Cytotoxic Illudane Sesquiterpenes from the Fungus Granulobasidium vellereum (Ellis and Cragin) Jülich

Article abstract of DOI:10.1021/acs.jnatprod.5b00500

Eight illudane sesquiterpenes were obtained from the wood-decomposing fungus Granulobasidium vellereum (Ellis and Cragin) Jülich; among them were the enantiomers of the known compounds illudin M (1) and dihydroilludin M (4) and the diastereomers of illudin M (2) and illudin S (3), as well as two previously undescribed illudanes (5, 6). The cytotoxicity of compounds 1-4 and 6 was evaluated against two tumor cell lines (Huh7 and MT4), which showed that compounds 1-3 had potent cytotoxic activity, whereas compounds 4 and 6 had no or only moderate effects at concentrations up to 400 μM. Surprisingly, both compounds 2 and 3 were about 10 times more potent than 1. When the chemical reactivity of 1 and 2 was tested, compound 2 was shown to have a substantially higher reaction rate when reacted both with 2 M HCl and with cysteine, indicating that the difference in cytotoxicity is probably due to chemical reactivity and not to enzymatic affinity.

Full text of DOI:10.1021/acs.jnatprod.5b00500

Article
pubs.acs.org/jnp
Cytotoxic Illudane Sesquiterpenes from the Fungus Granulobasidium
̈
vellereum (Ellis and Cragin) Julich
Christina Nord,† Audrius Menkis,‡ and Anders Broberg*,†
†Department of Chemistry and Biotechnology, Uppsala BioCenter, Swedish University of Agricultural Sciences, P.O. Box 7015,
SE-75007, Uppsala, Sweden
‡Department of Forest Mycology and Plant Pathology, Uppsala BioCenter, Swedish University of Agricultural Sciences, P.O. Box
7026, SE-75007, Uppsala, Sweden
S
* Supporting Information
ABSTRACT: Eight illudane sesquiterpenes were obtained from the
wood-decomposing fungus Granulobasidium vellereum (Ellis and
Cragin) Julich; among them were the enantiomers of the known
̈
compounds illudin M (1) and dihydroilludin M (4) and the
diastereomers of illudin M (2) and illudin S (3), as well as two
previously undescribed illudanes (5, 6). The cytotoxicity of compounds
1−4 and 6 was evaluated against two tumor cell lines (Huh7 and MT4),
which showed that compounds 1−3 had potent cytotoxic activity,
whereas compounds 4 and 6 had no or only moderate effects at
concentrations up to 400 μM. Surprisingly, both compounds 2 and 3
were about 10 times more potent than 1. When the chemical reactivity
of 1 and 2 was tested, compound 2 was shown to have a substantially
higher reaction rate when reacted both with 2 M HCl and with cysteine,
indicating that the difference in cytotoxicity is probably due to chemical
reactivity and not to enzymatic affinity.
against several types of cancer before being canceled in 2012
due to lack of efficacy.9
ungi from the phylum Basidiomycotina, one of the five
Fsubdivisions of Eumycota (true fungi), have been found to
produce a large variety of secondary metabolites, of which
many are of terpenoid origin. Most common among the
terpenes are the sesquiterpenes, which in the basidiomycetes
are mainly formed via the humulene protoilludane biosynthetic
pathway.1,2 Through rearrangements of the protoilludane
carbon skeleton a number of different classes of compounds
are formed. Some of the skeletal types have been described to
possess interesting biological properties, of which the illudane
skeletal type with its characteristic highly electrophilic
spirocyclopropyl moiety has been of particular interest due to
their high cytotoxicity.2
The naturally occurring illudanes illudin M and S were
originally obtained from the poisonous Jack-o’-lantern mush-
room Omphalotus olearius (prev. Clitocybe illudens).3 They have,
due to their extreme cytotoxicity, been the subject of numerous
studies in order to determine their mechanism of action. They
were found to be able to initiate DNA breakage in the cell
through a two-step reaction initiated by a Michael-type addition
of a thiol bionucleophile followed by a second nucleophilic
attack on the spirocyclopropane group.4−6 Unfortunately
illudins M and S have turned out to be too toxic for clinical
use against different cancer types. Instead the work has been
focused on finding analogues with more beneficial therapeutic
windows.6,7 The most successful was an analogue of illudin S
named irofulven,8 which was subjected to phase III clinical trials
Granulobasiodium vellereum (Ellis & Cragin) Julich is a
̈
saprotrophic wood decay fungus from which we have
previously obtained a large variety of sesquiterpenoid
metabolites,10−12 of which two have shown potent cytotoxic
activity.12,13 This paper describes work to continue the
identification and characterization of secondary metabolites
from G. vellereum and to assess their cytotoxic effects.
