capability to catalyze 7α/β-hydroxylation of DCA molecule was shown here for the first time. 3-Keto-DCA was accumulated as a major product from DCA by 9 fungal strains belonging to Gibberella, Scopulariopsis, Thielavia, Mucor, Rhizopus, Phycomyces, and Mortierella genera (Table 1). 3α-Hydroxysteroid dehydrogenases (3α-HSDHs) which are most likely responsible for the oxidation of the 3α-hydroxy group in DCA, were detected earlier in intestinal and nonintestinal bacteria [32,14]. Fungi from the Gibberella and Cunninghamella genera showed this activity towards LCA [26], but no 3α-HSDH activity towards DCA was reported to date from other filamentous fungi. When incubated with DCA, the strain of Curvularia lunata VKM F-644 produced several metabolites with the major one identified as 12-keto-LCA by TLC comparison with an authentic standard (Fig.2). This strain is known to effectively carry out the 11β-hydroxylation 9

of cortexolone and other pregnane steroids [33,25], and is capable of introducing a hydroxy group in position 14α of 3-ketoandrostanes as well [33].



The oxidation of the 12α-hydroxy group is an important reaction for the complete removal of the functionality at C-12 which is necessary for further chemoenzymatic route to UDCA [14]. So far, 12α-hydroxysteroid dehydrogenase (12α-HSDH) activity toward CA was reported mainly for anaerobic bacterial strains from the Clostridium and Eubacterium genera [14]. The presence of this activity along with 7β-hydroxylation of DCA was earlier demonstrated for an only fungal strain, i.e., Penicillium sp. TTUR 422 [28]. In that case, it was shown that the oxidation of the 12α-hydroxy group was possible only in the presence of UCA, i.e., after the occurrence of the hydroxylation reaction at C-7 [28]. Finally, our screening experiments highlighted the strains of Acremonium rutilum VKM F2853, Neurospora crassa VKM F-875, Backusella lamprospora VKM F-944 and Scopulariopsis coprophila VKM F-2719 that converted DCA to form four main additional products (X1-X4) which differed both from the available standards of BAs and from each other (Fig.3, Table 2). In order to isolate the individual compounds X1-X4, each 72 h culture broth (50 mL) of DCA transformation by the abovementioned four strains was extracted three times with an equal volume of EtOAc. After rotavaporation of the solvent, crude residues were fractionated by silica gel column chromatography with CHCl3-MeOH-CH3COOH (70:30:0.5) as a mobile phase. The fractions containing compounds X1-X4 were collected and evaporated under vacuum. The individual residues of compound X1 (18 mg), X2 (15.5 mg), X3 (16.5 mg) and X4 (23.5 mg) were obtained and an attempt to identify them by MS and NMR analyses was carried out.



3.2. Structural characterization of X1-X4 metabolites 3.2.1. ESI-MS analyses of X1-X4 metabolites Molecular masses of compounds X1-X4 were easily estimated by ESI-MS analysis. Metabolites X1, X3, and X4 showed an identical molecular weights (М 408) that exceeded of 10

16 atomic units that of DCA (М 392). The obtained data suggested the presence of one additional hydroxyl group in the structure of metabolites X1, X3 and X4, possibly differing in its regio- and/or stereo-position on the steroid skeleton as the three compounds showed a very different mobility during the TLC analyses (Fig. 3). As far as compound X2 concerns, the m.w. (М 406) indicated possible presence of one additional keto group. In order to identify the precise structure of metabolites X1-X4 and define the position of the additional substituents, 1

H- and 13C-NMR analyses were carried out.

3.2.2. 1H- and 13C-NMR analyses of X1-X4 metabolites 1

H- and 13C-NMR spectroscopic data of X1 metabolite are presented in Table 3.



Both data, compared with that of DCA, show great variation for the nuclei of the fivemembered ring suggesting that the location of the additional OH group should occur at carbon C-15. This hypothesis is supported by the observation of a strong long-range heterocorrelation in the HMBC spectrum between the nuclei H-14 and C-15. Moreover the chemical shifts of C14 and C-16 show strong downfield shift (+5.0 and +13.1 ppm, respectively) with respect to that of DCA in agreement with the effect exerted by a substituent on the nuclei in β position, while C-8 displays an upfield shift (-4.3 ppm) due to the γ-effect on the C-H carbon in 1,3-syn diaxial relationship with the substituent. These observations strongly support the position 15β of the new hydroxyl group (Fig. 4a).



