A novel β-amino acid in cytotoxic peptides from the cyanobacterium Tychonema sp.

Article abstract of DOI:10.1002/ejoc.200701033

The cyclic dodecapeptides tychonamide A (1) and B (2) isolated from the methanolic extract of the cyanobacterium Tychonema sp. contain the novel β-amino acid 3-amino-2,5,7-trihydroxy-8-phenyloctanoic acid (Atpoa). Compounds 1 and 2 have cytotoxic activity

Full text of DOI:10.1002/ejoc.200701033

FULL PAPER
DOI: 10.1002/ejoc.200701033
A Novel β-Amino Acid in Cytotoxic Peptides from the Cyanobacterium
Tychonema sp.
Christian Mehner,[a] Daniela Müller,[a] Anja Krick,[a] Stefan Kehraus,[a] Reik Löser,[b]
Michael Gütschow,[b] Armin Maier,[c] Heinz-Herbert Fiebig,[c] Reto Brun,[d] and
Gabriele M. König*[a]
Keywords: Natural products / Peptides / Cyanobacteria / Structure elucidation / Configuration determination
The cyclic dodecapeptides tychonamide A (1) and B (2) iso-
lated from the methanolic extract of the cyanobacterium
Tychonema sp. contain the novel β-amino acid 3-amino-
2,5,7-trihydroxy-8-phenyloctanoic acid (Atpoa). Compounds
1 and 2 have cytotoxic activity towards cancer cell lines
in a monolayer assay with mean IC50 values of 0.9 and
3.3 µgmL–1, respectively. Compound 2 was shown to be
active against tumor cell suspensions derived from
solid tumor xenografts in a clonogenic assay (mean IC50
2.4 µgmL–1). Additionally, antiprotozoal activity was ob-
served (Trypanosoma b. rhodesiense, IC50 0.1 µgmL–1) for
compound 1.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
compounds are thus further members of the β-amino acid
containing peptides, such as microcystins,[5] nodularins,[6]
nostophycin,[7] scytonemin A,[8] calophycin,[9] cryptophy-
cin,[10] puwainaphycins,[11] schizotrin A,[3] and pahayokolide
A and B.[4] Tychonamide A and B exhibit cytotoxic activity
towards several cancer cell lines (mean IC50 value of 0.9
and 3.3 µgmL–1, respectively). Tychonamide B inhibited
anchorage independent growth and in vitro colony forma-
tion of tumor cells in a concentration-dependent and
tumor-type selective manner (mean IC50 value of
2.4 µgmL–1). Additionally, antiprotozoal activity was ob-
served (Trypanosoma b. rhodesiense, IC50 value of
0.1 µgmL–1) for tychonamide A.
Introduction
Cyanobacteria are the source of a diverse array of com-
pounds of pharmaceutical interest. A specific feature of
their secondary metabolite spectrum is the occurrence of
cyclic peptides containing unusual amino acids, such as β-
amino acids.[1] Previously, we isolated three new cyclic hexa-
peptides with MptpB (Mycobacterium tuberculosis proteine
tyrosine phosphatase B) inhibitory activity: the brunsvic-
amides A–C from the cyanobacterium Tychonema sp. These
peptides are related to the sponge-derived mozamides and
therefore support the suggestion that secondary metabolites
of certain marine invertebrates are produced by associated
microorganisms.[2]
In the current study, the cytotoxic fraction of the same
cyanobacterial extract was investigated and yielded cyclic
dodecapeptides 1 and 2 (tychonamide A, B). They are
structurally related to some other cyclic peptides, for exam-
ple, schizotrin A, isolated from the cyanobacterium
Schizotrix sp.[3] and pahayokolide A from Lyngbya sp.[4]
Tychonamide A and B contain the novel β-amino acid 3-
amino-2,5,7-trihydroxy-8-phenyloctanoic acid (Atpoa). The
Results
The cyanobacterial strain Tychonema sp. was collected
from a pond of a sugar factory (waste water) near
Braunschweig (Germany). After cultivation in a photobio-
reactor, cells were lyophylized and subsequently extracted
with dichloromethane, ethyl acetate, and methanol. The
methanol extract was dissolved in water and repeatedly ex-
tracted with diethyl ether and butanol. The latter phase was
purified by repeated preparative reverse-phase HPLC to ob-
tain compounds 1 and 2 as white amorphous materials.
Compound 1 showed a positive ESI quasimolecular ion
signal at m/z = 1486.8 [M + H]+. Its molecular formula,
C73H107N13O20, is based on HRFTICR mass measure-
ments (calcd. 743.8950; observed 743.8951 for [M +
[a] Institute for Pharmaceutical Biology, University of Bonn,
Nussallee 6, 53115 Bonn, Germany
Fax: +49-228733250
E-mail: g.koenig@uni-bonn.de.
[b] Pharmaceutical Institute, Pharmaceutical Chemistry I, Univer-
sity of Bonn,
An der Immenburg 4, 53121 Bonn, Germany
[c] Oncotest GmbH, Institute of Experimental Oncology,
Am Flughafen 12-14, 79108 Freiburg, Germany
[d] Swiss Tropical Institute,
2 H]2+). Signals in the H NMR spectra were very broad,
which is indicative of the presence of different conformers
in solution. NMR spectra were thus measured at 75 °C and
1
Socinstrasse 57, 4002 Basel, Switzerland
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
1732
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1732–1739

Novel β-Amino Acid in Cytotoxic Peptides from Cyanobacterium
showed characteristic resonances for α-CH groups between the low-mass region: Pro at m/z = 70 (19%), Ile at m/z =
δ = 3.9–5.0 ppm and exchangeable NH resonances between
δ = 7.3–9.2 ppm, which provides evidence for the peptidic
nature of compound 1 (Scheme 1). Resonance signals be-
tween δ = 6.8–7.3 ppm evidenced the presence of aromatic
86 (2%), and MeHty (O-methylhomotyrosine) at m/z = 164
(4%). After assignment of all protons to their directly
1
bonded carbon atoms by H–13C HSQC, it was possible to
1
establish all amino acid residues by H–1H COSY, HMBC,
1
amino acids. In addition, the H NMR spectrum featured
HSQC–TOCSY, and TOCSY experiments and to deduce
fragments A–C (Figure 1). Proline was present in triplicate
as indicated by a HSQC–TOCSY spectrum that exhibited
three sharp signals at δ = 1.96, 2.80, and 3.71 ppm for an
N-acetyl group (72-CH3), an N-methyl group (73-CH3), and
a -OCH3 (30-CH3) functionality, respectively. The 13C
NMR spectrum revealed signals for 9 methyl, 20 methylene,
26 methine, and 18 quaternary carbon atoms.