RESULTS AND DISCUSSION
■
From the wood-decomposing fungus G. vellereum eight illudane
sesquiterpenes were isolated through a combination of solid-
phase extraction (SPE) and chromatographic techniques. Their
respective structures were elucidated with spectroscopic
techniques, and the cytotoxicity of compounds 1−4, 6, and
7,9-illudadiene-3,14-diol was evaluated against Huh7 and MT4
tumor cell lines.
Compound 1 had a molecular formula of C15H20O3
according to HRESIMS analysis. The 1H NMR data of
compound 1 were identical to the literature data of illudin
M,4,14 but the specific rotation of compound 1 (96°) did not
match that reported for illudin M (−284°).3 This indicates that
compound 1 might be an enantiomer of illudin M, and to
Received: June 5, 2015
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
through the oxidation of methyl-11 in illudin M to a CH2OH
group in illudin S. The HMBC bonding pattern of 3 was also
identical to what could be expected from illudin S, making it
likely that they are diastereomers.
The relative configuration of C-6 and C-7, though not of C-3,
could be deduced through diagnostic ROESY correlations
between H-7 and H2-11 (Figure 2). To be able to perform the
Mosher’s test on compound 3, the primary hydroxy group
CH2OH was selectively acetylated, since it might otherwise
react with the chiral Mosher’s reagent [α-methoxy-
(trifluoromethyl)phenylacetyl chloride] and interfere with the
result of the test. The resulting acetylated product (3a) was
then subjected to the Mosher’s test,15 from which the absolute
configuration of C-6 and C-7 then could be deduced as 6S,7R.
If compound 3 had a 3R configuration, it would be identical to
illudin S, which according to NMR and polarimetric data it is
not,3,18 and consequently the absolute configuration of
compound 3 must be 3S,6S,7R. Compound 3 was named
(3S,6S,7R)-illudin S.
According to HRESIMS analysis, compound 4 had a
molecular formula of C15H22O3. The 1H NMR data were
identical to those described for dihydroilludin M,14 another
illudane sesquiterpene also originally isolated from O. olearius.19
The specific rotation of compound 4 was +77° compared to
−35° for dihydroilludin M,19 indicating that compound 4 was
the enantiomer of dihydroilludin M. To further verify this
hypothesis, compound 4 was selectively oxidized at position C-
4.16 The 1H NMR and specific rotation of the oxidized product
were identical to those of compound 1 (the enantiomer of
illudin M), and accordingly it was concluded compound 4 was
the enantiomer of dihydroilludin M and named (3S,4S,7R)-
dihydroilludin M.
The molecular composition of compound 5 was determined
to be C15H22O2 from HRESIMS analysis, resulting in an
unsaturation index of five. Four sp2 carbons were identified
from the NMR data, indicating that 5 had a tricyclic structure.
The NMR data also showed that the structure included four
methyl groups, two quaternary carbons, two oxygen-linked
tertiary carbons, and three CH2 groups. From COSY and
HSQC data a −CH2(1)−CH2(2)− spin system was identified.
HMBC showed that both Me-10 and Me-11 had correlations to
C-6, C-7, and a protonated sp2-carbon (C-5). Both H-5 and H2-
7 had correlations to an sp2 carbon (C-4a) and an oxygen-
linked tertiary carbon (C-7a), forming the cyclopentene moiety
of the compound (Figure 3). C-7a also had a correlation to Me-
12, which in turn correlated to C-8 and C-2a. The latter
correlated also to both ends of the spin system, as well as to
Me-9. Me-9 correlated to two sp2 carbons (C-3 and C-4).
Finally H-4 correlated to C-4a and C-5, resulting in the
proposed structure of 5, which was named illudadiene A. The
relative configuration of 5 was determined through diagnostic
ROESY correlations (in DMSO-d6) between Me-12 and OH-7a
and from the latter to H-7α. H-7β had a ROESY cross-peak to
OH-8, giving the relative configuration of compound 5. If it
investigate this, the Mosher’s test was performed on compound
1.15 From the shift differences between the S-MTPA and R-
MTPA monoesters of 1 (Figure 1) it could be deduced that
compound 1 had a 3S,7R configuration and indeed was the
enantiomer of illudin M (3R,7S); compound 1 was named
(3S,7R)-illudin M.