The analysis of the proton spectrum fully confirms the 15β stereochemistry of the OH group. In fact, the coupling constants J(14, 15) = 5.8 Hz, J(15, 16a) = 7.7 Hz, J(15, 16b) = 2.0 Hz, J(16a, 17) = 9.0 Hz, J(16b, 17) = 10.2 Hz correspond, applying the generalized Karplus equation [34], to the following set of dihedral torsion angles, respectively: 49°, 20°, 101°, 23° and 153° (or alternatively 138°, 135°, 62°, 148° and 7°). The structure with OH-15β-oriented fits rather well with the first set of dihedral angles (while the fitting between the structure with OH-15αoriented and the second set of dihedral angles is not satisfactory). Moreover, by irradiation of H-15, strong NOEs are observed for the nuclei H-14, H-7eq, H-16a (but not for H-16b). The 11

irradiation of CH3-18 produces NOE enhancements on the nuclei H-20, H-8, H-11, H-12 and H-16b (weak) but not on H-15. Taken together, the NMR data discussed above allow to establish with certainty that X1 is the 3α,12α,15β-trihydroxy-5β-cholanoic acid (15β-hydroxy-DCA) (Fig. 4a). This compound was previously described by Carlström et al. [36]. They established its structure on the basis of the analysis of the low-field proton NMR spectrum which, although very partial, is in perfect agreement with our data. The 1H- and 13C-NMR spectroscopic data for compound X2 are presented in Table 4. The NMR data for the five-membered ring of compound X2 was very similar to that of compound X1: the vicinal coupling constants were almost identical and also the NOE enhancements were the same. The keto group at C-12 in compound X2 (Fig. 4b) was determined from the 1 H-13C long-range heterocorrelation spectrum which shows very clear correlations between the carbonyl carbon and the protons H-11a, H-11b, H-14 and CH3-18.

Based on the results of MS, 1H- and

13

C-NMR analyses, compounds X1 and X2 were

unequivocally identified as 3α,12α,15β-trihydroxy-5β-cholanoic acid (15β-hydroxy-DCA) and 3α,15β-dihydroxy-12-oxo-5β-cholanoic acid (15β-hydroxy-12-oxo-LCA), respectively. 15βhydroxy-12-oxo-LCA was found to be a new compound with mp 175-178°. Unfortunately, 1 H- and

13

C-NMR analyses of the unknown compounds X3 and X4 formed by

Backusella lamprospora VKM F-944 and Scopulariopsis coprophila VKM F-2719 strains, respectively, showed the presence of the mixtures of different trihydroxylated products that could not be separated by flash chromatography. Further investigations are currently ongoing in our labs to try to differentiate and isolate the compounds produced by these fungal strains in pure form. The results obtained in the characterization of metabolites X1 and X2 confirmed the presence of 15β-hydroxylating activity towards DCA both in Acremonium rutilum VKM F-2853 and Neurospora crassa VKM F-875. This activity was earlier described for Curvularia lunata NRRL-2380 [27] and Penicillium sp. ATCC 12556 [36]. In both cases, the introduction of an additional hydroxyl group at the 15β position of DCA was deduced by interpretation of MS and NMR spectra, but the complete NMR spectroscopic data were not reported. Moreover, it should 12