1
three almost-identical spin systems with typical H and 13C
NMR shifts and 2D correlations in the HMBC spectrum
for this amino acid.
1H–1H COSY correlations and HSQC–TOCSY data al-
lowed us to deduce the presence of a spin system beginning
with 2-CH (δC = 71.9 ppm and δH = 3.91 ppm) and contin-
uing through to 8-CH2 (δC = 43.6 ppm and δH = 2.63,
2.68 ppm). The HMBC correlations between 10/11-H (δH
=
7.18 ppm) and C-8 led us conclude that an aromatic moiety
was connected to this long aliphatic chain. On the basis of
the chemical shifts, one nitrogen-bearing methine carbon
atom, that is, C-3 (δ = 48.1 ppm), and three oxygen-bearing
carbon atoms, that is, C-2 (δC = 71.9 ppm), C-5 (δC
=
69.4 ppm), and C-7 (δC = 67.5 ppm), were identified.
Furthermore, HMBC correlations from 8-H2 (δH = 2.63,
2.68 ppm) to C-7 (δC = 67.5 ppm) and to C-6 (δC
41.5 ppm) and NOESY correlations from 5-H (δH
=
=
5.03 ppm) to the neighboring CH2 groups (6-CH2 with δC
= 41.5 ppm and δH = 1.58 ppm; 4-CH2 with δC = 35.2 ppm
and δH = 1.79 ppm) and to 3-CH (δC = 48.1 ppm and δH
=
4.04 ppm), as well as from 3-H to 2-H (δH = 3.91 ppm) were
indicative for the novel β-amino acid Atpoa (3-amino-2,5,7-
trihydroxy-8-phenyloctanoic acid, Figure 1). The second
amino acid in fragment A could easily be determined by
COSY, HMBC, and TOCSY experiments as N-acetyl-N-
methylleucine. The N-methyl group 73-CH3 (δC = 31.7 ppm
Scheme 1. Compounds 1 and 2.
From the fragmentation pattern obtained from
a
MALDI-TOF MS–MS experiment, the immonium ions and δH = 2.80 ppm) possessed, besides its correlation to C-
(+NH2=CH–R) of three amino acids could be detected in 66 (δC = 53.9 ppm), a correlation to the carbonyl group at
Figure 1. Structure of fragments A–C and selected HMBC and NOE correlations for compound 1 (single-headed arrow = HMBC;
double-headed arrow = NOESY).
Eur. J. Org. Chem. 2008, 1732–1739
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1733

G. M. König et al.
FULL PAPER
C-71 (δC = 170.0 ppm), which is part of an acetyl residue. COSY correlations to 43-H (δH = 4.38 ppm), 45-H2 (δH
=
1
A very prominent fragment at m/z = 1299 (62%) in the 1.19, 1.25 ppm), and 47-H3 (δH = 0.83 ppm). The H–1H-
MALDI-TOF MS–MS experiment suggested the loss of the COSY correlation between 45-H2 and 46-H3 (δH
=
N-acetyl-N-methylleucine (Mr = 187.2) from 1 by elimi- 0.84 ppm) completed the isoleucine residue. The NH func-
nation, which led to the assumption that N-acetyl-N-meth- tionality of serine (40-NH, δH = 7.87 ppm) had an NOE
ylleucine is attached through an ester linkage and probably correlation to the NH group of isoleucine (43-NH, δH
forms an exocyclic side chain.[4,8] The characteristic 1H 8.27 ppm). Finally, 47-H3 (δH = 0.83 ppm) showed an NOE
NMR shift of 5-H (δH = 5.03 ppm) of Atpoa was a strong correlation to the α-CH of proline (II) (49-H, δH
=
=
indication for the presence of an acyl moiety at this posi- 4.36 ppm), and 52-H2 (δH = 3.52, 3.83 ppm) of this proline
tion. It was thus concluded that N-acetyl-N-methylleucine residue showed an NOE correlation to the α-CH of proline
is attached to C-5 (δC = 69.4 ppm) through an ester bridge. (III) (54-H, δH = 4.61 ppm). These correlations connected
1
Because of the overlapping H NMR signals of 5-H and serine to isoleucine, which was extended by two proline resi-
66-H (δH = 5.03 and 5.00 ppm, respectively) an HMBC cor- dues and completed fragment B (Figure 1).
relation between 5-H and C-65 (δC = 170.4 ppm) could not
Fragment C finally consisted of two further amino acids.
be detected unambiguously. However, all experimental data The 59-CH2 (δC = 40.7 ppm and δH = 3.85, 3.93 ppm)
and comparison with literature data[3,4,8] led us to conclude methylene group had no 1H–1H-COSY correlations and no
that N-acetyl-N-methylleucine is bound through an ester correlations in the HSQC–TOCSY spectrum. An HMBC
functionality to C-5 of the β-amino acid Atpoa, which gives correlation between 59-H2 and the carbonyl group at C-58
rise to fragment A.
(δC = 166.2 ppm) clearly delineated the amino acid glycine.
1H and 13C NMR spectroscopic data indicated a second The 62-H2 (δH = 1.81, 1.97 ppm) protons offered H–1H-
aromatic moiety, which consisted of a 1,4-disubstituted COSY correlations to 61-H (δH = 4.26 ppm) and 63-H2 (δH
benzene ring as shown by characteristic resonances for 25/ = 2.10, 2.14 ppm). The 63-H2 protons had HMBC corre-
26-H (δH = 7.10 ppm) and 27/28-H (δH = 6.80 ppm). In the lations to the carbonyl group at C-64 (δC = 173.7 ppm),
HMBC experiment, a strong correlation between 25/26-H whereas 61-H exhibited an HMBC correlation to the car-
and C-23 (δC = 29.9 ppm), which is part of the 21-CH, 22- bonyl functionality at C-60 (δC = 170.6 ppm), which pro-
CH2, and 23-CH2 spin system, was detectable. An HMBC vided evidence for the glutamine residue. A short sequence
1
correlation between 30-H3 (δH = 3.71 ppm) and C-29 (δC
157.3 ppm) gave proof for a methoxy group to be connected through the heteronuclear long-range correlations between
to C-29. Thus, this aromatic amino acid could be deter- the CH2 group of glycine (C-59, δC = 40.7 ppm and δH
mined as O-methylhomotyrosine (MeHty). The 19-H2 (δH 3.85, 3.93 ppm) and the carbonyl functionalities of both
= 3.35 ppm) proton of proline (I) showed an NOE corre- glycine (C-58, δC = 166.2 ppm) and glutamine (C-60, δC
lation to 21-H (δH = 4.31 ppm) of MeHty, which made clear 170.6 ppm). NOE correlations between 62-H2 (δH = 1.81,
that both amino acids were attached to each other as illus- 1.97 ppm) and the NH group of glycine (59-NH, δH
=
consisting of glutamine and glycine could be assigned
=
=
=
trated in Figure 1. The spin system 32-CH, 33-CH, and 34- 7.99 ppm) confirmed the connection of these two amino ac-
CH3, with its characteristic shift of the β-33-CH group (δH ids and secured fragment C (Figure 1).
