Hydroxylation of steroids by Fusarium oxysporum, Exophiala jeanselmei and Ceratocystis paradoxa

Article abstract of DOI:10.1016/j.steroids.2011.06.010

The potential of Fusarium oxysporum var. cubense UAMH 9013 to perform steroid biotransformations was reinvestigated using single phase and pulse feed conditions. The following natural steroids served as substrates: dehydroepiandrosterone (1), pregnenolone (2), testosterone (3), progesterone (4), cortisone (5), prednisone (6), estrone (7) and sarsasapogenin (8). The results showed the possible presence of C-7 and C-15 hydroxylase enzymes. This hypothesis was explored using three synthetic androstanes: androstane-3,17-dione (9), androsta-4,6-diene-3,17-dione (10) and 3α,5α-cycloandrost-6- en-17-one (11). These fermentations of non-natural steroids showed that C-7 hydroxylation was as a result of that position being allylic. The evidence also pointed towards the presence of a C-15 hydroxylase enzyme. The eleven steroids were also fed to Exophiala jeanselmei var. lecanii-corni UAMH 8783. The results showed that the fungus appears to have very active 5α and 14α-hydroxylase enzymes, and is also capable of carrying out allylic oxidations. Ceratocystis paradoxa UAMH 8784 was grown in the presence of the above-mentioned steroids. The results showed that monooxygenases which effect allylic hydroxylation and Baeyer-Villiger rearrangement were active. However, redox reactions predominated.

Full text of DOI:10.1016/j.steroids.2011.06.010

Steroids 76 (2011) 1317–1330
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Hydroxylation of steroids by Fusarium oxysporum, Exophiala jeanselmei
and Ceratocystis paradoxa
Patrice C. Peart a, Kayanne P. McCook a, Floyd A. Russell a, William F. Reynolds b, Paul B. Reese a,
⇑
a Department of Chemistry, University of the West Indies, Mona, Kingston 7, Jamaica
b Department of Chemistry, University of Toronto, Toronto, Ontario, Canada M5S 3H6
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 May 2011
Received in revised form 26 June 2011
Accepted 28 June 2011
Available online 7 July 2011
The potential of Fusarium oxysporum var. cubense UAMH 9013 to perform steroid biotransformations was
reinvestigated using single phase and pulse feed conditions. The following natural steroids served as sub-
strates: dehydroepiandrosterone (1), pregnenolone (2), testosterone (3), progesterone (4), cortisone (5),
prednisone (6), estrone (7) and sarsasapogenin (8). The results showed the possible presence of C-7
and C-15 hydroxylase enzymes. This hypothesis was explored using three synthetic androstanes: andro-
stane-3,17-dione (9), androsta-4,6-diene-3,17-dione (10) and 3a,5a-cycloandrost-6-en-17-one (11).
Keywords:
These fermentations of non-natural steroids showed that C-7 hydroxylation was as a result of that posi-
tion being allylic. The evidence also pointed towards the presence of a C-15 hydroxylase enzyme.
The eleven steroids were also fed to Exophiala jeanselmei var. lecanii-corni UAMH 8783. The results
Fusarium oxysporum
Exophiala jeanselmei
Ceratocystis paradoxa
showed that the fungus appears to have very active 5
of carrying out allylic oxidations.
a and 14a-hydroxylase enzymes, and is also capable
Ceratocystis paradoxa UAMH 8784 was grown in the presence of the above-mentioned steroids. The
results showed that monooxygenases which effect allylic hydroxylation and Baeyer–Villiger rearrange-
ment were active. However, redox reactions predominated.
Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction
Fungi belonging to the Exophiala genus are described as black
yeasts due their yeast-like growth and melanin production [5].
Fusaria are filamentous fungi which are widely distributed in
plants and soil and are commonly found in tropical and sub-trop-
ical regions [1]. They are well known vascular wilt pathogens,
which exhibit a high degree of host-specialization. A variety of
plants are affected, including bananas, cotton, hops, peas, carna-
tions and tomatoes [2]. Fusarium oxysporum var. cubense UAMH
9013, isolated from the pseudostem of Musa sp., is the causative
agent of the Panama disease of banana crops [3]. There is only
one reported use of this particular subspecies of F. oxysporum for
biotransformation. In the previous study the double phase pulse
feeding protocol was used. Steroids were fed to the fungus at 60,
108, 132 and 156 h after inoculation [4]. It is now known that by
feeding the substrate during the log phase of growth it is possible
to induce the enzymes necessary to metabolize the xenobiote. This
could lead to improved yields; furthermore, new metabolites are
sometimes encountered. Hence, for this investigation feeding of
1–11 was done at 24, 36, 48 and 60 h after inoculation and the sin-
gle phase method was employed.
The strain used in our studies was isolated as an endophyte from
Zingiber officinale plantlets growing in tissue culture. Exophiala
jeanselmei has significant bioremediation potential due to its capa-
bility to degrade volatile organic compounds [6]. There is one pre-
vious report in which the fungus was utilized in steroid
transformation. In that study the double phase-single feed method
was used and this yielded mainly products of redox chemistry and
side chain degradation [7]. The potential for E. jeanselmei var. leca-
nii-corni UAMH 8783 to perform hydroxylations on 1–11 was ex-
plored using the single-phase pulse feed procedure – a feeding
protocol which is noted for giving products of hydroxylation (vide
supra).
Ceratocystis paradoxa, also known as Ceratosomella paradoxa,
Ophiostoma paradoxa, Thielaviopsis paradoxa and Chalara paradoxa,
has been found on every continent and is known to attack crops
such as sugarcane, pineapple, banana, coffee, and cocoa among
others [8]. The biotransformation potential of C. paradoxa UAMH
8784 was previously examined by Wilson et al. using steroid sub-
strates. The results showed that only redox products were formed
[4]. The biotransformation potential of the fungus was reinvesti-
gated, using the previously mentioned steroid substrates, with a
modification to the feeding protocol in an effort to induce cyto-
chrome P450 hydroxylating enzymes.
⇑
Corresponding author. Tel.: +1 876 927 1910; fax: +1 876 977 1835.
E-mail address: paul.reese@uwimona.edu.jm (P.B. Reese).
0039-128X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2011.06.010

1318
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
or by spraying with ammonium molybdate–sulfuric acid or meth-
2. Experimental
2.1. General procedure
anol–sulfuric acid reagents, followed by heating for colour devel-
opment. Petrol refers to the petroleum fraction boiling between
60 and 80°. Steroids 1–7 were obtained from Steraloids Inc. (Wil-
ton, NH, USA) and sarsasapogenin (8) was purchased from Aldrich
Chemical Co. Inc. (Milwaukee, WI, USA). The non-natural andros-
tanes 9–11 were prepared in the laboratory. Fusarium oxysporum
var. cubense UAMH 9013, Exophiala jeanselmei var. lecanii-corni
UAMH 8783 and C. paradoxa UAMH 8784 were isolated in Jamaica
[4,7]. They were identified by Professor Phyllis Coates Beckford of
the University of the West Indies, Mona. The classifications were
confirmed by Professor Lynne Sigler, Curator, University of Alberta
Microfungus Collection and Herbarium (UAMH), Edmonton, Alber-
ta, Canada. Cultures have been deposited with the UAMH.
Melting points were determined using a Reichert Hot Stage
melting point apparatus and are uncorrected. Infrared data was ac-
quired using a Perkin–Elmer Fourier transform infrared spectro-
photometer 1000 using sodium chloride disks. NMR spectra were
obtained on Bruker Avance 200 and 500 instruments as well as a
Varian Unity 500 spectrometer. Compounds were dissolved in
CDCl3 with tetramethylsilane as an internal standard, unless other-
wise stated. C-13 NMR data for new compounds and those not fully
characterised previously are reported in Tables 1 and 2, respec-
tively. Optical rotations were obtained using a Perkin–Elmer 241
MC polarimeter and solutions were prepared in CH2Cl2, unless sta-
ted otherwise. Fungal mycelia were homogenized using an IKA Ul-
tra-Turrax T25 homogenizer. Purifications were carried out by
column chromatography using silica gel (230–400 mesh) as the
stationary phase. Additionally, preparative thin layer chromatogra-
phy (PLC) was performed on glass backed plates with silica gel
2.1.1. Synthesis of 5a-androstane-3,17-dione (9)
Dehydroepiandrosterone (1) was hydrogenated, catalyzed by 5%
palladium on carbon. Subsequent oxidation of the C-3 alcohol
using Jones reagent gave 9 [9–11].
(60 Å, 250 lm and 1000 lm thicknesses). Thin layer chromato-
graphic (TLC) analyses utilized polyester backed silica gel plates.
Both the TLC and PLC plates were visualized under ultraviolet light
2.1.2. Synthesis of androsta-4,6-diene-3,17-dione (10)
Testosterone (3) was subjected to oxidation using Jones reagent
to give androst-4-ene-3,17-dione (18). The latter was treated with
chloranil [12] in tert-butyl alcohol to yield 10 [13,14].
Table 1
13C NMR resonance values (d) for novel steroids.
2.1.3. Synthesis of 3a,5a-cycloandrost-6-en-17-one (11)
Dehydroepiandrosterone (1) was converted to its 3b-tosylate,
followed by buffered solvolysis in aqueous sodium acetate to give
34
33.8
35
33.8
36
71.4
39
31.4
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-11
C-12
C-13
C-14
C-15
C-16
C-17
C-18
C-19
33.8
199.3
124.3
162.7
128.9
137.6
36.9
50.5
36.1
19.6
31.1
48.2
45.8
30.2
71.1
218.0
14.0
16.4
33.8
199.6
123.6
163.4
127.8
140.5
37.2
50.3
36.0
19.8
31.2
51.0
55.2
69.9
46.3
215.4
15.2
16.3
42.9
197.0
123.5
159.5
129.3
138.8
36.9
43.8
40.9
19.4
31.1
48.3
48.6
21.4
35.6
219.6
13.6
17.7
36.9
71.4
41.8
140.7
120.5
31.0
34.4
48.9
36.6
21.9
38.9
83.3
46.7
20.1
28.8
171.8
19.8
19.3
6b-hydroxy-3a,5a-cycloandrostan-17-one. The latter was sub-
jected to dehydration using triphenylphosphine [15] in carbon tet-
rachloride to afford 11 [16,17], along with 3b-chloroandrost-5-en-
17-one.
2.2. Fusarium oxysporum var. cubense
This fungus was maintained on potato dextrose agar slants at
28°. Five two week old slants were used to inoculate twenty
500 mL Erlenmeyer flasks, each containing 125 mL of liquid culture
medium. The medium was prepared from glucose (40 g/L), yeast
extract (5 g/L), ammonium nitrate (5 g/L), potassium di-hydrogen
phosphate (2 g/L) and magnesium sulfate heptahydrate (1 g/L) [4].
The steroid (1.0 g) was dissolved in EtOH and distributed across
19 flasks. One flask was fed only EtOH and this served as the con-
Table 2
13C NMR resonance values (d) for compounds that were not previously fully characterized.
14
37.2
17
38.6
21
37.2
22
35.8
25
35.7
26
34.6
30
38.6
31
37.0
32
34.0
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-11
C-12
C-13
C-14
C-15
C-16
C-17
C-18
C-19
C-20
C-21
31.6
71.6
42.1
140.5
121.1
32.2
31.8
49.8
36.5
20.1
31.4
50.6
57.8
71.1
46.1
216.4
15.1
19.5
32.4
72.5
43.1
142.3
122.4
33.2
33.2
51.9
37.2
22.1
40.8
43.5
59.4
26.9
25.6
57.8
12.6
19.9
70.9
23.8
34.3
200.3
126.5
168.1
73.1
38.1
29.7
53.7
38.0
20.6
36.4
42.9
50.4
23.3
30.5
81.7
11.1
19.6
33.9
199.5
123.9
170.8
32.7
32.2
35.4
53.8
38.7
20.6
36.6
44.4
58.6
72.6
42.7
78.8
12.6
17.5
32.8
199.5
123.8
171.0
34.0
32.1
34.9
53.8
38.6
20.6
38.8
42.5
62.8
73.3
37.7
55.8
13.9
17.5
69.1
23.8
33.5
200.2
124.3
169.6
32.3
32.3
36.9
62.3
38.1
211.8
51.7
51.3
48.3
23.4
33.6
84.3
15.6
17.2
73.2
64.2
38.1
211.9
44.5
46.4
28.7
31.8
35.3
53.7
35.8
20.6
31.5
50.8
57.7
70.2
46.2
216.7
15.4
11.5
31.8
71.1
38.0
44.5
28.4
35.4
32.3
54.1
35.7
20.4
31.4
50.9
58.1
70.6
46.0
216.5
15.5
12.4
33.9
199.6
123.7
163.7
128.0
140.3
37.7
50.8
36.1
20.3
36.3
43.9
48.3
23.0
30.4
81.3
11.0
16.3