According to HRESIMS analysis, compound 2 had a
molecular composition of C15H20O3. The NMR data for
compound 2 (Table 1) closely resembled those of illudin M
and compound 1. The HMBC spectrum showed that the
bonding pattern of compound 2 was identical to that of
compound 1, but since the 1D NMR data were not identical,
the structures must be diastereomeric. When stored over
extended periods of time (−18 °C, >1 year), compound 2
decomposed and formed a variety of degradation products, of
which the most abundant was the highly oxygenated compound
2a. Compound 2a had the same HMBC correlations as would
be expected for illudin B and illudin H,16,17 two illudane
sesquiterpenes obtained from the fungi Omphalotus olearius and
1
Omphalotus nidiformis, but the H NMR data of 2a did not
match those reported for these compounds,16,17 indicating that
compound 2a, illudin B, and illudin H are diastereomeric
compounds. The relative configuration of compound 2a could,
unlike that of compound 2, be determined from ROESY data
(Figure 2), displaying diagnostic correlations between Me-9
and H-5 and from the latter to Me-11. H-7 had correlations to
both Me-11 and Me-12, resulting in the relative configuration
of compound 2a. The relative configuration of compound 2 can
also be deduced from these data if it is assumed that the
configuration of C-3 and C-7 is retained in 2a compared to
compound 2 during the degradation process. To be able to
determine the absolute configuration of compound 2, the
Mosher’s test was performed,15 which demonstrated the
configuration to be 3S,7S for compound 2 (Figure 1).
Compound 2 was consequently named (3S,7S)-illudin M.
The molecular formula of compound 3 was determined to be
1
C15H20O4 from HRESIMS analysis. The H NMR data of
compound 3 (Table 1) closely resembled but were not identical
to those of illudin S,18 which differs from illudin M only
Figure 1. Chemical shift differences (in ppm) between the S-MTPA monoesters and the R-MTPA monoesters of 1, 2, and 3a (pyridine-d5).
B
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
1
Table 1. H and 13C NMR Spectroscopic Data (600 MHz H NMR and 150 MHz 13C NMR, MeOH-d4) for (3S,7S)-Illudin M
(2), (3S,6S,7R)-Illudin S (3), and Illudadienes A (5) and B (6)
2
3
5
6
pos.
1
δC, mult.
δH (J in Hz)
δC, mult.
δH (J in Hz)
δC, mult.
δH (J in Hz)
δC, mult.
δH (J in Hz)
9.1, CH2 1.08, ddd (9.9, 6.2, 4.0)
0.46, ddd (9.6, 6.8, 4.0)
5.9, CH2 0.99, ddd (9.6, 6.2, 4.6)
0.78, ddd (9.9, 6.8, 4.6)
33.5, C
9.0, CH2 1.08, ddd (9.9, 6.0, 3.8)
0.46, ddd (9.5, 6.7, 3.8)
5.9, CH2 0.99, ddd (9.5, 6.0, 4.7)
0.78, ddd (9.9, 6.7, 4.7)
33.5, C
12.3, CH2 0.95, m
0.80, m
9.7, CH2 1.00, ddd (9.7, 5.8, 3.8)
0.46, ddd (9.6, 6.7, 3.8)
6.5, CH2 0.90, ddd (9.6, 5.8, 4.5)
0.80, ddd (9.7, 6.7, 4.5)
33.8, C
2
5.7, CH2 0.89, m
0.81, m
2a
3
29.4, C
77.8, C
77.8, C
139.9, C
143.2, C
4
201.8, C
201.6, C
119.4, CH
139.7, C
137.7, CH
44.5, C
6.06, d (1.3)
5.39, s
120.9, CH
143.9, C
129.4, CH
51.4, C
5.99, d (1.6)
5.15, d (2.6)
4a
5
139.9, C
138.5, C
146.7, CH
49.9, C
6.43, s
4.43, s
143.0, CH
56.3, C
6.39, s
4.68, s
6
7
79.8, CH
75.5, CH
47.7, CH2 2.09, d (13.1)
1.60, d (13.1)
36.2, CH2 1.95, dd (13.1, 7.9)
1.45, dd (13.1, 9.2)
7a
8
136.2, C
135.1, C
140.4, C
134.8, C
87.6, C
54.3, CH
72.8, C
2.99, ddd (9.2, 7.8, 2.6)
75.7, C
9
24.9, CH3 1.27, s
28.4, CH3 1.11, s
21.0, CH3 1.13, s
24.9, CH3 1.29, s
16.2, CH3 1.11, s
69.8, CH2 3.44, d (10.8)
3.38, d (10.8)
19.4, CH3 1.53, d (1.3)
30.3, CH3 1.24, s
29.8, CH3 1.11, s
19.5, CH3 1.52, d (1.6)
24.9, CH3 1.11, s
10
11
70.5, CH2 3.34, m
12
14.3, CH3 1.71, s
14.3, CH3 1.71, s
16.8, CH3 0.94, s
19.7, CH3 1.00, s
Figure 2. Molecular models of compounds 2a, 3, 5, and 6 obtained with MM2 energy minimization displaying diagnostic ROESY correlations used
for structure determination. ROESY data for 2a and 3 were obtained in MeOH-d4, for compound 5 in DMSO-d6, and for compound 6 in acetone-d6.
