Two novel cyclic peptides with antifungal activity from the cyanobacterium Tolypothrix byssoidea (EAWAG 195)

Article abstract of DOI:10.1021/np000297e

Two novel cyclic tridecapeptides, tolybyssidins A (1) and B (2), were isolated from the culture medium of mass cultured cyanobacterium Tolypothrix byssoidea (EAWAG 195) by means of bioguided isolation. The gross structures of these peptides were determined by 1D and 2D NMR experiments and tandem mass spectrometry. Both peptides contain the nonnatural amino acid dehydrohomoalanine (Dhha) as well as proteinogenic amino acids albeit with D- or L-configuration. The compounds exhibit moderate antifungal activity against the yeast Candida albicans.

Full text of DOI:10.1021/np000297e

154
J . Nat. Prod. 2001, 64, 154-158
Tw o Novel Cyclic P ep tid es w ith An tifu n ga l Activity fr om th e Cya n oba cter iu m
Tolypoth r ix byssoid ea (EAWAG 195)
B. J aki, O. Zerbe, J . Heilmann, and O. Sticher*
Department of Applied BioSiences, Institute of Pharmaceutical Sciences, Swiss Federal Institute of Technology (ETH) Zurich,
CH-8057 Zu¨rich, Switzerland
Received J une 14, 2000
Two novel cyclic tridecapeptides, tolybyssidins A (1) and B (2), were isolated from the culture medium of
mass cultured cyanobacterium Tolypothrix byssoidea (EAWAG 195) by means of bioguided isolation. The
gross structures of these peptides were determined by 1D and 2D NMR experiments and tandem mass
spectrometry. Both peptides contain the nonnatural amino acid dehydrohomoalanine (Dhha) as well as
proteinogenic amino acids albeit with D- or L-configuration. The compounds exhibit moderate antifungal
activity against the yeast Candida albicans.
Cyanobacteria have received considerable attention in
the past as a source of natural products with unique
structures and biological activities.1 Malyngolide,2 majus-
culamide C,3 and kawaguchipeptin A and B4 are well-
described examples of bioactive agents from this group of
ubiquitous prokaryotic organisms. Herein, we report on the
isolation and structure determination of two novel cyclic
tridecapeptides with moderate antifungal activity, toly-
byssidins A and B, both isolated from the cyanobacterium
Tolypothrix byssoidea (Hass.) Kirch. (EAWAG 195).
Resu lts a n d Discu ssion
T. byssoidea (EAWAG 195) was isolated from a sample
collected in 1967 in Pokhara (Nepal) from a granite block
and was subsequently mass cultured. The culture medium
was filtered to remove cell mass. Thereafter, the cell-free
supernatant was subjected to an Amberlite XAD-2 column.
The Amberlite resin was subsequently eluted first with
the carbonyl signal of an acetate moiety at δc 173.6, one
signal of a carbon corresponding to a guanidine (δc 162.6,
C-6 of arginine), and 12 (amide-) carbonyl signals in the
MeOH and then DCM. The biologically active MeOH
range between 166.1 and 174.6 ppm. By using DEPT 135
extract was separated by open column chromatography
and DEPT 90 experiments, 27 methine, 10 methylene, and
using CHCl3 and MeOH as a step gradient mixture. The
17 methyl signals were identified. These signals, along with
final separation and purification of the two peptides was
three additional quaternary carbon signals at 130.3, 137.9,
carried out by reversed-phase HPLC with a MeCN/H2O
and 162.2 ppm, indicated that a total of 71 carbons were
gradient.
present in 1. From these data, 1 was undoubtedly identified
Tolybyssidin A was isolated as a white amorphous
as a peptide. Since the compound gave a negative response
powder. The MALDI mass spectrum showed a positive ion
in the ninhydrin test, 1 had to be either a cyclic peptide or
peak [M + Na]+ at 1488 m/z, indicating a molecular weight
an acyclic peptide with a protected N-terminus.
