L. Tichotová et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6062–6066
6065
Table 4
Cytostatic activities
antifungal action of the compounds, albeit not so dramatic. Only
the 4-methoxyphenyl compound 2g was comparable to 2f for most
species, with the exception of C. albicans ATCC44859, Candida
parapsilosis and T. asahii, for which somewhat higher MICs were
recorded. 4-methylphenyl substance 2h displayed the same
inhibitory effect as the 4-methoxyphenyl derivative 2g against
C. albicans ATCC44859, and there are no significant differences for
most other strains, with the exception of C. krusei ATCC 6258, Asper-
gillus fumigatus and Trichophyton mentagrophytes. On the whole,
however, the profile of activity of 2h is worse than those of 2f
and 2g, and the same applies to haloaryl compounds 2i–l, the
effects of which are even lower. A partial exception was the
4-dimethylaminophenyl substance 2m with better MIC values
against both C. krusei strains, C. parapsilosis and T. mentagrophytes.
As regards the derivatives with a different aryl than phenyl
(2o–r), their MICs were also higher than those of unsubstituted
styryl and phenyl derivatives 2a and 2f.
The in vitro activities of the most effective derivative 2f were
comparable to coruscanone A against C. albicans, C. tropicalis and
T. asahii, and better than those of the natural product against C.
krusei, C. glabrata and A. fumigatus. Hence, the deoxygenation of
the methoxy function combined with the removal of the styryl
double bond in the natural product has very little influence on
the biological activity, and the simple benzylidene analogue has
a broader spectrum of activity against Candida strains. Compound
2f (and, in some cases, coruscanone A as well) is in vitro clearly
superior to fluconazole against non-albicans Candida sp., namely
C. parapsilosis, C. krusei, and C. glabrata, and comparable to the drug
against C. albicans, C. tropicalis and C. lusitaniae. In line with the
conclusions by Clark et al, further substitution of the phenyl ring
generally led to a decrease in the activity. However, unlike the pre-
viously described7 methoxy derivatives, the drop in the activity for
the substituted styryl analogues described herein was much more
pronounced, as demonstrated by the loss of activity of 2b and 2d.
On the other hand, the activities of the substituted benzylidene
derivatives 2g–m were not far from that of unsubstituted 2f. The
high in vitro antifungal effect of 2f is a possible indication that
the compounds of the simple benzylidene series described in this
paper may bind in a different way to the target protein, since
methoxy compound 37 (Fig. 4) lacking the styryl double bond
was inactive. However, Michael addition of a nucleophile also
seems to be the major driving force behind the interaction with a
cellular target.
Compd L1210a
HL 60b
HeLa S3c
CCRF-CEMd HT-29e Colo 201f
2a
2b
2c
2d
2f
2g
2h
2i
NAg
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NAg
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
7.3 0.2
NA
NA
NA
NA
NA
NA
4.9 0.8
NA
NA
5.4 0.7
NA
13.5 1.1 3.7 0.1
3.6 0.2
4.4 0.2
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NT
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
2.6 0.2
3.8 0.2
2.7 0.2
6.1 0.4
1.6 0.1
1.8 0.3
7.5 0.5
4.3 0.1
2.5 0.3
4.2 0.2
2j
2k
2l
2m
2o
2q
2r
1.3 0.1
NA
NA
2.4 0.2
NT
CorAh
3.8 0.2 4.5 0.3 NTi
a
b
c
d
e
f
L1210 (ATCC CCL 219)—mouse lymphocytic leukemia cells.
HL-60 (ATCC CCL 240)—human promyelocytic leukemia cells.
HeLa S3 (ATCC CCL 2.2)—human epithelial carcinoma cells.
CCRF-CEM (ATCC CCL 219)—human leukemic T-cell lymphoblasts.
HT-29 (ATCC HTB 38)—human colon adenocarcinoma cells.
Colo 201 (ECACC 87091201)—human colon adenocarcinoma cells.
NA = not active at 10
Coruscanone A.
NT = not tested.
g
h
i
lmol/L.
compounds on the proliferation of HEK293 cells (human embryo-
nal kidney cells). With the exception of coruscanone A and com-
pounds 2a, 2c and 2k, for which a cytostatic-like action was
recorded (a mild decrease in proliferation, followed by a slow rise
and plateauing), no effect was observed. This, taken together with
the results of screening for cytostatic activity shown in Table 4
could be an indication that, compared to the previously prepared
derivatives,7 the cytotoxicity of the desmethoxy compounds de-
scribed herein may be reduced.
In summary, we have described reproducible conditions for the
Knoevenagel condensation of cyclopent-4-ene-1,3-dione with a
variety of aromatic and unsaturated aldehydes, and prepared a ser-
ies of desmethoxy analogues of the antifungal natural product, cor-
uscanone A. Evaluation of antifungal activity showed that some of
the compounds retained antifungal effect on the level of the natu-
ral product. Similar to previously described derivatives of corusca-
none A, further substitution of the phenyl or styryl moiety as well
as switch from phenyl to a naphthyl or heterocyclic ring led to a
decrease or complete loss of the biological effect. The most active
compound 2f was comparable to fluconazole as the drug standard
against C. albicans, C. tropicalis and C. lusitaniae, and was superior to
the drug against C. parapsilosis, C. krusei, and C. glabrata. Even
though Michael addition could be the reason for the binding of
the compound to its cellular target, it is notable that further
screening experiments indicated reduced cytotoxicity of the com-
pound in comparison to coruscanone A.
All compounds were also subjected to screening on a panel of
cancer cell lines, including three types of leukemic cells and three
solid tumors19; cell viability was quantified using XTT standard
spectrophotometric assay20 (Table 4). A majority of the new com-
pounds possessed moderate IC50 values (micromolar range)
against the CCRF-CEM line, and four of them also against HeLa
cells. The most active derivative 2f inhibited the growth of just
the CCRF-CEM cells, while moderate IC50 values against all leuke-
mic lines used in this study were recorded for coruscanone A.
In addition, all compounds were subjected to a preliminary
kinetic cell-based morphological screening.21 In this experiment,
impedance readout was used for monitoring the effect of the
Acknowledgments
This work was supported by the Ministry of Education, Youth
and Sports of the Czech Republic (Project No. 1M0508) and by
the Czech Science Foundation (Project No. P207/10/2048). L.T.
and E.M. also acknowledge partial financial support from Charles
University Grant Agency (Projects Nos. 98908 and SVV-263-001).
O
O
O
O
Me
O Me
3
2f
Supplementary data
not active
active
Supplementary data associated with this article can be found, in
Figure 4. Comparison of 3 and 2f.