= 4.15 ppm and δC = 65.9 ppm), belongs to a threonine
Connection of the three fragments was achieved by ana-
residue. The protons 37-H (δH = 5.69 ppm) and 38-H3 (δH lyzing MALDI-TOF MS–MS measurements. Two amino
= 1.88 ppm) possessed an HMBC correlation to C-36 (δC acid sequences, that is, -Pro-Pro-Ile-Ser-Dhb-Thr-MeHty
= 129.8 ppm). In addition, 37-H showed an HMBC corre- (m/z = 770) and -Pro-(Atpoa-AcMeLeu)-Gln-Gly- (m/z =
lation to the carbonyl functionality at C-35 (δC
=
717), were identified as outlined in Table 1. Both of these
163.9 ppm). Thus, the unusual amino acid Dhb (dehy- sequences start with a proline residue at the N-termini,
drobutyrine) was assigned. The methyl group of threonine which is in accordance with the proposal of Schilling et
(34-H3, δH = 1.11 ppm) had NOE correlations to the NH al.[12] that states that cyclic and proline-containing peptides
functionality of O-methylhomotyrosine (21-NH, δH
7.92 ppm) and to the methyl group of Dhb (38-H, δH
=
=
favorably break in such a manner that the proline residue
is positioned at the N-terminated end of the resulting linear
1.88 ppm). This way, the fragment could be extended in the peptide. Both sequences taken together give the linear pep-
N-terminal direction to form the short sequence -Dhb-Thr- tide -Pro-(Atpoa-AcMeLeu)-Gln-Gly-Pro-Pro-Ile-Ser-Dhb-
MeHty-Pro(I)-. The α-40-CH (δC = 56.3 ppm and δH
=
Thr-MeHty-. The only possibility to get a cyclic structure
4.25 ppm) proton showed HMBC correlations to the oxy- from this linear peptide was the connection of the carboxyl
gen atom bearing C-41 (δC = 61.3 ppm) and to the carbonyl group of glycine and the amine group of proline (III). The
functionality at C-39 (δC = 169.5 ppm), which is indicative presence of further fragment ions (see Table 1), for example,
of a serine residue. The 36-NH resonating at δ = 9.14 ppm a fragment for -Gly-Pro-Pro- (m/z = 252), confirmed this
showed clear HMBC correlations to both the carbonyl proposal. Thus, the planar structure of compound 1 was
carbon atoms of serine (C-39, δC = 169.5 ppm) and Dhb completed.
(C-35, δC = 163.9 ppm), in which the intensity of the corre-
For the assignment of the absolute configuration by chi-
lation between 36-NH and C-39 was much stronger. Thus, ral GC–MS and chiral HPLC, the unusual amino acid O-
it was obvious that serine was attached to Dhb as shown in methylhomotyrosine was synthesized in its  and  enantio-
Figure 1. The 44-H (δH = 2.12 ppm) proton showed 1H–1H- mers as described previously by Yamada et al.[13] This two-
1734
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1732–1739

Novel β-Amino Acid in Cytotoxic Peptides from Cyanobacterium
Table 1. MALDI-TOF-PSD fragment ions of 1.
Fragment
m/z (%)[b]
[M[a] – 187 (AcMeLeu)]
[M – 191 (MeHty)]
[M – 101 (Thr) – 191 (MeHty)]
[M – 83 (Dhb) – 101 (Thr) – 191 (MeHty)]
[M – 87 (Ser) – 83 (Dhb) – 101 (Thr) – 191 (MeHty)]
[83 (Dhb) +101 (Thr) + 191 (MeHty) + 97 (Pro) + H]+
[97 (Pro) + 434 (Atpoa-AcMeLeu) + 128 (Gln) + 57 (Gly) + H]+
[97 (Pro) + 434 (Atpoa-AcMeLeu) + 128 (Gln) + H]+
[97 (Pro) + 434 (Atpoa-AcMeLeu) + H]+
1299 (62)
1295 (10)
1194 (100)
1111 (19)
1024 (22)
473 (4)
717 (4)
660 (17)
532 (1)
[434 (Atpoa-AcMeLeu) + 128 (Gln) + H]+
[57 (Gly) + 97 (Pro) + 97 (Pro) + 113 (Ile) + 87 (Ser) + 83 (Dhb) + 101 (Thr) + 191 (MeHty) + 97 (Pro) + H]+
[97 (Pro) + 97 (Pro) + 113 (Ile) + 87 (Ser) + 83 (Dhb) + 101 (Thr) + 191 (MeHty) + H]+
[97 (Pro) + 97 (Pro) + 113 (Ile) + 87 (Ser) + 83 (Dhb) + 101 (Thr) + H]+
[97 (Pro) + 97 (Pro) + 113 (Ile) + 87 (Ser) + 83 (Dhb) + H]+
[97 (Pro) + 97 (Pro) + 113 (Ile) + 87 (Ser) + H]+
[97 (Pro) + 97 (Pro) + 113 (Ile) H]+
563 (1)
924 (7)
770 (19)
579 (4)
478 (20)
395 (16)
308 (3)
195 (3)
381 (2)
298 (2)
211 (3)
[97 (Pro) + 97 (Pro) + H]+
[97 (Pro) + 113 (Ile) + 87 (Ser) + 83 (Dhb) + H]+
[97 (Pro) + 113 (Ile) + 87 (Ser) + H]+
[97 (Pro) + 113 (Ile) +H]+
[128 (Gln) + 57 (Gly) +H]+
186 (1)
252 (4)
155 (8)
[57 (Gly) + 97 (Pro) + 97 (Pro) + H]+
[57 (Gly) + 97 (Pro) + H]+
[a] M = [1485 + H]+. [b] Values were rounded down.