of 1465 for 1. The 1H NMR spectrum in methanol-d3
Sequential resonance assignment was performed largely
displayed 12 amide proton signals between 9.50 and 7.23
following the protocol developed by Wu¨thrich and co-
ppm that completely vanished upon exchange into fully
workers.5 A first step interpretation of the 80 ms TOCSY
deuterated methanol. In addition, signals occurred in the
and DQF-COSY spectra led to the classification of a total
aromatic region coding for one phenyl group; also, one
number of 13 different spin systems corresponding to the
triplet, seven doublets, and eight additional signals (each
following amino acids: Phe, Val, Thr, Arg, Leu, Ile, Ac-
3H), whose multiplicities could not be determined due to
Thr, Pro, and Dhha. However, since the sequential reso-
overlapping, were found between 0.74 and 1.25, corre-
nance assignment procedure usually requires the primary
sponding to 17 aliphatic methyl groups. The distribution
sequence to be known in advance, the amino acid identi-
of signals with doublets in the amide-proton region and
fications were mainly based upon complete 13C assignments
signals covering the complete range between 5 and 0.7 ppm
of the amino acids and comparison of these values with
was a clear indication that the investigated compound was
the random coil carbon shifts. In almost all cases very close
correspondence was observed with deviations being smaller
than 1 ppm. The presence of arginine instead of ornithine
a peptide. Surprisingly for the case of a peptide a singlet
at 2.03 ppm appeared indicative of an OCOCH3 group.
Further spectroscopic evidence supporting the classifica-
was established by confirming the guanidine residue via
tion of the compound as a peptide was derived from the
long-range correlations between H2-5 (δH 3.09, m, 2H) and
C-6 (δc 162.6) as well as by COSY and HMBC correlations
linking H-2, H2-3, H2-4, and H2-5 of the arginine moiety.
13C NMR and DEPT 135 spectra. These spectra revealed
* To whom correspondence should be addressed. Tel: ++41 1 635 6050.
Fax: ++41 1 635 6882. E-mail: sticher@pharma.anbi.ethz.ch.
1
The detailed H and 13C assignments are given in Table 1.
10.1021/np000297e CCC: $20.00
© 2001 American Chemical Society and American Society of Pharmacognosy
Published on Web 01/13/2001

Cyclic Peptides from Tolypothrix
J ournal of Natural Products, 2001, Vol. 64, No. 2 155
Ta ble 1. 1H (500 MHz) and 13C (125 MHz) NMR Data of 1 in CD3OH
amino acid
Phe
1H J (Hz)
13C
amino acid
Pro
1H J (Hz)
13C
1
2
3
174.6 (s)
57.6 (d)
37.3 (t)
1
2
3
173.7 (s)
62.2 (d)
30.6 (t)
26.1 (t)
49.2 (t)
4.55 (m)
2.99 (dd, 7.6, 13.3)
3.19 (dd, 7.6, 13.2)
4.39 (dd, 6.2 each)
1.94 (m)
2.21 (m)
2.07 (m)
3.64 (m)
4
137.9 (s)
130.5 (d)
129.7 (d)
128.0 (d)
4
5
5,9
6,8
7
7.29a
7.30a
3.92 (m)
7.23a
NH
1
2
3
4
8.75 (d, 9.7)
Dhha
166.1 (s)
130.3 (s)
134.1 (d)
13.3 (q)
Arg
Leu
Thr2
Ile1
1
2
3
4
5
6
NH
1
2
3
4
5
6
NH
1
2
3
4
174.6 (s)
54.2 (d)
30.3 (t)
27.7 (t)
40.5 (t)
162.2 (s)
4.31 (m)
1.65 (m)
1.51 (m)
3.09 (m)
6.61 (q, 7.0)
1.31 (d, 7.0)
9.50 (s)
NH
8.23 (d, 8.0)
Val1
1
2
3
4
5
NH
171.0 (s)
61.5 (d)
31.6 (d)
19.9 (q)
19.4 (q)
174.5 (s)
53.4 (d)