between the two compounds was that methyl-11 was oxidized
to an CH2OH in compound 6 and that C-8 was oxidized in
compound 5. Consequently compound 6 was named
illudadiene B. The relative configuration of 6 was deduced
through ROESY correlations from H-7α to Me-12 and Me-10
(Figure 2). H2-11 correlated to H-7β and H-7a, giving the
Figure 3. Diagnostic HMBC correlations of compound 5 (MeOH-d4).
relative configuration of compound 6. If it is again assumed that
7,9-illudadiene-3,14-diol, radulol, and compounds 5 and 6 share
the configuration on C-8, the absolute configuration of
assumed that compound 5 is produced by the same enzymatic
machinery as 7,9-illudadiene-3,14-diol and radulol, they are
likely to share the configuration on C-8, resulting in a 7aS,8S
compound 6 would be 5S,7aS,8R.
configuration of 5.
Radulol and 7,9-illudadiene-3,14-diol were identified by
HRESIMS data showed that compound 6 had the molecular
formula C15H22O2, as with compound 5. The 1H and 13C NMR
data of compound 6 were also similar to those of compound 5
comparison of NMR, MS, and polarimetric data with literature
values.13,20,21 7,9-Illudadiene-3,14-diol has previously been
(Table 1), and 2D NMR data showed that the only difference
isolated from the fungi Agrocybe aegerita and Russula
C
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. Cytotoxicity of Compounds 1−6 and 7,9-Illudadiene-3,14-diol against the Two Tumor Cell Lines Huh7 and MT4
(Mean Values, Samples in Duplicate)
CC50 (μM)
a
cell line
1
2
3
4
5
6
7,9-illudadiene-3,14-diol
INX-189
b
Huh7
MT4
1.3
0.38
0.098
0.023
160
45
n.t.
n.t.
>400
>400
240
110
3
3
0.12
0.014
a
b
Positive control.22 Not tested.
delica,20,21 whereas radulol has been obtained from the fungus
Radulomyces confluens.13
increased reactivity observed for compound 2 and subsequently
for the increase in cytotoxicity.
Compounds 2 and 3 were found to have very potent
cytotoxic effects against the two tested tumor cell lines Huh7
and MT4 with CC50 values (the concentration that results in
50% cell viability compared to untreated control cells) of 0.38
μM (Huh7) and 0.014 μM (MT4) for compound 2 and 0.098
μM (Huh7) and 0.023 μM (MT4) for compound 3,
respectively (Table 2). Surprisingly, compound 1 was found
to be substantially less cytotoxic than its diastereomer 2, with
CC50 values of 1.3 μM (Huh7) and 0.12 μM (MT4), whereas
compounds 4, 6, and 7,9- illudadiene-3,14-diol, as expected,
due to the lack of possibility to facilitate Michael-type reactions,
had no or moderate activity at concentrations up to 400 μM
(Table 2).