step procedure involved diastereoselective aza-Michael ad- 30 between compound 1 and 2, which corresponds to a
dition of enantiopure 1-phenylethylamine to the commer- -CH2O- group. The molecular formula was therefore
cially available 3-(4-methoxybenzoyl)acrylic acid, followed C72H105N13O19. In the NMR spectra, no signal for a meth-
by hydrogenolysis of the carbon–heteroatom bonds in the oxy group was discernible and an obvious change in the
benzylic positions (for details, see Supporting Information). aromatic region of the 1H NMR spectrum could be ob-
The absolute configuration of the amino acids was deter- served. Therefore, we concluded that in compound 2, which
mined as: -serine, -threonine, -glutamine, -O-methyl- was named tychonamide B, the O-methylhomotyrosine of
homotyrosine, -proline, and -allo-isoleucine. /-N-meth- 1 is replaced by a homophenylalanine residue.
ylleucine resulting from N-acetyl-N-methylleucine, which is
Both compounds were tested in a cytotoxicity assay
deacetylated during hydrolysis, could not be separated by against 37 human cancer cell lines. Compound 1 inhibited
chiral GC–MS. Thus, we determined the absolute configu- the growth of all 37 cell lines with similar potency; the mean
ration by chiral HPLC and showed that the  isomer is part IC50 value was 0.9 µgmL–1 and the IC50 values ranged from
of compound 1. A strong NOE correlation between the NH 0.3 to 1.3 µgmL–1. Compound 2 was more selective than 1
group (36-NH, δH = 9.14 ppm) and the olefinic proton of and the IC50 values ranged from 1.0 (LXF H460; lung) to
Dhb (37-H, δH = 5.69 ppm) indicated the E configuration 14.9 µgmL–1 (PRXF LNCAP; prostate). The mean IC50
for the ∆36,37 double bond. Thus, with the exception of the value was 3.3 µgmL–1. Inhibition of clonogenicity of tumor
four chiral centers in Atpoa, the structure was defined. The cells was evaluated in additional tumor models by using
determination of the configuration of Atpoa would include a clonogenic assay. The antiproliferative activity of 2 was
several derivatization steps and excessive NMR spectro- evaluated in cell suspensions prepared from 22 human
scopic analysis.[7,8] These investigations were beyond the tumor xenografts of 11 different tumor types. In addition,
scope of this study but represent issues of future research.
2 was tested in a preparation of hematopoietic stem cells
Compound 2 (Scheme 1) showed a positive ESI quasi- as a model system for nonmalignant tissue. Compound 2
molecular ion at m/z = 1456.8 [M + H]+. Its HRFTICR inhibited anchorage-independent growth and in vitro col-
mass (calcd. 728.8903, observed 728.8898 for [M + 2 H]2+
)
ony formation in a concentration-dependent and tumor-
led us to conclude that there is a mass difference of type-selective manner. The mean IC50 value in this assay
Table 2. Antiprotozoal activity of compounds 1 and 2.[a]
Compound
T. b. rhod.
T. cruzi
L. don.
P. falc.
Cytotox· L6
IC50 [µgmL–1] (drug[b]
)
IC50 [µgmL–1] (drug[c]
)
IC50 [µgmL–1] (drug[d]
)
IC50 [µgmL–1] (drug[e]
)
IC50 [µgmL–1] (drug[f])
1
2
0.116 (0.003)
0.241 (0.002)
3.51 (0.563)
5.48 (0.512)
1.76 (0.101)
2.38 (0.116)
1.47 (0.078)
1.81 (0.078)
4.11 (0.009)
4.83 (0.009)
[a] T. b. rhod. = Trypanosoma brucei rhodesiense (strain STIB 900), T. cruzi = Trypanosoma cruzi (strain Tulahuen C4), L. don. =
Leishmania donovani (strain MHOM-ET-67/L82), P. falc. = Plasmodium falciparum (strain K1). [b] Melarsoprol. [c] Benznidazole. [d]
Miltefosine. [e] Chloroquine. [f] Podophyllotoxin.
Eur. J. Org. Chem. 2008, 1732–1739
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1735

G. M. König et al.
FULL PAPER
was 2.4 µgmL–1 and the IC50 values ranged from 0.2
(MEXF 989, melanoma) to 4.0 µgmL–1 (PAXF 736, pan-
creas). Colony formation of hematopoietic stem cells was
inhibited by 2 with an IC50 value of 2.8 µgmL–1.
Tychonamide A and B were tested against several proto-
zoans: Plasmodium falciparum, Trypanosoma cruzi, Leish-
mania donovani, and sleeping sickness causing Trypanosoma
brucei rhodesiense. Both compounds were active against all
tested protozoans, and 1 (IC50 value towards T. b. rhod.
0.1 µgmL–1) seems to be more potent than 2 (see Table 2).
Discussion
The most unusual structural feature of compounds 1 and
2 is the presence of a novel β-amino acid. One of the natural
functions of such β-amino acids could be the enhancement
of the proteolytic stability towards peptidases,[14] which
supports the ecological function of these secondary meta-
bolites. The most common β-amino acids in cyanobacterial
peptides is Adda [(2S,3S,8S,9S)-3-amino-9-methoxy-2,6,8-
trimethyl-10-phenyldeca-(4E,6E)-dienoic acid], which is
part of the hepatotoxic microcystins and nodularins. Adda,
Ahoa [(2S,3R,5R)-3-amino-2,5-dihydroxy-8-phenyloctanoic
acid[7]], Ahda [(2S,3R,5S)-3-amino-2,5,9-trihydroxy-10-
phenyldecanoic acid[8]], and the novel amino acid Atpoa
are all aromatic β-amino acids, in contrast to, for example,
Figure 2. Proposed biosynthesis of cyanobacterial aromatic β-
amino acids on the basis of literature.[5,15] KR = ketoreductase, MT
= methyltransferase, DH = dehydratase, AMT = aminotransferase,
H = hydroxylation.
Aound
(3-amino-2,7,8-trihydroxy-10-methyl-5-oxyunde-
canoic acid[3]), which is aliphatic. For Adda, the biosyn-
thetic pathway and the corresponding gene cluster was de-
scribed.[5,15] Most unusual is the phenylacetate starter unit,
which is prolonged with four malonyl CoA units to form
Adda. On the basis of this biosynthesis, a similar pathway
(phenylacetate as starter prolonged with three-to-four mal-
onyl CoA units) can be suggested for Ahoa, Ahda, and At-
poa (Figure 2). The main difference lies in the hydroxyla-
tion and the methylation pattern on one hand and the
length of the side chain on the other.
