1806
L. T. Maillard et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1803–1807
Table 2
hydrophobicity balance was already proved to be important in
such series to obtain anti-Candida activity.8 Finally, some activity
was recovered for the substituted imidazo[2,1-b]thiazolyl deriva-
tive (compound 1k). This compound showed a relative high clogD
(5.19), which suggested the partial influence of this parameter to
obtain activity in this series.
In the second series (series B, compounds 2a–h, Scheme 2, Ta-
ble 2), we replaced the phenol moiety at C-4 of the thiazole ring
in 1a by different aminoalkyl substituents. When the amino group
was protected with a Z group, no significant anti-Candida activity
was detected (compounds 2a–d). Among the unprotected deriva-
tives (compounds 2e–f and 2h), compound 2e showed an interest-
ing activity against several Candida spp., with a selectivity oriented
towards non-Candida albicans species. In particular, a high activity
was detected against the C. krusei 6258 strain, which presents a
multi-resistant profile against several reference drugs. To deter-
mine the selectivity of 2-hydrazino-1,3-thiazoles towards fungi,
the most active compounds 1d, 1e, 1k and 2e were finally evalu-
ated for their cytotoxic activity against mouse fibroblast cells.
Cytotoxicity of selected compounds to mouse fibroblast (NIH/3T3) cell line
Concentration (
l
g/mL)a
0.5
1
31.3
62.5
1d
1e
1k
2e
83.7 1.3
91.9 10.7
94.6 6.3
93.6 4.0
69.2 0.2
80.6 8.2
90.0 0.1
92.6 1.8
45.3 2.4
70.6 3.2
64.1 2.1
36.2 4.1
52.7 3.9
50.2 4.6
59.5 2.9
32.4 1.2
a
Values represent mean standard deviation of triplicate determinations.
Compounds 1a–k, 2a–f and 2h were evaluated for their anti-
fungal activity against 8 Candida spp. strains (C. albicans, C. glabra-
ta, C. krusei and Candida parapsilosis) and compared with five
reference drugs, that were, amphotericin B, itraconazole, fluconaz-
ole, voriconazole and caspofungin (Table 1). Several resistant
strains to one or several reference drugs (see Table 1) were chosen
to evaluate the potential activity of synthesized compounds
against multi-resistant Candida spp. Prior testing, each Candida
spp. isolate was subcultured on a qualified medium to ensure pur-
ity and optimal growth. The susceptibility assays were determined
by the microbroth dilution method performed in sterile flat-bot-
tom 96-well microplates (Difco Laboratories, Detroit, USA) as de-
scribed in CLSI guidelines, M27-A3 document.19 Compounds 1d,
1e, 1k and 2e that achieved the strongest anti-Candida activity
were also analyzed to estimate their cytotoxic effects (Table 2).
Cytotoxicity was evaluated using the MTT cell proliferation/viabil-
ity assay on mouse 3T3/NIH fibroblast cells, which were originally
obtained from the American Type Culture Collection (ATCC, USA).
The level of cellular MTT reduction was quantified as previously
described in literature with small modifications.20–22
The cytotoxic profile (Table 2) showed a weak toxicity at 0.5 lg/
mL for all tested compounds, with more than 83% of cell viability.
In conclusion, we have synthesized a series of hydrazino-1,3-
thiazoles based on compound 1a. Their in vitro anti-Candida activ-
ity was evaluated against 8 strains of multi-resistant Candida spp.
Preliminary structure activity relationships have shown that sub-
stitution of the hydrazine by a hydrogen bond donating heterocy-
cle enhanced the anti-Candida activity. In particular, introduction
of an indolyl substituent instead of the indanyl led to interesting
compounds, especially active against C. albicans strains. Moreover,
replacement of the phenol moiety in compound 1a by a (S)-2-ami-
noethyl chain (compound 2e) largely enhanced the antifungal
activity, with a selectivity oriented towards non Candida albicans
species. Based on these results, derivatives 1e and 2e represent
good starting points for the development of novel anti-Candida
spp. agents. Synthesis of next generation compounds combining
modifications on both sides of the scaffold will be reported in
due course.