To test if the difference in cytotoxicity between compounds
1 and 2 could be explained with chemical reactivity, the
compounds were subjected to diluted aqueous HCl and then in
a second experiment to cysteine in aqueous solution, using
experimental conditions similar to previous studies on illudin
M.4,5 Both compounds 1 and 2 reacted with 2 M HCl in the
same way as described for illudin M, with the chloride acting as
a nucleophile on the cyclopropyl moiety as the rate-
determining step followed by a Michael-type addition of
water, resulting in two isomeric phenolic products.4 The
reaction rate of compound 1 (k = 4.1 × 10−3 min−1, t1/2 = 170
min) was similar to that described for illudin M (k = 4.7 × 10−3
min−1, t1/2 = 147 min),4 whereas compound 2 (k = 12 × 10−3
min−1, t1/2 = 38 min) reacted noticeably faster. The reaction of
illudin M with cysteine and other thiol nucleophiles has been
shown to be highly pH dependent, with a maximum rate at pH
5−6 for the reaction between illudin M and cysteine. Therefore,
the reactions between cysteine and compounds 1 and 2 were
carried out in a phosphate buffer at pH 5.4. Compound 1
reacted in a similar though not identical way to that described
for illudin M, forming a 2:1 mixture of two isomeric reaction
products (1a and 1b, respectively) with a t1/2 of 27 min (Figure
3).5 Compound 2 reacted much faster with a t1/2 of 5.6 min and
formed mainly one of the two expected reaction products (20:1
ratio) (Figure 4, compound 2b). Together this indicates that
the cis configuration of OH-3 and OH-7 as observed for
compounds 2 and 3 compared to the trans configuration in
compound 1 and illudin M is likely responsible for the
EXPERIMENTAL SECTION
■
General Experimental Procedures. The optical rotation of the
compounds was measured with PerkinElmer 341 or 343 polarimeters
(λ 589 nm, path length 10.0 cm, 20 °C). The UV absorption maxima
were obtained with a Hitachi U-2001 spectrophotometer, whereas a
PerkinElmer Lambda 2 UV/vis spectrophotometer was used to
monitor the decrease of the absorption bands at 318 nm. A Bruker
Avance III 600 MHz NMR spectrometer (5 mm QXI probe, 5 mm
CryoProbe, or 5 mm SmartProbe) or a Bruker DRX400 NMR
spectrometer (5 mm QNP probe) was used to obtain the NMR data,
with the chemical shifts reported relative to the residual solvent signal
of MeOH-d4 (δH 3.31; δC 49.00). HRESIMS data were obtained with a
Bruker maXis Impact ESI UHR Q-TOF with Na formate as calibrant
(positive mode). Preparative HPLC was performed on a Gilson 305/
306 system, with a Gilson 118 UV/vis detector (254 nm). Energy
minimization modeling of the compounds was performed employing
the MM2 force field using ChemBio3D Ultra version 11.0
(CambridgeSoft Corp.).
Fungus and Cultivation. Isolation and identification of a fungal
culture of G. vellereum strain olrim243 is described in a previous
study.10 In the present study, supernatants of G. vellereum were
produced by growing cultures in 500 mL Erlenmeyer flasks each
containing 250 mL of liquid Hagem medium.23 Five agar plugs 0.5 ×
0.5 cm in size with established fungal mycelia from an actively growing
colony were aseptically inoculated in each flask and incubated on a
rotary shaker at 120 rpm at room temperature (ca. 21 °C) for average
periods of 4 weeks. After cultivation, supernatants were filtered to
obtain cell-free samples.
Extraction and Isolation. Cell-free filtrates of G. vellereum were
extracted through SPE columns [C18 (EC); 50 mL filtrate/per 1 g
packing material; International Sorbent Technology, Hengoed, UK]
using aqueous 95% MeCN as eluent. The extracts were dried in a
vacuum centrifuge and redissolved in aqueous 40% MeCN before
being fractionated on a preparative HPLC system (Reprosil-Pur ODS-
3, C18, 5 μm, 100 × 20 mm and guard column 30 × 20 mm, Dr Maisch
GmbH, Ammerbuch, Germany) using a linear gradient (10−95%
aqueous MeCN in 10 min, followed by a hold at 95% MeCN for 10
min, 10 mL/min).
Radulol was obtained pure from the first chromatography step,
whereas the other compounds had to be purified further as described
below. Fractions containing crude compounds 1, 2, 4−6, and 7,9-
illudadiene-3,14-diol were individually pooled and rechromatographed
using preparative HPLC (column as above, 13.2 mL/min) and at
isocratic conditions with either 30% aqueous MeCN for compounds 1,
2, 5, 6, and 7,9-illudadiene-3,14-diol or 20% aqueous MeCN for
compound 4. Crude compound 3 was rechromatographed using a
linear gradient with 7.5% to 40% aqueous MeCN in 30 min (column
as above, 13.2 mL/min). The maximum yields obtained for the
compounds were 11.7 (1), 15.0 (2), 1.9 (3), 48.5 (4), 0.3 (5), 0.4 (6),
1.0 (7,9-illudadiene-3,14-diol), and 0.4 (radulol) mg/L filtrate.
