Biotransformation of dehydroepiandrosterone with Macrophomina phaseolina and β-glucuronidase inhibitory activity of transformed products

Abstract

The biotransformation of dehydroepiandrosterone (1) with Macrophomina phaseolina was investigated. A total of eight metabolites were obtained which were characterized as androstane-3,17-dione (2), androst-4-ene-3,17-dione (3), androst-4-ene-17β-ol-3-one (4), androst-4,6-diene-17β-ol-3-one (5), androst-5-ene-3β,17β-diol (6), androst-4-ene-3β-ol-6,17-dione (7), androst-4-ene-3β,7β,17β−triol (8), and androst-5-ene-3β,7α,17β-triol (9). All the transformed products were screened for enzyme inhibition, among which four were found to inhibit the β-glucuronidase enzyme, while none inhibited the α-chymotrypsin enzyme.

Keywords::

• Dehydroepiandrosterone
• microbial transformation
• Macrophomina phaseolina
• β-glucuronidase inhibition
• α-chymotrypsin inhibition

Introduction

Microbial transformation has tremendous applications in a variety of fields, ranging from organic synthesis to in vivo metabolic studies of drugs¹. Several processes have been developed for the commercial production of steroidal drugs by microbial transformation^2,3.

Dehydroepiandrosterone (1) (DHEA) is an essential hormone of human body. It is produced in the adrenal glands. DHEA was the first steroid of its type to be isolated from the urine of a normal man, and identified as the most abundant steroid in human body⁴. DHEA and its metabolites have been found to strengthen the immune system in animals^5–8. These metabolites are also synthesized de novo in the brain, and can also be regarded as neuroactive steroids⁹. Microbial transformation of DHEA (1) has been previously studied with Rhizopus stolonifer¹⁰, Mucor piriformis¹¹, Fusarium moniliforme¹², Botryodiplodia malorum¹³, Colletotrichum lini¹³, and Fusarium culmorum¹⁴, Penicillium citreo-viride¹⁵, Whetzelinia sclerotiorum¹⁶, Phanerochaete chrysosporium¹⁶, Mucor plumbeus¹⁶, Rhizomucor tauricus¹⁷, Chaetomium sp¹⁸., Aspergillus tamarii¹⁹, Penicillium lilacinum²⁰, Penicillium griseopurpureum²¹, and Penicillium glabrum²¹, as well as by cell culture of plant, Anisodus tanguticus²². Apart from this, the metabolism of DHEA has also been studied in specific regions of the aging brain from Alzheimer’s and non-demented patients²³, human liver S9 fractions²⁴, and rat liver²⁵, and hippocampal cells²⁶, as well as rodent brain cell lines²⁷. In vivo metabolism of DHEA has also been studied in men²⁸.

Both enzymes, chosen for the study are clinically important. β-Glucuronidase is an exoglycosidase enzyme that catalyzes the cleavage of glucuronosyl-O-bonds. It belongs to the class hydrolases. Glucuronidation is a biological process to eliminate different xenobiotics and endogenous substances which are not water soluble. β-Glucuronidase has the opposite role, as it catalyzes the hydrolysis of these glucuronides. This hydrolysis may generate free toxins. This process thus increases the endogenous exposure of organs to carcinnogens. Inhibition of the β-glucuronidase enzyme reduces the tumor induction in colon and rectum. In animal model studies, the use of a beta glucuronidase inhibitors with conjugate carcinogen significantly reduces the number of tumors formed in rat colon²⁹.

α-Chymotrypsin enzyme causes cystic fibrosis by catalyzing the activation of the epithelial sodium channel (EnaC³⁰). It also causes the on-set of inflammatory arthritis by catalyzing the cleavage of interleukin-1β (IL-1β) precursor into IL-1β³¹. α-Chymotrypsin also helps the replication of HIV and HCV viruses, by catalyzing the cleavage of their protein precursors³². α-Chymotrypsin, a protease, is also responsible of digestion of protein content in our food. Thus selective inhibition of β-glucuronidase by particular compounds does not interfere in the normal digestion of protein present in food.