In an initial screening program to determine potential anti-
Candida activities of 2-hydrazino-1,3-thiazoles, compound 1a was
chosen as the first hit for further derivatization (MIC of 16
lg/mL
against fluconazole/voriconazole resistant C. krusei strain).
a
Replacement of the indanyl nucleus by various aliphatic or aro-
matic carbocycles or heterocycles was initially studied (series A
compounds). Surprisingly when the indanyl nucleus was replaced
by the structurally related 1,3-benzodioxol-5-yl moiety (com-
pound 1b, Scheme 1, Table 1), the anti-Candida activity was lost.
Introduction of a cyclohexyl or a naphtyl ring was previously used
to design 2-hydrazino-1,3-thiazoles with good anti-Candida activi-
ties.7 In our case, insertion of a cyclohexyl (compounds 1c) or a
naphthyl ring (compound 1d) led to only a weak activity enhance-
ment. These disappointing results led us to switch to other modu-
lations consisting in the introduction of various aza-heterocycles.
The best activities were obtained when the indanyl substituent
was replaced by an indole ring: compound 1e presented a good
anti-Candida activity, with a selectivity oriented towards C. albi-
cans. In particular, this compound showed similar or better activity
than reference drugs for several Candida strains. We assumed that
the biological activity was enhanced by the presence of a NH
hydrogen bond donor in the carbocyclic part. This hypothesis
was confirmed when we introduced an imidazo-[1,2-a]pyridinyl
substituent (compound 1f), which is usually viewed as an aza-
indolyl isostere.23,24 Imidazo[1,2-a]pyridine does not have any
hydrogen bond donating group making 1f inactive toward all Can-
dida strains. That was also the case when indole was replaced by
benzofurane (compound 1g). Similar results were obtained with
series containing a five-membered heteroaromatic substituent:
the pyrrole derivative 1h was more active than its furyl counter-
part 1i. Surprisingly, although imidazole had a hydrogen bond
donor, compound 1j was totally inactive whatever the strain con-
sidered. This might be due to a higher hydrophilicity of the mole-
cule (clogD = 1.53) compared to compound 1h (clogD = 2.03). The
Acknowledgments
The authors are grateful to G. Tambutet (Université Montpellier
2, France), C. Lamoureux and M. Aussedat (Faculté de Pharmacie de
Montpellier, France) for technical assistance.
Supplementary data
Supplementary data associated with this article can be found,
References and notes
1. Rodloff, C.; Koch, D.; Schaumann, R. Eur. J. Med. Res. 2011, 16, 187.
2. Pfaller, M. A. Am. J. Med. 2012, 125, S3.
3. Mathew, B. P.; Nath, M. ChemMedChem 2009, 4, 310.
4. Munro, C. A. Expert Rev. Anti-Infect. 2012, 10, 115.
5. Perlin, D. S. Drug Resist. Updat. 2007, 10, 121.
6. Chimenti, F.; Bizzarri, B.; Bolasco, A.; Secci, D.; Chimenti, P.; Granese, A.;
Carradori, S.; D’Ascenzio, M.; Lilli, D.; Rivanera, D. Eur. J. Med. Chem. 2011, 46,
378.
7. Chimenti, F.; Bizzarri, B.; Maccioni, E.; Secci, D.; Bolasco, A.; Fioravanti, A.;
Chimenti, P.; Granese, A.; Carradori, S.; Rivanera, D.; Lilli, D.; Zicari, A.; Distinto,
S. Bioorg. Med. Chem. Lett. 2007, 17, 4635.
8. Secci, D.; Bizzarri, B.; Bolasco, A.; Carradori, S.; D’Ascenzio, M.; Rivanera, D.;
Mari, E.; Polletta, L.; Zicari, A. Eur. J. Med. Chem. 2012, 53, 246.
9. De Logu, A.; Saddi, M.; Cardia, M. C.; Borgna, R.; Sanna, C.; Saddi, B.; Maccioni, E.
J. Antimicrob. Chemother. 2005, 55, 692.
10. Turan-Zitouni, G.; Ozdemir, A.; Kaplancikli, Z. A.; Benkli, K.; Chevallet, P.;
Akalin, G. Eur. J. Med. Chem. 2008, 43, 981.