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1319
trol. A solution of 10% of the total mass of the substrate was fed
24 h after inoculation. Then 20, 30 and 40% of the substrate were
fed at 36, 48 and 60 h after inoculation, respectively. The fermen-
tation was allowed to proceed for an additional 5 d. The mycelia
were filtered from the broth. The latter was extracted with ethyl
acetate (2 Â 1000 mL). The mycelia were placed in ethyl acetate
(500 mL) and homogenized, warmed, and then filtered. The organic
extracts were dried using sodium sulfate, and were filtered. The
solvent was removed in vacuo.
The second metabolite, isolated in fractions eluted with ace-
tone/dichloromethane (1:9 v/v), was 3b,20-dihydroxypregn-5-ene
(17) (46.3 mg), Rf = 0.49, acetone/dichloromethane (1:4 v/v), which
resisted crystallization, [
a
]
À47.3° (c 1.54, MeOH); IR: mmax 3252,
D
2930, 1434, 1372, 1020 cmÀ1
;
HREIMS: m/z: 318.2559 M+
(C21H34O2 requires 318.2559); 1H NMR (CD3OD): d 0.78 (3H, s, H-
18), 1.04 (3H, s, H-19), 1.13 (3H, d, J = 6 Hz, H-21), 3.31 (1H, m,
w/2 = 3 Hz, H-3a), 3.63 (1H, m, w/2 = 8 Hz, H-20), 5.35 (1H, d,
J = 4 Hz, H-6).
2.2.1. Biotransformation of 3b-hydroxyandrost-5-en-17-one (1)
The mycelial and broth extracts were combined (1.2 g) and
chromatographed. Elution with acetone/dichloromethane (1:49 v/
v) allowed for the recovery of 1 (435.9 mg). Further elution yielded
the first metabolite, 3b,17b-dihydroxyandrost-5-ene (12)
(111.9 mg), Rf = 0.37, acetone/dichloromethane (1:19 v/v), which
2.2.3. Biotransformation of 17b-hydroxyandrost-4-en-3-one (3)
Extraction of the mycelia afforded an off-white crystalline solid
(673.5 mg), while the broth gave an orange–brown gum
(341.4 mg). TLC analysis of the extract from the mycelia showed
the presence of 3 only, and as such only the broth extract was puri-
fied. The first metabolite, eluted with acetone/dichloromethane
(1:49 v/v), was androst-4-ene-3,17-dione (18) (26.5 mg),
Rf = 0.70, acetone/dichloromethane (1:19 v/v), which crystallized
crystallized from acetone as plates, m.p. 188–189°, [
0.45) (lit. [28] m.p. 180–183°, [ À56°); IR: mmax 3202, 2933,
1704, 1447, 1229 cmÀ1 1H NMR: d 0.79 (3H, s, H-18), 1.05 (3H,
a]
À27.4° (c
D
a
]
D
;
from acetone as needles, m.p. 173–174°, [
[22] m.p. 176–178°, [ +182.5°); IR: mmax 2937, 1733, 1436,
1170 cmÀ1 1H NMR: d 0.91 (3H, s, H-18), 1.20 (3H, s, H-19), 5.74
(1H, s, H-4).
The second transformed compound, purified by PLC using ace-
tone/dichloromethane (3:97 v/v), was 17b-hydroxy-5 -andro-
a] +147.7° (c 0.75) (lit.
D
s, H-19), 3.55 (1H, m, w/2 = 15 Hz, H-3a), 3.67 (1H, t, J = 7.5 Hz,
a]
D
H-17), 5.37 (1H, d, J = 5 Hz, H-6).
;
Elution with acetone/dichloromethane (1:19 v/v) yielded a sec-
ond metabolite, 3b-hydroxyandrost-5-ene-7,17-dione (13)
(24.3 mg), Rf = 0.32, acetone/dichloromethane (1:19 v/v), which
resisted crystallization, [
À38.9° (c 0.19) (lit. [19] m.p. 229–
232°); IR: mmax 3418, 2935, 1732, 1668, 1455 cmÀ1
0.89 (3H, s, H-18), 1.23 (3H, s, H-19), 3.69 (1H, m, w/2 = 15.5 Hz,
H-3 ), 5.75 (1H, s, H-6).
Further percolation with acetone/dichloromethane (1:19 v/v)
yielded the third transformed compound, 3b,15 -dihydroxyand-
a
a
]
D
stan-3-one (19) (2 mg), Rf = 0.54, acetone/dichloromethane (1:19
v/v), which crystallized from acetone as plates, m.p. 161–163°,
;
1H NMR: d
[a] +15.4° (c 0.06) (lit. [23] m.p. 177–182°, [a] +33.5°); IR: mmax
D D
a
3426, 1733, 1436, 1170 cmÀ1
(3H, s, H-19), 3.67 (1H, t, J = 7.5 Hz, H-17
Elution with acetone/dichloromethane (1:9 v/v) allowed for the
recovery of a third metabolite, 15 -hydroxyandrost-4-ene-3,17-
;
1H NMR: d 0.78 (3H, s, H-18), 1.05
a).
a
rost-5-en-17-one (14) (3 mg), Rf = 0.08, acetone/dichloromethane
a
(1:19 v/v), which crystallized from acetone as cubes, m.p. 195–
dione (20) (16.4 mg), Rf = 0.14, acetone/dichloromethane (1:19 v/
196°, [
a
]
+13.3° (c 0.08); IR: mmax 3385, 2930, 1734, 1655,
v), which crystallized from dichloromethane as plates, m.p. 178–
D
1458 cmÀ1
;
HREIMS: m/z: 304.2029, M+ (C19H28O3 requires
180°, [
a] +147.7° (c 0.44) (lit. [24] m.p. 200–201°, [a] +220°);
D
D
304.2038); 1H NMR: d 0.94 (3H, s, H-18), 1.08 (3H, s, H-19), 3.00
(1H, d, J = 5 Hz, H-16), 3.03 (1H, d, J = 5 Hz, H-16), 3.57 (1H, m,
w/2 = 15 Hz, H-3a), 4.44 (1H, q, J = 10 Hz, H-15a), 5.41 (1H, d,
IR: mmax 3418, 2946, 1732, 1651, 1451, 1230 cmÀ1 1H NMR: d
;
0.96 (3H, s, H-18), 1.22 (3H, s, H-19), 4.42 (1H, q, J = 8 Hz, H-
15b), 5.76 (1H, s, H-4).
J = 5 Hz, H-6).
Further elution with acetone/dichloromethane (1:9 v/v) yielded
Elution with acetone/dichloromethane (1:4 v/v) allowed for the
recovery of 3b,7b-dihydroxyandrost-5-en-17-one (15) (7.3 mg),
Rf = 0.07, acetone/dichloromethane (1:19 v/v), which resisted crys-
a fourth biotransformed compound, 6
en-17-one (21) (2 mg), Rf = 0.18, acetone/dichloromethane (1:9 v/
a,17b-dihydroxyandrost-4-
v), which resisted efforts of crystallization, [
a
]
+13.3° (c 0.08);
tallization, [
IR:
max 3390, 2933 1732, 1661, 1455 cmÀ1; 1H NMR: d 0.91 (3H, s,
H-18), 1.08 (3H, s, H-19), 3.57 (1H, m, w/2 = 12 Hz, H-3 ), 3.95 (1H,
m, w/2 = 10.8 Hz, H-7 ), 5.32 (1H, t, J = 5 Hz, H-6).
The final metabolite, obtained in fractions eluted with acetone/
dichloromethane (1:4 v/v), was 3b,7 -dihydroxyandrost-5-en-17-
one (16) (315.9 mg), Rf = 0.09, acetone/dichloromethane (1:19 v/
v), which crystallized from acetone as needles, m.p. 98–100°, [
À60° (c 0.82) (lit. [18] m.p. 182–183°, [ À72°); IR: mmax 3395,
2935, 1732, 1373 cmÀ1 1H NMR: d 0.86 (3H, s, H-18), 1.00 (3H,
a]
D +51.4° (c 0.18) (lit. [20] m.p. 213–215°, [
a]
D +63°);
IR: mmax 3402, 2941, 1732, 1661, 1231 cmÀD1
; HREIMS: m/z:
m
304.2043, M+ (C19H28O3 requires 304.2038); 1H NMR: d 0.85 (3H,
a
s, H-18), 1.40 (3H, s, H-19), 3.67 (1H, t, J = 8.5 Hz, H-17
(1H, t, J = 3 Hz, H-6b), 5.83 (1H, s, H-4).
a), 4.36
a
The final metabolite, also isolated using acetone/dichlorometh-
ane (1:9 v/v), was 15 ,17b-dihydroxyandrost-4-en-3-one (22)
(14 mg), Rf = 0.08, acetone/dichloromethane (1:9 v/v), which crys-
tallized as needles from acetone, m.p. 93–94°, [ D +23.4° (c 0.15);
IR: mmax 3391, 2925, 1661, 1651, 1455 cmÀ1
HREIMS: m/z:
304.2035, M+ (C19H28O3 requires 304.2038); 1H NMR: d 0.83 (3H,
a
a
a]
a]
D
a]
;
D
;
s, H-19), 3.58 (1H, m, w/2 = 15.8 Hz, H-3
a), 3.98 (1H, t, J = 5 Hz,
s, H-18), 1.21 (3H, s, H-19), 3.91 (1H, t, J = 9 Hz, H-17
a), 4.14 (1H,
H-7b), 5.65 (1H, dd, J = 2, 5.5 Hz, H-6).
dt, J = 3.5, 9.5 Hz, H-15b), 5.74 (1H, s, H-4).
2.2.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
The mycelial extract was a crystalline solid (775 mg) and the
broth gave a brown gum (172 mg). Both extracts were purified sep-
arately. Column chromatography of the mycelial extract using ace-
2.2.4. Biotransformation of pregn-4-ene-3,20-dione (4)
TLC analysis revealed that the extracts of the mycelia (817 mg)
contained mostly fed compound. This was recrystallized from
dichloromethane to yield progesterone (4) (581.4 mg). The
mother-liquor was mixed with the broth extract (354 mg) and
chromatographed. Elution with dichloromethane allowed for the
recovery of the first metabolite, 20-hydroxypregn-4-en-3-one
(23) (2.8 mg), Rf = 0.46, acetone/dichloromethane (1:19 v/v), which
tone/dichloromethane (1:49 v/v) afforded
2
(667.5 mg).
Purification of the broth extract using acetone/dichloromethane
(1:49 v/v) yielded the first product of transformation, 17b-
hydroxyandrost-4-en-3-one (3) (5 mg). Rf = 0.50, acetone/dichlo-
romethane (1:19 v/v), which resisted crystallization, [
(c 0.08) (lit. [21] m.p. 152–153°, [ +110.7°); IR: mmax 3920,
2932, 1687, 1055 cmÀ1 1H NMR: d 0.80 (3H, s, H-18), 1.20 (3H,
s, H-19), 3.66 (1H, t, J = 8.2 Hz, H-17 ), 5.74 (1H, s, H-4).
a
]
+102.5°
resisted crystallization, [
a
]
+13.3° (c 0.07) (lit. [25] m.p. 140–
D
D
a
]
141°); IR: mmax 3441, 2930, 1674, 1450, 1229 cmÀ1 1H NMR: d
;
D
;
0.74 (3H, s, H-18), 1.21 (6H, s, H-19, H-21), 3.75 (1H, q, J = 5 Hz,
H-20), 5.75 (1H, s, H-4).
a

