New biologically active epidioxysterols from Stereum hirsutum

Article abstract of DOI:10.1016/j.bmcl.2007.08.072

From the fungus Stereum hirsutum have been isolated and identified two new epidioxysterols 1, 4, together with two known ones 2 and 3. Their structures were elucidated on the basis of spectroscopic analysis and chemical reactions. Epidioxysterols 1-4 have been shown to possess a significant activity against Mycobacterium tuberculosis.

Full text of DOI:10.1016/j.bmcl.2007.08.072

Bioorganic & Medicinal Chemistry Letters 17 (2007) 6330–6334
New biologically active epidioxysterols from Stereum hirsutum
Francesca Cateni,a,* Bojan Doljak,b Marina Zacchigna,a Marko Anderluh,c
Andrej Piltaver,d Giuditta Scialinoe and Elena Banfie
aDepartment of Pharmaceutical Sciences, University of Trieste, P.zle Europa 1, 34127 Trieste, Italy
bDepartment of Pharmaceutical Biology, Faculty of Pharmacy, University of Ljubljana, Ljubljana, Slovenia
cDepartment of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Ljubljana, Ljubljana, Slovenia
dInstitute for the Systematics of Higher Fungi, Zofke Kvedrove 24, Ljubljana, Slovenia
eDepartment of Biomedical Sciences, Microbiology Section, University of Trieste, Via A. Fleming 22, 34127 Trieste, Italy
Received 16 July 2007; revised 29 August 2007; accepted 30 August 2007
Available online 2 September 2007
Abstract—From the fungus Stereum hirsutum have been isolated and identified two new epidioxysterols 1, 4, together with two
known ones 2 and 3. Their structures were elucidated on the basis of spectroscopic analysis and chemical reactions. Epidioxysterols
1–4 have been shown to possess a significant activity against Mycobacterium tuberculosis.
Ó 2007 Elsevier Ltd. All rights reserved.
Stereum hirsutum is a basidiomycete involved in esca,
one of the most destructive diseases in grapevine1 and
seems to play an important role in the wood deteriora-
tion process.2 From the culture medium of this fungus,
different new acetylenic compounds,3 tricyclic sesquiter-
penes,4 chromene, and aromatic aldehydes derivatives5
have been isolated and identified.
sources have been described for antimicrobial activity,7,8
we studied the antibacterial, antifungal, and antituber-
cular activity of the complex mixture and of the pure
epidioxysterols and detected a significant activity against
Mycobacterium tuberculosis.
This paper describes the structural elucidation of these
compounds and the antitubercular activity against
M. tuberculosis.
Recently, we reported the isolation and biological inves-
tigation of active principles from S. hirsutum on the ba-
sis of inhibitory potency on thrombin.6 A bioassay
oriented fractionation of the extract of S. hirsutum has
led to the isolation of complex mixtures of diacylglyce-
rophospholipids (DAGPs) and diacylglycerols (DGs)
responsible for the activity of the extract. Moreover,
docking studies on thrombin were performed in order
to clarify the binding mode of some isolated com-
pounds. In the course of our investigations on biologi-
cally active chemical substances from S. hirsutum,
preliminary experiments indicated that most of the anti-
microbial activity was associated with a fraction con-
taining steroid derivatives; this was examined in detail
and four epidioxysterols (1–4) (Fig. 1) were isolated
and identified. Since few epidioxysterols from different
Stereum hirsutum was collected in September 2005 in
northeast Slovenia. Within 24 h of collection, it was
dried in an air-flow chamber at 30–35 °C and then
stored at ꢀ20 °C. A mushroom specimen was deposited
for evidence. The freshly frozen mushroom (1 kg) was
disrupted into small pieces and homogenized with
100 mL of 50% (v/v) methanol. The homogenate was
exposed to ultrasound for 10 min, macerated at room
temperature for 12 h, and again exposed to ultrasound
for 10 min.
