M. Isaka et al. / Tetrahedron Letters 55 (2014) 469–471
471
tween H-5 of Pro-2 and H-2 of Ile suggested the Ile–Pro-2 connec-
tion. HMBC correlations from NH of Val to C-1 of Pro-2, and from
N-CH of N-Me-Gln-2 to C-1 of Val and C-2 of N-Me-Gln indicated
the connections of Pro-2–Val–N-Me-Gln. Finally, the ester linkage
of N-Me-Gln–Hmp to form a cyclodecadepsipetide was revealed
by the HMBC correlation from H-2 of Hmp to C-1 of N-Me-Gln-2.
The structure of 3 is most closely related to SCH 378161, which
possesses Pro instead of pipecolic acid (Pip) in 1.
Supplementary data
3
References and notes
1
2
3
4
.
.
.
.
The absolute configurations of Hmp and the amino acid resi-
dues of 1–3 were determined by HPLC analysis of the acid hydro-
lyzate using a ligand-exchange-type chiral column. Thus, SCH
2
17048 (1) was hydrolyzed in 6 M HCl at 110 °C for 15 h. After
cooling, the aqueous solution was extracted with Et O. The Et O
2
2
solution and the aqueous layer were concentrated in vacuo and
separately subjected to HPLC analysis. For the determination of
the absolute configuration of the Hmp residue, standard samples
5. The fungus used in this study was isolated from an elephant dung sample
collected from the Sakaerat Environmental Research Station, Nakhon
Ratchasima Province, Thailand, by Dr. Nigel L. Hywel-Jones. The living
culture was deposited in the BIOTEC Culture Collection as BCC 7069 on
September 21, 1999. Results from the LSU and ITS gene sequences indicated
that this fungus belongs to the family Pleosporaceae of the order Pleosporales,
but it is not assignable to any genera. These sequencing data were deposited in
GenBank (accession No. KF486914 for LSU gene, and KF482067 for ITS gene).
of four Hmp isomers were prepared from
L
- and
, 400 MHz) spectroscopic data of
O extract were identical with those of (2S,3R)-Hmp/
D-Ile, and L- and
9
1
D
-allo-Ile. The H NMR (CDCl
3
the Et
2
(
(
2R,3S)-Hmp, and they were different from those of (2S,3S)-Hmp/
2R,3R)-Hmp. HPLC analysis using the chiral column established
2
6
6. SCH 217048 (1): Colorless solid; mp 199–201 °C; ½
a
ꢁ
ꢂ112 (c 0.20, MeOH); UV
D
10
(MeOH) k (log ) 214 (4.20) nm; IR (ATR) 3407, 3328, 1750, 1676, 1631,
e
m
max
max
ꢂ1
that the hydrolyzate contained (2R,3S)-Hmp. The analysis of
the aqueous layer of the hydrolyzate, employing standard - and
-amino acids, revealed the presence of Gly, -Pip, -Pro, N-Me-
Val, -Val, (2S,3S)- -Ile, and
-Phe.11 By acid hydrolysis, the N-Me-
Gln residue should be converted into N-Me-Glu. Due to the signif-
icant retention time flipping of standard N-Me- -Glu and N-Me-
+
1
1
447 cm ; HRESI MS m/z 1159.6386 [M+Na] (calcd for C57
H
88
N
10
O
14Na,
L
159.6374); 1H NMR (500 MHz, DMSO-d
13
6
) and C NMR (125 MHz, DMSO-d )
6
D
L
L
D
-
spectroscopic data were consistent with those reported in the literature.
26
D
7.
SCH 218157 (2): Colorless solid; mp 187–189 °C; ½
a
ꢁ
ꢂ116 (c 0.21, MeOH); UV
L
L
L
(
1
MeOH) kmax (log
ꢂ1
e) 214 (4.20) nm; IR (ATR) mmax 3421, 1749, 1670 sh, 1631,
+
449 cm ; HRESI MS m/z 1136.6719 [M+H] (calcd for C H N O ,
57 90 11 13
1
1
3
L
D
-
6
1136.6714); H NMR (500 MHz, DMSO-d ) and C NMR (125 MHz, DMSO-
d ) spectroscopic data were consistent with those reported in the literature.
6
Glu, they could not be assigned during the analysis. The absolute
configurations of N-Me-Glu and N-Me-Gln in 1 were determined
by application of Marfey’s method.12 The acid hydrolyzate of 1
26
D
8
9
.