41.8 (t)
25.9 (d)
23.1 (q)
22.8 (q)
4.37a
4.43 (m)
1.67 (m)
1.59 (m)
0.92a
1.85 (m)
0.97a
0.95 (d, 7.1)
8.14 (d, 7.7)
0.91a
8.0 (d 7.4)
Thr(Ac)
1
2
3
4
5
6
NH
1
2
172.9 (s)
58.0 (d)
71.6 (d)
20.0 (q)
173.6 (s)
22.8 (q)
172.4 (s)
59.9 (d)
68.1 (d)
17.7 (q)
4.39 (dd, 7.2 each)
5.15 (m)
1.15 (d, 6.3)
4.65 (dd, 8.3 each)
4.23 (m)
1.21 (d, 6.2)
8.44 (d, 8.3)
NH
2.03 (s)
7.82 (d, 7.2)
Thr1
171.0 (s)
60.9 (d)
68.7 (d)
20.4 (q)
1
2
3
4
173.3 (s)
59.0 (d)
39.6 (d)
27.5 (t)
4.05 (dd, 7.2 each)
4.32a
1.25 (d, 6.3)
8.29 (d, 7.2)
4.43 (dd, 9.7 each)
1.83 (m)
1.10 (m)
3
4
NH
1.36 (m)
5
6
NH
1
2
0.88 (t, 7.1)
0.82 (d, 5.6)
8.55 (d, 9.7)
15.1 (q)
12.1 (q)
Val2
1
2
3
4
5
NH
173.3 (s)
60.1 (d)
32.2 (d)
20.2 (q)
18.5 (q)
Ile2
173.2 (s)
58.1 (d)
40.4 (d)
4.46a
4.10 (dd, 6.2 each)
1.85 (m)
2.12(m)
3
0.97a
0.77 (d, 6.6)
7.82 (d, 7.2)
4
1.01 (m)
1.34 (m)
27.6 (t)
14.7 (q)
12.2 (q)
5
6
0.82a
0.74 (d, 6.9)
7.23 (6.2)
NH
Val3
1
2
3
4
172.3 (s)
58.7 (d)
31.4 (d)
19.9 (q)
19.4 (q)
4.28 (dd, 6.6, 7.7)
2.13 (m)
0.92a
5
0.91a
NH
7.90 (d, 6.6)
a
Multiplicity of the signals is unclear due to overlapping
These spin systems were subsequently linked sequen-
tially via (i,i+1) HN,HR NOEs. NOEs were observed for Phe
NH/Ile2 R-H, Dhha NH/Phe R-H, Thr1 NH/Thr(Ac) R-H,
Val2 NH/Thr1 R-H, Val3 NH/Val2 R-H, Arg NH/Pro R-H, Leu
NH/Arg R-H, Thr2 NH/Leu R-H, Ile1 NH/Thr2 R-H, and Ile2
NH/Ile1 R-H. Furthermore, amide protons typically dis-
played intraresidual and interesidual (i,i-1) (HN,C′) HMBC
correlations supporting the sequential assignment. Such
HMBC correlations were observed for Phe R-H/Phe CdO,
Dhha â-H/Dhha CdO, Val1 R-H/Dhha CdO, Val1 R-H/Val1
CdO, Thr(Ac) R-H/Thr(Ac) CdO, Thr1 R-H/Thr(Ac) CdO,
Thr1 R-H/Thr1 CdO, Val2 R-H/Thr1 CdO, Val2 R-H/Val2
CdO, Val3 R-H/Val2 CdO, Val3 R-H/Val3 CdO, Pro R-H/
Val3 CdO, Pro R-H/Pro CdO, Arg R-H/Pro CdO, Arg R-H/
Arg CdO, Leu R-H/Arg CdO, Leu R-H/Leu CdO, Thr2 R-H/
Leu CdO, Thr2 R-H/Thr2 CdO, Ile1 R-H/Thr2 CdO, Ile1
R-H/Ile1 CdO, Ile2 R-H/Ile1 CdO, Ile2 R-H/Ile2 CdO, and
Phe R-H/Ile2 CdO. These data, together with a molecular
weight of 1465 as determined by MALDI mass spectrom-
etry, are consistent with the molecular formula C71H116
-
N
16O17. No HMBC correlations between Thr(Ac) R-H and
Val1 CdO as well as no NOEs between Thr(Ac) NH and
Val1 R-H could be found. However, the ring connection
between Val1 and Thr(Ac) could be confirmed by an NOE
observed between Val1 H3-5 and Ac-Thr H3-6 and with the
analysis of the tandem mass spectrum of 1 showing the
fragment ion [M + H - Dhha-Val-Thr(Ac)-Thr-Val-Val]+
at m/z 842.
The following MS fragments that helped to establish the
amino acid sequence of the molecule 1 were detected in
the tandem mass spectrum: m/z (rel int) 1466.5 [M + H]+
(72), 1423 [M + H - (CH3)2CH]+ (25), 1323 [M + H - Thr-
(Ac)]+ (73), 1284 [M + H - Dhha-Val]+ (7), 1222 [M + H -
Thr(Ac)-Thr]+ (6), 1123 [1423 - H - Val-Val-Thr]+ (39),
1119 [1423 - (CH3)2CH-Phe-Ile]+ (1), 1024 [M + H - Ile-
Phe-Dhha-Val]+ (100), 907 [1423 + H - (CH3)2CH-Phe-
Ile-Ile-Thr]+ (11), 842 [M + H - Dhha-Val-Thr(Ac)-Thr-

156 J ournal of Natural Products, 2001, Vol. 64, No. 2
J aki et al.