1320
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
The second product of transformation, purified by PLC using
acetone/dichloromethane (1:49 v/v), was 17b-hydroxyandrost-4-
en-3-one (3) (3.5 mg).
(1H, d, J = 7.5 Hz, H-21), 3.77 (1H, d, J = 7.5 Hz, H-20), 5.72 (1H, s,
H-4).
The third transformed compound isolated, eluted with acetone/
dichloromethane (1:9 v/v), was 15a-hydroxypregn-4-ene-3,20-
2.2.6. Biotransformation of 17
3,11,20-trione (6)
a,21-dihydroxypregna-1,4-diene-
dione (24) (83.9 mg), Rf = 0.18, acetone/dichloromethane (1:19 v/
v), which crystallized as needles from dichloromethane, m.p.
TLC analysis revealed that both broth and mycelial extracts con-
tained significant amounts of the fed compound. Subsequently,
they were combined and subjected to recrystallization to give
prednisone (6) (626.7 mg). The mother-liquors were concentrated
under reduced pressure to give an orange–brown gum (600 mg),
which was purified by column chromatography. The lone metabo-
lite, further purified by PLC using acetone/dichloromethane (2:3 v/
237-238°, [
3486, 2943, 1732, 1694, 1435, 1360 cmÀ1
H-18), 1.21 (3H, s, H-19), 2.13 (3H, s, H-21), 2.82 (1H, m, w/
2 = 11 Hz, H-17 ), 4.10 (1H, dt, J = 4, 9 Hz, H-15b), 5.73 (1H, s, H-4).
The final metabolite, purified via PLC using acetone/dichloro-
methane (1:19 v/v), was 15 ,20-dihydroxypregn-4-en-3-one (25)
(4 mg), Rf = 0.12, acetone/dichloromethane (3:17 v/v), which re-
a] +74.5° (c 0.77) (lit. [25] m.p. 226–232°); IR: mmax
D
;
1H NMR: d 0.71 (3H, s,
a
a
v), was 17
(20.6 mg), Rf = 0.08, acetone/dichloromethane (1:9 v/v), which
crystallized from acetone as plates, m.p. 98–101°, [ +62.3° (c
0.22) (lit. [4] [ +38°); IR: mmax 3391, 2937, 1699, 1661, 1615,
1242 cmÀ1
a, 20S,21-trihydroxypregna-1,4-diene-3,11-dione (27)
sisted crystallization, [
a
]
+22.9° (c 0.14); IR: mmax 3402, 2933,
a]
D
D
1659, 1450, 1230 cmÀ1; HREIMS: m/z: 332.2359 M+ (C21H32O3 re-
quires 332.2351); 1H NMR: d 0.76 (3H, s, H-18), 1.20 (3H, s, H-
19), 1.23 (3H, d, J = 6.5 Hz, H-21), 3.70 (1H, t, J = 6.5 Hz, H-20),
4.07 (1H, dt, J = 3.5, 9 Hz, H-15b), 5.74 (1H, s, H-4).
a]
D
;
HREIMS: m/z: 360.1933 M+ (C21H28O5 requires
360.1937); 1H NMR: d 0.83 (3H, s, H-18), 1.45 (3H, s, H-19), 3.77
(3H, bs, H-20, H-21), 6.09 (1H, bs, H-4), 6.20 (1H, dd, J = 2,
10.5 Hz, H-2), 7.70 (1H, d, J = 10.5 Hz, H-1).
2.2.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
The mycelial extract was a crystalline solid (635 mg), which by
TLC analysis was shown to contain only the fed substrate. The
extraction of the broth afforded a red–brown gum (189 mg), which
was subjected to column chromatography. Elution with acetone/
dichloromethane (1:99 v/v) allowed for the isolation of a lone
metabolite, 3,17b-dihydroxyestra-1,3,5(10)-triene (28) (56 mg),
Rf = 0.51, acetone/dichloromethane (1:4 v/v), which crystallized
from acetone as plates, m.p. 179–181°, [
(lit. [26] m.p. 178–179.5°, [ +70.5°); IR: mmax 3202, 2933, 1608,
1496, 1281, 1054 cmÀ1 1H NMR: d 0.78 (3H, s, H-18), 2.81 (2H, t,
J = 4.0 Hz, H-6), 3.74 (1H, t, J = 7.7 Hz, H-17 ), 6.56 (1H, m, w/
a] +64.6° (c 1.83, MeOH)
D
a]
D
;
a
2 = 7 Hz, H-4), 6.63 (1H, m, w/2 = 14 Hz, H-2), 7.15 (1H, d,
J = 14 Hz, H-1).
2.2.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
No metabolites were obtained in this fermentation.
2.2.9. Biotransformation of 5a-androstane-3,17-dione (9)
The broth and mycelial extracts were combined (1.32 g) and
subjected to column chromatography. Elution with acetone/
dichloromethane (1:49 v/v) allowed for the recovery of the fed
compound (665.4 mg). Further elution gave the first transformed
product, 17b-hydroxy-5
The second compound, eluted in acetone/dichloromethane
(1:49 v/v), was 3b-hydroxy-5 -androstan-17-one (29) (11.5 mg),
Rf = 0.51, acetone/dichloromethane (2:23 v/v), which crystallized
from dichloromethane as needles, m.p. 171–172° [ +90.5° (c
1.01) (lit. [9] m.p. 177–182°, [ +33.5°); IR: mmax 3470, 2930,
2361, 1731 cmÀ1 1H NMR: d 0.83 (3H, s, H-18), 0.86 (3H, s, H-
a-androstan-3-one (19) (10.5 mg).
a
a]
D
a]
D
;
19), 3.59 (1H, m, w/2 = 16 Hz, H-3).
Elution with acetone dichloromethane (1:19 v/v) allowed for the
recovery of the third product of transformation, 15a-hydroxy-5a-
2.2.5. Biotransformation of 17
a
,21-dihydroxypregn-4-ene-3,11,20-
androstane-3,17-dione (30) (52.9 mg), Rf = 0.35, acetone/dichloro-
trione (5)
methane (1:19 v/v), which crystallized from dichloromethane as
The broth and mycelial extracts were combined (864 mg) and
chromatographed. Elution with acetone/dichloromethane (1:9 v/
v) allowed for the recovery of the fed substrate (208 mg). Percola-
tion with acetone in dichloromethane (1:4 v/v) led to the isolation
needles, m.p. 159–161°, [a]D +103.3° (c 1.31); IR: mmax 3383, 2948,
1732, 1446, 1268, 1233, 1096 cmÀ1; HREIMS: m/z: 304.2040 M+
(C19H28O3 requires 304.2038); 1H NMR: d 0.92 (3H, s, H-18), 1.06
(3H, s, H-19), 2.10 (1H, d, J = 6.5 Hz, H-16), 2.96 (1H, dd, J = 8,
19 Hz, H-16), 4.39 (1H, td, J = 1.5, 6.5 Hz, H-15b).
of the lone transformed compound, 17a,20S,21-trihydroxypregn-
4-ene-3,11-dione (26) [18] (379.8 mg), Rf = 0.14, acetone/dichloro-
The final metabolite, also isolated using acetone/dichlorometh-
methane (1:9 v/v), which crystallized as needles from acetone, m.p.
ane (1:19 v/v), was 3b,15
(7.5 mg), Rf = 0.14, acetone/dichloromethane (1:19 v/v), which crys-
tallized from dichloromethane as needles, m.p. 69–72°, [ D +49.2°
(c 0.20); IR:
max 3307, 2926, 1732, 1651, 1455, 1373 cmÀ1; HREIMS:
a-dihydroxy-5a-androstan-17-one (31)
100–103°, [a] +104.1° (c 1.03); IR: mmax 3419, 2937, 1699, 1668,
D
1348, 1220 cmÀ1; HREIMS: m/z: 362.2099 M+ (C21H30O5 requires
362.2093); 1H NMR: d 0.80 (3H, s, H-18), 1.40 (3H, s, H-19), 3.72
a]
m

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1321
m/z: 306.2196 M+ (C19H30O3 requires 306.2195); 1H NMR: d 0.85
(3H, s, H-19), 0.90 (3H, s, H-18), 2.10 (1H, d, J = 8 Hz, H-16), 2.98
2.3.1. Biotransformation of 3b-hydroxyandrost-5-en-17-one (1)
Chromatography of the broth extract (718 mg) with ethyl ace-
tate/petrol (1:3 v/v) yielded the fed compound (31.3 mg), while
(1H, J = 8.6 Hz, H-16), 3.61 (1H, m, w/2 = 13.8 Hz, H-3
td, J = 1.5, 6.5 Hz, H-15b).
a), 4.38 (1H,
elution with ethyl acetate/petrol (65:35 v/v) gave 3b,7
a-dihydr-
oxyandrost-5-en-17-one (16) (39.9 mg).
Elution of the mycelial extract (291 mg) with ethyl acetate/petrol
(1:3 v/v) led to the isolation of untransformed substrate (1)
(32.6 mg). Further elution with ethyl acetate/petrol (3:7 v/v) gave
2.2.10. Biotransformation of androsta-4,6-diene-3,17-dione (10)
Androsta-4,6-diene-3,17-dione (10) (500 mg) was dissolved in
ethyl acetate (20 mL) and fed to the growing fungus. The mycelia
and broth extracts were combined (375 mg) and purified by chro-
matography. Elution with dichloromethane allowed for the recov-
ery of the fed compound (241 mg). The first transformed
compound, also isolated in fractions eluted with dichloromethane,
3
a
,17b-dihydroxy-5b-androstane (36) (27.9 mg), Rf = 0.29, ethyl
acetate/petrol (1:1 v/v), which crystallized from acetone as needles,
m.p. 171–173°, [ D +6.2° (c = 0.7, CH2Cl2) (lit. [28] m.p. 228–229°,
D +26°); IR:
max 3351, 2935, 2865 cmÀ1; 1H NMR: d 0.75 (3H, s,
a]
[a
]
m
H-18), 0.96 (3H, s, H-19), 3.66 (1H, t, J = 9 Hz, H-17
a
), 4.05 (1H, t,
was 5a-androstane-3,17-dione (9) (11.7 mg).
J = 5 Hz, H-3b).
17b-Hydroxyandrosta-4,6-dien-3-one (32) (4.2 mg), the second
metabolite, was purified by PLC using acetone/dichloromethane
(1:24 v/v), Rf = 0.32, acetone/dichloromethane (1:19 v/v). This
Elution with ethyl acetate/petrol (3:2 v/v) yielded 3
dihydroxyandrost-5-ene (37) (31.8 mg), Rf = 0.59, ethyl acetate/
petrol (9:1 v/v), which did not crystallize, [
À16.6° (c = 0.3,
CH2Cl2) (lit. [29] m.p. 196–199°, [ D À71° (CHCl3)); IR:
2927 cmÀ1 1H NMR: d 0.81 (3H, s, H-18) 1.05 (3H, s, H-19), 3.67
a
,17b-
a
]
D
compound resisted efforts of crystallization, [
a]
+35.3° (c 0.09);
a]
mmax 3340,
D
IR: mmax 3421, 2941, 1653, 1616, 1457, 1224 cmÀ1; HREIMS: m/z
(rel. int.): 287.2007 (100) M+ (C19H26O2 requires 287.2005); 1H
NMR: d 0.85 (3H, s, H-18), 1.13 (3H, s, H-19), 3.70 (1H, t,
;
(1H, t, J = 9 Hz, H-17), 4.05 (1H, t, J = 5 Hz, H-3), 5.32 (1H, s, H-6).
J = 8.5 Hz, H-17
6.11 (1H, d, J = 2 Hz, H-6).
Elution with acetone/dichloromethane (1:49 v/v) yielded a
third metabolite, 16b-hydroxyandrosta-4,6-diene-3,17-dione (33)
(7.8 mg), Rf = 0.14, acetone/dichloromethane (1:19 v/v), which re-
a
), 5.68 (1H, s, H-4), 6.10 (1H, d, J = 2 Hz, H-7),
2.3.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
The resulting extracts were combined (741.8 mg) and subjected
to chromatography. Elution with ethyl acetate/petrol (1:4 v/v)
yielded the fed compound (11.4 mg). Elution with ethyl acetate/
petrol (2:3 v/v) gave 3
a,14a-dihydroxy-5a-pregnan-20-one (38)
sisted crystallization, [a]D +5.1° (c 0.24); IR: mmax 3391, 2942, 1732,
(19 mg), Rf = 0.43, ethyl acetate/petrol (2:3 v/v), which crystallized
1652, 1455, 1228 cmÀ1; HREIMS: m/z (rel. int.): 301.1806 (100) M+
(C19H24O3 requires 301.1798); 1H NMR: d 1.06 (3H, s, H-18), 1.14
from methanol as plates, m.p. 208–209°, [
a]
D
+87.9° (c = 0.48,
MeOH); IR: mmax 3455, 2933, 1715 cmÀ1; HRMS (EI): m/z (rel. int.
%) 255 (95), 283 (37), 298 (100), 316 (19), 334 (3); Calc. For M+
(C21H34O3): 334.2508, Found: 334.2513; 1H NMR: d 0.74 (3H, s,
H-19), 0.81 (3H, s, H-18), 2.14 (3H, s, H-21), 3.23 (1H, t,
(3H, s, H-19), 4.45 (1H, t, J = 8 Hz, H-15a), 5.72 (1H, s, H-4), 6.13
(1H, dd, J = 1.5, 7.5 Hz, H-7), 6.20 (1H, dd, J = 2.5, 9.5 Hz, H-6).
A fourth transformed compound, obtained by elution with ace-
tone in dichloromethane (7:93 v/v), was 15
4,6-diene-3,17-dione (34) (49.9 mg), Rf = 0.18, acetone/dichloro-
methane (1:19 v/v), which crystallized from dichloromethane as
a
-hydroxyandrosta-
J = 8.7 Hz, H-17
Further elution with ethyl acetate/petrol (1:1 v/v) led to the iso-
lation of ,5 ,14 -trihydroxypregnan-20-one (39) (20.8 mg),
Rf = 0.33, ethyl acetate/petrol (9:1 v/v), which crystallized from
a), 4.07 (1H, t, J = 2.8 Hz, H-3b).
3a
a
a
needles, m.p. 189–190°, [
a
]
+110.3° (c 0.78); IR: mmax 3419,
D
2940, 1738, 1652, 1615, 1356, 1273, 1083 cmÀ1; HREIMS: m/z
(rel. int.): 301.1811 (100) M+ (C19H24O3 requires 301.1798); 1H
NMR: d 1.00 (3H, s, H-18), 1.15 (3H, s, H-19), 2.20 (1H, dd, J = 6.5,
19.5 Hz, H-16), 3.02 (1H, dd, J = 8, 19.5 Hz, H-16), 4.57 (1H, q,
J = 8 Hz, H-15b), 5.70 (1H, s, H-4), 6.16 (1H, dd, J = 3, 10 Hz, H-6),
6.85 (1H, dd, J = 2, 10 Hz, H-7).
methanol as cubes, m.p. 188–189°, [
a
;
] +34.6° (c = 1.86, CH2Cl2);
D
IR: mmax 3398, 2928, 1694 cmÀ1
HRMS: Calc. for MNa+
(C21H34O4Na): 373.2349, Found: 373.2370; 1H NMR: d 0.74 (3H,
s, H-18), 0.91 (3H, s, H-19), 2.13 (3H, s, H-21), 3.27 (1H, t,
J = 17 Hz, H-17), 4.01 (1H, t, J = 2.6 Hz, H-3).
3a,7a-Dihydroxypregn-5-en-20-one (40) (8 mg) was obtained
The final metabolite, isolated using acetone/dichloromethane
when the column was eluted with ethyl acetate/petrol (3:2 v/v),
(1:4 v/v), was
1a-hydroxyandrosta-4,6-diene-3,17-dione (35)
Rf = 0.21, ethyl acetate/petrol (9:1 v/v). The metabolite crystallized
(19.2 mg), Rf = 0.12, acetone/dichloromethane (1:19 v/v), which
crystallized as needles from dichloromethane, m.p. 194–196°,
from methanol asplates, m.p. 145–147°, [
IR: m
a
]D À21.4° (c = 1.0, CH2Cl2);
max 3389, 2933, 1697 cmÀ1; HRMS: m/z (rel. int. %) 281 (26), 296
[
a
]
+53.2° (c 0.44); IR: mmax 3419, 2942, 1738, 1652, 1455, 1374,
(100), 314 (42); Calc. for M+ – H2O (C21H30O2): 314.2246, Found:
314.2249; 1H NMR: d 0.66 (3H, s, H-18), 1.01 (3H, s, H-19), 2.16
(3H, s, H-21), 2.59 (1H, t, J = 9.3 Hz, H-17), 3.85 (1H, d, J = 4.7 Hz, H-
7b), 4.12 (1H, bs, H-3b), 5.65 (1H, d, J = 5.1 Hz, H-6).
D
1271, 1059 cmÀ1; HREIMS: m/z (rel. int.): 301.1812 (100) M+
(C19H24O3 requires 301.1798); 1H NMR: d 0.96 (3H, s, H-18), 1.15
(3H, s, H-19), 4.19 (1H, t, J = 7.5 Hz, H-1b), 5.80 (1H, s, H-4), 6.22
(1H, dd, J = 1.5, 10 Hz, H-7), 6.24 (1H, dd, J = 2.5, 10 Hz, H-6).
2.3.3. Biotransformation of 17-hydroxyandrost-4-en-3-one (3)
The combined organic extract (982.3 mg) was purified using col-
umn chromatography. Elution with acetone/petrol (1:4 v/v) gave
the fed compound (61.3 mg). Further purification of mixed fractions
from this column using a solvent system of acetone/dichlorometh-
2.2.11. Biotransformation of 6b-hydroxy-3
one (11)
a,5a-cycloandrostan-17-
No metabolites were obtained in this fermentation.
ane (1:4 v/v) led to the isolation of 3
(41) (50 mg), Rf = 0.20, ethyl acetate/petrol (9:1 v/v), which crystal-
lized from methanol as cubes, m.p. 187–188°, [ +2° (c = 0.14,
CH2Cl2) (lit. [30] m.p. 193–195°); IR: mmax 3386, 2935 cmÀ1
NMR (CD3OD): d 0.76 (3H, s, H-18), 0.96 (3H, s, H-19), 3.65 (1H, t,
J = 8.5 Hz, H-17 ), 3.95 (1H, t, J = 5 Hz, H-3b).
a,5a,17b-trihydroxyandrostane
2.3. Exophiala jeanselmei var. lecanii-corni
a]
D
;
1H
Exophiala jeanselmei var. lecanii-corni was maintained on potato
dextrose agar (PDA) slants. A medium comprised of sucrose (10 g/
L), yeast extract (4 g/L), bactopeptone (5 g/L), MgSO4Á7H2O (1 g/L)
and KH2PO4 (1 g/L) was used for the fermentations [5]. Incubations
and extractions were conducted in a similar manner to those used
for Fusarium oxysporum var. cubense.
a
2.3.4. Biotransformation of pregn-4-ene-3,20-dione (4)
Work up of the fermentation gave mycelial and broth extracts
which were chromatographed separately. Elution of the mycelial