Tychonamide A (1) showed cytotoxic potency with a
mean IC50 value of 0.9 µgmL–1 in a panel consisting of 37
tumor cell lines. However, antitumor selectivity, that is,
differential antitumor activity, was only slightly pro-
nounced, which indicates a rather unspecific and nonselec-
tive activity. Tychonamide B (2) showed less cytotoxic po-
tency (mean IC50 = 3.3 µgmL–1), but selectivity was more
pronounced, that is, for nonsmall-cell lung cancer, mela-
noma, and ovarian cancer. Furthermore, 2 displayed ac-
more sensitive than the mean IC50 of all models tested.
Moreover, these sensitive tumor models were on average
about sevenfold more sensitive than hematopoietic stem
cells as representative model system for nonmalignant tis-
sue, which is indicative of the satisfactory therapeutic index
of the compound.
Compound 1 showed pronounced antiprotozoal activity
towards Trypanosoma brucei rhodesiense in the nanomolar
range (IC50 value of about 100 ngmL–1). This result has to
be interpreted with care because of the cytotoxicity of the
compound. The antitrypanosomal activity is, however, ap-
proximately 40-fold higher than the cytotoxicity [IC50 value
of 4.1 µgmL–1; rat skeletal myoblasts (L6-cells)].
Experimental Section
General Procedure: HPLC was performed with a Merck–Hitachi
tivity and selectivity in clonogenic assays against tumor cell system equipped with an L-6200A pump, an L-4500A photodiode
array detector, a D-6000A interface with D-7000 HSM software,
and a Rheodyne 7725i injection system to get the first separation
step and the chiral HPLC runs. A Waters system, controlled by
Waters millenium software, consisting of a 717 plus autosampler,
600 controller pump with in-line degasser, and a 996 photodiode
array detector was then used to purify peptide 1 and 2. 1H, 13C,
COSY, HSQC, HMBC, NOESY, and TOCSY NMR spectra were
recorded in [D6]DMSO by using either a Bruker Avance 500 DRX
or a 600 DRX spectrometer operating at 500 or 600 MHz for pro-
ton and at 125 or 150 MHz for 13C. Spectra were calibrated to
suspensions from 22 solid tumors grown in semisolid me-
dium. Those tumor cells, which represent 11 different tumor
types, were freshly prepared from solid human tumor xeno-
grafts grown in nude mice. Cells that show anchorage inde-
pendent growth in semisolid medium contain, to a certain
extent, tumor stem cells that are considered to be responsi-
ble for the metastatic and infiltrative potential of a
tumor.[16–18] Tychonamide B (2) selectively inhibited colony
formation of melanoma cells. On the basis of IC50 values,
these sensitive tumor models were on average about sixfold residual solvent signals with resonances at δH/C = 2.5/39.5 ppm
1736 Eur. J. Org. Chem. 2008, 1732–1739
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Novel β-Amino Acid in Cytotoxic Peptides from Cyanobacterium
Table 3. 13C and 1H NMR spectral data for 1 in [D6]DMSO at
75 °C.
([D6]DMSO). Chiral GC–MS analyses were performed with a Per-
kin–Elmer Turbomass mass spectrometer by using an Alltech Cap-
illary Chirasil-Val column. UV and IR spectra were measured with
Perkin–Elmer Lambda 40 and Perkin–Elmer Spectrum BX instru-
ments. Optical rotations were obtained by using a Jasco DIP 140
polarimeter. The high-resolution mass spectra were recorded with
a Bruker Apex II FTICR (7 tesla magnet). The MS–MS MALDI-
TOF spectra were recorded with an Applied Biosystems 4700 Pro-
teomics Analyzer 7021 with an α-cyanocinnamic-acid matrix.
Amino acid
No
δH [ppm][a]
δC [ppm]
(mult., J [Hz])
[b]
Atpoa
1
–
2
3
3.91
4.04
7.30
1.79
5.03
1.58
3.67
2.63, 2.68
–
7.18
7.18
7.23
71.9
48.1
–
3-NH[c]
4
35.2
69.4
41.5
67.5
43.6
138.8
128.9
128.9
127.5
127.5
Biological Material: The cyanobacterial strain (for taxonomy see
ref.[19]) was isolated from a sample collected from a pond of a sugar
factory in Wierthe, Germany, and deposited in the Culture Collec-
tion of Algae and Protozoa (CCAP 1462/13) in Oban (Scotland,
United Kingdom). The cyanobacterium was cultivated in a 25-L
photobioreactor (Planktotec system Pluto) with sterile filtered BG-
11 medium (Sigma Aldrich, C3061 BG-11) over a period of
9 months. During cultivation at 25 °C, the culture was constantly
illuminated with white fluorescent light (Osram L 58W/11–869).
Oxygen was supplied by a continuous stream of sterile air from the
lower end of the bioreactor. The pH was titrated to a value between
7 and 10 by using an adequate volume of carbon dioxide.
5
6
7
8
9
10
11
12
13
7.23
14
15
7.15
–
125.3
[b]
-Proline (I)
16
17
18
4.28
1.86
1.75, 1.81
59.7
27.8
23.9
Isolation Procedure: The lyophilized biomass from a 42-L culture
(240 g) was exhaustively extracted with dichloromethane (8 L),
ethyl acetate (6 L), and methanol (4 L) in three steps. The dried
methanol extract amounted to 20.6 g. The 1H NMR and MALDI-
TOF spectroscopic measurements of this extract indicated the pres-
ence of several peptidic compounds. 12 g of the methanol extract
were dissolved in water and repeatedly extracted with diethyl ether
(1.5 L) and butanol (4.5 L), which resulted in 3.0 g of butanol ex-
tract with a high peptidic content. The crude butanol extract was
fractionated by reverse-phase HPLC (Macherey–Nagel, Nucleodur
100, C18, 5 µm, 250ϫ10 mm) by using a gradient elution (MeOH/
H2O, 2:3 to MeOH over 30 min, 2 mLmin–1) to yield six fractions.
Further reverse-phase HPLC separation of fraction 4 (307.1 mg)
(Knauer, Eurospher 100, C18, 5 µm, 250ϫ8 mm, MeOH/H2O,
72:28, 1 mLmin–1) led to the isolation of 74 mg of compound 1
and 23 mg of compound 2 (0.62 and 0.19% of the crude methanol
extract, respectively).