Ta ble 2. 1H (500 MHz) and 13C (125 MHz) NMR Data of 2 in CD3OH
amino acid
Val1
1H J (Hz)
13C
amino acid
Phe
1H J (Hz)
4.54 (m)
2.58 (dd, 8.7, 16.2)
3.17 (dd, 8.9, 13.8)
13C
1
2
3
4
5
NH
173.9 (s)
59.2 (d)
32.6 (d)
19.9 (q)
15.2 (q)
1
2
3
173.3 (s)
56.9 (d)
40.6 (t)
4.43a
1.81 (m)
0.84 (d, 5.7)
0.84 (d, 5.7)
8.58 (d, 9.6)
4
138.8 (s)
130.5 (d)
131.3 (d)
129.8 (d)
5,9
6,8
7
NH
1
2
3
4
5
7.20a
7.30
7.29 (m)
8.78 (d, 8.7)
Thr1
1
2
3
4
175.0 (s)
61.2 (d)
68.8 (d)
14.3 (q)
Val5
173.8 (s)
61.3 (d)
32.2 (d)
19.4 (q)
14.6 (q)
4.40a
4.03 (m)
1.22 (d, 5.8)
8.30 (d, 7.4)
3.97 (dd, 5.9, 8.5)
2.1 (m)
0.94a
NH
0.94a
NH
1
2
3
4
7.76 (d, 8.5)
Ile
1
2
3
4
175.9 (s)
59.2 (d)
32.8 (d)
27.8 (t)
Dhha
166.2 (s)
130.9 (s)
134.0 (d)
13.3 (q)
4.41a
1.64 (m)
1.16 (m)
1.49 (m)
6.64 (q, 8.3)
1.29 (d, 7.3)
9.51 (s)
NH
5
6
NH
1
2
3
4
NH
0.92 (t, 7.6)
0.87 (d, 7.1)
8.04 (d, 9.1)
19.9 (q)
12.2 (q)
Thr2
173.9 (s)
61.2 (d)
66.5 (d)
15.9 (q)
Arg
1
2
3
175.1 (s)
56.6 (d)
32.7 (t)
4.62 (dd, 8.8 each)
4.00 (m)
1.20 (d, 5.8)
8.48 (d, 8.8)
4.06 (dd, 8.7 each)
1.66 (m)
1.85 (m)
4
5
1.53 (m)
2.80 (dd, 6.0, 13.0)
27.7 (t)
40.4 (t)
6
159.9 (s)
NH
1
2
7.39 (d, 8.7)
Val2
1
2
3
4
5
NH
175.9 (s)
59.7 (d)
30.4 (d)
18.5 (q)
15.1 (q)
Tyr
174.3 (s)
58.2 (d)
38.1 (t)
4.40a
1.91 (m)
0.88 (d, 7.1)
0.94a
4.05 (m)
3
2.76 (dd, 7.6, 13.2)
3.28a
4
5
6
7
8
9
NH
1
2
129.8 (s)
130.6 (d)
116.3 (d)
157.1 (s)
115.3 (d)
130.6 (d)
8.12 (d, 8.0)
7.08 (d, 8.3)
6.80 (d, 8.5)
6.72 (d, 8.3)
7.02 (d, 8.4)
8.93 (d, 8.5)
Val3
1
2
3
4
5
NH
173.8 (s)
62.3 (d)
31.6 (d)
19.5 (q)
14.6 (q)
Met
172.4 (s)
58.2 (d)
34.8 (t)
4.23a
4.71a
2.12 (m)
3
1.81 (m)
2.04 (m)
2.55a
0.95a
0.95a
4
32.3 (t)
15.8 (q)
7.92 (d, 7.6)
2.60a
5
NH
2.10 (s)
8.99 (d, 8.4)
Val4
1
2
3
4
5
NH
176.2 (s)
61.6 (d)
30.3 (d)
19.4 (q)
14.7 (q)
4.10 (dd, 7.5 each)
1.64 (m)
0.80 (d, 6.1)
0.76 (d, 6.4)
7.25 (d, 7.5)
a
Multiplicity of the signals is unclear due to overlapping.