.
Pleosporin A (3): Colorless solid; mp 124–126 °C; ½
a
m
+
ꢁ
ꢂ125 (c 0.20, MeOH); UV
(
MeOH) kmax (log ) 214 (4.23) nm; IR (ATR) max 3423, 1750, 1671, 1631,
e
1447 cmꢂ1
HRESI MS m/z 1122.6555 [M+H] (calcd for C56 88N11O13
1122.6558); see Table
;
H
,
1
13
was derivatized with N
a
-(2,4-dinitro-5-fluorophenyl)-
L-alanina-
1
for
H
NMR (500 MHz, CDCl
3
)
and
C NMR
mide (FDAA). HPLC analysis using a reverse-phase column revealed
1
3
the presence of N-Me-L-Glu and the absence of N-Me-D-Glu.
1
0. Chiral column: Phenomenex Chirex 3126 (D)-penicillamine, 4.6 ꢀ 250 mm;
flow rate 0.5 mL/min, detection UV 235 nm. The Et O extract from the acid
hydrolysis solution was analyzed using 15% 2-propanol in 2 mM aqueous
CuSO as the mobile phase. The retention times of (2S,3R)-Hmp and (2R,3S)-
The HPLC chromatogram (chiral column) of the acid hydroly-
zate (aqueous layer) of SCH 218157 (2) was indistinguishable from
that of 1, which indicated that both hydrolyzates possessed the
same amino acids. Chiral column HPLC analysis of the hydrolyzate
2
4
Hmp were 107 and 176 min, respectively. The Et
Hmp.
2
O extract contained (2R,3S)-
1
1. Three mobile phase conditions were employed due to the large retention time
differences of the standard amino acids: (1) 5% MeCN in 2 mM aqueous CuSO
Gly (t 12.6 min), -Pip (t 16.7 min), -Pip (t 25.4 min), -Pro (t 18.9 min),
Pro (t 45 min), N-Me- -Val (t 19.6 min), N-Me- -Val (t 34 min), and
3 min); (2) 5% 2-propanol in 2 mM aqueous CuSO -Val (t 25 min),
0 min), -allo-Ile (t 44 min), -Ile (t 52 min), -allo-Ile (t
88 min); (3) 15% 2-propanol in 2 mM aqueous CuSO -Ile (t
m), mobile phase MeCN/(0.05% TFA
O) = 45:55, flow rate 0.5 mL/min, UV detection at 340 nm. Retention times
of the FDAA-derivatized standard amino acids, N-Me- -Glu and N-Me- -Glu,
were 8.1 and 7.4 min, respectively. The derivatized hydrolyzate contained N-
of 3 indicated the absence of
L-Pip, and relatively higher peak
4
,
intensity for -Pro than other amino acids when compared with
L
R
L
R
D
R
L
R
D-
the chromatograms for 1 and 2.
R
L
R
D
R
L-Val (t
R
8
4
4
,
L
R
D
-Val (t
R
Compounds 1–3 exhibited antimalarial activity against Plasmo-
L
R
L
R
D
R
71 min), and
D-Ile
dium falciparum K1 with IC50 values of 1.6, 6.4, and 1.6 lg/mL,
(t
R
4
,
L
R
29 min), L-Phe
1
4
respectively. These compounds were also subjected to other bio-
1
logical assays: antitubercular activity (Mycobacterium tuberculosis
1
5
15
13. Column: NovaPak C18 (3.9 ꢀ 150 mm, 4
l
H37Ra), antifungal activity (Candida albicans),
antibacterial
in H
2
15
activity (Bacillus cereus), and cytotoxicity to cancer cell-lines
L
D
15
(
KB, MCF-7, and NCI-H187)16 and nonmalignant Vero cells. They
were inactive in these assays at 50 g/mL.
l
1
In conclusion, a new antimalarial cyclodepsipeptide, pleosporin
A (3), was isolated from an elephant dung fungus and its structure
was elucidated. The absolute configurations of 3 and the known
analogs 1 and 2 were determined by chiral column HPLC analysis
and Marfey’s method. Although the antimalarial activities of 1–3
were much weaker than the standard drug, dihydroartemisinin
1
(
IC50 0.0044 lg/mL in our assay system), their good selectivity in-
dex suggested that they could serve as new antimalarial leads.
Acknowledgment
Financial support from the National Center for Genetic Engi-
neering and Biotechnology (BIOTEC) is gratefully acknowledged.