19
20
21
21-NH
22
23
24
25
26
27
28
29
30
31
32
32-NH
33
34
35
36
36-NH
37
38
39
40
40-NH
41
42
43
43-NH
44
45
46
47
48
49
50
3.35, 3.47
–
4.31
7.92
1.92
2.52, 2.59
–
7.10 (d, 7.8)
7.10 (d, 7.8)
6.80 (d, 7.8)
6.80 (d, 7.8)
46.3
[b]
-O-Methyl-
homotyrosine
50.9
–
32.4
29.9
133.0
128.9
128.9
113.5
113.5
157.3
–
3.71 (s)
–
4.29
7.64
4.15
1.11 (d, 6.6)
–
–
54.7
[b]
-Threonine
58.8
–
65.9
19.4
163.9
129.8
–
124.5
12.7
169.5
56.3
–
Compound 1: White amorphous solid. [α]2D4 = –2.4 (c = 2.48,
Dhb
MeOH). UV (MeOH): λ (ε, –1 cm–1) = 203 (79238), 228 sh
9.14
(28970), 250 sh (6034), 273 sh (2127) nm. IR (ATR): ν = 3321,
˜
2958, 2360, 2341, 1733, 1618, 1511, 1449, 1244, 1202, 1024 cm–1.
1H and 13C NMR spectroscopic data see Table 3. HRMS (FTICR):
calcd. for C73H109N13O20, [M + 2 H+] 743.8950; found 743.8951.
5.69 (q, 7.3)
1.88 (d, 7.3)
–
4.25
7.87
3.70, 3.78
–
4.38
8.27
2.12
1.19, 1.25
0.84
0.83
–
4.36
2.16
2.09
3.52, 3.83
–
4.61
-Serine
Compound 2: White amophous solid. [α]2D4 = –7.7 (c = 1.52,
MeOH). UV (MeOH): λ (ε, –1 cm–1) = 203 (81991), 228 sh
61.3
[b]
(19681), 250 sh (6234) nm. IR (ATR): ν = 3327, 2934, 2362, 2340,
˜
-allo-Isoleucine
1620, 1541, 1453, 1247, 1202, 1024 cm–1. HRMS (FTICR): calcd.
for C72H107N13O19, [M + 2 H+] 728.8898; found 728.8903.
54.7
–
34.8
25.5
11.0
Chiral GC-MS: Peptide 1 (3.5 mg) was dissolved in HCl (6 , 1 mL)
and heated for 16 h at 110 °C in a closed vial. After this procedure,
the solvent was removed by a stream of nitrogen. The dried hydrol-
ysate was treated with isopropyl alcohol (500 µL) and acetyl chlo-
ride (150 µL) at 110 °C for 1 h to create the isopropyl esters of the
amino acids. After removal of the solvent, the dry residues were
acylated [trifluoroacetic anhydride (0.4 mL) in CH2Cl2 (0.4 mL);
heating for 15 min at 110 °C]. The reaction mixture was dried and
dissolved in EtOAc (1 mL). A 1 µL portion of this solution was
analyzed by GC–MS by using an Alltech Capillary Chirasil-Val
column (25 mϫ0.25 mm; 0.16 µm; program rate: column tempera-
ture held at 50 °C for 3 min; 50 °C–180 °C at 4 °Cmin–1; flow:
0.6 mLmin–1; injector temperature: 250 °C).
14.0
[b]
-Proline (II)
-Proline (III)
60.3
28.4
24.7
51
52
53
54
55
56
57
46.7
[b]
57.6
27.1
23.9
45.5
1.76, 2.22
1.89
3.51
Eur. J. Org. Chem. 2008, 1732–1739
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1737

G. M. König et al.
FULL PAPER
conditions (500 µg peptide; 6  HCl; 88 °C for 12 h) was done to
minimize epimerization. The HPLC chromatogram of the latter ex-
periment showed that the ratio N-Me--Leu to N-Me--Leu was
shifted in favor of N-Me--Leu (ratio: 12:1). Consequently it was
concluded that the  form is present in 1.
Table 3. (Continued)
Amino acid
Glycine
No
δH [ppm][a]
(mult., J [Hz])
δC [ppm]
58
59
59-NH
60
61
61-NH
62
63
64
64-NH2
–
166.2
40.7
–
170.6
52.2
–
28.1
31.3
173.7
–
170.4
53.9
36.5
24.1
22.4
21.1
170.0
21.0
31.7
3.85, 3.93
7.99
–
4.26
7.65
Antitumor Testing. Monolayer Assay: A modified propidium iodide
assay was used to determine the cytotoxic activity of the com-
pounds against human tumor cell lines. The test procedure was
described elsewhere.[20] Cell lines tested were derived from patient
tumors engrafted as a subcutaneously growing tumor in NMRI nu/
nu mice, or obtained from the American Type Culture Collection,
Rockville, MD, USA, National Cancer Institute, Bethesda, MD,
USA, or Deutsche Sammlung von Mikroorganismen und Zellkul-
turen, Braunschweig, Germany. Briefly, human tumor cells lines
were grown at 37 °C in a humidified atmosphere (95% air, 5% CO2)
in monolayer cultures in RPMI 1640 medium supplemented with
10% FCS and phenol red (PAA, Cölbe, Germany). Cells were tryp-
sinized and maintained weekly. Cells were harvested from exponen-
tially growing cultures by trypsination, counted and plated in 96-
well flat-bottomed microplates (140 µL cell suspension, 5ϫ103 to
10ϫ103 cells per well). After a 24 h recovery to allow cells to re-
sume exponential growth, 10 µL of culture medium (6 control wells
per plate) or medium containing the test drug were added to the
wells. Each drug concentration was plated in triplicate. After 4 d
of incubation the culture medium was replaced by fresh medium
containing 6 µgmL–1 of propidium iodide. Microplates were then
kept at –18 °C for 24 h, to give a total cell kill. After thawing of
the plates, fluorescence was measured by using the Cytofluor 4000
microplate reader (Perseptive Biosystems) (excitation 530 nm, emis-
sion 620 nm). The amount of viable cells was proportional to the
fluorescence intensity.
-Glutamine
1.81, 1.97
2.10, 2.14
–
n.d.[d]
–
N-acetyl-N-methyl- 65
leucine
66
67
68
69
70
71
72
73
5.00
1.63
1.45
0.88 (d, 6.6)
0.84
N-Acetyl
N-Methyl
–
1.96 (s)
2.80 (s)
[a] The multiplicity of many 1H resonances was not determined due
to broad and overlapping signals. [b] Chemical shifts of the remain-
ing unassigned carbonyl carbon atoms: 169.7, 170.3, 170.5, 170.9,
171.0, 171.2, 171.7 ppm. [c] The assignment of the chemical shifts
of the amide protons was done at room temperature (apart from
36-NH). [d] n.d. = not determined.