Val-Val]+ (6), 722 [M + H - Phe-Ile-Ile-Thr-Leu-Arg]+ (4),
699 [M - Thr-Leu-Arg-Pro-Val-Val-Thr]+ (5).
Val1 NH/Met R-H, Thr1 NH/Val1 R-H, Ile NH/Thr1 R-H,
Thr2 NH/Ile R-H, Val2 NH/Thr2 R-H, Val3 NH/Val2 R-H,
Val4 NH/Val3 R-H, Phe NH/Val4 R-H, Val5 NH/Phe R-H,
Dhha-NH/Val5 R-H, and Tyr NH/Arg R-H.
Tolybyssidin B (2) was obtained as a yellow-white solid.
MALDI MS and 13C NMR data were consistent with the
molecular formula C72H115N16O16S. The 1H NMR data
indicated 2 to be a peptide as well. Due to the limited
quantity of the isolated material, carbon signals of 2 had
to be exclusively detected indirectly through HSQC and
The ring connection between Dhha and Arg could be
unambiguously confirmed by analysis of the tandem mass
spectrum of 2 with the fragment ion [M - H - Phe-Val]+
at m/z 1244 and the fragment ion of this fragment [1244
- Dhha-Arg]+ at m/z 1004. The following fragments,
supporting the sequence of the molecule 2, were detected
in the tandem mass spectrum: m/z (rel int) 1473 [M -
H2O]+ (3), 1447 [M - H - CH(CH3)2]+ (24), 1445 [M - H
- CH(OH)CH3]+ (100), 1383 [M - H - OH(C6H4)CH2]+
(10), 1372 [1473 - Thr]+ (30), 1347 [1447 - Val]+ (17), 1310
[M + H - Val-Dhha]+ (10), 1284 [1383 - Val]+ (15), 1245
1
HMBC correlations. The detailed H and 13C assignments
are given in Table 2. The NMR spectroscopic identification
of 2 was performed analogously to 1. In this process the
spin systems of the amino acid residues were again
identified by interpretation of the TOCSY and COSY
spectra and sequentially linked through NOEs and HMBC
correlations. NOEs were observed for Met NH/Tyr R-H,

Cyclic Peptides from Tolypothrix
J ournal of Natural Products, 2001, Vol. 64, No. 2 157
Amberlite XAD-2 resin (nonpolar, surface area 330 m2/g).
Subsequently, the column was eluted with MeOH.
Isola tion of 1 a n d 2. The MeOH extract (0.8 g) obtained
from 90 L of microscopically cell-free culture medium was
applied to an open column (4 × 100 cm, Si Gel). Elution was
carried out with CHCl3/MeOH as step gradient to obtain 10
fractions (20-100 mg). Bioactive fraction 6 (80 mg, eluted with
CHCl3/MeOH, 25:75) was subjected to reversed-phase HPLC
(UV detection, 220 nm) using 1:1 MeCN/H2O as an eluent to
yield four fractions (5-30 mg). Bioactive fraction 3 (30 mg)
was rechromatographed with reversed-phase HPLC with a
80:20 MeCN/H2O eluent to yield 1 (8 mg) and 2 (3 mg) as pure
compounds.
HP LC An a lysis of th e Ma r fey Der iva tives of 1 a n d 2.
To the acid hydrolysate (6 N HCl, 16 h, 110 °C) of a 200 µg
portion of 1 and 2, respectively, were added 100 µL of 1-fluoro-
2,4-dinitrophenyl-5-L-alanine amide (L-FDAA) in Me2CO (10
mg/mL) and 200 µL of 1 M NaHCO3, and the reaction mixture
was kept at 80 °C for 3 min. Then 100 µL of 2 N HCl and 400
µL of 50% MeCN were added and analyzed by reversed-phase
HPLC: Lichrosorb RP-18 (250 × 4 mm); gradient elution from
95:5:0.1 MeCN/H2O/TFA to 40:60:0.1 MeCN/H2O/TFA in 60
min; UV detection 340 nm, flow-rate 1.5 mL/min. Retention
times of the standard amino acids (min): L-Leu (41.42), D-Leu
(45.31), L-Phe (41.76), D-Phe (44.51), L-Ile (40.70), D-Ile (44.94),
L-Val (37.07), D-Val (40.56), L-Pro (31.44), D-Pro (32.78), L-Arg
(27.82), D-Arg (28.68), L-Thr (26.84), D-Thr (28.42), L-Thr(Ac)
(25.27), D-Thr(Ac) (26.05), L-Met (37.07), D-Met (38.25), L-Tyr
(48.56), D-Tyr (50.96).