1322
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
extract (399.2 mg) with ethyl acetate/petrol (15:85 v/v) yielded
the fed compound (4) (35 mg). Percolation of the broth extract
Chromatography on the mycelial extract using ethyl acetate/
petrol (15:85 v/v) led to the isolation of the fed substrate (214 mg).
(418.5 mg) with ethyl acetate/petrol (35:65 v/v) yielded 14a-hy-
droxy-5b-pregnane-3,20-dione (42) (5.1 mg), Rf = 0.32, ethyl ace-
tate/petrol (1:1 v/v), which crystallized from acetone as cubes,
2.3.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
No metabolites were obtained in this fermentation.
m.p. 130–131°, [a] +38.9° (c = 0.39, CH2Cl2) (lit. [31] m.p. 165–
D
168°,
[
a
]
+128.4° (c = 0.5, MeOH)); IR: mmax 3510, 2940,
1H NMR: d 0.82 (3H, s, H-19), 0.88 (3H, s, H-18), 2.14
1703 cmÀ1D
;
2.3.9. Biotransformation of 5a-androstane-3,17-dione (9)
The fermentation yielded broth (867.8 mg) and mycelial
(256.2 mg) extracts which were separately subjected to column
chromatography. Purification of the broth extract led to the isola-
(3H, s, H-21), 3.24 (1H, t, J = 8 Hz, H-17
Continued elution with ethyl acetate/petrol (35:65 v/v) led to the
isolation of 3 ,14 -dihydroxy-5 -pregnan-20-one (38) (31 mg).
a).
a
a
a
tion of five metabolites. The first analogue, 3a-hydroxy-5a-andro-
stan-17-one (44) (8.9 mg), was isolated when the column was
eluted with ethyl acetate in petrol (1:3 v/v), Rf = 0.40, ethyl ace-
tate/petrol (1:1 v/v). It crystallized from acetone as cubes, m.p.
2.3.5. Biotransformation of 17a,21-dihydroxypregn-4-ene-3,11,20-
trione (5)
The broth (746.9 mg) and mycelial (342.7 mg) extracts which
were combined and purified by chromatography. The fed substrate
was isolated (238.6 mg) when the column was eluted with ethyl
acetate/petrol (1:1 v/v). Further elution with ethyl acetate/petrol
160–163°, [
a] +36.8° (c = 0.19, CH2Cl2) (lit. [33] m.p. 181–182°,
D
[a]
+97° (c = 1.20, EtOH)); IR: mmax 3423, 2923, 1740 cmÀ1
;
1H
NMR: d 0.81 (3H, s, H-19), 0.86 (3H, s, H-18), 4.05 (1H, br s, H-3b).
Continued elution with this solvent system yielded 3 ,17b-
dihydroxy-5 -androstane (45) (13.8 mg), Rf = 0.32, ethyl acetate/
petrol (1:1 v/v), which crystallized from acetone as needles, m.p.
215–218°, [ +142.2° (c = 0.18, CH2Cl2) (lit. [27] m.p. 220–
223°); IR: mmax 3423, 2922 cmÀ1
0.79 (3H, s, H-18), 3.64 (1H, t, J = 9 Hz, H-17
Elution with ethyl acetate/petrol (2:3 v/v) yielded 11
xy-5 -androstane-3,17-dione (46) (35.2 mg), Rf = 0.22, ethyl ace-
tate/petrol (1:1 v/v), which crystallized from dichloromethane as
cubes, m.p. 157–160°, [ +34.5° (c = 0.59, CH2Cl2) (lit. [34] m.p.
D
a
(7:3 v/v) gave 17a,20S,21-trihydroxypregn-4-ene-3,11-dione (26)
a
(200 mg).
a]
D
;
1H NMR: d 0.73 (3H, s, H-19),
), 4.04 (1H, br s, H-3b).
-hydro-
a
a
a
a
]
D
192–194°, [
a
]
+66° (c = 0.1)); IR: mmax 3449, 2926, 1737 cmÀ1
;
D
1H NMR: d 0.90 (3H, s, H-18), 1.15 (3H, s, H-19), 4.03 (1H, td,
J = 10, 5.1 Hz, H-11b).
This was followed by 3
(47) (109.8 mg), Rf = 0.14, ethyl acetate/petrol (1:1 v/v), which
crystallized from acetone as prisms, m.p. 209–211°, [ +66.9°
a,9a-dihydroxy-5a-androstan-17-one
a
]
D
2.3.6. Biotransformation of 17a,21-dihydroxypregna-1,4-diene-
3,11,20-trione (6)
The fermentation yielded broth (548.8 mg) and mycelial
(352 mg) extracts. Chromatography of the former using ethyl ace-
tate/petrol (1:1 v/v) led to the isolation of the untransformed ste-
roid
(427.3 mg).
Further
elution
yielded
17a,20S,21-
trihydroxypregna-1,4-diene-3,11-dione (27) (8.2 mg).
2.3.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
This fermentation produced broth (93.9 mg) and mycelial
(419.5 mg) extracts. Chromatography of the former yielded the fed
compound, 7 (13.3 mg), eluted with ethyl acetate in petrol (15:85
v/v). Percolation with ethyl acetate/petrol (3:7 v/v) led to the isola-
tion of 3,6b-dihydroxyestra-1,3,5(10)-trien-17-one (43) (6.9 mg),
Rf = 0.38, ethyl acetate/petrol (9:1 v/v). Metabolite 43 crystallized
from acetone as plates, m.p. >270°, [
(lit. [32] m.p. 265–270°); IR:
max 3854, 3422, 2933, 1753 cmÀ1; 1H
NMR: d 0.92 (3H, s, H-18), 4.68 (1H, br s, H-6 ), 6.72 (1H, d,
J = 10 Hz, H-2), 6.87 (1H, s, H-4), 7.15 (1H, d, J = 10 Hz, H-1).
a] +17.1° (c = 0.69, CH2Cl2)
D
m
a

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1323
(c = 0.46, CH2Cl2); IR:
m
max 3441, 2927, 1733 cmÀ1; HRMS: m/z (rel.
sium sulfate heptahydrate (4 g/L), zinc sulfate heptahydrate
(8 mg/L), iron(III) chloride hexahydrate (6.6 mg/L) [37]. Incuba-
tions and extractions were carried out as previously described for
Fusarium oxysporum var. cubense. However, after the final feed
the flasks were shaken for another 10 d.
int. %) 237 (38), 255 (76), 270 (100), 288 (54), 306 (4); Calc. For
M+(C19H30O3): 306.2195, Found: 306.2195; 1H NMR: d 0.86 (3H,
s, H-18), 0.93 (3H, s, H-19), 4.03 (1H, br s, H-3b).
The final analogue isolated with this solvent system was
3a,11a-dihydroxy-5a-androstan-17-one (48) (69.8 mg), Rf = 0.12,
ethyl acetate/petrol (1:1 v/v), which crystallized from acetone as
2.4.1. Biotransformation of 3b-hydroxyandrost-5-en-17-one (1)
The fermentation yielded a combined organic extract (3.82 g)
which was chromatographed. Elution with ethyl acetate/petrol
(4:1 v/v) afforded the fed substrate (118.2 mg), while elution with
ethyl acetate/petrol (1:1 v/v) gave 3b,17b-dihydroxyandrost-5-ene
(12) (148 mg). Continued elution with ethyl acetate/petrol (7:3 v/
v) yielded 3b-hydroxyandrost-5-ene-7,17-dione (13) (16 mg).
cubes, m.p. 182–183°, [a] +30.4° (c = 0.21 CH2Cl2) (lit. [35] m.p.
D
193–194°, [
a
]
À47° (c = 0.4)); IR: mmax 3405, 2924, 1733 cmÀ1
;
D
1H NMR: d 0.86 (3H, s, H-18), 0.94 (3H, s, H-19), 3.90 (1H, m, w/
2 = 11 Hz, H-11b), 4.01 (1H, br s, H-3b).
Column chromatography on the mycelial extract using ethyl
acetate/petrol (15:85 v/v) led to the isolation of the fed compound
2.4.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
(64.7 mg) along with
3
a
-hydroxy-5a-androstan-17-one (44)
Biotransformation produced
a combined organic extract
(16.8 mg) and 3 ,17b-dihydroxy-5
a
a-androstane (45) (33.4 mg).
(1.42 g). Purification by column chromatography, ethyl acetate/
petrol (1:4 v/v) afforded the fed substrate (370.1 mg). Continued
percolation with ethyl acetate/petrol (1:1 v/v) gave 3b-hydroxy-
13,17-secoandrost-5-ene-17,13a-lactone (52) (41.9 mg), Rf = 0.18,
ethyl acetate/petrol (1:1 v/v), which crystallized from ethyl acetate
2.3.10. Biotransformation of androsta-4,6-diene-3,17-dione (10)
Substrate 10 (500 mg) was fed and broth (342 mg) and mycelial
(143 mg) extracts were produced. Two metabolites were isolated
from the broth extract during elution with ethyl acetate/petrol
as needles, m.p. 227–228°, [
227–230°, [ À93.9 (c 0.1 in CHCl3)); HRMS (EI): m/z (rel. int.):
304.2037 [M+] (29), C19H28O3 requires 304.2038; IR: mmax 3398,
2928, 1700 cmÀ1 1H NMR: d 0.98 (3H, s, H-19), 1.33 (3H, s, H-
18), 3.54 (1H, m, w/2 = 6 Hz, H-3 ), 5.34 (1H, s, H-6);
a]
À91.7° (c = 0.0033) (lit. [38] m.p.
D
(1:3 v/v). The first was 3
a-hydroxy-6a,7a-epoxy-5a-androstan-
a]
D
17-one (49) (49.8 mg), Rf = 0.18, ethyl acetate/petrol (1:1 v/v),
which crystallized from methanol as plates, m.p. 208–211°, [a]
D
;
+12.9° (c = 0.99, MeOH); IR: mmax 3456, 2930, 1729 cmÀ1; HRMS:
m/z (rel. int. %) 253 (30), 268 (100), 286.2 (13), 304 (12); Calc. for
M+ (C19H28O3): 304.2038, Found: 304.2039; 1H NMR: d 0.78 (3H,
s, H-19), 0.92 (3H, s, H-18), 2.75 (1H, bd, J = 4 Hz, H-6b), 3.16
(1H, dd, J = 2, 3.9 Hz, H-7b), 4.14 (1H, quintet, J = 3 Hz, H-3b).
a
2.4.3. Biotransformation of 17b-hydroxyandrost-4-en-3-one (3)
The combined organic extract (2.27 g) was chromatograped
with ethyl acetate/petrol (1:4 v/v) to afford androst-4-ene-3,17-
This was followed by 3
(25 mg).
The compound 3
a
,17b-dihydroxy-5b-androstane (36)
dione (18) (30 mg) and 3
a-hydroxy-5a-androstan-17-one (44)
(14.6 mg). Further percolation with ethyl acetate/petrol (1:3 v/v)
gave the fed steroid (491 mg), followed by 3a,17b-dihydroxy-5a-
androstane (45) (11.2 mg).
a
,6b,7
a
-trihydroxy-5 -androstan-17-one (50)
a
(14.4 mg) was isolated during elution with ethyl acetate/petrol (1:2
v/v), Rf = 0.12, ethyl acetate/petrol (1:1 v/v), which crystallized from
acetone as needles, m.p. 207–209°, [a]D +21.3° (c = 0.15, Me2CO); IR:
2.4.4. Biotransformation of pregn-4-ene-3,20-dione (4)
mmax 3397, 1684 cmÀ1; HRMS (ES): Calc. for MH+ (C19H31O4):
323.2216, Found: 323.2222; 1H NMR: d 0.65 (3H, s, H-19), 0.81 (3H,
s, H-18), 1.83 (1H, m, w/2 = 14 Hz, H-8), 3.51 (1H, t, J = 2.7 Hz, H-
The fermentation gave broth extract (0.75 g), and mycelial
(0.67 g) extracts. Column chromatography of the broth extract,
eluting with ethyl acetate/petrol (1:4 v/v), afforded androst-4-
7b), 3.76 (1H, t, J = 2.7 Hz, H-6
Purification of the mycelial extract led to the isolation of one
analogue, -hydroxy-6 ,7 -epoxy-5 -androstan-17-one (49)
a), 4.07 (1H, quintet, J = 2.7 Hz, H-3b).
ene-3,17-dione (18) (9 mg), and 5a-pregnane-3,20-dione (53)
3a
a
a
a
(6.1 mg), which was obtained during elution with ethyl acetate/
petrol (30:70 v/v).
2.3.11. Biotransformation of 3a,5a-cycloandrost-6-en-17-one (11)
Incubation with 11 (980 mg) gave broth (188.9 mg) and myce-
lial (324.4 mg) extracts, which were chromatographed separately.
Purification of the broth extract led to the isolation of the 6b,7
a-
dihydroxy-3 ,5 -cycloandrostan-17-one (51) (4.3 mg), which
a
a
was obtained when the column was eluted with ethyl acetate/pet-
rol (3:2 v/v), Rf = 0.42, ethyl acetate (1 v). This metabolite did not
crystallize, [
a] +27.1° (c = 0.81, CH2Cl2) (lit. [36] m.p. 172–174°,
D
[
a]
D
+95°); IR: mmax 3448, 2933, 1727, 1375 cmÀ1 1H NMR: d
;
0.30 (1H, dd, J = 5, 8.1 Hz, H-3), 0.55 (2H, t, J = 4.4 Hz, H-4), 0.95
(3H, s, H-18), 1.10 (3H, s, H-19), 1.36 (1H, s, H-9), 3.25 (1H, d,
J = 2.7 Hz, H-6a), 3.89 (1H, br s, H-7b).
The mycelial extract yielded only the fed material (264 mg),
during elution with ethyl acetate/petrol (1:19 v/v).
2.4. Ceratocystis paradoxa
The fungus was maintained on potato dextrose agar slants at
28°. The medium used was a modification of the Salmink medium
which contained sucrose (20 g/L), corn steep solids (4 g/L), yeast
extract (2 g/L), potassium-di-hydrogen phosphate (6 g/L), magne-