Dry extract (10 g) was redissolved in methanol, mixed
with 5 g silica gel, and evaporated in vacuo. The dry ex-
tract was chromatographed on silica gel using CH2Cl2,
CH2Cl2/MeOH (5:1), and MeOH to give three fractions
[F-1: 1.8 g (18%); F-2: 2.05 g (20.5%); F-3: 4.85 g
(48.5%)]. The fraction eluted with CH2Cl2 was concen-
trated in vacuo and submitted to column chromatogra-
phy over silica gel using CH2Cl2/n-hexane (9:1) and
CH2Cl2/MeOH (5:0.5) as eluents to afford two fractions
Keywords: Stereum hirsutum; Epidioxysterols; Mycobacterium tuber-
culosis; Antitubercular activity.
*
Corresponding author. Tel.: +39 040 5583722; fax: +39 040 52572;
e-mail: cateni@units.it
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.08.072

F. Cateni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6330–6334
6331
21
22
from the observation of a quasi-molecular ion-peak in
high-resolution (HR) FAB-MS. In the EI-MS spectrum
of 1 the molecular ion at m/z 412 was observed, while the
fragments at m/z 394 and 380 were due to the loss of
H2O and O2 from the molecular ion, respectively. The
fragment at m/z 300 was due to the loss of the side chain
C8H15 from the molecular ion. IR absorption bands at
3380, 2962, and 1650 cmꢀ1 indicated the presence of hy-
droxyl group, alkyl moiety, and double bonds,
respectively.
24
26
20
18
23
12
25
H
11
17
13
14
19
27
16
1
9
6
2
15
10
8
H
7
O
5
3
HO
4
O
1
Detailed 1H and 13C NMR data for compounds 1–4 are
available as a Supplementary Table. 1H NMR spectrum
of 1 showed signals due to five methyl groups: three sec-
ondary methyls [dH 1.03, 3H, d, J = 6.6 Hz (H3-21); dH
0.85, 3H, d, J = 6.6 Hz (H3-26); dH 0.87, 3H, d,
J = 6.6 Hz (H3-27)] and two tertiary methyls [dH 0.82,
28
21
22
26
20
24
18
23
12
25
H
11
17
15
1
13
14
19
3H, s (H3-18); dH 1.11, 3H, s (H3-19)]. The H NMR
27
16
aided spectrum, with the 13C NMR spectral data, sug-
gested the presence of three olefins (two disubstituted
and one trisubstituted ones). Two-dimensional NMR
spectra [1H–1H COSY, 1H-detected heteronuclear multi-
ple quantum coherence (HMQC), and HMBC] indi-
cated that one double bond is located between the C-6
and C-7 positions, one between the C-9 and C-11 posi-
tion, and the remaining one between the C-22 and C-
23 positions. Olefin protons [dH 6.60, 1H, d (H-7); dH
6.27, 1H, d (H-6)] with cis-coupling (J = 8.5 Hz), to-
gether with two oxygenated quaternary carbons on
C-5 (dC 82.2) and C-8 (dC 78.3) were suggestive of the
presence of a peroxide structure. The HMBC spectrum
1
9
2
10
8
H
7
O
5
3
HO
6
4
O
2
28
21
22
26
20
24
18
23
12
25
H
17
11
13
14
19
27
16
1
9
1
of 1 afforded long-range H–13C correlations shown in
2
15
10
8
Figure 2.
H
H
7
O
5
3
HO
6
The relative configuration of 1 was determined based on
the following evidence. NOESY analysis focused on the
ring junctures of the steroidal skeleton confirmed that 1
4
O
3
29
28
21
22
26
20
24
18
CH3
23
12
25
H
17
11
13
14
19
CH3
27
16
1
9
H
H
2
H
15
10
8
H
H
H
HO
7
O
O
5
3
HO
H
H
H
6
4
O
O
4
Figure 2. Selected HMBC correlations for 1 and 2.
Figure 1. Chemical structures of epidioxysterols 1–4.
H
18
12
19
17
[F-4: 250 mg (13.9%); F-5: 1.38 g (76.7%)]. Successive
purification of F-4 by reverse-phase flash chromatogra-
phy, using MeOH, followed by reverse-phase RP-18 pre-
parative TLC, using CH3CN/MeOH (7:3) afforded
almost pure 1 [40 mg, 16% (purity 94%)]; 2 [60 mg,
24% (purity 94%)]; 3 [36 mg, 14.4% (purity 92%)]; 4
[50 mg, 20% (purity 98%)], respectively.