(3S,7R)-Illudin M (1): light yellow, amorphous solid; [α]D 96 (c 0.1
in methanol); HRESIMS m/z 249.1489 (M + H)+ (calcd for
C15H21O3, 249.1485).
(3S,7S)-Illudin M (2): light yellow, amorphous solid; [α]D 59 (c 0.2
in methanol); UV λmax (MeOH) nm (log ε) 224, 243, 307 (4.29, 4.25,
Figure 4. Reaction products 1a, 1b, and 2b.
D
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
3.71); NMR data, see Table 1; HRESIMS m/z 249.1482 (M + H)+
(calcd for C15H21O3, 249.1485).
(3S,6S,7R)-Illudin S (3): colorless oil; [α]D 34 (c 0.2 in methanol);
UV λmax (MeOH) nm (log ε) 232, 240, 213 (4.05, 4.03, 3.46); NMR
data, see Table 1; HRESIMS m/z 265.1429 (M + H)+ (calcd for
C15H21O4, 265.1434).
(3S,4S,7R)-Dihydroilludin M (4): white, amorphous solid; [α]D 77
(c 0.3 in methanol); UV λmax (MeOH) nm (log ε) 204, 251 (3.68,
3.68); NMR data, see Table 1; HRESIMS m/z 273.1460 (M + Na)+
(calcd for C15H22NaO3, 273.1461).
Illudadiene A (5): white, amorphous solid. [α]D −6 (c 0.1 in
methanol); UV λmax (MeOH) nm (log ε) 204, 248 (3.84, 4.08); NMR
data, see Table 1; HRESIMS m/z 257.1511 (M + Na)+ (calcd for
C15H22NaO2, 257.1512).
Selective Acetylation of Compound 3. Acetic anhydride (2.0
μL, 21 μmol) was added to a stirred solution of 3 (3.8 mg, 14 μmol) in
pyridine (500 μL) at 0 °C and was left stirring for 24 h. The solvent
was then evaporated, after which the crude product was redissolved in
50% aqueous MeCN and purified with preparative HPLC (column as
above; 10−90% aqueous MeCN in 10 min, followed by a hold at 90%
MeCNaq in 10 min, 10 mL/min). The product was pooled and dried,
yielding 0.94 mg (19%) of the desired product (3a).
Oxidation of Compound 4. To a stirred solution of 4 (27 mg,
0.11 mmol) in 10 mL of dichloromethane were added 0.04 mL of
pyridine (0.49 mmol) and 20 mg (0.50 mmol) of NBS at room
temperature. All the starting material had been consumed after 3 h
according to TLC analysis (silica gel 60 F254, mobile phase; 1:4
EtOAc−hexane), after which 2 mL of 2-propanol was added to the
mixture. The solvent was then evaporated, resulting in a yellow oil,
which was dissolved in 50% aqueous MeCN and purified on
preparative HPLC (isocratic conditions, aqueous 22.5% MeCN;
same column as above, 13.2 mL/min), yielding compound 1.
Formation of MTPA Esters of Compounds 1, 2, and 3a. The
(S)-MTPA esters were formed by treating compounds 1, 2, and 3a
(0.8−2.9 μmol) with (R)-(−)-MTPA-Cl (12- to 24-fold excess) in 500
μL of pyridine-d5 for 72 h at room temperature. Analogously, the (R)-
MTPA esters were formed by treating compounds 1, 2, and 3a with
(S)-(+)-MTPA-Cl (6- to 24-fold excess). All MTPA esters were
analyzed by NMR spectroscopy without further purification.
Cytotoxicity Assay. MT4 (T-cell line, a kind gift from Prof.
Yamamoto, Yamaguchi University, Japan) and Huh7 (hepatocarcino-
ma cell line, ReBlikon GmbH, Germany) cell lines were passaged into
96-well microplates (2 × 104 cells/well) followed by the addition of
the test substances the next day, in duplicate samples. The number of
viable cells was assessed after 6 days by using a soluble formazan
(XTT) assay.24 The compound INX-189 was used as positive
control.22
Illudadiene B (6): white, amorphous solid; [α]D 59 (c 0.06 in
methanol); UV λmax (MeOH) nm (log ε) 204, 250 (3.86, 4.40); NMR
data, see Table 1; HRESIMS m/z 257.1512 (M + Na)+ (calcd for
C15H22NaO2, 257.1512).