1324
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
(4.5 mg), Rf = 0.25, ethyl acetate/petrol (1:1 v/v). The latter crystal-
lized from ethyl acetate as needles, m.p. 199–200°, [ +85
(c = 0.005) (lit. [39] m.p. 198–201°, [ +144°); IR: mmax 1737,
1670 cmÀ1 1H NMR: d 0.60 (3H, s, H-18), 0.99 (3H, s, H-19).
The 1H NMR spectrum of 13 showed that the olefinic proton (H-
6) was now a singlet. This indicated that the protons at C-7 were no
longer present, which was supported by the appearance of a new
non-protonated carbon at d 201.1. The chemical shift value also
suggested that the new carbonyl moiety was conjugated. This
transformed product was determined to be 3b-hydroxyandrost-
5-ene-7,17-dione (13) [19].
a]
D
a]
D
;
The 13C NMR spectrum of 14 displayed the presence of a new
methine at d 71.6. The corresponding proton at d 4.44 was coupled
to H-14 (d 1.37) and H-16 (d 2.14, 2.97). Therefore, it was deter-
mined that oxygen insertion had occurred at C-15. The T-ROESY
data was used to establish that the stereochemistry of the hydroxyl
2.4.5. Biotransformation of 17a,21-dihydroxypregn-4-ene-3,11,20-
trione (5)
The experiment yielded broth (1.52 g) and mycelial (1.62 g) ex-
tracts. Column chromatography of the broth extract with ethyl ace-
tate provided the fed steroid (57.2 mg) and 17a,20S,21-
group was
H-15 (d 4.41) and H-18 (d 0.92). This metabolite was deduced to
be 3b,15 -dihydroxyandrost-5-en-17-one (14).
a. This was based on the couplings observed between
trihydroxypregn-4-ene-3,11-dione (26) (30.3 mg).
a
2.4.6. Biotransformation of 17a,21-dihydroxypregna-1,4-diene-
1H NMR data of 15 showed the presence of a new proton reso-
nating at d 3.95, which suggested that a methylene had been
hydroxylated. Furthermore, this proton showed a correlation to
H-8 (d 1.58) in the COSY spectrum, an indication that the new alco-
hol was located at C-7. The HMBC spectrum corroborated this as
the proton was coupled to C-5 (d 143.8), C-6 (d 125.5) and C-14
(d 51.2). In the T-ROESY spectrum H-7 was coupled to H-9 (d
1.10) and H-1 (d 1.45); therefore, the stereochemistry of the hydro-
xyl group was b. This compound was assigned as 3b,7b-dihydroxy-
androst-5-en-17-one (15) [20].
The spectral information for 16 was similar to that of 15, how-
ever, there was a significant difference in the resonance frequency
of the new methine. From the 2D data it was evident that oxygen
insertion had occurred at C-7 as both H-6 (d 5.65) and H-8 (d 1.70)
were coupled to H-7 (d 3.98) in the COSY spectrum. Using the T-
ROESY correlations it was observed that H-7 was coupled to H-8
and H-15b (d 1.59), thus making the stereochemistry of the hydro-
3,11,20-trione (6)
Incubation gave a combined broth and mycelial extract (2.01 g),
which was chromatographed with ethyl acetate/petrol (1:1 v/v) to
give the fed substrate (274 mg). Continued elution with ethyl ace-
tate gave 17a,20S,21-trihydroxypregna-1,4-diene-3,11-dione (27)
(66 mg).
2.4.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
The experiment provided broth (578.3 mg) and mycelial
(483.5 mg) extracts. Purification of the mycelial extract with ethyl
acetate/petrol (4:6 v/v) provided the fed substrate (247.5 mg). Per-
colation with ethyl acetate/petrol (1:1 v/v) afforded 3,17b-dihydr-
oxyestra-1,3,5(10)-triene (28) (58.1 mg).
2.4.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
Incubation with the substrate provided no transformed
metabolites.
xyl moiety
a. Therefore, this metabolite was determined to be
3b,7 -dihydroxyandrost-5-en-17-one (16) [18].
a
2.4.9. Biotransformation of 5a-androstane-3,17-dione (9)
Incubation provided a combined broth and mycelial extract
(839.2 mg). The fed substrate (110 mg) was recovered from ethyl
acetate/petrol (1:9 v/v). Continued elution with the same solvent
3.1.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
Compounds 3 and 17 were generated in this fermentation.
The 13C NMR spectrum of 3 revealed that carbons resonating at
d 209.6 and 31.8 in fed substrate were no longer present. This
metabolite possessed only 19 carbons, indicating the loss of the
side chain. Additionally, there was a new methine at d 81.7 and a
new non-protonated carbon at d 199.6, pointing towards the pres-
ence of a conjugated ketone. From this data it was deduced that
there had been a Baeyer–Villiger oxidation of the C-20 ketone, fol-
lowed by hydrolysis of the resulting acetate ester to give an alcohol
at C-17. Along with this there was oxidation of the hydroxyl group
at C-3 and concomitant rearrangement of the C-5,6 double bond to
C-4,5. This compound was determined to be 17b-hydroxyandrost-
4-en-3-one (3) [21].
system afforded 3
a
-hydroxy-5a-androstan-17-one (44) (53.5 mg)
and 3b-hydroxy-5
a
-androstan-17-one (29) (59.9 mg). Percolation
with ethyl acetate/petrol (1:4 v/v) yielded 3a,17b-dihydroxy-5a-
androstane (45) (323.8 mg).
2.4.10. Biotransformation of androsta-4,6-diene-3,17-dione (10)
The combined broth and mycleial extract (454.4 mg) was chro-
matographed with ethyl acetate/petrol (3:17 v/v) to provide the fed
substrate (170.9 mg). Further elution with ethyl acetate/petrol (1:4
v/v) gave 17b-hydroxyandrosta-4,6-dien-3-one (32) (113.3 mg).
There was a new multiplet at d 3.39 in the 1H NMR data of 17.
The 13C NMR spectrum showed the loss of the carbonyl signal at d
209.6, which was accompanied by the appearance of a new
methine at d 70.8. This indicated that reduction of the C-20 ketone
had occurred. This was supported by the significant upfield shift in
the resonance for C-17. There were no other significant changes in
the NMR data, and as such this congener was assigned as 3b,20-
dihydroxypregn-5-ene (17).
2.4.11. Biotransformation of 3a,5a-cycloandrost-6-en-17-one (11)
TLC analysis of the mycelial and the broth extracts showed that
the neither contained products of transformation.
3. Results
3.1. Biotransformation with Fusarium oxysporum var. cubense
3.1.1. Biotransformation of 3b-hydroxyandrost-5-en-17-one (1)
Metabolites 12–16 were obtained in this feeding.
3.1.3. Biotransformation of 17b-hydroxyandrost-4-en-3-one (3)
Compounds 18–22 were produced in this fermentation.
The 13C NMR data of the least polar metabolite (12) showed a
disappearance of the carbon resonating at d 221.2, and this was
accompanied by the appearance of a new methine at d 81.9. It
was determined that C-17 carbonyl was reduced to the alcohol.
There were no other significant changes in the NMR data, and as
such, compound 12 was deduced to be the 3b,17b-dihydroxyand-
rost-5-ene [18].
The IR spectrum of 18 did not show a stretch corresponding to
the presence of a hydroxyl group. This indicated that the C-17 hy-
droxyl was no longer present. Examination of the 1H NMR spec-
trum revealed a loss of the multiplet at d 3.66, further supporting
this claim. There was a new peak at d 220.5 which was concomi-
tant with the loss of the methine at d 81.9 in the 13C NMR spectrum