Thus the inhibition of both of these enzymes is of clinical significance. During the current study, the substrate 1 was screened against a number of clinically important enzymes and the above two enzymes were selected based on the result of screening experiments.

In continuation of our work on biotransformation of steroidal drugs^33–36, we screened DHEA (1) against several enzymes, and it was found to be active against β-glucuronidase. We therefore carried out the biotransformation of 1 to produce its transformed products. Small scale screening showed that Macrophomina phaseolina has the capacity to transform 1 into polar metabolites. Large scale fermentation of 1 with M. phaseolina produced eight metabolites 2–9, which were characterized as known compounds by detailed spectral analyses.

Experimental

General

DHEA was purchased from Acros Organics (NJ, USA). Precoated plates (silica gel, Merck, PF₂₅₄) were utilized for thin layer chromatography-based purification. Column chromatography (CC) was performed by using silica gel (E. Merck, Germany). ¹H- and ¹³C-NMR spectra were recorded in deuterated methanol and chloroform on Bruker Avance-NMR spectrometers with solvent signal as the internal standard. The chemical shifts (δ values) are given in parts per million, and the coupling constants (J values) in Hertz. JEOL JMS-600H Mass spectrometer (Japan) was used for recording EI-MS; in m/z (rel. %).

Microorganisms and culture medium

Macrophomina phaseolina (KUCC 730) was obtained from Karachi University Culture Collection (KUCC), grown on sabouraud dextrose-agar at 25°C, and stored at 4°C. The medium for M. phaseolina was prepared by dissolving glycerol (40.0 mL), KH₂PO₄ (20.0 g), glucose (40.0 g), yeast extract (20.0 g), NaCl (20.0 g), and peptone (20.0 g) into distilled H₂O (4.0 L).

Fermentation of DHEA (1) and purification of transformed products

Biotransformation was carried out by a two-stage fermentation protocol. In the first stage, the medium was equally distributed in 40 Erlenmeyer flasks (capacity 250 mL) 100 mL each, plugged with cotton swab, and autoclaved at 121°C. Spores of M. phaseolina were transferred from already grown slants to 10 flasks, and kept on a shaker at 25°C. The second stage started when enough growth was achieved in the 10 seed flasks. The mycelia were transferred to the remaining flasks, and kept on shaker for 2 days. A positive control was left without inoculation. After enough fungal growth was achieved in all the flasks, a solution of DHEA (800 mg) in 20 mL acetone (20 mg/0.5 mL each flask), was evenly distributed among all the flasks (including the positive control), except one (negative control). The flasks were again left on shaker for 6 days. The biomass was then filtered and washed thoroughly with dichloromethane (DCM). The filtrate (medium) was extracted with DCM three times. The DCM solution was dried over Na₂SO₄ (anhydrous), and concentrated under reduced pressure. The resulting gum (2.3 g) was subjected to repeated CC to purify the metabolites. Spectroscopic techniques led to the identification of compounds 2–9 (Figure 1).

Figure 1.  Biotransformation of dehydroepiandrosterone (1)by Macrophomina phaseolina.

Spectral data

The ¹H- and ¹³C-NMR chemical shifts of compounds 2–9 are presented in Tables 1 and 2, respectively. Other data is given below:

Table 1.  ¹H-NMR data of compound 2–9 at 300, 400, 500 MHz; in CDCl₃ and CH₃OH; δ in ppm, J and W_1/2 in Hz