be noted that expression of 15β-hydroxylase activity toward BAs was not hitherto described for any species belonging to the Acremonium and Neurospora genera. Earlier, the 15βhydroxylation of LCA with formation of 15β-hydroxy-LCA was demonstrated for a Cunninghamella blakesleeana strain [37]. Hydrophilicity measurements and in vitro cholesterol solubilization tests showed that this hydroxyderivative was as effective as UDCA in cholesterol solubilization [37]. It is reasonable to propose similar features of 15β-hydroxy-DCA and 15βhydroxy-12-oxo-LCA produced by A. rutilum VKM F-2853 and N. crassa VKM F-875. These compounds may also be used as precursors for 15β-hydroxy-LCA production. The identification of metabolite X2 as 15β-hydroxy-12-oxo-LCA indicates the additional presence of 12α-HSDH activity towards 15β-OH-DCA in N. crassa VKM F-875. Noteworthy, no 12α-HSDH activity by this strain was observed towards DCA. Probably, the activity was induced only in the presence of the 15β-hydroxylated derivative. Similar suggestions were mentioned above on the induction of 12α-HSDH by 7β-hydroxylated derivative of DCA in Penicillium sp. TTUR 422 [28]. Figure 5 summarizes the set of products obtained from DCA by using selected fungal strains in this study.

4.

Conclusions

In this work, the capability of a large set of different fungal strains to perform structural modifications of DCA has been investigated. A few strains have shown to be capable to fully degrade DCA, whereas, more interestingly, several others converted DCA to valuable compounds, such as CA, UCA, and 12-keto-LCA, which can be exploited for further chemoenzymatic production of UDCA. The novel data have shown the occurrence of 7α-, 7β- and 15β-hydroxylase activities, as well as 3α- and 12α-hydroxysteroid dehydrogenase activities towards DCA in different fungal strains. For many organisms, these DCA-transforming activities were shown in this work for the first time. Fusarium merismoides VKM F-2310 has been identified as the most promising biocatalyst for the selective 7β-hydroxylation reaction of DCA and the scale-up of this biotransformation is currently under investigation in our labs. Moreover, Acremonium rutilum F-2853 and Neurospora crassa F-875 exhibited high 15β-hydroxylase and 12α-hydroxysteroid

13

dehydrogenase activities with formation of 15β-OH-DCA and 12-keto-15β-OH-DCA as the major products, respectively. These results confirm the broad biocatalytic potential of filamentous fungi towards bile acids, and could be suitable for preparative-scale exploitation of the selected fungal strains for the production of valuable cholanic acids.

ACKNOWLEGMENTS We would like to thank Fausto Polentini and Dr. Gianluca Galdi (Prodotti Chimici e Alimentari, S.p.A.) for providing us samples of pure bile acids derivatives. Dr. Anita Larovere (ICRM-CNR) is also thanked for her help in products isolation. The Cariplo Foundation and the Landau Network-Centro Volta are gratefully acknowledged for a fellowship to V. K. V. Kollerov, T. Lobastova, N. Deshcherevskaya and M. Donova are grateful to Russian Science Foundation (Project No. 142400169) for support of their work.

14

REFERENCES

[1] Wollam J, Antebi A. Sterol regulation of metabolism, homeostasis, and development. Annu Rev Biochem 2011;80:885-916. [2] Lee JY, Arai H, Nakamura Y, Fukiya S, Wada M, Yokota A. Contribution of the 7βhydroxysteroid dehydrogenase from Ruminococcus gnavus N53 to ursodeoxycholic acid formation in the human colon. J Lipid Res 2013;54:3062-3069. [3] Begley M, Gahan G, Hill C. The interaction between bacteria and bile. FEMS Microbiol Rev 2005;29:625-651. [4] Hofmann AF. The continuing importance of bile acids in liver and intestinal disease. Arch Intern Med 1999;159:2647-2658. [5] Ridlon J, Kang D, Hylemon P. Bile salt biotransformations by human intestinal bacteria. J Lipid Res 2006;47:241-259. [6] Kim DH, Larson AC. Deoxycholate bile acid directed synthesis of branched Au nanostructures for near infrared photothermal ablation. Biomaterials 2015;56:154-64. [7] Bansal S, Singh M, Kidwai S, Bhargava P, Singh A, Sreekanth V, Singh R, Bajaj A. Bile acid amphiphiles with tunable head groups as highly selective antitubercular agents. Med. Chem. Commun 2014;5:1761-1768. [8] García LG, Antonela ZM, Simonetti L, Longhi SA, Baldessari A.Enzymatic synthesis of bile acid derivatives and biological evaluation against Trypanosoma cruzi. Bioorg Med Chem 2015 May 30. pii: S0968-0896(15)00445-9. doi: 10.1016/j.bmc.2015.05.035. [9] Jampilek J, Brychtova K. Azone analogues: classification, design, and transdermal penetration principles. Medicinal research reviews 2012;32: 907-47. [10] Marples BA, Stretton RJ. Antifungal utility of bile acids. Patent US 4681876A:1987. [11] Setchell KDR, Heubi JE. Defects in bile acid biosynthesis--diagnosis and treatment. J Pediatr Gastroenterol Nutr 2006;43:S17–22; [12] Crosignani A, Stechell KDR, Invernizzi P, Larghi A, Rodrigues CMP, Podda M. Clinical pharmacokinetics of therapeutic bile acids. Clin Pharmacokinet 1996;30:333–358.