H
H
H3C
H
H
2
H
1
H
14
6
3
7
4
H
5
8
HO
H
O
O
NOE
Compound 1 was obtained as colorless amorphous solid
and was shown to have the molecular formula C27H40O3
Figure 3. Important NOE correlations and long-range coupling (bold)
for 1 and 2.

6332
F. Cateni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6330–6334
has the same relative configuration as that of 2 (Fig. 3).
The NOE correlation between H-14 and H-12a, which
showed long-range coupling (W-shape) with the angular
methyl group (H-18) in the COSY spectrum, as shown
in the bold lines in Figure 3, exhibited the trans config-
uration between H-14 and H-18. The relative configura-
tion of H-17 was assigned as an a-orientation due to the
NOE correlation between H-12 a and H-17. The a-ori-
entation of the epidioxy group was determined by the
NOE correlations between H-18 and H-7 (olefinic pro-
ton). The relative configuration of the angular methyl
group at C-10 was indicated by the NOE correlation be-
tween H-6 (olefinic proton) and H-19, and the stereo-
chemistry at C-3 was deduced from the J values of H-
3 (J = 5.0, 11.2 Hz).
due to the loss of O2 and of the side chain C7H13 from
the molecular ion. IR absorption bands at 3379, 2960,
and 1650 cmꢀ1 indicated the presence of hydroxyl
group, alkyl moiety, and double bonds, respectively.
1H NMR spectrum of 2 showed signals due to six
methyl groups: four secondary methyls [dH 1.03, 3H,
d, J = 6.6 Hz (H3-21); dH 1.09, 3H, d, J = 6.8 Hz (H3-
28); dH 0.85, 3H, d, J = 6.6 Hz (H3-26); dH 0.87, 3H,
d, J = 6.6 Hz (H3-27)] and two tertiary methyls [dH
0.82, 3H, s (H3-18); dH 1.11, 3H, s (H3-19)]. The chem-
ical shift values of carbons in 2 were essentially identical
to those of 1, except for C-24. The side chain was as-
1
signed through the H–1H COSY from H3-20 to H3-
28. HMBC correlations of the olefinic methine proton
H-22 (d 5.22 ppm) with C-20; C-23 and C-24 and those
of H-23 (d 5.26 ppm) with C-20; C-22; C-24; confirmed
these connectivities. The nature of the double bond
was assumed to be trans by the coupling constant value
J = 15.2 Hz observed between H-22 and H-23. The ste-
reochemistry of C-20 chiral center in the side chain
was assigned by NOESY analysis while the configuration
Compound 2 was obtained as a white amorphous solid
and was shown to have the molecular formula
C28H42O3 from the observation of a quasi-molecular
ion-peak in high-resolution (HR) FAB-MS. In the EI-
MS spectrum of 2 the molecular ion at m/z 426 was
observed, while the fragments at m/z 394 and 329 were
Scheme 1.

F. Cateni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6330–6334
6333
of the methyl group at C-24 was assumed to be R by
comparison of the C-26, C-27, and C-28 proton chemi-
cal shifts with those observed for the epimer.9 The chem-
ical conversions from 2 to the known diol 5 were carried
out to establish the absolute configuration of 2, as
shown in Scheme 1.10,11 Catalytic hydrogenation of 2
Compound 4 was obtained as a white amorphous solid
and was shown to have the molecular formula
C29H46O3 from the observation of a quasi-molecular
ion-peak in high-resolution (HR) FAB-MS. In the EI-
MS spectrum of 4 the molecular ion at m/z 443 was
observed, while the fragments at m/z 411 and 410 were
due to the loss of O2 and H2O from the molecular ion,
respectively. The fragment at m/z 303 was due to the loss
of the side chain C10H19 from the molecular ion. IR
absorption bands at 3380, 2962, and 1645 cmꢀ1 indi-
cated the presence of hydroxyl group, alkyl moiety,
and double bonds, respectively.
in the presence of 10% palladium on carbon in ethyl ace-
25
tate gave 5 {½aꢁ +13.5° (c 0.05, CHCl3)}. The authentic
D
25
diol 5 {½aꢁ +10.4°} was independently prepared from
D
ergosterol in two steps with photosensitized oxidation,11
in which the desired epidioxysterol 6, identical to 2, was
obtained as a minor product, followed by catalytic
hydrogenation. The spectral data of 5 derived from 2
were identical to those of authentic 5, including the sign
of its optical rotation value. These findings together with
the spectral analysis of 6 confirmed the structure of 2.