Position	Compounds
	1	2	3	4	5	6	7	8	9
1	1.05, 1.83	1.34, 1.98	2.01, 2.46	1.54, 2.03	2.41, 2.56	1.05, 1.80	1.79, 2.41	1.49, 1.62	1.14, 1.84
2	1.22, 1.44	2.28, 2.34	2.35, 2.42	2.30, 2.41	1.70, 1.98	1.44, 1.48	1.14, 2.27	1.83, 1.98	1.48, 1.77
3	3.51 m W1/2 = 24.6	—	—	—	—	3.44 m W1/2 = 25.7	4.34 br sW1/2 = 23.8	3.43 ddd J = 15.5, 4.0, 2.0	3.47 m
4	2.21, 2.30	2.08, 2.26	5.73 s	5.71 s	5.65 s	2.18, 2.25	6.18 s	5.24 s	2.25, 2.35
5	—	1.53	—	—	—	—	—	—	—
6	5.36 d J = 4.8	1.36, 1.41	2.29, 2.40	2.26, 2.39	6.08 s	5.29 d J = 5.4	—	2.22, 2.28	5.55 d J = 4.4
7	1.66, 2.01	0.97, 1.79	1.25, 1.85	0.98, 1.85	6.08 s	1.77, 1.95	1.75, 2.04	3.71 d J = 8.0	3.75 t J = 3.6
8	2.14	1.58	1.72	1.72	2.23 t J = 10.8	1.45	2.37	1.47	1.49
9	0.99	0.77	0.97	0.91	1.19	0.87	0.96	1.03	1.30
10	—	—	—	—	—	—	—	—	—
11	1.42, 1.65	1.37, 1.67	1.45, 1.67	1.38, 1.57	1.36, 1.60	1.42, 1.55	1.41, 1.67	1.61, 1.91	1.52, 1.62
12	1.65, 1.84	1.24, 1.83	1.11, 1.97	1.06, 1.83	1.14, 1.87	1.03, 1.78	1.27, 1.86	1.11, 1.94	1.05, 1.81
13	—	—	—	—	—	—	—	—	—
14	1.23	1.27	1.28	0.97	1.17	0.90	1.33	1.06	1.43
15	1.46, 1.93	1.48, 1.92	1.55, 1.95	1.25, 1.62	1.48, 1.79	1.21, 1.53	1.57, 1.98	0.96, 1.78	1.25, 1.75
16	2.04, 2.42	2.06, 2.41	1.68, 2.42	1.44, 2.07	1.46, 2.13	1.40, 1.98	2.11, 2.47	1.45, 1.95	1.50, 2.00
17	—	—	—	3.63 s	3.68 ddJ = 6.4, 7.6	3.56 t J = 8.4	—	3.55 tJ = 8.5	3.60 tJ = 8.4
18	0.87	0.87 s	0.89 s	0.77 s	0.82 s	0.69	0.90 s	0.75 s	0.74 s
19	1.01	1.02 s	1.19 s	1.17 s	1.10 s	0.96	1.19 s	1.07 s	1.00 s

Table 2.  ¹³C-NMR data of compounds 1–9, σ in ppm.

C	Compounds
	1	2	3	4	5	6	7	8	9
1	37.2	38.4	35.7^a	35.7^a	33.9^a	37.2	33.7^a	21.9	38.1
2	31.5^a	38.1	33.9	33.9	33.9^a	31.5	40.1	30.8	32.1
3	71.6	211.8	199.3	199.5	199.5	71.4	68.4	72.1	71.9
4	42.1	44.6	124.2	123.9	123.7	42.0	120.1	127.4	42.9
5	140.8	46.6	170.3	171.2	163.6	140.9	170.4	144.2	146.6
6	121.0	28.6	32.6	32.8	128.0	121.3	199.1	42.6	124.8
7	31.3^a	30.5	31.3	31.5	140.3	31.3	36.3	74.0	65.4
8	30.7	34.9	35.1	35.7^a	37.7	31.9	33.8^a	41.2	39.1
9	50.0	53.8	53.8	53.9	50.8	50.2	53.8	50.1	43.6^a
10	36.1	35.8^a	38.6	38.7	36.1	37.2	39.1	37.7^a	38.5
11	20.0	20.7	20.3	20.6	20.3	20.6	20.3	32.3	21.5
12	31.2^a	31.5	30.8	36.4	36.3	36.5	31.2	37.8^a	37.5
13	47.0	47.8	50.8	42.8	43.8	42.3	47.5	44.3	43.5^a
14	51.6	51.2	49.7	50.5	48.3	51.3	50.6	52.3	45.3
15	21.5	21.8	21.7	23.3	22.9	23.4	21.8	38.3	24.2
16	35.5	35.8^a	35.8^a	30.5	30.4	30.1	35.7	26.6	30.6
17	220.8	221.2	220.4	81.6	81.3	81.6	219.9	82.3	82.5
18	13.3	13.8	13.7	11.0	10.9	10.9	13.7	11.5	11.4
19	19.5	11.5	17.4	17.4	16.3	19.4	18.3	19.5	18.7