15

[13] Podda M, Zuin M, Dioguardi ML, Festorazzi S, Dioguardi N. A combination of ursodeoxycholic acid and chenodeoxycholic acid is more effective than either alone in reducing biliary cholesterol saturation. Hepatology 1982;2:334–339. [14] Eggert T, Bakonyi D, Hummel W. Enzymatic routes for the synthesis of ursodeoxycholic acid. J Biotechnol 2014;191:11-21. [15] Monti D, Ferrandi EE, Zanellato I, Hua L, Polentini F, Carrea G, Riva S. One-pot multienzymatic synthesis of 12-ketoursodeoxycholic acid: subtle cofactor specificities rule the reaction equilibria of five biocatalysts working in a row. Adv Synth Catal 2009;351:1303-1311. [16] Philipp B. Bacterial degradation of bile salts. Appl Microbiol Biotechnol 2011;89:903– 915. [17] Mahato S, Mukherjee E, Banerjee S. Advances in microbial biotechnology of bile acids. Biotech Adv 1994;12:357–361. [18] Sharma R, Prichard D, Majer F, Byrne AM, Kelleher D, Long A, Gilmer JF. Ursodeoxycholic acid amides as novel glucocorticoid receptor modulators. J Med Chem 2011;54:122-130. [19] Moriarty RM, David NE, Mahmood NA. Synthetic bile acid compositions and methods. US 20140038932 A1:2014. [20] Peterlik M. Role of bile acid secretion in human colorectal cancer. Wien Med Wochenschr 2008;158:539-541. [21] Neugebauer J. Detergents: an overview. Methods Enzymol 1990;182:239-253. [22] Leppik RA, Park RJ, Smith MG. Aerobic catabolism of bile acids. Appl Environ Microbiol 1982;44:771–776. [23] Lobastova TG, Khomutov SM, Vasiljeva LL, Lapitskaya MA, Pivnitsky KK, Donova MV. Synthesis of 3β-hydroxy-androsta-5,7-dien-17-one from 3β-hydroxyandrost-5-en-17-one via microbial 7α-hydroxylation. Steroids 2009;74:233–237. [24] Donova MV, Egorova OV. Microbial steroid transformations: current state and prospects. Appl Microbiol Biotechnol 2012;94:1423-47.