1H and 13C NMR spectra of 4 were quite similar to
those of 3, suggesting that differences between 3 and 4
were due to the presence of an ethyl group in 4 instead
of a methyl group at the C-24 position. The relative con-
figuration of 3 and 4 was determined on the basis of
NOESY analysis as reported above for compound 1.
Compound 3 was obtained as a white amorphous solid
and was shown to have the molecular formula
C28H45O3 from the observation of a quasi-molecular
ion-peak in high-resolution (HR) FAB-MS. In the EI-
MS spectrum of 3 the molecular ion at m/z 428
(MꢀH)+ was observed, while the fragment at m/z 410
was due to the loss of H2O from the molecular ion. IR
absorption bands at 3381, 2962, and 1644 cmꢀ1 indi-
cated the presence of hydroxyl group, alkyl moiety,
and double bonds, respectively.
All the isolated compounds 1–4 were evaluated for anti-
bacterial activity against a set of Gram-positive and
Gram-negative reference strains, Staphylococcus aureus
ATCC 25923, Escherichia coli ATCC 25922, Pseudomonas
aeruginosa ATCC 27753; for antifungal activity against
11 Candida spp. clinical isolates; for antitubercular
activity against the reference strain M. tuberculosis
H37Rv. Antimicrobial activity was always evaluated
by reference methods.12–14 Ciprofloxacin was chosen as
a standard in both antibacterial and antitubercular
activity measurements, as it is an antibiotic employed
in the treatment of a wide range of infections and as it
is a drug known to have an excellent activity against
most Gram-negative and Gram-positive bacteria. Mico-
nazole was chosen as a standard in antifungal activity
measurements. Antitubercular activity was evaluated
by MRA, a recently developed, one-week duration, mi-
cro-dilution Resazurin assay.12 The minimum inhibitory
concentration, MIC, was defined as the lowest drug con-
centration that prevented Resazurin color change from
blue to pink and was determined by visual inspection
twice in duplicate experiments; viable counting from con-
trol wells and from test wells performed into agar plates
confirmed bactericidal and bacteristatic activity of the
compounds. Isoniazid and Rifampicin were always in-
cluded as a standard in antitubercular activity measure-
ments, having a MIC of 0.05 lg/mL and of 0.1 lg/mL,
respectively. The antimicrobial activity of F-4 and of the
pure compounds 1–4 is reported in Table 1.
1H NMR spectrum of 3 showed signals due to six
methyl groups: four secondary methyls [dH 1.03, 3H,
d, J = 6.6 Hz (H3-21); dH 1.09, 3H, d, J = 6.8 Hz (H3-
28); dH 0.85, 3H, d, J = 6.6 Hz (H3-26); dH 0.87, 3H,
d, J = 6.6 Hz (H3-27)] and two tertiary methyls [dH
1
0.82, 3H, s (H3-18); dH 1.11, 3H, s (H3-19)]. H NMR
aided spectrum, with 13C NMR spectral data, suggested
the presence of two disubstituted olefins. Olefin protons
[dH 6.60, 1H, d (H-7); dH 6.27, 1H, d (H-6)] with cis-cou-
pling (J = 8.5 Hz), together with two oxygenated quater-
nary carbons on C-5 (dC 82.0) and C-8 (dC 79.3) were
suggestive of the presence of a peroxide structure. The
1H–1H COSY spectrum of 3 revealed the following cor-
relations for H-1-H-4, H-6-H-7, H-9-H-12, H-14-H-22,
and H-24-H-28. The HMBC spectrum of 3 afforded
1
long-range H–13C correlations shown in Figure 4. Be-
sides, the photosensitized oxidation of ergosterol led to
a major product which spectral data were identical to
those of 3 (Scheme 1).
No antibacterial nor antifungal activity was detected.
Epidioxysterols 1 and 2 exhibited a killing activity with
MIC of 16 lg/mL against M. tuberculosis H37Rv refer-
ence strain, while compounds 3 and 4 exhibited a MIC
of 64 lg/mL.