Note: ^aInterchangeable δ values.

CDCl₃ at 75 MHz (2, 8).

CDCl₃ at 100 MHz (4, 7).

CDCl₃ at 150 MHz (5, 6).

CD₃OD at 125 MHz (3).

CD₃OD at 150 MHz (9).

Androstane-3,17-dione (2). Colorless crystalline solid. 129–130°C (lit. 130–132°C³⁷). ¹H-NMR (CDCl₃, 400 MHz): Table 1, ¹³C-NMR (CDCl₃, 150 MHz): Table 2. HREI-MS: m/z (rel. int. %) 288.2084 (54, M⁺, C₁₉H₂₈O₂, Calcd 288.2089), 244 (22), 217 (29), 199 (4), 161 (9), 147 (13), 124 (36), 79 (76), 55 (100).

Androst-4-ene-3,17-dione (3). Colorless crystalline solid. M. P. 168–169°C (lit. 170–171°C³⁸). ¹H-NMR (CDCl₃, 400 MHz): Table 1, ¹³C-NMR (CDCl₃, 75 MHz): Table 2. HREI-MS: m/z (rel. int. %) 286.1941 (70, M⁺, C₁₉H₂₆O₂, Calcd 286.1967), 244 (51), 201 (22), 148 (50), 124 (100), 107 (61), 91 (73), 55 (90).

Androst-4-ene-17β-ol-3-one (4). Colorless crystalline solid. M. P. 148–150°C (lit. 150–151°C³⁸). ¹H-NMR (CDCl₃, 400 MHz): Table 1, ¹³C-NMR (CDCl₃, 100 MHz): Table 2. HREI-MS: m/z (rel. int. %) 288.2084 (100, M⁺, C₁₉H₂₈O₂, Calcd 288.2089), 255 (5), 185 (4), 145 (15), 105 (30), 91 (43), 81 (51), 55 (49).

Androst-4,6-diene-17β-ol-3-one (5). Colorless crystalline solid. 197–199°C (lit. 199–200°C³⁹). ¹H-NMR (CDCl₃, 400 MHz): Table 1, ¹³C-NMR (CDCl₃, 75 MHz): Table 2. HREI-MS: m/z (rel. int. %) 286.1933 (56, M⁺, C₁₉H₂₆O₂, Calcd 286.1967), 253 (9), 227 (11), 161 (20), 136 (100), 107 (55), 91 (73), 55 (56).

Androst-5-ene-3β,17β−diol (6). Colorless crystalline solid. M. P. 176–177°C (lit. 179–182°C³⁹). ¹H-NMR (CDCl₃, 300 MHz): Table 1, ¹³C-NMR (CDCl₃, 75 MHz): Table 2. HREI-MS: m/z (rel. int. %) 290.8234 (61, M⁺, C₁₉H₃₀O₂, Calcd 290.8257), 272 (38), 239 (24), 205 (37), 179 (39), 145 (57), 119 (60), 105 (100), 91 (96), 55 (94).