[1445 - Val - Thr]+ (5), 1244 [M - H - Phe-Val]+ (10),
1217 [1447 - Met-Val]+ (4), 1183 [1284 - Thr]+ (<1), 1052
[1183 - Met]+ (1), 1004 [1217 + H - Thr-Ile]+ (<1).
Analysis of the acid hydrolysate as Marfey derivatives6
indicated two amino acids (Pro and Leu) of 1 to be D and
10 amino acids (Val1, Val2, Val3, Thr(Ac), Thr1, Thr2, Arg,
Ile1, Ile2, and Phe) to be L. All amino acid residues of 2
were determined to be L.
Retention times of the amino acids of 1: D-Leu (45.31), L-Phe
(41.76), L-Ile (40.70), L-Val (37.07), D-Pro (32.78), L-Arg (27.82),
L-Thr (26.84), L-Thr(Ac) (25.27).
Retention times of the amino acids of 2: L-Phe (41.98), L-Ile
(40.96), L-Val (36.86), L-Arg (27.98), L-Thr (26.84), L-Met
(37.07), L-Tyr (48.56).
Tolybyssidins A (1) and B (2) inhibit the growth of the
yeast Candida albicans at a concentration of 32 µg/mL for
1 and 64 µg/mL for 2, respectively. A MIC value of 8 µg/
mL was detected for the reference compound miconazole.
Exp er im en ta l Section
P r ep a r a tion of th e Acetyl Der iva tive of Th r (Acetyl-
Th r eon in e, Th r (Ac)). Thr (5.6 mg), anhydrous pyridine (0.5
mL), and acetic anhydride (0.5 mL) were kept dark and at
room temperature for 18 h. The reaction mixture was diluted
with 2 mL of water and stored at 4 °C for 1 h. Then it was
applied to a reversed-phase cartridge, which had been washed
with 10 mL of water. Pyridine, acetic anhydride, and non-
acetylated Thr were eluted with water, followed by the elution
of acetylated compound with CHCl3. Evaporation of CHCl3
under reduced pressure below 30 °C yielded 2.5 mg of the
acetyl derivative, which was completely dried in a vacuum-
drying oven at 25-28 °C. The identity of Thr(Ac) was proven
Gen er a l Exp er im en ta l P r oced u r es. Optical rotations
were measured in MeOH on
a Perkin-Elmer model 241
polarimeter. IR spectra were recorded in KBr pellets on a
Perkin-Elmer 2000 FT infrared spectrophotometer. UV spectra
were obtained in methanol on a UVIKON 930 spectrophotom-
eter. Melting points were measured on a Mettler FP 5. MALDI
MS spectra were measured on
a Perseptive Biosystems
Voyager Elite spectrometer and tandem MS spectra on a
Finnigan LCQ Ion Prep spectrometer. NMR spectra were
recorded on a Bruker DRX-500 spectrometer operating at a
basic 1H frequency of 500 MHz at 298 K, using the solvent
line (CD3OH, 1H δ 3.31, 13C δ 49.0) for referencing. For
homonuclear proton 2D experiments (DQF-COSY,7 TOCSY,8
and NOESY9) standard experiments were performed with
suppression of the methanol OH line by low-power presatu-
ration. [1H,13C] HSQC10 and HMBC11 experiments utilized
pulsed-field gradients for coherence selection.12 HPLC separa-
tions were performed with a Merck-Hitachi pump connected
to a Rheodyne HPLC injector, a Merck variable wavelength
monitor, and Knauer HPLC columns (Hypersil ODS, 5 µm,
250 × 16). Si gel (Si gel 60, 40-60 µm, Merck) was applied for
open column chromatography. For TLC controls, RP-18 F254
precoated sheets (0.25 mm, Merck) were used. All solvents
were of HPLC grade.
Or ga n ism s a n d Cu ltu r e Con d ition s. T. byssoidea (Hass.)