16

[25] Kollerov VV, Fokina VV, Sukhodolskaya GV, Shutov AA, Donova MV. 11βHydroxylation of 6α-fluoro-16α-methyl-deoxycorticosterone 21-acetate by filamentous fungi. Appl Biochem Microbiol 2015;51:169–179. [26] Kollerov V, Monti D, Deshcherevskaya N, Lobastova T, Ferrandi E, Larovere A, Gulevskaya S, Riva S, Donova M. Hydroxylation of lithocholic acid by selected actinobacteria and filamentous fungi. Steroids 2013;78:370-378. [27] Hayakawa S. Microbial transformation of bile acids. A unified scheme for bile acid degradation, and hydroxylation of bile acids. Z Allg Mikrobiol 1982;22:309–326. [28] Okamura A, Matsui H. Bile acid converting microorganism and process for preparing bile acid. US Patent 5,989,855:1999. [29] Lobastova TG, Gulevskaya SA, Sukhodolskaya GV, Turchin KV, Donova MV. Screening of mycelial fungi for 7a- and 7b-hydroxylase activity towards dehydroepiandrosterone. Biocatal Biotransform 2007;25:434–42. [30] Jork H, Funk W, Fischer W, Wimmer H. Thin-layer chromatography. Reagents and detection methods. FRG: Weinheim; 1990. [31] Macdonald I. Detection of bile acids with Komarowsky´reagent and group specific dehydrogenases. J Chromatogr 1977;136:348-352. [32] Prabha V, Ohri M. Bacterial transformation of bile acids. World J Microbiol Biotechnol 2006;22:191-196. [33] Kollerov VV, Shutov AA, Fokina VV, Sukhodolskaya GV, SA Gulevskaya Donova MV. Bioconversion of C19- and C21-steroids with parent and mutant strains of Curvularia lunata. Appl Biochem Microbiol 2010;46:198–205. [34] Haasnoot CAG, de Leeuw FAAM, Altona C. The relationship between proton-proton NMR coupling constants and substituent electronegativities - I : An empirical generalization of the Karplus equation. Tetrahedron 1980;36:2783-2792. [35] Waterhouse D, Barnes S, Mucio D. Nuclear magnetic resonance spectroscopy of bile acids. Development of two-dimensional NMR methods for the elucidation of proton resonance assignment for five common hydroxylated bile acids, and their parent bile acid, 5β-cholanic acid. J Lipid Res 1985;26:1068–1078.

17

[36] Carlstrom K, Kirk D, Sjovall J. Microbial synthesis of 1β- and 15β-hydroxylated bile acids. J. Lipid Res 1981;22:1225-1234. [37] Kulprecha S, Nihira T, Yamada K, Yoshida T, Nilubol N, Taguchi H. Transformation of lithocholic acid to a new bile acid, 3α, 15β-dihydroxy-5β-cholanic acid by Cunninghamella blakesleeana ST22. Appl Microbiol Biotechnol 1985;22:211-216.

18

Fig. 1. a) OH COOH

COOH 12

3

7

OH

HO

OH

HO

Cholic acid (CA)

Chenodeoxycholic acid (CDCA)

b) OH COOH

HO

Deoxycholic acid (DCA)

HO

COOH

HO

Lithocholic acid (LCA)

COOH

OH

Ursodeoxycholic acid (UDCA)

19

Fig. 2

12-keto-LCA

DCA

1

2

20

Fig. 3

X2

DCA

X4 X1 X3

1

2

3

4

21

Fig. 4

a) 18 19

21

11 12

20 17

10

1

23

8 9

14 15

7

16 3

18

b) 19

21

11 12

10

1

20 17

23

8 9

14 7

15 16

3

22

Fig. 5

23

Table 1. DCA transformation by fungal strains Order

Family

Genus

Species/Strain No.

DCA Productsa, b conver UC CA Others sion A

Ascomycota Bipolaris australiensis Dothideale s

Pleosporace ae

Curvular ia

VKM F-3040 geniculata VKM F-958 inaequalis VKM F-3289 lunata VKM F644 protuberata VKM F-3708

Drechsle ra

avenacea VKM F3284 awamori VKM F808 flavus VKM F1024 niger VKM F-212 oryzae VKM F-44

Aspergill us

Eurotiales Trichocoma ceae

sydowii VKM F441 sydowii VKM F2268 wentii VKM F797 sp. MTOC F-129 sp. MTOC F-246 sp. MTOC F-261

Paecilom variotii VKM F1296 yces brevicompactum VKM F-1127 camemberti VKM F-2531 chrysogenum var. chrysogenum VKM F-1078 citrinum VKM F2350 Penicilli decumbens VKM um F-1077 hispanicum VKM F-2179

-

-

-

-

+

-

-

+ n.d. c

+

-

-

+ n.d.

++

-

- ++ (12-keto-LCA)

-

-

-

-

-

-

-

-

-

-

-

-

++

-

-

full degradation

-

-

-

-

-

-

-

-

+

-

-

+ n.d.

++

-

-

full degradation

+

-

-

+ n.d.

-

-

-

-

++

++

+

-

+

-

-

+ n.d.