Androst-4-ene-3β-ol-6,17-dione (7). Colorless crystalline solid. 190–192°C (lit. 193–196°C⁴⁰). ¹H-NMR (CDCl₃, 400 MHz): Table 1, ¹³C-NMR (CDCl₃, 100 MHz): Table 2. HREI-MS: m/z (rel. int. %) 302.1853 (52, M⁺, C₁₉H₂₆O₃, Calcd 302.1882), 287 (14), 260 (14), 233 (47), 170 (14), 109 (44), 81 (84), 55 (100).

Androst-4-ene-3β,7β,17β-triol (8). Colorless crystalline solid. M. P. 232–233°C (lit. 230.3–231.8°C³⁸). ¹H-NMR (CD₃OD, 500 MHz): Table 1, ¹³C-NMR (CD₃OD, 125 MHz): Table 2. HREI-MS: m/z (rel. int. %) 306.2186 (6, M⁺, C₁₉H₃₀O₃, Calcd 306.2195), 288 (100), 273 (4), 154 (22), 119 (23), 107 (35), 91 (45), 55 (67).

Androst-5-ene-3β,7α,17β-triol (9). Colorless crystalline solid. M. P. 256–257°C (lit. 258–259°C³⁹). ¹H-NMR (CD₃OD, 400 MHz): Table 1, ¹³C-NMR (CD₃OD, 150 MHz): Table 2. HREI-MS: m/z (rel. int. %) 306.2186 (30, M⁺, C₁₉H₃₀O₃, Calcd 306.2195), 246 (27), 203 (15), 167 (43), 150 (34), 124 (100), 105 (52), 77 (55), 65 (96), 55 (50).

β-Glucuronidase inhibition assay

β-Glucuronidase inhibitory activity was evaluated by a biochemical assay, based on the measurement of the absorbance of p-nitrophenol at 405 nm, produced from the enzymatic hydrolysis of chromogenic substrate^36,41. The reaction mixture, consisting of 5 µL of test compound solution, 185 µL of 0.1 M acetate buffer (pH 7.0) and 10 µL β-glucuronidase solution, was kept for 30 min at 37°C. The total reaction volume being 250 µL. 50 µL of p-nitrophenyl-β-D-glucuronide was added to each plate. A multiplate reader was used to read the absorbance at 405 nm. IC₅₀ Values were calculated by measuring the effect of varying concentrations of test compounds by using the EZ-Fit Enzyme Kinetic program. D-Saccharic acid-1, 4-lactone was used as a standard inhibitor of β-glucuronidase enzyme.

α-Chymotrypsin Inhibition assay

The α-chymotrypsin inhibition assay was performed in 50 mM Tris-HCl buffer pH 7.6 with 10 mM CaCl₂ as mentioned by Carnell et al. (1988) with the slight modification. The enzyme (12 Units/mL prepared in buffer) with the test compounds (0.5 mM) prepared in DMSO (final concentration 7%), was incubated for 25 min at 30°C. The reaction was started by the addition of the substrate N-succinyl-L-phenylalanine-p-nitroanilide (SPpNA; 0.4 mM prepared in the buffer). The change in absorbance by released p-nitroanilide was monitored at 410 nm. The positive control without test compound and the negative control without enzyme or with standard inhibitor were run in parallel. The IC₅₀ was determined as mentioned above. The % inhibition was based upon initial velocity and calculated as:

Table 3.  β-Glucuronidase inhibitory activity of compounds 1–9.

Compound	IC50 (µM) ± SEM^c
1	77.9 ± 1.95
2	373.5 ± 9.57
3	430 ± 7.13
4	NA^a
5	NA^a
6	NA^a
7	221.6 ± 12.5
8	191.4 ± 1.17
9	NA^a
D-Saccharic acid-1,4-lactone^b	48.4

^aNot active.

^bStandard used.

^cStandard error mean.

Results and discussion

Biotransformation of DHEA (1) (C₁₉H₂₈O₂) with M. phaseolina is being carried out for the first time. Compound 1 was incubated for 6 days and a total of eight metabolites 2–9 were isolated as the major metabolites.

