3-Phenyl-5-methyl-2H,5H-furan-2-ones: Tuning antifungal activity by varying substituents on the phenyl ring

Article abstract of DOI:10.1016/S0960-894X(00)00376-0

A series of racemic 3-phenyl-5-methyl-2H,5H-furan-2-ones related to a natural product, (-)incrustoporine, was synthesized, and their antifungal activity evaluated. The key structural feature, furanone ring, was closed via H₂SO₄-mediated cyclization of 2-phenylpent-4-enoic acids. The compounds displayed antifungal activity, especially against filamentous fungi. Expressed as the minimum inhibition concentration (MIC) in μmol/L, the activity of the most promising derivative against Absidia corymbifera matched that of ketoconazole (31.25 μmol/L). In terms of μg/mL, the substance was more active (7.6 μg/mL) than this standard antifungal drug (16.6 μg/mL). (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00376-0

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1893±1895
3-Phenyl-5-methyl-2H,5H-furan-2-ones: Tuning Antifungal
Activity by Varying Substituents on the Phenyl Ring
Milan Pour,a,* Marcel Spulak,a Vojtech Balsanek,a
Jirõ Kunes,a Vladimõr Buchtab and Karel Waissera
aLaboratory of Structure and Interactions of Biologically Active Molecules, Department of Inorganic and Organic Chemistry,
Â
Â
Â
Faculty of Pharmacy, Charles University, Heyrovskeho 1203, CZ-500 05 Hradec Kralove, Czech Republic
Â
Department of Biological and Medical Sciences, Faculty of Pharmacy, Charles University, Heyrovskeho 1203,
b
 Â
CZ-500 05 Hradec Kralove, Czech Republic
Received 16 May 2000; revised 26 June 2000; accepted 27 June 2000
AbstractÐA series of racemic 3-phenyl-5-methyl-2H,5H-furan-2-ones related to a natural product, ( )incrustoporine, was synthe-
sized, and their antifungal activity evaluated. The key structural feature, furanone ring, was closed via H2SO4-mediated cyclization
of 2-phenylpent-4-enoic acids. The compounds displayed antifungal activity, especially against ®lamentous fungi. Expressed as the
minimum inhibition concentration (MIC) in mmol/L, the activity of the most promising derivative against Absidia corymbifera
matched that of ketoconazole (31.25 mmol/L). In terms of mg/mL, the substance was more active (7.6 mg/mL) than this standard
antifungal drug (16.6 mg/mL). # 2000 Elsevier Science Ltd. All rights reserved.
As there are just a few structurally di€erent groups of
antifungal agents, which is especially true about the
systemic ones,1 identi®cation of new lead structures and
further development of novel antifungal drugs is an
important goal of current pharmaceutical research. In
particular, it appears highly desirable to continue the
process of drug discovery for the treatment of serious
mycoses caused by opportunistic fungal pathogens, such
as the members of Mucorales, Aspergillus, Fusarium,
Trichosporon and non-albicans Candida species, which
are characterized with decreased susceptibility to current
antifungal drugs.2
lead structures, we were interested in preparing simple
analogues of ( )incrustoporine. We speculated that,
similar to griseofulvin,6 compounds related to ( )incrusto-
porine could also display activity against pathogenic
fungi targeting humans as their hosts. In this letter, we
wish to disclose the design, preparation and biological
evaluation of a series of 3-phenyl-5-methyl-2H,5H-
furan-2-ones, the goal of which was a preliminary
assessment of the suitability of this class of compounds
for further development as potential antimycotics.
At the initial stage of our research, we aimed to prepare
compounds with various substituents with di€erent elec-
tronic in¯uence on the phenyl ring, and a simple alkyl
moiety at C(5) of the furanone ring. In the natural product
lead, the ethyl group at C(5) is attached to a centre of
chirality. To introduce this stereochemical element, Yajima
and Mori4 used a rather costly 1,(2S)-epoxybutane in a
moderate-yielding epoxide ring opening step (53%) of
their sequence. As the possibility of biological evaluation
against human pathogenic fungi providing negative results
(and thus adding little value to the synthetic endeavours)
could not be ruled out, we designed a series of simple 5-
methyl derivatives 5. We envisaged an easy, high-yielding
route to these compounds from their saturated precursors
4, which in turn could be easily derived from the cycli-
zation of 2-phenylpent-4-enoic acids 3 with di€erent
substituents attached to the phenyl ring (Fig. 1).
In 1995, Zapf et al. reported3 the isolation of an inter-
esting fungal metabolite from the family of butenolides,
( )incrustoporine 6. The natural product was found to
possess activity against a wide array of phytopathogenic
fungi as well as some cytotoxic activity.3 Following the
isolation, two total syntheses were published, by Yajima
and Mori4 in 1996, and later by Rossi5 in 1999. The
structure of the compound indicates that it is a rela-
tively lipophilic, small, rigid molecule of low structural
complexity, which renders it a possible target for further
development as a potential drug. As part of a medicinal
chemistry program focused on the search for new anti-
mycotic agents based on optimizing natural product
*Corresponding author. Fax:+42-49-5210002; e-mail: pour@faf.cuni.cz
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00376-0