-

-

-

-

+

-

-

+ n.d.

-

-

-

-

-

-

-

-

+

-

-

+ n.d.

+

-

-

+ n.d.

+

-

-

+ n.d. 24

Helotiales

Sclerotiniac eae

Talarom yces Sclerotin ia

indonesiae VKM F-905 ochrochloron VKM F-1702 onobense VKM F2183 raistrickii VKM F-2387 simplicissimum VKM F-1149 thomii VKM F3026 luteus VKM F304 sclerotiorum VKM F-879 eguptiacum VKM F-199

+

+

-

-

-

-

-

-

+

+

-

-

-

-

-

-

-

-

-

-

-

-

-

-

+

+

-

-

+

-

-

+ n.d.

+

-

-

+ n.d.

-

-

-

-

+

-

-

+ n.d.

+

-

+

-

+

-

-

+ n.d.

+

-

-

+ n.d.

-

-

-

-

-

-

-

-

-

-

-

-

roseum VKM F1458

-

-

-

-

rutilum VKM F2853

++

+

- ++ (15β-hydroxyDCA)

strictum VKM F1336

+

-

-

+ n.d.

++

+

-

++ n.d.

+

-

+

-

-

-

-

-

-

-

-

-

-

-

-

-

arxii VKM F2717

Acremon ium

biseptum VKM F2899 cereale VKM F1542 cereale VKM F3033 domschii VKM F2819 felinum VKM F1018

felinum VKM F1300 felinum VKM F1327

Hypocreal es

Hypocreace ae

culmorum VKM F-1017 culmorum VKM F-2303 graminearum VKM F-2306 heterosporum VKM F-2747 lateritium VKM F-2308

25

Fusariu m

merismoides VKM F-1181 merismoides VKM F-2310 merismoides VKM F-3993 moniliforme VKM F-670 oxysporum VKM F-845 oxysporum VKM F-931 proliferatum VKM F-136 sarcochroum VKM F-2315 semitectum VKM F-1938 solani VKM F142 solani VKM F2316 sporotrichioidesV KM F-2317 viride VKM F2735 wolgoense VKM F-2763 sp. MTOC F-159 sp. MTOC F-160 sp. MTOС F-161 sp. MTOC F-3051

Gibberel la

Clavicipitac eae Incertae sedis Nectriaceae

Sepedoni um Trichode rma Beauveri a Metarhiz um Trichoth ecium Nectria

fujikuroi VKM F1014 zeae VKM F-2598 zeae VKM F-2599 zeae VKM F2600 ampullosporum VKM F-2821 viride VKM F2430 bassiana VKM F2533

anisopliae VKM F-1490 roseum VKM F843 cosmariospora VKM F-2862

+

+

-

-

++

++

-

-

+

+

-

-

+

-

+

-

-

-

-

-

-

-

-

-

+

-

+

-

-

-

-

-

-

-

-

-

-

-

-

-

+

+

-

-

-

-

-

-

++

-

-

full degradation

++

++

-

-

++

++

-

+ n.d.

-

-

-

-

-

-

-

-

++

++

+

-

+

-

-

+ n.d.

++

++

-

-

++

+

- ++ (3-keto-DCA)

+

+

-

+

-

-

-

-

-

-

-

-

+

-

-

+ n.d.

+

+

-

-

-

-

-

-

+

-

-

+ n.d. 26

Doratom yces

Microascal Microascac es eae Scopular iopsis

Sordariale s

Chaetomiac eae

Sordariacea e Trichospha Trichosphae eriales riaceae -

Chaetom idium Myceliop htora Thielavia Neurosp ora Nigrospo ra Spicaria

purpureofuscus VKM F-2519 asperula VKM F760 brevicaulis VKM F-406 brevicaulis VKM F-2409 brevicaulis var.glabra VKM F-2411 brumptii VKM F415 coprophila VKM F-2719 viridae VKM F2430 pilosum VKM F1851 lutea VKM F2876 ovispora VKM F1734 crassa VKM F875 oryzae VKM F1939 fumoso-rosea VKM F-881

++

-

-

++ n.d.

+

-

-

+ n.d.