The M⁺ of compound 2 (C₁₉H₂₈O₂) appeared at m/z 288.2084 (calcd 288.2089) in the HREI-MS, which is the same as compound 1. The compound showed no florescence under UV light. This suggested that the compound is devoid of an α, β-unsaturated carbonyl system. The ¹³C- and ¹H-NMR spectra indicated the absence of the double bond between C-5/C-6, and a C-3 -OH in 2, and the appearance of a new carbonyl signal at δ 211.8 for C-3 ketone. This observation was further supported by the HMBC of H-1 (δ 1.34, 1.98), H-2 (δ 2.28, 2.34), and H-4 (δ 2.08, 2.26) with the C-3 carbonyl (δ 211.8). The α-stereochemistry at C-5 was deduced from NOESY interactions between H-5 (δ 1.53) and H-9 (δ 0.77). Metabolite 2 was thus identified as 5α-androstane-3,17-dione. The compound has been previously prepared by the in vitro metabolism of 7α -3H-testosterone by human mandibular bone⁴².

Compound 3 had a molecular composition C₁₉H₂₆O₂, deduced from HREI-MS (M⁺ = m/z 286.1941, calcd 286.1967), 2 a.m.u. less than 1. A new olefinic signal at δ 5.73 (s, H-4) appeared in the ¹H-NMR spectrum, while the signal for methine proton, geminal to hydroxyl, was absent. An additional conjugated ketonic carbonyl signal at δ 199.3 appeared in the ¹³C-NMR. This suggested the oxidative isomerization of homoallylic alcohol at C-3. This inference was supported by the HMBC of H-4 (δ 5.73), with C-2 (δ 33.9), C-6 (δ 32.6), and C-19 (δ 17.4). This metabolite was thus characterized as androst-4-en-3,17-dione (3). Androst-4-en-3,17-dione was previously obtained by the metabolism of DHEA by rat testicular homogenates⁴³ and microbial transformation with Penicillium glabrum²¹.

Metabolite 4 had a molecular formula C₁₉H₂₈O₂ as inferrred from the HREI-MS which showed an M⁺ at m/z 288.2084 (calcd 288.2089). The ¹H-NMR of 4 showed an olefinic proton singlet at δ 5.71 (s, H-4). The chemical shift was similar to that in compound 3. An additional hydroxyl-bearing methine singlet was also appeared at δ 3.63, assigned to C-17 proton. This assignment was further confirmed by HMBC correlations of H-18 (δ 0.77), H-14 (δ 0.97), and H-16 (δ 1.44, 2.07) with C-17 (δ 81.6). H-17 (δ 3.63) showed NOESY correlation with H-14 (δ 0.97), indicating its α-orientation. The ¹³C-NMR of 4 lacked a ketonic carbonyl at C-17, but showed another carbonyl carbon peak at δ 199.5, characteristic of an α, β-unsaturated carbonyl group. The compound was therefore identified as testosterone, an important human body hormone. Testosterone has been already reported from various sources such as microbial transformation of DHEA by Rhizopus stolonifer¹⁰, in vitro rat liver²⁵, and rat hippocampal cell cultures²⁷, and in human brain cells²³.

The molecular composition of metabolite 5 was found to be C₁₉H₂₆O₂ (M⁺ m/z 286.1933, calcd 286.1967) from the HREI-MS analysis. Three olefinic proton signals appeared at δ 5.65 (1H, s, H-4), 6.08 (1H, s, H-6), and 6.08 (1H, s, H-7). A hydroxyl-bearing methine proton also appeared at δ 3.68 (q, J = 7.6 Hz, H-17). The ¹³C-NMR spectrum showed a ketone carbonyl signal at δ 199.5 (C-3). A proton resonated at δ 5.65 showed HMBC correlations with C-2 (δ 33.9), C-10 (δ 36.1), and C-6 (δ 128.0), and thus assigned to H-4. A two-proton signal at δ 6.08 showed HMBC correlations with C-4 (δ 123.7), C-6 (δ 128.0), C-8 (δ 37.7), C-9 (δ 50.8), and C-14 (δ 48.3). Therefore this signal was assigned to C-6 and C-7 protons. The compound was thus characterized as androst-4,6-diene-17β-ol-3-one. Compound 5 has been earlier synthesized from DHEA in two steps⁴⁴.