1894
M. Pour et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1893±1895
All unsaturated lactones 5a±j7 9 prepared via the above
procedures are summarized in Table 1.
Figure 1.
The preparation of the unsubstituted 3-phenyl-5-
methyl-2H,5H-furan-2-one 5a is outlined as a repre-
sentative example in Scheme 1. Thus, phenylacetic acid
was ®rst converted into the corresponding methyl ester.
Following the deprotonation with LDA, the ester eno-
late was quenched with allyl bromide to a€ord the
methyl ester of 2-phenylpent-4-enoic acid. The car-
boxylic group was then liberated by hydrolysis, and
subjected to a proton-mediated cyclization onto the
terminal double bond to a€ord both diastereomers of 3-
phenyl-5-methyltetrahydrofuran-2-one. Following some
experimentation, we found that the reaction proceeded
best in concentrated sulphuric acid at 0 ꢀC, giving
almost quantitative yield of a mixture of trans and cis
isomers in the ratio of 2:1, as determined by NMR.
Finally, the double bond was introduced by enolization
of the mixture of saturated lactones followed by treat-
ment with phenylselenenyl chloride, oxidation and
spontaneous selenoxide elimination to yield the target
compound. Throughout the sequence, yields in all steps
did not fall below 80% for each target molecule.
Scheme 2.
The target compounds and their saturated precursors
were evaluated for their in vitro antifungal activity
against a set of eight clinical isolates of human patho-
genic fungi (Candida albicans ATCC 44859, Candida
krusei E28, Candida tropicalis 156, Candida glabrata 20/
I, Trichosporon beigelii 1188, Aspergillus fumigatus 231,
Absidia corymbifera 272, Trichophyton mentagrophytes
445) using the microdilution format of the NCCLS
M27-A guidelines.10 In order to have a standard for
comparison, we also synthesized natural incrustoporine
in the racemic form by the route of Yajima and Mori,4
and ketoconazole was used as a standard antifungal
drug. The saturated lactones 4a±j displayed no anti-
fungal activity at all (>1000 mmol/L), but the target
compounds 5a±j and the racemic natural product 6 were
found to be antifungally active in general. While the
activity of racemic incrustoporine itself was moderate to
low, and the compound was only marginally more e-
cient than its 5-methyl counterpart 5b, the halogenated
analogues displayed appreciable fungistatic e€ect, in
particular against ®lamentous fungi (Table 1).
The activity of the most potent 3-(3,4-dichlorophenyl)-
5-methyl-2H,5H-furan-2-one 5e against the strains of
Aspergillus fumigatus (AF) and Absidia corymbifera
(AC) read after 24 h is higher than that of ketoconazole.
After 48 h, MIC of 5e against AF increases as compared
to ketoconazole, but the compound remains as ecient
as this antifungal drug against AC. Expressed in mg/mL,
even after the period of 48 h, furanone 5e is more active
against AC (7.6 mg/mL) than ketoconazole (16.6 mg/mL).
The activity of the substances against yeast strains is not
so signi®cant, and generally increases with time.
Scheme 1.
In conclusion, the results show that the structure of
( )incrustoporine represented by the unsaturated ®ve-
membered lactone ring may serve as a potential lead to
the preparation of analogues possessing antifungal
activity against human pathogens. Antifungal activity
appears to be linked to the presence of the double bond
conjugated with the carbonyl group of the lactone
function, and can be signi®cantly increased by sub-
stituting the phenyl ring with halogens. Encouraged by
these results, we are continuing to explore this class of
antifungal compounds and our progress will be reported
in due course.
A minor modi®cation had to be used in the preparation
of the p-methoxy derivative 5c, as its saturated pre-
cursor 4c could not be prepared by the sulphuric acid
induced cyclization of 2-(p-methoxyphenyl)pent-4-enoic
acid 3c due to the competing sulfonation of the phenyl
ring.
Acid 3c was therefore subjected to reaction with hydrogen
bromide, and the intermediate bromide was immediately
cyclized with sodium carbonate in methanol (Scheme 2).