++

-

+ ++ (3-keto-DCA)

+

+

-

+ n.d.

++

++

-

+ n.d.

+

-

-

+ n.d.

++

-

+ ++ (3-keto-DCA)

-

-

-

-

+

-

-

+ n.d.

-

-

-

-

++

-

- ++ (3-keto-DCA)

++

+

+ ++ (15β-hydroxy12-oxo-LCA)

-

-

-

-

-

-

-

-

Zygomycota and Mucorales Absidia

Mucorales

Cunningham Cunningh ellaceae amella

blakesleeana VKM F-1721

++

-

-

full degradation

coerulea VKM F-833

+

+

+

-

+

+

-

-

+

-

+

+ n.d.

++

-

-

full degradation

++

-

-

full degradation

++

-

-

full degradation

++

-

-

full degradation

++

-

-

full degradation

-

-

-

-

+

+

-

+ n.d.

echinulata VKM F- 470 echinulata VKM F-662 japonica VKM F-663 japonica VKM F-957 japonica VKM F-1205 sp. MTOC F-5

sp. MTOC F-6

Gongronel la Actinomuc or

butleri VKM F1033 elegans VKM F492

27

Backusella lamprospora VKM F-944

Mucoraceae

Mucor

Rhizomuc or Rhizopus Mycotyphac eae Phycomycet aceae

Mortierella les Mortierellac eae

Entomopht horales

Ancylistacea e

Benjamini ella Phycomyc es

Mortierell a

Conidiobo lus

circinelloides VKM F-1315 flavus VKM F1249 racemosus VKM F-541 miehei VKM F1365 stolonifer VKM F-34 stolonifer VKM F-401 poitrasii VKM F-1367

blakesleeanus VKM F-828 elongata VKM F-1614 geminifera VKM F-1631 humilis VKM F1652 isobellina VKM F-525 verticellata VKM F-1612 thromboides VKM F-2529

++

-

-

++ n.d.

-

-

-

-

++

-

- ++ (3-keto-DCA)

-

-

-

-

++

-

-

full degradation

++

-

- ++ (3-keto-DCA)

++

++

+

++ n.d.

++

-

-

full degradation

++

-

- ++ (3-keto-DCA)

++

-

- ++ (3-keto-DCA)

-

-

-

-

-

-

-

-

++

+

+

++ n.d.

++

-

- ++ (3-keto-DCA)

-

-

-

-

Basidiomycota Polyporale Fomitopsida Sporotrich pruinosum VKM F-1764 s ceae um a Ursocholic acid (UCA), cholic acid (CA), 3-keto-DCA, and 12-keto-LCA were identified by TLC using authentic standards; UCA and CA identity were confirmed by MS and 1H-NMR analysis; 15β-hydroxy-DCA and 15β-hydroxy-12-oxo-LCA were identified by MS, 1H- and 13 b

C-NMR analyses.

Product conversions: «-» - no product detected; «+» - less than 15 % ; «++» - over 20 %. For

selected reactions, conversions of DCA into UCA and CA were determined by HPLC analyses (see Experimental for details) c

n.d.: the structures of the products were not determined due either to their low amount or to

the concomitant formation of several products.

28

Table 2. Identification of metabolites X1-X4 formed by selected fungal strains from DCAa Strain

Metabolite

Acremonium rutilum VKM F-2853

X1

Yield,%

M.w.

(molar)

[M+]

45-50

408

Structure

15β-hydroxyDCA

Neurospora crassa VKM F-875

X1

20-25

408

15β-hydroxy-

X2

35-40

392

DCA 15β-hydroxy12-oxoLCA

Backusella lamprospora VKM F-944

X3

40-45

408

n.d.b

Scopulariopsis coprophila VKM F-719

X4

35-40

408

n.d.b

a

Biotransformations were carried out for 72 h in the presence of 1 g/L initial DCA

concentration (for details see Experimental). b

n.d.: not determined (see text for details).