Reaction of Compounds 1 and 2 with Cysteine. To a solution
of 1 (1.2 mg, 4.7 μmol) in 5 mL of acetate buffer pH 5.4 (50 mM) was
added L-cysteine (5.7 mg, 47 mmol) at room temperature. The
reaction was monitored both spectrophotometrically at 318 nm and
through LC-HRESIMS analysis (aqueous 10−90% MeCN, 10 min and
then a hold at 90% for 10 min; Reprosil-Pur ODS-3, C18, 3.5 μm, 125
× 4.6 mm, Dr Maisch GmbH, Ammerbuch, Germany) during a time
period of 3 h. The reaction mixture was left overnight and was then
extracted by SPE (C18 (EC), 300 mg). The SPE column was eluted
with MeCN (1 mL), and the resulting crude product was purified by
preparative HPLC (aqueous 10−90% MeCN, 10 min and then a hold
at 90% MeCN for 10 min; column as above, 10.0 mL/min), resulting
in the isolation of two isomeric reaction products (1a and 1b) in a 2:1
ratio. Compound 1a: NMR data, see Table S3; HRESIMS m/z
370.1688 (M + H)+ (calcd for C18H28NO5S, 370.1683). Compound
1b: NMR data, see Table S3; HRESIMS m/z 370.1687 (M + H)+
(calcd for C18H28NO5S, 370.1683). Compound 2 was subjected to the
same reaction procedure as compound 1, with 2.0 mg (8.1 μmol) of
compound 2 and 9.8 mg (81 μmol) of L-cysteine, resulting in one
reaction product (2b). Compound 2b: NMR data, see Table S3;
HRESIMS m/z 370.1685 (M + H)+ (calcd for C18H28NO5S,
370.1683).
Reaction of Compounds 1 and 2 with HCl. Compound 1 (0.64
mg, 2.6 μmol) was added to 10 mL of 2 M HCl at room temperature.
The reaction was monitored both spectrophotometrically at 318 nm
and through LC-HRESIMS analysis (aqueous 10−90% MeCN, 10 min
and then a hold at 90% for 10 min; Reprosil-Pur ODS-3, C18, 3.5 μm,
125 × 4.6 mm, Dr Maisch GmbH, Ammerbuch, Germany). The
reaction mixture was left overnight before being neutralized with
NaOH and passed through a SPE-column (C18 (EC), 1 g). After
elution of the SPE column with MeCN (1 mL) the resulting crude
product was purified by preparative reversed-phase HPLC (aqueous
10−90% MeCN, 10 min and then a hold at 90% MeCN for 10 min;
column as above, 10.0 mL/min), resulting in two isomeric products
(1c and 1d). Compound 1c: NMR data, see Table S3; HRESIMS m/z
307.1071 (M + Na)+ (calcd for C15H21ClNaO3, 307.1071).
Compound 1d: NMR data, see Table S3; HRESIMS m/z 307.1068
(M + Na)+ (calcd for C15H21ClNaO3, 307.1071). Compound 2 was
subjected to the same reaction procedure as compound 1, with 1.8 mg
(7.0 μmol) of compound 2, resulting in the formation of two isomeric
reaction products (2c and 2d). Compound 2c: NMR data, see Table
S3; HRESIMS m/z 307.1075 (M + Na)+ (calcd for C15H21ClNaO3,
307.1071). Compound 2d: NMR data, see Table S3; HRESIMS m/z
3071073 (M + Na)+ (calcd for C15H21ClNaO3, 307.1071).
Formation of Degradation Product 2a. Pure compound 2 was
stored in −18 °C for a prolonged period of time (>1 year) and was
then rechromatographed by preparative HPLC at isocratic conditions
with 30% aqueous MeCN (column as above, 13.2 mL/min), yielding a
5:2 mixture of two different decomposition products, of which the
most abundant was compound 2a. NMR data, see Table S1;
HRESIMS m/z 305.1362 (M + Na)+ (calcd for C15H22NaO5,
305.1359).
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.5b00500.
1
1H and 13C NMR spectra for compounds 1−6 and H
NMR spectra for compounds 1a−d, 2a−d, 3a, and the
(S/R)-MTPA esters of 1, 2, and 3a, along with tabulated
1H NMR data for 1a−d, 2a−d, 3a, 5, 6, and the (S/R)-
MTPA esters of 1, 2, and 3a (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*Tel: +46 18 672217. Fax: +46 18 673476. E-mail: Anders.
Broberg@slu.se.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
C. Åhgren, Medivir AB, Huddinge, Sweden, is gratefully
acknowledged for performing the cytotoxicity assays. The
NMR-based metabolome platform, SLU, is acknowledged for
its financial support of C.N. Financial support to A.M. from the
Carl Tryggers Foundation is also gratefully acknowledged.