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1325
of the fed compound. This confirmed that oxidation of the C-17
alcohol had occurred. This compound was deduced to be and-
rost-4-ene-3,17-dione (18) [22].
H-14 (d 1.22) and H-16 (d 1.58, 2.76). In the T-ROESY spectrum
H-15 (d 4.10) showed correlations to H-8 (d 1.74) and H-18 (d
0.71). Therefore, the stereochemistry of the hydroxyl moiety was
The 1H NMR spectrum of the second product of bioconversion
(19) did not display any signal that corresponded to olefinic pro-
tons. This suggested that the C-4,5 double bond was no longer
present. Information from 13C NMR analysis further supported this
claim, as the resonances representing the olefinic carbons at d
171.7 and 124.2 were missing from the spectrum. There was also
a downfield shift of 12 ppm in the resonance value for C-3. This
a
. This compound was determined to be 15
a-hydroxypregn-4-
ene-3,20-dione (24) [25].
The 1H NMR spectrum of the most polar analogue (25) exhib-
ited two new peaks at d 3.70 and 4.07, suggesting that it was a
dihydroxy analogue. The proton at d 3.70 was coupled to H-17 (d
1.64) and H-21 (d 1.23) in the COSY spectrum, which was an indi-
cation that there had been a reduction of the C-20 carbonyl. This
was further supported by information from the 13C NMR spectrum,
where there was the disappearance of the peak at d 209.3. The
COSY data displayed correlations between the proton resonating
at d 4.07 with H-14 (d 1.09) and H-16 (d 1.78, 2.26). Therefore it
was deduced that the second hydroxyl group was located at C-
15. In the T-ROESY spectrum H-15 (d 4.07) was coupled to H-8 (d
1.74) and H-18 (d 0.76), and thus the stereochemistry of the hydro-
analogue was determined to be 17b-hydroxy-5
one (19) [19].
a-androstan-3-
The 1H NMR data for the third metabolite (20) displayed the
loss of the proton at d 3.66 in the spectrum of the fed substrate.
Additionally, a new quartet was observed at d 4.42. In the 13C
NMR spectrum there was a new methine at d 70.4 and a new
non-protonated carbon at d 215.8. In the HMBC spectrum the
new methine was correlated to both C-14 (d 57.2) and C-16 (d
46.3). From this it was determined that hydroxylation had oc-
curred at C-15. The stereochemistry of the alcohol was derived
from the T-ROESY data where H-15 (d 4.42) showed couplings to
H-8 (d 1.96) and H-18 (d 0.96). The non-protonated carbon was
formed by the oxidation of the C-17 alcohol to the ketone. The
xyl was assigned as
dihydroxypregn-4-en-3-one (25).
a. This congener was identified as 15a,20-
3.1.5. Biotransformation of 17a,21-dihydroxypregn-4-ene-3,11,20-
trione (5)
Transformation of this steroid gave a single metabolite (26). The
13C NMR spectrum revealed the presence of a new methine reso-
nating at d 73.2, and the concomitant loss of the carbonyl at d
211.5 in the fed substrate. This was indicative of reduction of the
C-20 carbonyl moiety. Downfield shifts in the resonances for both
C-17 and C-21 further supported that there was now an alcohol at
compound was deduced to be 15a-hydroxyandrost-4-ene-3,17-
dione (20) [24].
The 13C NMR data of the fourth transformed product (21) re-
vealed the presence of a new methine resonating at d 73.1. This
was concurrent to the loss of a methylene at d 33.2 which corre-
sponded to C-6 in the fed compound. HMBC data showed that
the methine proton was coupled to C-4 (d 126.5), C-8 (d 29.7)
and C-10 (d 38.0). Based on this it was determined that oxygen
insertion had taken place at C-6. The stereochemistry of the alcohol
was b, as H-6 (d 4.36) was correlated to H-7b (d 1.23) and H-8 (d
2.02) in the T-ROESY spectrum. Therefore, this compound was des-
C-20. This analogue was determined to be 17a,20S,21-trihydroxy-
pregn-4-ene-3,11-dione (26) [4].
3.1.6. Biotransformation of 17a,21-dihydroxypregna-1,4-diene-
3,11,20-trione (6)
This fermentation yielded only steroid 27. There was a new
peak resonating at d 4.82 in the 1H NMR spectrum and this pointed
to the fact that a new hydroxyl group was present. Inspection of
the 13C NMR data displayed the disappearance of the carbonyl moi-
ety at d 211.7, which coincided with the appearance of a new
methine at d 72.7. It was concluded that the C-20 ketone under-
went reduction to give the corresponding alcohol. Compound 27
ignated to be 6a,17b-dihydroxyandrost-4-en-17-one (21).
The final congener (22) appeared to have been a monohydroxy-
lated derivative of testosterone, as the 13C NMR spectrum dis-
played a new methine resonating at d 72.6. Notably, there was
an upfield shift in the resonance value for C-17, indicating that
there was change in ring D. This was corroborated by the COSY
data which showed that this new proton was coupled to both H-
14 (d 0.98) and H-16 (d 1.48, 2.08). Hydroxylation had taken place
was deduced to be 17
dione [4].
a,20S,21-trihydroxypregna-1,4-diene-3,11-
at C-15. The stereochemistry was resolved as being
a, due to the
signal enhancements observed between H-15 (d 1.37) and H-16b
(d 1.48), and H-18 (d 0.85) in the T-ROESY spectrum. Based on this
information, this analogue was deduced to be 15a,17b-dihydroxy-
androst-4-en-3-one (22).
3.1.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
Compound 28 was the only product of this fermentation. The IR
spectrum did not display a stretch corresponding to a ketone. This
was indicative that the C-17 carbonyl was no longer present. This
was corroborated by information obtained from the 13C NMR data
in which the resonance corresponding to the C-17 ketone was ab-
sent. The spectrum revealed the appearance of a new methine at d
81.9. Thus, it was deduced that the C-17 carbonyl had been re-
duced to the hydroxyl group. This metabolite was determined to
be 3,17b-dihydroxyestra-1,3,5(10)-triene (28) [26].
3.1.4. Biotransformation of pregn-4-ene-3,20-dione (4)
Steroids 3 and 23–25 were formed during this fermentation.
The IR spectrum of the first metabolite (23) displayed a stretch
at 3441 cmÀ1, pointing towards the presence of a hydroxyl group.
This was substantiated by the 13C NMR data in which the carbonyl
moiety which resonated at d 209.3 was no longer present. Addi-
tionally, there was a new signal at d 70.2, which was indicative
of reduction of C-20 ketone to the alcohol. These conclusions were
further supported by an upfield shifts in the signals for C-17 and C-
21. This compound was determined to be 20-hydroxypregn-4-en-
3-one (23) [25].
3.1.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
No transformed compounds were isolated.
3.1.9. Biotransformation of 5a-androstane-3,17-dione (9)
Metabolites 19 and 29–31 were formed in this transformation.
Investigation of the spectral data for the second metabolite re-
vealed that this analogue was 17b-hydroxyandrost-4-en-3-one (3).
The stretch at 3486 cmÀ1 displayed in the IR spectra of the third
congener (24) indicated the presence of a hydroxyl group. This was
corroborated by the presence of a new methine peak at d 73.3. The
position of oxygen insertion was established through the COSY
data as being at C-15. The new methine proton was coupled to
Inspection of the spectral data of the first compound revealed
that it was 17b-hydroxy-5a-androstan-3-one (19), which was dis-
cussed earlier in this article.
The second transformed product (29) had spectral similarities
to compound 19, however, the new peak in the 1H NMR spectrum
was a multiplet at d 3.66. This proton was coupled to both H-2 (d
1.41, 1.81) and H-4 (d 1.30, 1.59), which suggested that the new

1326
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
alcohol was located at C-3. The carbonyl at d 211.3 in the fed com-
pound was not present in the 13C NMR spectrum, further support-
ing this deduction. Accompanying this was the presence of a new
methine at d 71.2. The T-ROESY spectrum showed that H-3 (d
ported this view. The 1H NMR spectrum exhibited a new peak at
d 4.57. The COSY data showed that this proton was coupled to H-
16 (2.20, 3.02), indicating that hydroxylation had occurred at C-
15. Information obtained from the T-ROESY experiment revealed
that the stereochemistry of the new hydroxyl functionality was
3.60) was coupled to H-4
a (d 1.59) and H-5a (d 1.13), and as such
the stereochemistry of the hydroxyl group was deduced to be b.
a. This was based on the correlation of H-15 (d 4.57) with H-8 (d
This analogue was found to be 3b-hydroxy-5a-androstan-17-one
2.59), H-16b (d 3.02) and H-18 (d 1.00). This analogue was deter-
(29) [9].
mined to be the previously unreported 15a-hydroxyandrosta-
The IR spectrum of the third metabolite (30) had a stretch at
3383 cmÀ1, which indicated the presence of a hydroxyl group.
The 13C NMR spectrum exhibited a new methine at d 70.2 along
with the loss of a methylene at d 21.7 (C-15). The proton corre-
sponding to this methine was coupled to H-16 (d 2.10, 2.96) in
the COSY data, indicative of oxygen insertion at C-15. H-15 (d
4.39) was correlated to H-8 (d 1.85) and H-18 (d 0.92) in the T-
ROESY spectrum; therefore, the stereochemistry of the hydroxyl
4,6-diene-3,17-dione (34).
The final metabolite (35) had a molecular formula of C19H25O3,
which was determined from the HREIMS, where there was a
molecular ion peak at m/z = 301.1798. This pointed towards the
formation of a monohydroxy analogue. There was a new methine
at d 71.4, which was coupled to H-19 (d 1.15) in the HMBC spec-
trum. This placed the site of oxygen insertion in ring A or B. The
COSY data revealed that the proton of this new methine was corre-
lated to H-2 (d 2.62, 2.90), confirming that there was a hydroxyl
group at C-1. The stereochemistry was deduced as being
1 (d 4.19) was coupled to H-2b (d 2.62), H-11b (d 1.46) and H-19
(d 1.15). This compound was determined to be 1 -hydroxyandro-
sta-4,6-diene-3,17-dione (35), which was hitherto unreported in
the literature.
group was deduced to be
15 -hydroxy-5 -androstane-3,17-dione (30).
The 1H NMR spectrum of the final transformed compound (31)
a. This congener was determined to be
a
a
a, as H-
displayed two new peaks at d 3.61 and 4.38, evincing that this was
a dihydroxy analogue. Investigation of the 13C NMR spectrum dis-
played that the ketone with resonance at d 211.3 of compound 9
was no longer present. It was realized that this C-3 carbonyl had
been reduced to the alcohol. There was also the loss of the methy-
lene at d 21.8 (C-15), and along with this was a new methine at d
70.6. The HMBC data showed that H-14 (d 1.32) and H-16 (d
2.10, 2.98) were coupled to this carbon, signifying that the alcohol
was in ring D; therefore, hydroxylation had occurred at C-15. The
T-ROESY data was used to establish the stereochemistry of the hy-
a
3.1.11. Biotransformation of 3a,5a-cycloandrost-6-en-17-one (11)
No transformed products were obtained from this fermentation.
3.2. Biotransformation with Exophiala jeanselmei var. lecanii corni
3.2.1. Bioconversion of 3b-hydroxyandrost-5-en-17-one (1)
droxyl as
16b (d 2.10) and H-18 (d 0.90). This metabolite was 3b,15
droxy-5 -androstan-17-one (31).
a
, because H-15 (d 4.38) was coupled to H-7b (d 1.78), H-
Fermentation of this steroid yielded 16, 36 and 37.
Metabolite 16 was previously discussed.
a
-dihy-
a
For 36, DEPT experiments showed that the number of double
bond equivalents had decreased from six to four. Thus the C–C
and C–O double bonds had been reduced. The C-17 signal had
moved upfield to 82.4 ppm. The stereochemistries at positions 3
and 5 in this compound were determined from comparison with
NMR data compiled by⁄⁄ Blunt and Stothers [33]. There was a
change in the stereochemistry at C-3 which suggested the follow-
ing. The hydroxyl group was first oxidized to the ketone. Double
bond isomerization then led to formation of the conjugated ketone.
Reduction of the carbon–carbon double bond of the enone system
from the b-face generated the 5b-androstane. The latter was subse-
3.1.10. Biotransformation of androsta-4,6-diene-3,17-dione (10)
This transformation yielded compounds 9 and 32–35.
The 1H NMR spectrum for the first product of transformation (9)
did not display any peaks in the olefinic region. This was supported
by information from the 13C NMR spectrum. There were no other
significant changes in the NMR data of this analogue. This com-
pound was determined to be
[10,11].
5a-androstane-3,17-dione (9)
The absorption at 3421 cmÀ1 in the IR spectrum of the second
analogue (32) was indicative of the presence of a hydroxyl moiety.
In the 13C NMR spectrum the disappearance of the carbonyl at d
219.5 in 10 was noted along with the appearance of a new methine
at d 81.3. This was interpreted as a reduction of the C-17 ketone
having occurred. As there were no other significant changes in
the spectral data, compound 32 was identified as 17b-hydroxyan-
drosta-4,6-dien-3-one.
quently reduced at C-3 from the top face to give 3a,17b-dihydroxy-
5b-androstane (36).
The 13C NMR spectrum of 37 showed that the double bond was
still present, the carbonyl in 1 was absent, and there was a carbon
with a chemical shift at 82.4 ppm. It was, therefore, assumed that
the 17-ketone had been reduced. The stereochemistry at C-3 was
confirmed by NMR data [37]. The proton attached to this carbon
had a downfield chemical shift of 4.05 ppm and a coupling con-
stant of 5 Hz. This is characteristic of protons attached to C-3 which
are in an equatorial position (b stereochemistry). This leads to
equatorial–equatorial and equatorial–axial interactions which
have relatively small coupling constants (2–5 Hz). The compound
The 1H NMR data of the third metabolite (33) revealed the
appearance of a new peak at d 4.45 which pointed towards the
presence of a hydroxyl moiety. The information from the 13C
NMR spectrum supported this, as there was a new methine reso-
nating at d 71.1. This was accompanied by the loss of the methy-
lene signal at d 35.6 (C-16) in 10. In the COSY spectrum this
methine proton was coupled to H-15 (d 2.10, 2.16). From this it
was deduced that oxygen insertion had occurred at C-16. The T-
ROESY data displayed a coupling between H-16 (d 4.45) and H-
14 (d 1.70); and as such the stereochemistry of the hydroxyl group
was deduced to be b. This metabolite was determined to be the
hitherto unreported 16b-hydroxyandrosta-4,6-diene-3,17-dione
(33).
was determined to be 3a,17b-dihydroxyandrost-5-ene (37).
3.2.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
This steroid was converted into metabolites 38–40.
In metabolite 38 the C5-C6 double bond present in 2 had disap-
peared, and there was the insertion of a hydroxyl group which re-
sulted in the appearance of a signal at 86.1 ppm. 13C and DEPT
NMR experiments revealed that this was an additional non-proton-
ated carbon and that a methine group had disappeared. There was
an upfield shift in C-8, C-13 and C-15 of 6.5, 10.2 and 4.2 ppm,
respectively. There was no NOE correlation between H-19 (d
0.83) and H-5 (d 1.80), and so it was deduced that H-5 possessed
HREIMS data of the fourth congener (34) displayed a molecular
ion at m/z = 301.1798, which corresponded to a molecular formula
of C19H25O3. This suggested that compound 10 had undergone
monohydroxylation. The IR absorbance at 3419 cmÀ1 further sup-