29

]Table 3. 1H and

13

C NMR spectroscopic data for compound X1 (3α,12α,15β-trihydroxy-5β-

cholanoic acid)

Proton 1a 1b 2a 2b 3 4a 4b 5 6a 6b 7a 7b 8 9 11a+11b 12 14 15 16a 16b 17 18 19 20 21 22a 22b 23a 23b

δH (ppm) 0.98 1.78 1.41 1.59 3.53 1.82 1.48 1.41 1.27 1.96 1.22 1.75 1.85 1.95 1.49 3.86 1.44 4.16 2.37 1.33 1.78 0.93 0.96 1.53 1.00 1.79 1.34 2.33 2.23

JHH (Hz) td (14.4, 3.6) m m m td (11.2, 4.79) m m m m m m m qd (11.4, 3.3) m m t (2.9) dd (11.2, 5.8) ddd (7.7, 5.8, 2.0) ddd (14.3, 9.0, 7.7) ddd (14.3, 10.2, 2.0) q (ca. 9) s s m d (6.6) m m m m a Measured at 500 MHz in [2H4]-methanol.

Carbon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

δC (ppm) 36.6 31.1 72.6 37.3 43.6 28.4 26.6 33.0 34.9 35.5 29.6 74.6 47.2 54.1 70.4 41.6 48.0 15.8 23.7 36.5 17.7 32.3 32.0 178.1

30

Table 4. 1H and 13C NMR spectroscopic data for compound X2 (3α,15β-dihydroxy-12-oxo-5βcholanoic acid)

Proton 1a 1b 2a 2b 3 4a 4b 5 6a 6b 7a 7b 8 9 11a 11b 14 15 16a 16b 17 18 19 20 21 22a 22b 23a 23b

a

δH (ppm) 1.07 1.69 1.28 1.65 3.53 1.68 1.50 1.49 1.33 2.02 1.20 1.87 2.27 1.84 2.57 1.99 1.13 4.18 2.49 1.44 1.97 1.29 1.07 1.42 0.86 1.84 1.35 2.24 2.08

JHH (Hz) td (14.3, 3.3) m m m m (br) m m m m m m m m m dd (13.1, 12.3) dd (12.3, 4.7) ddd (11.4, 5.4) ddd (7.4, 4.5, 1.8) ddd (14.7, 9.6, 7.4) ddd (14.6, 9.9, 1.8) q (ca. 9.5) s s m d (6.6) m m ddd (13.8, 10.4, 4.9) ddd (13.8, 10.1, 6.1)

Carbon 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

δC (ppm) 36.5 31.0 72.1 37.2 43.1 28.2 26.5 32.9 45.5 36.7 39.0 217.8 58.8 64.6 71.1 41.9 48.1 14.5 23.3 37.3 19.4 33.6 36.5 183.4

Measured at 500 MHz in [2H3]-methanol.

31

Legends to Figures

Fig. 1. Chemical structure of the most abundant primary (a) and secondary (b) bile acids.

Fig. 2. DCA transformation by Curvularia lunata VKM F-644. TLC chromatogram: 1, broth sample (72 h; ~10 µg of steroids in a spot); 2, standard sample of 12-keto-LCA (3 µg) . Fig. 3. Metabolites obtained from DCA by selected fungal strains: X1, Acremonium rutilum VKM F-2853; X2, Neurospora crassa VKM F-875; X3, Backusella lamprospora VKM F-944; X4, Scopulariopsis coprophila VKM F-2719. TLC chromatogram, 72 h; ~30γ of steroids in each spot. Fig. 4. Chemical structure and numbering of compounds X1 (a) and X2 (b).

Fig. 5. The structures of DCA and products formed in selected biotransformation reactions by filamentous fungi. 1, 7β-hydroxylation; 2, 7α-hydroxylation; 3, 3α-OH-dehydrogenation; 4, 12α-OH-dehydrogenation; 5, 15β-hydroxylation; 6, 12α-OH-dehydrogenation.

32

HIGHLIGHTS:

Broad biocatalytic potency of filamentous fungi was revealed toward deoxycholic acid Strains with 7α/β-, 15β-hydroxylase and 3α/12α-HSDH activities were discovered Fusarium merismoides VKM F-2310 produced UCA with highest yield Acremonium rutilum and Neurospora crassa strains exhibited high 15β-hydroxylase activity

33