M. Pour et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1893±1895
Table 1. In vitro antifungal activities of compounds 5a±j, (Æ)6 and ketoconazole (k)
1895
Strain
Time
(h)
Compound No. Ð MIC (mmol/L)a
5a
H
5b
p-Me
5c
p-MeO
5d
p-Cl
5e
m,p-Cl
5f
p-F
5g
m-F
5h
p-Br
5i
m-Br
5j
p-NO2
6
k
TMb
CA
CT
72
120
500
1000
250
500
125
250
31.25
62.5
31.25
62.5
250
250
125
125
31.25
62.5
62.5
125
62.5
125
125
250
0.98
1.95
24
48
1000
>1000
1000
>1000
1000
1000
31.25
125
31.25
500
62.5
250
62.5
250
7.81
62.5
31.25
125
250
500
250
500
0.12
0.12
24
48
>1000
>1000
>1000
>1000
1000
>1000
250
500
>500
>500
250
500
500
1000
250
500
250
500
500
500
500
>500
15.63
15.63
CK
CG
TB
24
48
>1000
>1000
>1000
>1000
>1000
>1000
250
500
>500
>500
500
500
500
1000
250
250
250
500
500
500
500
>500
3.91
3.91
24
48
>1000
>1000
>1000
>1000
>1000
>1000
500
1000
>500
>500
500
500
>1000
>1000
500
1000
250
500
500
1000
500
>500
0.24
0.49
24
48
>1000
>1000
>1000
>1000
1000
>1000
125
250
31.25
>500
125
250
250
500
125
250
62.5
250
500
1000
250
500
0.12
0.12
AF
AC
24
48
1000
>1000
1000
>1000
1000
1000
62.5
125
7.81
62.5
125
125
250
500
62.5
250
62.5
250
250
1000
250
500
15.63
15.63
24
48
1000
>1000
1000
>1000
1000
>1000
125
250
7.81
31.25
62.5
62.5
500
500
125
125
62.5
125
250
250
250
250
31.25
31.25
aMinimum inhibitory concentration.
bTM=Trichophyton mentagrophytes; CA=Candida albicans; CT=Candida tropicalis; CK=Candida krusei; CG=Candida glabrata; TB=Trichos-
poron beigelii; AF=Aspergillus fumigatus; AC=Absidia corymbifera.
7. Compound 5a was described by DeShong and co-workers:
Acknowledgements
DeShong, P.; Sidler, D. R.; Rybczynski, P. J.; Slough, G. A.;
Rheingold, A. L. J. Am. Chem. Soc. 1988, 110, 2575.
8. Compounds 5b, 5c, 5d and 5j were reported to have been
obtained by the decomposition of the corresponding manga-
nacycles: DeShong, P.; Sidler, D. R.; Rybczynski, P. J.; Ogilvie,
A. A. J. Org. Chem. 1989, 54, 5432. However, neither the
This work was ®nancially supported by Charles Uni-
versity (project No. 171/1997/B) and by the Ministry of
Education of the Czech Republic (project No. VS 97124
and LB 98233). The following undergraduate students
participated in this project: Petr Koudelka, Veronika
Mysakova, Radan Schiller, Hana Sebestianova and
Filip St'astny. The mass spectra were recorded by Dr.
V. Vorõsek, Department of Clinical Biochemistry,
Charles University Medical Faculty Hospital, Hradec
Kralove, Czech Republic.
yields, nor the spectral data and physical constants are pre-
sented in this paper.
9. All compounds were fully characterized by spectral meth-
ods, and their purity checked by elemental analysis. Repre-
sentative data of the most potent derivative: compound 5e: mp
108±111 ꢀC; IR (CHCl3) nmax 1323, 1472, 1758, 2988,
1
3026 cm
;
1H NMR (300 MHz, CDCl3) d 7.99 (1H, d,
J=1.9 Hz, Ar), 7.73 (1H, dd, J1=8.4 Hz, J2=1.9 Hz, Ar), 7.60
(1H, d, J=1.9 Hz, H4), 7.48 (1H, d, J=8.4 Hz, Ar), 5.17 (1H,
qd, J1=6.9Hz, J2=1.9 Hz, H5), 1.53 (3H, d, J=6.9 Hz, CH3);
13C NMR (75 MHz, CDCl3) d 170.71, 150.15, 133.18, 132.65,
130.39, 129.17, 128.95, 128.61, 126.05, 76.87, 18.98; LRMS:
References and Notes
ꢁ
+
1. Kaufman, C. A.; Carver, P. L. Drugs 1997, 53, 539.
2. Graybill, J. R. Clin. Infect. Dis. 1996, 22 (Suppl. 2), S166.
3. Zapf, S.; Anke, T.; Sterner, O. Acta Chem. Scand. 1995, 49,
233.
242 (M
H, 100), 225 (4), 214 (3), 199 (33), 171 (94), 150 (7),
136 (31), 126 (7), 115 (9), 99 (13), 86 (6), 74 (13), 63 (9), 50
(14). Anal. calcd for C11H8Cl2O2: C, 54.35; H, 3.32. Found: C,
54.56; H, 3.25.
4. Yajima A., Mori K. Liebigs Ann. Chem. 1996, 1091.
5. Rossi, R.; Bellina, F.; Biagetti, M.; Mannina, L. Tetra-
hedron: Asymmetry 1999, 10, 1163.
10. National Committee For Clinical Laboratory Standards.
(1997). Reference method for broth dilution antifungal sus-
ceptibility testing of yeasts. Approved standard M27-A.
NCCLS 1997, Villanova, PA.
6. De Carli, L.; Larizza, L. Mutation Research 1988, 195, 91.