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1327
a
-stereochemistry. Also H-3 had a coupling constant of 2.8 Hz,
indicating that it was equatorial. Steroid 38 was, therefore, the no-
vel 3 ,14 -dihydroxy-5 -pregnan-20-one.
3.2.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
The 1H NMR spectrum of the single metabolite (43) possessed a
new peak at 4.68 ppm. The 13C NMR spectrum contained the car-
bonyl signal at 218.5 ppm and a new resonance at 66.5 ppm. DEPT
experiments revealed that a methylene carbon had been converted
to a methine. In the COSY data, there was a correlation between the
proton on the newly hydroxylated carbon and H-7. HMBC data also
showed a relation between C-4 and that position. Also, there was a
downfield shift of 9.2 ppm in the value of C-7, and so it was con-
cluded that the benzylic position C-6 had been hydroxylated.
a
a
a
For compound 39 13C NMR data showed that there were three
hydroxylated carbons at 66.6, 78.6, 87.0 ppm, and that the double
bond had been reduced. The DEPT 90 NMR experiment revealed
that the number of methine carbons had decreased from six to
four, and that two of them were now non-protonated carbons.
The HMBC data indicated that the carbons with signals at 78.6
and 87.0 ppm were both correlated to H-19 (0.91 ppm) and H-18
(0.74 ppm), respectively. Therefore, it was inferred that carbons 5
NOESY experiments displayed a correlation between H-14
a (d
and 14 had been hydroxylated to give the metabolite 3
trihydroxypregnan-20-one (39).
The DEPT 135 NMR experiment for analogue 40 revealed that a
methylene carbon present in 2 had been converted to a methine.
a
,5
a
,14
a
-
1.59) and H-6 (d 4.68), and so the hydroxyl group was determined
to have b-stereochemistry. The new compound, 43, was therefore
3,6b-dihydroxyestra-1,3,5(10)-trien-17-one [32].
The COSY spectrum showed
a
correlation between H-6
3.2.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
(5.65 ppm) and the proton on the newly hydroxylated carbon
(4.12 ppm). There was also an NOE correlation between H-8
(d1.50) and H-7 (d 3.85), as well as H-3 (d 4.12) and H-21 (d
No transformed compounds were observed.
3.2.9. Biotransformation of 5a-androstane-3,17-dione (9)
2.16). Therefore, it was inferred that the 7 substituent was an
droxyl and there was now an inversion in the stereochemistry at C-
3 as a result of redox chemistry. The product was determined to be
a
-hy-
Steroids 44–48 were produced in this fermentation.
In the least polar metabolite, 44, the 13C NMR spectrum re-
vealed that there was reduction of one of the ketones, leading to
the appearance of a new peak at 66.4 ppm; while in the 1H NMR
data a proton resonating at 4.05 ppm, attached to a carbon bearing
a hydroxyl group was observed. The presence of hydroxyl and car-
bonyl groups was supported by IR data, which contained absorp-
tions at 3423 and 1740 cmÀ1, respectively. It was deduced that
the C-3 ketone had been reduced from the b-face, as the signal
for the attached proton was relatively far upfield (4.05 ppm). Also,
the signal for C-4 had decreased by 8.7 ppm and this supported the
3a,7a-dihydroxypregn-5-en-20-one (40).
3.2.3. Biotransformation of 17b-hydroxyandrost-4-en-3-one (3)
The only metabolite identified from this fermentation was
3
a
,5
and the 3-ketone had been reduced from the
tively. This was followed by hydroxylation at C-5. The hydroxyls at
positions 3 and 5 were determined to possess -stereochemistry.
a
,17b-trihydroxyandrostane (41) [30]. The C4–C5 double bond
a
and b faces, respec-
a
conclusion that C-3 had been reduced to give 3a-hydroxy-5a-
androstan-17-one (44) [34].
H-3 had a chemical shift of 3.95 Hz and a coupling constant of
5 Hz. These showed that the proton was equatorial. The OH group
at C-5 did not appear to have an effect on the methyl protons at C-
Metabolite 45 was produced by the simple reduction of the C-
17 ketone in compound 44. There were peaks at 4.00 ppm and
82.0 ppm in the 1H and 13C NMR spectra, respectively. DEPT exper-
iments showed that the number of double bond equivalents had
decreased from six to four. It was assumed that both ketones had
been reduced. This was supported by the observation of the signal
at 82.0 ppm in the 13C NMR spectrum, which was indicative of the
presence of a hydroxyl group attached to C-17. This compound
19, and so it was deduced that they were on the
molecule.
a face of the
3.2.4. Biotransformation of pregn-4-ene-3,20-dione (4)
Compounds 38 and 42 were generated in this fermentation.
The 1H NMR spectrum of 42 showed that there was reduction of
the C4–C5 double bond. The 13C NMR spectrum revealed that both
carbonyls were still intact. The DEPT 90 experiment showed that
the product had one less methine group than the fed compound,
while the 13C NMR spectrum indicated the presence of a (non-pro-
tonated) carbon resonating at 85.5 ppm which had an HMBC corre-
lation with H-18 (d 0.88). There was an upfield shift of 4.3 and
9.2 ppm for C-13 and C-15 resonances, respectively. These all led
to the conclusion that C-14 had been hydroxylated, and so the
was, therefore, 3a,17b-dihydroxy-5a-androstane (45) [30].
The 13C NMR spectrum of compound 46 showed that both car-
bonyls were still present with resonances at 211.9 and 219.2 ppm.
There was also a new methine at 68.6 ppm and one less methylene.
The 1H NMR spectrum revealed the presence of a triplet of doublets
at 4.03 ppm and this signal showed a COSY correlation with peaks
for H-9 (d 0.90) and H-12 (d 1.28, 2.12). Downfield shifts of 11.5
and 5.6 ppm were noticed for C-12 (d 43.0) and C-9 (d 60.0),
respectively, and so it was concluded that C-11 had been hydrox-
ylated. NOESY experiments showed a correlation between H-11
(d 4.03) and the protons on the two angular methyl groups; and
new compound was 14a-hydroxy-5b-pregnane-3,20-dione (42).
This metabolite was previously obtained from the fermentation
of Thamnostylum piriforme with 5b-pregnane-3,20-dione [31].
3
a
,14
a
-Dihydroxy-5a-pregnan-20-one (38) was previously
so the hydroxyl group was assigned
46 was, therefore, identified as 11
3,17-dione [34].
a
a
-stereochemistry. Steroid
discussed.
-hydroxy-5 -androstane-
a
Analogue 47 was a monohydroxylated derivative of compound
44. The DEPT NMR data showed that one less methine was present,
and that there was an additional non-protonated carbon at
75.9 ppm. HMBC data revealed a correlation between H-19 (d
0.93) and the newly hydroxylated carbon. When the 13C NMR spec-
trum of 44 was compared with that of metabolite 47, a slight in-
crease (4.5 ppm) was noticed in the chemical shift of C-10. Also,
there was an increase of 2.2 and 4.7 ppm in the resonance values
of C-8 and C-11, respectively. HMBC data showed a correlation be-
tween H-12 (d 1.60) and the carbon at 75.9 ppm. It was, therefore,
3.2.5. Biotransformation of 17
trione (5)
Cortisone (5) was converted to a single compound, 17
trihydroxypregn-4-ene-3,11-dione (26), which was previously
a
,21-dihydroxypregn-4-ene-3,11,20-
,20S,21-
a
discussed.
3.2.6. Biotransformation of 17a,21-dihydroxypregna-1,4-diene-
3,11,20-trione (6)
Incubation with prednisone (6) produced one analogue, identi-
fied as 17 ,20S,21-trihydroxypregna-1,4-diene-3,11-dione (27),
and which was previously discussed.
a
deduced that C-9 had been functionalized to give 3
droxy-5 -androstan-17-one (47).
a,9a-dihy-
a

1328
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
For metabolite 48 there was a strong resemblance between its
lence of six. The FTIR spectrum contained a ketone stretch at
1700 cmÀ1. The 1H NMR spectrum of compound 52 revealed pro-
tons on two angular methyl groups, resonating at d 0.98 and d
1.33, indicative of H-18 and -19, respectively. The absence of the
protons belonging to C-21 in the fed substrate and the presence
of only 19 peaks in the carbon spectrum implied that the C-17 side
chain had been degraded. Two signals at d 83.3 and 171.8 were ob-
served in the 13C NMR spectrum and these were representative of
the presence of a lactone moiety. Additionally, DEPT analysis
showed that there was one less methine signal compared with
those from 2.
13C NMR spectrum and that of compound 46. The major difference
was that there was now only one carbonyl present at C-17
(221.4 ppm), and so it was assumed that the C-3 ketone had been
reduced. DEPT experiments also showed that there was now an
additional methine group at 68.4 ppm. COSY data showed a corre-
lation between H-9 and the proton resonating at 3.90 ppm. The
HSQC experiment revealed that the latter was attached to the car-
bon at 68.4 ppm. There was also a correlation between H-12 (d
1.22, 2.09) and this proton and so it was assigned as H-11. As in
the case of 46, NOE data showed that the angular methyl protons
and H-11 all possessed the same stereochemistry. Hence, it was
concluded that C-11 had been hydroxylated in the
a-position.
Metabolite 48 was 3 ,11 -dihydroxy-5 -androstan-17-one [35].
a
a
a
Compound 44 was previously discussed.
3.2.10. Biotransformation of Androsta-4,6-diene-3,17-dione (10)
The products of transformation were 36, 49 and 50.
The 13C and 1H NMR spectra of metabolite 49 showed the disap-
pearance of the 4,5 and 6,7 C–C double bonds. There was a signal at
4.15 ppm, which indicated that there was a proton attached to a
carbon bearing a hydroxyl group. Further inspection of the 13C
NMR data revealed that the C-17 ketone had remained while the
one at C-3 had been reduced. Two carbon resonances were seen
at d 55.8 and d 53.5, and this suggested that an epoxide function-
ality was present. COSY experiments confirmed that the protons
attached to these carbons were adjacent, and so it was realized that
the 6,7 double bond had been epoxidized. The NOESY data showed
a correlation between H-9 (d 0.97) and H-5 (d 1.98); thus H-5 was
assigned
therefore, 3
,17b-Dihydroxy-5b-androstane
discussed.
a
a
-stereochemistry. The biotransformed steroid was,
-hydroxy-6 ,7 -epoxy-5 -androstan-17-one (49).
(36) was previously
a
a
a
3a
The other analogue (50) was a product of hydration of the 6,7-
epoxide ring in steroid 49. There were carbons bearing hydroxyl
groups with chemical shifts at 66.3, 71.3 and 76.8 ppm. There
was also a carbon with a signal at 220.0 ppm which is characteris-
tic of the C-17 ketone. COSY data showed three bond coupling be-
tween H-6 (d 3.51) and -7 (d 3.76). Therefore, the C-3 carbonyl had
been reduced. Opening of the epoxide ring accompanied by nucle-
ophilic attack by a water molecule yielded the hitherto unreported
3.3.3. Biotransformation of 17b-hydroxyandrost-4-en-3-one (3)
The incubation of 3 yielded androst-4-ene-3,17-dione (18) [22],
3a-hydroxy-5a-androstan-17-one (44) [34] and 3a,17b-dihy-
droxy-5 -androstane (46) [37], which were previously discussed.
a
3.3.4. Biotransformation of pregn-4-ene-3,20-dione (4)
This substrate was converted to metabolites 18 and 53.
Androst-4-ene-3,17-dione (18) was previously discussed.
The 13C NMR spectrum of 53 showed that the resonances corre-
sponding to C-4 and C-5 (at d 124.2 and 171.4) in the fed com-
pound had disappeared. The stereochemistry of the proton at C-5
3a,6b,7a-trihydroxy-5a-androstan-17-one (50).
3.2.11. Biotransformation of 3
a
,5a-cycloandrost-6-en-17-one (11)
The 13C NMR spectrum of the sole product of transformation
(51) revealed that the cyclopropane ring was still present, while
the IR data showed an absorption at 1727 cmÀ1, indicative of the
presence of a carbonyl group. The 6,7 double bond was absent,
and there were two new peaks at 70.1 and 77.2 ppm. HSQC and
COSY data showed that the hydrogen atoms on these oxygenated
carbons were vicinal. It was, therefore, deduced that the 6,7-double
bond had been first epoxidized, and then subsequently hydrated to
was assigned
associated with C-18 (d 13.8) and C-19 (d 11.8). Compound 53
was identified as 5 -pregnane-3,20-dione, based on NMR data
compiled by Blunt and Stothers [39].
a-stereochemistry because of the low shift values
a
3.3.5. Biotransformation of 17a,21-dihydroxypregn-4-ene-3,11,20-
trione (5)
17a,20S,21-Trihydroxypregn-4-ene-3,11-dione (26) [4] was
yield 6b,7
a-dihydroxy-3a,5a-cycloandrostan-17-one (51) [36].
previously discussed.
3.3. Ceratocystis paradoxa
3.3.6. Biotransformation of 17a,21-dihydroxypregna-1,4-diene-
3,11,20-trione (6)
17a,20S,21-Trihydroxypregna-1,4-diene-3,11-dione (27) [4]
was previously described.
3.3.1. Biotransformation of 3b-hydroxyandrost-5-en-17-one (1)
Fermentation with 1 yielded 3b,17b-dihydroxyandrost-5-en-
17-one (12) [18] and 3b-hydroxyandrost-5-ene-7,17-dione (13)
[19], which were previously discussed.
3.3.7. Biotransformation of 3-hydroxyestra-1,3,5(10)-trien-17-one (7)
3,17b-Dihydroxyestra-1,3,5(10)-triene (28) [26] was previously
discussed.
3.3.2. Biotransformation of 3b-hydroxypregn-5-en-20-one (2)
Fermentation of 2 afforded a product of Baeyer–Villiger type
rearrangement,
3b-hydroxy-17a-oxa-D-homo-androst-5-en-17-
one (52) [38]. HRMS analysis of 52 (M+ = 304.2037) gave a molec-
3.3.8. Biotransformation of 3b-hydroxy-5b,25S-spirostane (8)
Incubation of 8 did not yield any transformed compounds.
ular formula of C19H28O3, which denoted a double bond equiva-