Derivatization of the amino acids was performed by using 5–10 mg
of - and -amino acids (for Ile and Thr the configurations -, -
allo, - and -allo were taken) and processed as described above.
Before being analyzed by GC–MS, the solutions of standard amino
acids were diluted with EtOAc (1:50). The retention times of the
N-trifluoroacetylisopropyl ester of the amino acids were compared
with those of the derivatized standards. In the case of Pro, the
sample was spiked with both standards (because of the small differ-
ence in the retention times: -Pro 19.98 min, -Pro 20.05 min) to
show that the  enantiomer is present. -Thr and Gly could not be
separated from the complex amino acid mixture of the hydrolysate
with this method. By spiking experiments and comparison of the
mass spectra it could be shown that both amino acids were present
in the hydrolysate and eluted with tR = 14.80 min. [Retention times
for the standards: -Thr 14.26 min, -Thr 14.80 min, -allo-Thr
17.34 min, -allo-Thr 17.88 min, Gly 14.97 min, -Ile 16.50 min, -
Ile 17.03 min, -allo-Ile 16.06 min, -allo-Ile 16.69 min, N-Me-Leu
17.36 min (L and D not to separate), -Ser 17.03 min, -Ser
17.53 min, -Pro 19.98 min, -Pro 20.09 min, -Gln 28.00 min, -
Gln 28.38 min, -O-MeHtyr 37.82 min, -O-MeHtyr 38.11 min;
compound 1: -Thr 14.80 min, glycine 14.80 min, -allo-Ile
15.97 min, N-Me-Leu 17.28 min, -Ser 17.40 min, -Pro 20.09 min,
-Gln 27.87 min, -O-MeHtyr 37.73 min].
Clonogenic Assays with Human Tumor Xenografts and Hematopoi-
etic Stem Cells: Effects of the test compound on clonogenicity of
tumor cells were investigated in a clonogenic assay. Tumor xeno-
grafts were derived from patient tumors engrafted as a subcutane-
ously growing tumor in NMRI nu/nu mice obtained from Onco-
test’s breeding facility. Details of the test procedure have been de-
scribed earlier.[21] Briefly, solid human tumor xenografts were re-
moved from mice under sterile conditions, mechanically disaggre-
gated and subsequently incubated with an enzyme cocktail con-
sisting of collagenase type IV (41 UmL–1), DNase I (125 UmL–1),
hyaluronidase type III (100 UmL–1), and dispase II (1.0 UmL–1)
in RPMI 1640-Medium at 37 °C for 45 min. Cells were passed
through sieves of 200 and 50 µm mesh size and washed twice with
sterile PBS-buffer. The percentage of viable cells was determined in
a Neubauer-hemocytometer using trypan blue exclusion. The bot-
tom layer consisted of 0.2 mL per well Iscove’s Modified Dulbec-
co’s Medium (Invitrogen), supplemented with 20% (v/v) fetal calf
serum (Sigma), 0.01% (w/v) gentamicin (Invitrogen) and 0.75%
(w/v) agar (BD Biosciences). 1.5ϫ104 to 4ϫ104 cells were added
to 0.2 mL of the same culture medium supplemented with 0.4%
Chiral HPLC: Compound 1 (600 µg) was hydrolyzed in HCl (6 ) (w/v) agar and plated in 24-multiwell dishes onto the bottom layer.
at 110 °C for 16 h. The hydrolysate was dried with a stream of
nitrogen. The residue was dissolved in the mobile phase used for
the chiral HPLC (2 m CuSO4 in H2O/MeCN, 95:5; flow rate
1 mLmin–1). The chiral HPLC was performed by using a Chirex
3126 (D)-penicillamine column (Phenomenex; 250ϫ4.60 mm). In
this experiment the retention times of the standards were 40.15 (N-
Me--Leu) and 59.89 min (N-Me--Leu). In the chromatogram of
the hydrolysate, two signals were obtained: one strong signal be-
longing to the  enantiomer and a smaller one belonging to the 
enantiomer (ratio: 4:1). We proposed that an epimerization process
took place during hydrolysis. Therefore, hydrolysis under milder
The test compounds were applied by continuous exposure (drug
overlay) in 0.2 mL of culture medium. Every dish included six un-
treated control wells and drug-treated groups in triplicate at 6 con-
centrations. Cultures were incubated at 37 °C and 7.5% CO2 in a
humidified atmosphere for 7–20 d and monitored closely for colony
growth using an inverted microscope. Within this period, in vitro
tumor growth led to the formation of colonies with a diameter of
Ͼ50 µm. At the time of maximum colony formation, counts were
performed with an automatic image analysis system (OMNICON
3600, Biosys GmbH). 24 h prior to evaluation, vital colonies were
stained with a sterile aqueous solution of 2-(4-iodophenyl)-3-(4-
1738
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1732–1739

Novel β-Amino Acid in Cytotoxic Peptides from Cyanobacterium
nitrophenyl)-5-phenyltetrazolium chloride (1 mgmL–1, 100 µL per
well).
Supporting Information (see footnote on the first page of this arti-
cle): Proposed MS elimination pattern of the exocyclic amino acid,
description of the synthesis of the enantiomers of O-methylhomo-
tyrosine, detailed cytotoxicity data, and NMR spectroscopic data
and purity data of compound 1 and 2.
For testing hematopoietic stem cells, buffy coats or samples of hu-
man umbilical cord blood were diluted 2- to threefold with PBS
containing 0.1% (w/v) BSA. Peripheral blood mononuclear cells
were enriched from the respective samples by Ficoll Paque density
gradient centrifugation and washed twice with PBS containing
0.1% (w/v) BSA. The resulting cell suspension was stored in ali-
quots in freezing medium at –80 °C. Aliquots were thawed for test-
ing as appropriate. The colony forming test was performed using
6-well plates and HSC-CFU (Miltenyi Biotec) as culture medium.
1.25ϫ104 to 1.5ϫ104 cells per mL of the above mentioned prepa-
ration were seeded in a final volume of 1.0 mL per well. Solutions
of the test article were added directly to the medium. Every 6-well
dish contained one triplicate of one test concentration of one test
compound. The other 3 wells of the test plate were filled with
2.0 mL of sterile water to ensure that maximum humidity is at-
tained during the subsequent incubation period. Two 6-well plates
received one triplicate of untreated cells each, which served as un-
treated control for the whole experiment. Cultures were incubated
at 37 °C and 7.5% CO2 in a humidified atmosphere for 14 d. Col-
ony growth was evaluated by eye using an inverted microscope. For
detailed results see the Supporting Information.