Kirch., designated strain EAWAG 195, was isolated from a
sample collected in Nepal, 1967. The culture is deposited at
the Culture Collection of Algae at the Swiss Federal Institute
for Water Resources and Water Pollution Control (EAWAG),
Du¨bendorf, Switzerland. The cyanophyte was cultivated in 10
L bottles containing a modified inorganic culture medium. The
cultures were illuminated continuously with fluorescent lamps
(Philips TLM/33 Rs 40 W) at 29 µmol/(s‚m2), aerated with a
mixture of 2% CO2 in air, and incubated at a temperature of
24 ( 1 °C. The cyanobacterial cultures were harvested after
25-30 days. The supernatant was separated from the cells
by filtration and adsorbed on a column filled with 250 g of
1
by H NMR spectroscopy measured in CD3OD. In comparison
1
with the H NMR spectrum of non-acetylated Thr, the deriva-
tized compound showed the expected additional CH3 signal at
1.80 ppm, and a double acetylation was excluded.
An tifu n ga l Assa y. The MIC determinations for 1 and 2
were performed as described previously.13 The yeast C. albi-
cans (ATCC 26790) was applied as a test organism.
Tolybyssid in A (1): white, amorphous solid; [R]20 -13°
D
(c 0.2, MeOH); mp 177-187 (dec); UV (MeOH) λmax 207 nm;
IR (KBr) ∼3400, 1735, 1640 (br), 1510, 1370; 1H and 13C NMR
data, see Table 1; MS data (tandem MS) see Results and
Discussion, MALDI MS m/z 1488 [M + Na]+.
Tolybyssid in B (2): yellow-white, waxy solid; [R]20 -18°
D
(c 0.3, MeOH); mp 127-137 (dec); UV (MeOH) λmax 222, 278
1
nm; IR (KBr) ∼3400, 1640 (br), 1510; H and 13C NMR data,
see Table 2; MS data (tandem MS) see Results and Discussion,
MALDI MS m/z 1514 [M + Na]+.
Ack n ow led gm en t. We thank Dr. Hans-Rudolf Bu¨rgi, Dr.
Marianne Bosli, and Frank Sunder (EAWAG, Du¨bendorf,
Switzerland) for providing and cultivating the cyanobacterial
material, Dr. Peter J ames and Manfredo Quadroni (ETH
Biochemstry Department, Mass Spectral Service) as well as
Dr. Ernst Schro¨der (Finnigan MAT, Bremen) for recording
mass and tandem mass spectra, and Dr. Engelbert Zass for
performing literature searches.

158 J ournal of Natural Products, 2001, Vol. 64, No. 2
J aki et al.
(8) Braunschweiler, L.; Ernst, R. R. J . Magn. Reson. 1983, 53, 521-528.
Bax, A.; Davis, D. G. J . Magn. Reson. 1985, 65, 355-360.
(9) J eener, J .; Meier, B. H.; Bachmann, P.; Ernst, R. R. J . Chem. Phys.
1979, 71, 4546.
(10) Bodenhausen, G.; Ruben, D. J . Chem. Phys. Lett. 1980, 69, 185-189.
(11) Bax, A.; Summers, M. J . Am. Chem. Soc. 1986, 108, 2093-2094.
(12) Keeler, J .; Clowes, R. T.; Davis, A. L.; Laue, E. D. Methods Enzymol.
1994, 239, 145-207.
Refer en ces a n d Notes
(1) Patterson, G. M. L.; Larsen, L. K.; Moore, R. E. J . Appl. Phycol. 1994,
6, 151-157.
(2) Cardellina, J . H.; Moore, R. E.; Arnold, E. V.; Clardy, J . J . Org. Chem.
1979, 44, 4039-4042.
(3) Carter, D. C.; Moore, R. E.; Mynderse, J . S.; Niemczura, W. P.; Todd,
J . S. J . Org. Chem. 1984, 49, 236-241.
(4) Ishida, K.; Matsuda, H.; Murakami, M.; Yamaguchi, K. J . Nat. Prod.
1997, 60, 724-726.
(13) Rios, J . L.; Recio, M. C.; Villar, A. J . Ethnopharmacol. 1988, 23,
(5) Wu¨thrich, K. NMR of Proteins and Nucleic Acids; Wiley: New York,
1986.
127-149.
(6) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591-596.
(7) Piantini, U.; Sørensen, O. W.; Ernst, R. R. J . Am. Chem. Soc. 1982,
104, 6800-6801.
NP000297E