REFERENCES
■
(1) Ayer, W. A.; Browne, L. M. Tetrahedron 1981, 37, 2199−2248.
(2) Abraham, W. R. Curr. Med. Chem. 2001, 8, 583−606.
(3) McMorris, T. C.; Anchel, M. J. Am. Chem. Soc. 1965, 87, 1594−
1600.
E
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(4) McMorris, T. C.; Kelner, M. J.; Chadha, R. K.; Siegel, J. S.; Moon,
S.; Moya, M. M. Tetrahedron 1989, 45, 5433−5440.
(5) McMorris, T. C.; Kelner, M. J.; Wang, W.; Moon, S.; Taetle, R.
Chem. Res. Toxicol. 1990, 3, 574−579.
(6) McMorris, T. C.; Kelner, M. J.; Wang, W.; Estes, L. A.; Montoya,
M. A.; Taetle, R. J. Org. Chem. 1992, 57, 6876−6883.
(7) Schobert, R.; Knauer, S.; Seibt, S.; Biersack, B. Curr. Med. Chem.
2011, 18, 790−807.
(8) McMorris, T. C.; Kelner, M. J.; Wang, W.; Yu, J.; Estes, L. A.;
Taetle, R. J. Nat. Prod. 1996, 59, 896−899.
(9) Williams, R. Expert Opin. Invest. Drugs 2013, 22 (12), 1627−
1644.
(10) Nord, C. L.; Menkis, A.; Vasaitis, R.; Broberg, A. Phytochemistry
2013, 90, 128−134.
(11) Nord, C. L.; Menkis, A.; Lendel, C.; Vasaitis, R.; Broberg, A.
Phytochemistry 2014, 102, 197−204.
(12) Nord, C. L.; Menkis, A.; Vasaitis, R.; Broberg, A. Molecules 2014,
19, 14195−14203.
(13) Fabian, K.; Lorenzen, K.; Anke, T.; Johansson, M.; Sterner, O.
Z. Naturforsch. C 1998, 53, 939−945.
(14) Kinder, F. R.; Bair, K. W. J. Org. Chem. 1994, 59, 6965−6967.
(15) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092−4096.
(16) Arnone, A.; Cardillo, R.; Nasini, G.; de Pava, O. V. J. J. Chem.
Soc., Perkin Trans. 1 1991, 4, 733−737.
(17) Burgess, M. L.; Zhang, Y. L.; Barrow, K. D. J. Nat. Prod. 1999,
62, 1542−1544.
(18) Bradshaw, A. P. W.; Hanson, J. R.; Sadler, I. H. J. Chem. Soc.,
Perkin Trans. 1 1982, 2445−2448.
(19) Singh, P.; Nair, M. S. R.; McMorris, T. C.; Anchel, M.
Phytochemistry 1971, 10, 2229−2230.
̌
(20) Stran
Budes nsky, M.; Ubik, K.; Kohoutova,
Chem. Commun. 1992, 57, 590−603.
́
sky, K.; Semerdzieva, M.; Otmar, M.; Prochaz
́
ka, Z.;
́
̌
̌
í
̌
́
́
J.; Streinz, L. Collect. Czech.
(21) Clericuzio, M.; Han, F.; Pan, F.; Pang, Z.; Sterner, O. Acta
Chem. Scand. 1998, 52, 1333−1337.
(22) McGuigan, C.; Madela, K.; Aljarah, M.; Gilles, A.; Brancale, A.;
Zonta, N.; Chamberlain, S.; Vernachio, J.; Hutchins, J.; Hall, A.; Ames,
B.; Gorovits, E.; Ganguly, B.; Kolykhalov, A.; Wang, J.; Muhammad, J.;
Patti, J. M.; Henson, G. Bioorg. Med. Chem. Lett. 2010, 20, 4850−4854.
(23) Stenlid, J. Can. J. Bot. 1985, 63, 2268−2273.
(24) Weislow, O. S.; Kiser, R.; Fine, D. L.; Bader, J.; Shoemaker, R.
H.; Boyd, M. R. J. Natl. Cancer Inst. 1989, 81, 577−586 [published
erratum appears in J. Natl. Cancer Inst. 1989, 81, 963].
F
DOI: 10.1021/acs.jnatprod.5b00500
J. Nat. Prod. XXXX, XXX, XXX−XXX