P.C. Peart et al. / Steroids 76 (2011) 1317–1330
1329
3.3.9. Biotransformation of 5
Three analogues were generated: 3
17-one (44) [34], 3 ,17b-dihydroxy-5 -androstane (45) [39] and
3b-hydroxy-5 -androstan-17-one (29) [39], all of which were pre-
a
-androstane-3,17-dione (9)
presence of a 6,7-double bond. Allylic hydroxylation was also
observed.
a
-hydroxy-5 -androstan-
a
a
a
a
4.3. Ceratocystis paradoxa
viously discussed.
It appears from the results of the various fermentations that the
modified feeding protocol led to the induction of mono-oxygenase
enzymes. This is the first report of a hydroxylation reaction being
performed by a Ceratocystis or Thielaviopsis sp. The redox enzymes,
however, seem to be more active than the desired mono-oxygen-
3.3.10. Biotransformation of androsta-4,6-diene-3,17-dione (10)
The reduced metabolite, 17b-hydroxyandrosta-4,6-dien-3-one
(32), which was previously discussed was produced.
ases. The 3b-hydroxy-D
5-steroid substrates (1 and 2), under the
3.3.11. Biotransformation of 3a,5a-cycloandrost-6-en-17-one (11)
new feeding protocol, were able to induce different mono-oxygen-
ases, one which effected 7-hydroxylation. Another enzyme carried
out Baeyer–Villiger rearrangement leading to the ring D lactone.
Incubation of 11 did not yield any transformed compounds.
4. Discussion
The 3-keto-D
4-steroids (3 and 4) underwent mainly redox trans-
formations, more specifically Michael type reduction of the C4–
C5 double bond. Redox chemistry was also observed at C-3 and
C-17. The presence of the side chain in progesterone (5), however,
induced a cytochrome P450 enzyme which led to the degradation
of the side chain via Baeyer–Villiger rearrangement, as was seen
with pregnenolone. The presence of the C-11 carbonyl in cortisone
(6) and prednisone (7) apparently inhibited the ring A redox en-
zymes. With substrates containing flattened rings (estrone, 8, and
androsta-4,6-diene-3,17-dione, 10), only reduction of the C-17 ke-
4.1. Fusarium oxysporum var. cubense
The results showed that dehydroepiandrosterone (1) was pre-
dominantly functionalized in the C-7 position. This was similar
to previous results obtained by Wilson [5]. However, in this inves-
tigation oxygen insertion at C-15 was also observed. Bioconversion
of pregnenolone (2) afforded testosterone (3) and 3b,20-dihydr-
oxypregn-5-ene (17), while previous workers had obtained proges-
terone and 3b,7
the
4-3-ketosteroids, testosterone (3) and progesterone (4), gave
similar results to those of Wilson, i.e., both steroids were converted
to 15 -hydroxy metabolites. Additionally, in the previous study
12b,15 -dihydroxypreg-4-ene-3,20-dione was isolated. Wilson
et al. identified two metabolites from both cortisone and predni-
sone, namely, 17 ,20,21-dihydroxypregn-4-ene-3,11-diione (26)
and 17 ,20,21-trihydroxypregna-1,4-diene-3,11,-dione (27),
a-dihydroxypregn-5-en-20-one. Fermentations of
tone occurred. 5
chemistry at C-3 and C-17 apparently by two different enzymes.
The 3 -hydroxyl product, from the same fermentation, may have
a-Androstane-3,17-dione (9) underwent redox
D
a
a
a
been generated by the same enzyme which reduced C-17 carbon-
yls to 17b alcohols. Not surprisingly the cyclosteroid (11) was unaf-
fected by the fungus.
a
a
Acknowledgments
respectively. In the current study only compounds 26 and 27 were
isolated. The fermentation of estrone (7) in our hands had pro-
Partialfunding was providedby the Universityof theWest Indies,
Mona/Inter-American Development Bank (UWI/IDB) programme.
PCP and KPM are grateful to Tanaud International B.V. (Jamaica)
and the University of the West Indies respectively for the granting
of postgraduate scholarships. PCP, KPM and FAR acknowledge the
Chemistry Department for the granting of Departmental Awards
and Tutorial Assistantships. The authors thank Professors Phyllis
Coates Beckford (University of the West Indies, Mona) and Lynne
Sigler (University of Alberta) for identification of the fungi.
duced
1,3,5(10)-triene (28). Wilson et al. had isolated this compound
along with 3,15 -dihydroxyestra-1,3,5(10)-trien-17-one. In the
previous study better yields with the
4-3-keto-steroids were ob-
a single transformed product, 3,17b-dihydroxyestra-
a
D
tained, although the number of metabolites was less. For the
remaining steroids there was an improvement in the yields, but
similarly the number of metabolites found in this study was re-
duced when compared to those of Wilson.
The results from these experiments made it evident that Fusar-
ium oxysporum var. cubense possesses the ability to hydroxylate ste-
roids in the C-7 and C-15 positions. Investigations were performed
using three synthetic steroids to determine the effect of altering
the functionalities in rings A and B on the ability of the fungus to bio-
transform androstanes. It was shown that oxygen insertion at C-15
was not dependent on a carbon–carbon double bond being present
in either ring A or B. It was also not affected by the presence of an ex-
tended conjugation. However, C-7 hydroxylation appeared to occur
only if the position was allylic. With the absence of a binding group
at C-3, no transformation was observed, which indicated that it
might be necessary for acceptance in to the active site of the en-
zymes and, therefore, for bioconversion to occur.
References
[1] Kranz J, Schmutterer H, Koch W. Diseases, Pests and Weeds. Chichester,
UK: Wiley; 1977. p. 203–4.
[2] Deacon JW. Introduction to Modern Mycology. Oxford, UK: Blackwell; 1988. p.
193–5.
[3] Robinson RK. Ecology of Fungi. UK: London University Press; 1967. p. 77–9.
[4] Wilson MR, Gallimore WA, Reese PB. Steroid transformations with Fusarium
oxysporum var. Cubense and Colletotrichum musae. Steroids 1999;64:834–43.
[5] Roy P, Nayak KK, Pandey NK. Characterization of a novel yeast synthesizing
melanin-like pigment. J Gen Microbiol 1989;135:3385–91.
[6] Cheng Q, Kinney KA, Whitman CP, Szaniszlo PJ. Characterization of two
polyketide synthase genes in Exophiala lecanii-corni, a melanized fungus with
bioremediation potential. Bioorg Chem 2004;32:92–108.
[7] Porter RBR, Gallimore WA, Reese PB. Steroid transformations with Exophiala
jeanselmei var. Lecanii-corni and Ceratocystis paradoxa. Steroids
1999;64:770–9.
[8] Cook A. Diseases of Tropical and Subtropical Field, Fiber and Oil Plants. New
York: Macmillan; 1981. p. 379.
4.2. Exophiala jeanselmei var. lecanii corni
Exophiala jeanselmei var. lecanii corni appears to have a 5-
hydroxylase enzyme which is activated by an oxygen functionality
at position 3. A 14-hydroxylase enzyme also seems to be present
and is triggered when 3-keto-D4 steroids with side chains at C-
17 are present. Cortisone (6) and prednisone (7), which both pos-
[9] Mainwaring WI. The mechanism of action of androgens. Monogr Endocrinol
1977;10:1–178.
[10] Matsuo J, Iida D, Yamanaka H, Mukaiyama T. N-tert-Butylbenzenesulfenamide-
catalyzed oxidation of alcohols to the corresponding carbonyl compounds
with N-chlorosuccinimide. Tetrahedron 2003;59:6739–50.
[11] Bowers A. Steroids. CLXIII. Studies in cyanosteroids. II. The C5-cyano analogs
of
dihydrotestosterone
and
dihydroprogesterone.
J
Org
Chem
sess a 17
11 -Hydroxylation was also evident when there were similar
functionalities at C-3 and C-17, while epoxidation occurred in the
a-hydroxyl group were not hydroxylated, however.
1961;26:2043–7.
a
[12] Agnello EJ, Laubach GD. The dehydrogenation of corticosteroids with chloranil.
J Am Chem Soc 1960;82:4293–9.

1330
P.C. Peart et al. / Steroids 76 (2011) 1317–1330
[13] Rao PN. Manganese dioxide oxidation: the optional introduction of
bond with simultaneous cleavage of dihydroxyacetone or 17, 20-glycol side
chains in
4-3-ketosteroids. J Org Chem 1961;26:2149–52.
D
6-double
[27] Li H, Liu H, Ge W, Huang L, Shan L. Synthesis of 7a-hydroxy-
dehydroepiandrosterone and 7b-hydroxy-dehydroepiandrosterone. Steroids
2005;70:970–3.
[28] Fieser LF, Dominguez XA. Configuration of steroid bromoketones. II. 4b-
D
[14] Sondheimer F, Amendolla C, Rosenkranz G. Steroids. LI.1
Am Chem Soc 1953;75:5932–5.
D
4,6-Dien-3-ones. J
Bromotestane-17b-ol-3-one acetate and 2b-bromocholestane-3-one.
J Am
[15] Hanson JR, Hitchcock PB, Reese PB, Truneh A. Steroidal allylic and homoallylic
rearrangements during halogenation with triphenylphosphine and carbon
tetrachloride. J Chem Soc Perkin Trans I 1988:1469–72.
Chem Soc 1953;75:1704–7.
[29] Tweit RC, Dodson RM. Thiosteroids. III. The transannular addition of hydrogen
disulfide to the steroid A ring. J Am Chem Soc 1959;81:4409–12.
[16] Butenandt A, Grosse W. Über eine einfache Darstellung des aus Männerharn
isolierten Chlorketons C19H27OCl (Dehydro-androsteryl-chlorids). Ber Chem
Ges 1936;69:2776–8.
[30] Hanson JR, Liman DM, Nagaratnam S. Regio- and stereo-chemical effects in the
hydroboration of
1997;8:282–3.
D
2-steroidal allylic and homoallylic alcohols. J Chem Res (S)
[17] Butenandt A, Suranyi LA. Über einige Derivate des i-Androstans. Chem Ber
1942;75B:591–7.
[31] Hu S-H, Genain G, Azerad R. Microbial transformation of steroids: contribution
to 14 -hydroxylations. Steroids 1995;60:337–52.
a
[18] Choudhary MI, Sultan S, Yaqoob M, Musharraf SG, Yassin A, Shaheen F, et al.
Microbial transformation of cortisol and prolyl endopeptidase inhibitory
activity of its transformed products. Nat Prod Res 2003;17:215–20.
[19] Silvestre SM, Salvador JAR. Allylic and benzylic oxidation reactions with
sodium chlorite. Tetrahedron 2007;63:2439–45.
[32] Kirdani RY, Slaunwhite Jr W. The epimeric 6-hydroxyestrones. J Org Chem
1961;26:2148–9.
[33] Caglioti L, Cainelli G, Maina G, Selva A. Hydroboration. V: a new synthesis of
androsterone and 19-norandrosterone. Tetrahedron 1964;20:957–61.
[34] Blunt JW, Clark IM, Evans JM, Jones ERH, Meakins GD, Pinhey JT.
ˇ
´
[20] Pouzar V, Cerny I, Hill M, Bicikova M, Hampl R. Derivatives of 16
a-hydroxy-
Microbiological hydroxylation of steroids. Part III.
A
convenient,
dehydroepiandrosterone with an additional 7-oxo or 7-hydroxy substituent:
synthesis and gas chromatography/mass spectrometry analysis. Steroids
2005;70:739–49.
microbiological route to 15-oxygenated
1971:1136–8.
[35] Bell AM, Browne JW, Denny WA, Jones ERH, Kasal A, Meakins GD.
Microbiological hydroxylation of steroids. Part VI. Hydroxylation of simple
5a-androstanes. J Chem Soc (C)
[21] Faramarzi MA, Yazdi MT, Jahandar H, Amini M, Monsef-Esfahani HR. Studies
on the microbial transformation of androst-1, 4-dien-3, 17-dione with
Acremonium strictum. J Ind Microbiol Biotechnol 2006;33:725–33.
[22] Vatèle J-M. 2-(Prenyloxymethyl)benzoyl (POMB) group: a new temporary
protecting group removable by intramolecular cyclization. Tetrahedron
2007;63:10921–9.
mono- and di-oxygenated 5
a-androstanes and of 3-oxoestranes with the
fungus Aspergillus ochraceus. J Chem. Soc Perkin Trans I 1972:2930–6.
[36] Cambie RC, Thomas PW, Hanson JR. Epoxidation of 3
en-17-one. J Chem Soc Perkin Trans I 1975:323–6.
[37] Takai S. Studies on Cerato-ulmin, a wilt toxin of Ceratocystis ulmi (Buism.).
Phytopathol Z 1978;91:147–58.
a, 5-cyclo-5a-androst-6-
ˇ
´
[23] Cerny I, Fajkoš J, Pouzar V. Chemical synthesis of 15b-hydroxytestosterone and
its derivatives using a (4-methoxyphenyl)methyl protecting group. Steroids
1996;61:58–64.
[38] Kolek T, Szpineter A, Swizdor A. Baeyer–Villiger oxidation of DHEA,
pregnenolone, and androstenedione by Penicillium lilacinum AM111. Steroids
2008;73:1441–5.
[24] Bartmanska A, Dmochowska-Gladysz J, Huszcza E. Steroids’ transformations in
Penicillium notatum culture. Steroids 2005;70:193–8.
[39] Blunt JW, Stothers JB. 13C NMR spectra of steroids – a survey and commentary.
Org Magn Res 1977;9:439–64.
[25] Pollio A, Pinto G, Greca MD, Fiorentino A, Previtera L. Biotransformations of
progesterone by Chlorella spp.. Phytochemistry 1996;42:685–8.
[26] Lamm AS, Chen ARM, Reynolds WF, Reese PB. Steroid hydroxylation by
Whetzelinia sclerotiorum, Phanerochaete chrysosporium and Mucor plumbeus.
Steroids 2007;72:713–22.