CH3
CH3
From the obtained data, it is possible to deduce that kill-
ing activity of the new epidioxysterols is specific for
M. tuberculosis and that the MIC of F-4, containing
the complex mixture of epidioxysterols 1–4, is not due
to a synergic effect of the isolated compounds 1–4. 5a,
8a-Epidioxysterols were previously isolated from an edi-
ble mushroom Lentinula edodes15 and from sea hare
H
H
H
H
H
HO
O
H
H
H
O
Figure 4. Selected HMBC correlations for 3 and 4.

6334
F. Cateni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6330–6334
Table 1. Antimicrobial activity (MIC, lg/mL) of fraction 4 and of purified epidioxysterols 1–4; reference drugs are also reported
Mycobacterium
Staphylococcus
Escherichia coli
Pseudomonas aeruginosa
Candida spp.
(11 strains)
tuberculosis H37Rv
aureus ATCC 25923
ATCC 25922
ATCC 27753
F-4
1
16
16
16
64
64
0.5
—
>512
>512
>512
>512
>512
0.5
>512
>512
>512
>512
>512
0.25
—
>512
>512
>512
>512
>512
1
>512
>512
>512
>512
>512
—
2
3
4
Ciprofloxacin
Miconazole
—
—
0.25–32
Aplysia juliana.16 To our knowledge, compounds 2 and
3 have been isolated and identified from Pellia epiphylla
(L.) Corda, a common thalloid European liverwort,
while compounds 1 and 4 are new.17 Nevertheless, this
is the first example of the isolation and structure elucida-
tion of epidioxysterols from S. hirsutum.
6. Doljak, B.; Cateni, F.; Anderluh, M.; Procida, G.; Zilic, J.;
Zacchigna, M. Drug Dev. Ind. Pharm. 2006, 32, 635.
7. Abourriche, A.; Charrouf, M.; Chaib, N.; Bennamara, A.;
Bontemps, N.; Francisco, C. Il Farmaco 2000, 55, 492.
8. Saludes, J. P.; Garson, M. J.; Franzblau, S. G.; Agui-
naldo, A. M. Phytother. Res. 2002, 16, 683.
9. Rubinstein, I.; Goad, L. J.; Clague, A. D. H.; Mulheirn, L.
J. Phytochemistry 1976, 15, 195.
10. Hallsworth, A. S.; Heubest, H. B.; Wrigley, T. I. J. Chem.
Soc. 1957, 1969.
Supplementary data
11. Iguchi, K.; Shimura, H.; Yang, Z.; Yamada, Y. Steroids
1993, 58, 410.
12. Banfi, E.; Scialino, G.; Monti-Bragadin, C. J. Antimicrob.
Chemother. 2003, 52, 796.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2007.
08.072.
13. Barry, A. L. In Antibiotics in Laboratory Medicine;
Lorian, V., Ed.; Williams & Wilkins: Baltimore, 1980, p 1.
14. National Committee for Clinical Laboratory Standards,
Reference method for broth dilution antifungal susceptibility
testing of yeast. Approved standard M-27. NCCLS,
Villanova, PA, USA, 1997.
15. Yaoita, Y.; Amemiya, K.; Ohnuma, H.; Furumura, K.;
Masaki, A.; Matsuki, T.; Kikuchi, M. Chem. Pharm. Bull.
1998, 46, 944.
References and notes
1. Chiarappa, L. Phytopathology 1959, 49, 510.
2. Larignon, P.; Dubas, B. Eur. J. Plant Pathol. 1997, 103, 147.
3. Dubin, G. M.; Fkyerat, A.; Tabacchi, R. Phytochemistry
2000, 53, 571.
4. Yun, B. S.; Lee, I. K.; Cho, Y.; Cho, S. M.; Yoo, I. D.
J. Nat. Prod. 2002, 65, 786.
16. Miyamoto, T.; Honda, M.; Sugiyama, S.; Higuchi, R.;
Komori, T. Liebigs Ann. Chem. 1988, 589.
17. Cullmann, F.; Becker, H. Phytochemistry 1988, 47, 237.
5. Yun, B. S.; Cho, Y.; Lee, I. K.; Cho, S. M.; Lee, T. H.;
Yoo, I. D. J. Antibiot. 2002, 55, 208.