Acknowledgments
The authors thank Dr. Christian Richter, Johann Wolfgang Goethe
University, Frankfurt, Germany, for the measurement of the NMR
spectra with the 600 MHz spectrometer. Olaf Papendorf is thanked
for isolating the cyanobacterium. Financial support from BMBF
(Project No. 03F0415A) is gratefully acknowledged.
[1] K. Harada, Chem. Pharm. Bull. 2004, 52, 889–899.
[2] D. Müller, A. Krick, S. Kehraus, C. Mehner, M. Hart, F. C.
Küpper, K. Saxena, H. Prinz, H. Schwalbe, P. Janning, H.
Waldmann, G. König, J. Med. Chem. 2006, 49, 4871–4878.
[3] I. Pergament, S. Carmeli, Tetrahedron Lett. 1994, 35, 8473–
8476.
[4] T. An, T. K. S. Kumar, M. Wang, L. Liu, J. O. Lay Jr., R. Li-
yanage, J. Berry, M. Gantar, V. Marks, R. E. Gawley, K. S.
Rein, J. Nat. Prod. 2007, 70, 730–735.
[5] D. Tillett, E. Dittmann, M. Erhard, H. von Döhren, T. Börner,
B. A. Neilan, Chem. Biol. 2000, 7, 753–764.
Antiprotozoal Activity: Antiplasmodial activity was determined
against the K1 strain of Plasmodium falciparum by using a modified
[3H] hypoxanthine incorporation assay. Infected human erythro-
cytes were exposed to serial drug dilutions in microtiter plates for
48 h at 37 °C in a gas mixture with reduced oxygen and elevated
CO2. [3H] hypoxanthine was added to each well and after further
incubation for 24 h the wells were harvested on glass fiber filters
and counted in a liquid scintillation counter. From the sigmoidal
inhibition curve the IC50 value was calculated. Chloroquine was
used as positive control.
[6] A. Sandström, C. Glemarec, J. A. O. Meriluoto, J. E. Eriksson,
J. Chattopadhyaya, Toxicon 1990, 28, 535–540.
[7] K. Fujii, K. Sivonen, T. Kashiwagi, K. Hirayama, K. Harada,
J. Org. Chem. 1999, 64, 5777–5782.
[8] G. L. Helms, R. E. Moore, W. P. Niemczura, G. M. L. Pat-
terson, J. Org. Chem. 1988, 53, 1298–1307.
[9] S. S. Moon, J. L. Chen, R. E. Moore, G. M. L. Patterson, J.
Org. Chem. 1992, 57, 1097–1103.
[10] G. Trimurtulu, I. Ohtani, G. M. L. Patterson, R. E. Moore,
T. H. Corbett, F. A. Valeriote, L. Demchik, J. Am. Chem. Soc.
1994, 116, 4729–4737.
[11] J. M. Gregson, J. L. Chen, G. M. L. Patterson, R. E. Moore,
Tetrahedron 1992, 48, 3727–3734.
[12] B. Schilling, W. Wang, J. S. McMurray, K. F. Medzihradsky,
Rapid Commun. Mass Spectrom. 1999, 13, 2174–2179.
[13] M. Yamada, N. Nagashima, J. Hasegawa, S. Takahashi, Tetra-
hedron Lett. 1998, 39, 9019–9022.
[14] F. von Nussbaum, P. Spiteller, “ß-Amino acids in Nature” in
Highlights in Bioorganic Chemistry: Methods and Applications
(Eds.: C. Schmuck, H. Wennemers), VCH, Weinheim, 2004, pp.
63–89.
Activity against Trypanosoma brucei rhodesiense (strain STIB 900)
was evaluated according Räz et al.[22] Parasites were grown axeni-
cally in culture medium supplemented with horse serum. Following
a 3-day exposure to test compounds, the viability of tryptomastig-
ote parasites was quantified using the dye Almar Blue by monitor-
ing the reductive environment of living cells. Fluorescence develop-
ment was expressed as percentage of the control, and IC50 values
were calculated. Melarsoprol was included as positive control.
Activity against Trypanosoma cruzi was determined according to
Bruckner et al.[23] The strain Tulahuen C4 of T. cruzi, which had
been transfected with the galactosidase lac-Z gene, was cultivated
for 4 d on rat skeletal myoblasts (5% CO2, 37 °C) in the presence of
test compounds. For measurement of the IC50 value the substrate
chlorophenol red-ß--galactopyranoside was added. The color re-
action that developed during the following 2–4 h because of the
β-galactosidase activity (change of color from yellow to red) was
quantified photometrically employing an ELISA reader. As a posi-
tive control, benznidazole was included in each test series.
[15] B. S. Moore, Nat. Prod. Rep. 2005, 22, 580–593.
[16] T. Reya, S. J. Morrison, M. F. Clarke, I. L. Weissman, Nature
2001, 414, 105–111.
[17] M. Al-Hajj, M. S. Wicha, A. Benito-Hernandez, S. J. Mor-
rison, M. F. Clarke, Proc. Natl. Acad. Sci. USA 2003, 100,
3983–3988.
[18] P. A. Beachy, S. S. Karhadkar, D. M. Berman, Nature 2004,
432, 324–331.
[19] D. Müller, Dissertation, Rheinische Friedrich-Wilhelm-Uni-
versität Bonn, Germany, 2006.
[20] W. T. Dengler, J. Schulte, D. P. Berger, R. Mertelsmann, H. H.
Fiebig, Anticancer Drugs 1995, 6, 522–532.
[21] H. H. Fiebig, A. Maier, A. M. Burger, Eur. J. Cancer 2004, 40,
802–820.
[22] B. Räz, M. Iten, I. Grether-Bühler, R. Kaminsky, R. Brun,
Acta Trop. 1997, 68, 139–147.
Evaluation of antileishmanial activity was carried out in mouse
peritoneal macrophages. The ratio of infection with Leishmania
donovani (strain MHOM-ET-67/L82) was determined microscopi-
cally after exposure to test compounds, incubation and staining
with Giesma. IC50 values were calculated by linear regression. Mil-
tefosine was used as positive control.
[23] F. S. Bruckner, C. L. M. J. Verlinde, A. C. La Flamme, W. C.
Van Voorhis, Antimicrob. Agents Chemother. 1996, 40, 2592–
2597.
Cytotoxicity was evaluated in rat skeletal myoblasts (L6-cells) by
using podophyllotoxin as positive control.
Received: November 2, 2007
Published Online: February 15, 2008
Eur. J. Org. Chem. 2008, 1732–1739
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1739