Metabolite 6 had a molecular composition C₁₉H₃₀O₂ (HREI-MS, M⁺ m/z 290.8234, calcd 290.8257). The ¹H-NMR spectrum showed an additional peak for hydroxyl-bearing methine proton at δ 3.56 (t, J = 8.4 Hz, H-17). The olefinic proton signal at δ 5.29 (d, J = 5.4 Hz) was assigned to C-6 proton due to its HMBC correlations with C-4 (δ 42.0), C-8 (δ 31.0), and C-10 (δ 37.2). The ¹³C-NMR spectrum exhibited no peak for ketone carbonyl, which indicated the reduction of C-17 carbonyl into an−OH. The H-17 (δ 3.56) showed NOESY correlations with H-14 (δ 0.90), indicating its α-orientation. The compound 6 was identified as androst-5-ene-3β,17β-diol, an endocrine regulator of the immune response. Compound 6 has been previously reported as a product of microbial transformation of DHEA by Mucor piriformis¹¹.

Metabolite 7 had a composition C₁₉H₂₆O₃ deduced from the HREI-MS (M⁺ m/z 302.1853, calcd 302.1882). A close examination of ¹H-NMR spectrum of 7 indicated an OH-bearing methine proton signal at δ 4.34 (br.s., W_1/2 = 23.8 Hz, H-3). It showed COSY cross peaks with the downfield C-4 olefinic proton (δ 6.18). This indicated an α, β-unsaturated system in the vicinity. The ¹³C-NMR spectrum showed a conjugated ketone carbonyl at δ 199.1. This carbonyl was positioned at C-6, based on its HMBC with H-7 (δ 1.75, 2.04) and H-8 (δ 2.37). The C-4 olefinic proton showed COSY interaction with H-3 (δ 4.34) and the HMBC correlations with C-5 (δ 170.4), C-6 (δ 199.1), and C-10 (δ 39.1). H-3 (δ 4.34) showed NOESY correlation with α-oriented H-1 (δ 1.79), which was in turn NOE correlated with α-oriented H-9 (δ 0.96), indicating an α-orientation of H-3. Compound 7 was characterized as androst-4-ene-3β-ol-6,17-dione.

The molecular composition of metabolite 8 as deduced from HREI-MS (M⁺ m/z 306.2186, calcd 306.2195) was C₁₉H₃₀O₃, showing an addition of 18 a.m.u. to substrate 1. This may be due to the hydroxylation at a methylene carbon along with the reduction of the C-17 carbonyl into an -OH. This inference was further supported from three hydroxyl-bearing methine proton signals at δ 3.42 (m, W_1/2 = 20.8 Hz, H-3), 3.71 (d, J = 8.0 Hz, H-7), and 3.55 (t, J = 8.5 Hz, H-17), and the disappearance of the ketonic carbonyl peak in the ¹³C-NMR spectrum. The double bond had also been shifted to C-4/C-5 (δ 5.24, s, H-4). The ¹³C-NMR also displayed two extra hydroxyl-bearing methine carbon peaks at δ 74.0, and 82.3. Their positions were deduced from the key HMBC and COSY correlations. The stereochemistry at C-7 was deduced by NOESY interactions between H-7 and H-9 and H-14, as α. The orientation of H-17 was also deduced to be α, based on the fact that it showed a NOESY correlation with α-oriented H-14. The compound was identified as androst-4-ene-3β,7β,17β−triol.

Metabolite 9 had a composition C₁₉H₃₀O₃ (HREI-MS, M⁺ m/z 306.2186, calcd 306.2195). The ¹H-NMR spectrum of 9 possessed three hydroxyl-bearing methine peaks at δ 3.47 (m, H-3, W_1/2 = 20.5 Hz), 3.75 (t, J = 3.6 Hz, H-7), and 3.60 (t, J = 8.4 Hz, H-17). An olefinic proton at δ 5.55 (d, J = 4.4 Hz, H-6) was also observed. The olefinic proton (δ 5.55) showed HMBC correlations with the alcoholic methine carbon (δ 65.4), C-4 (δ 42.9), and C-10 (δ 38.5). The hydroxyl proton (δ 3.75) showed COSY correlation with the C-6 olefinic proton (δ 5.55), and HMBC correlations with C-5 (δ 146.6), C-6 (124.8), C-8 (δ 39.1), and C-9 (δ 43.6). It indicated that C-7 had been hydroxylated, while the C-17 carbonyl, which was present in compound 1, was reduced to an -OH. This was further inferred from the absence of a carbonyl signal, and appearance of two new hydroxyl-bearing carbon signals (δ 65.5, and 82.5) in the ¹³C-NMR spectrum. The H-7 (δ 3.75) showed NOESY cross peaks with H-8 (δ 1.49), further confirming its β-orientation. Finally the compound was characterized as androst-5-en-3β,7α,17β-triol. Compound 9 is also reported as a product of microbial transformation of DHEA by Mucor piriformis¹¹.

The variety of transformations observed in the current study have also been reported with other fungi, such as Cunninghamella elegans, Rhizopus stolonifer, Fusarium lini,⁴⁵ etc. Usually different types of reactions are catalyzed by various enzymes in a whole cell system, leading to a range of different metabolites.

Biotransformed products 2–9 were evaluated for their inhibitory potential against β-glucuronidase and α-chymotrypsin enzymes.

β-Glucuronidase enzyme is expressed in most mammalian tissues and is found in the lysosome. It is a tetrameric glycoprotein, which catalyzes the hydrolytic removal of the β-glucuronosyl residues, present at the non-reducing end of oligosaccharides, thus degrading the glycosaminoglycans⁴⁶. Deficiency of this enzyme causes Sly syndrome, also known as mucopolysaccharidosis type VII⁴⁷. This important enzyme has been reported to be inhibited by several classes of compounds, including steroids⁴⁸, flavonoids⁴⁹, and terpenes⁵⁰. Compounds 7 and 8 significantly inhibited the β-glucuronidase enzyme. The trihydroxy compound 8 showed the best activity among the transformed products, but was almost three times less active than the substrate.

Biological importance of α-chymotrypsin has been discussed earlier in the introduction. Based on its role in the cause of different disorders, the inhibition of this particular enzyme is therapeutically important. Biotransformed products 2–9 were tested for their chymotrypsin inhibitory activity, but none of the metabolites showed any significant inhibition of the enzyme.

β-Glucuronidase activity was reported to be inhibited by various steroidal glucuronosides⁵¹. In the current study, the aglycons have been tested against the particular enzyme, showing moderate activities. Efforts were made to discover some inhibitors of α-chymotrypsin in the form of steroids having a variety of functionalities. Compounds 3 and 8 have the same flattened ring A like in 4 and 5 but both show moderate activity. The difference in other functional groups might have caused the loss of activity in compounds 4 and 5.

Conclusion

In conclusion, the biotransformation of DHEA (1) with M. phaseolina provided an efficient route to the production of several biologically important steroids. The study provides new route for the synthesis of compounds 2, 5, 7, and 8. Compounds 7 and 8 were found to be significant inhibitors of the β-glucuronidase enzyme. However the metabolites did not show any inhibition of the α-chymotrypsin enzyme.

Declaration of interest

Two of the authors, S. Zafar and Dr. N. T. Khan, acknowledge the Higher Education Commission, Pakistan, for providing financial support through the HEC indigenous Ph. D. scholarship program.