Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, May 2006, p. 1710-1714, Vol. 50, No. 5
0066-4804/06/$08.00+0 doi:10.1128/AAC.50.5.1710-1714.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650204, People's Republic of China,1 National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences, School of Pharmacy, The University of Mississippi, University, Mississippi 386772
Received 27 October 2005/ Returned for modification 1 December 2005/ Accepted 7 February 2006
|
|
|---|
|
|
|---|
-lanosterol demethylase in ergosterol biosynthesis pathway), allylamines (inhibiting squalene epoxidase in ergosterol synthesis pathway), echinocandins (inhibiting synthesis of ß-1,3-glucan, which is a required component of the cell wall of many fungi), and flucytosine (incorporated into RNA and thus inhibiting DNA synthesis) (27). Representatives from the five classes are amphotericin B (AMB), fluconazole (FLU), terbinafine, caspofungin, and flucytosine, respectively. However, each drug class suffers one or more major shortcomings (for example, significant dose-limiting toxicities for AMB, rapid development of resistance for the azole drugs, limitation of the use to dermatophytoses for terbinafine, and lack of effectiveness in cryptococcosis for caspofungin) (20, 23, 24). This has highlighted the need to discover new antifungal agents, preferably with new modes of action. C-27 steroidal saponins are an important class of natural products which are probably best known as starting materials for the synthesis of steroidal hormones. A C-27 steroidal saponin is composed of a C-27 aglycone moiety (or referred to as a C-27 steroidal sapogenin) and a sugar chain of one or more monosaccharides. These compounds are further classified as two submajor classes: spirostanol saponins with a hexacyclic ABCDEF-ring system, as shown for the compounds in Fig. 1, and furostanol saponins with an open F-ring skeleton, which are considered biosynthetic precursors of the former. According to conservative statistics, there are at least 90 plant families containing such compounds, particularly in monocotyledons (Liliaceae, Convallariaceae, Dioscoreaceae, Agavaceae, Smilacaceae, and Amaryllidaceae, etc.) (31). The antifungal activity of steroidal saponins, particularly against agricultural pathogens, has been known for a long time (5, 7, 34, 35), while other reported activities for this class of compounds include antitumor, hypoglycemic, immunoregulatory, and cardiovascular disease prevention and treatment (31). Although saponins are generally considered to possess detergent-like surfactant properties, the classic viewpoint that saponins make poor drug candidates has been challenged in recent years. Several steroidal saponin-based drugs have been seen on the market. For example, "Di-ao-xin-xue-kang," which has an ingredient composition of several steroidal saponins from Dioscorea panthaica administered orally, has dominated the market of drugs for the treatment and prevention of cardio- and cerebrovascular diseases in China for a decade (6, 18, 38). Another steroidal saponin-based drug, "Chuan-shan-long injection," made from Dioscorea nipponica and used for rheumatism may be administered intraperitoneally (19). In addition, numerous patents on the therapeutic potentials of steroidal saponins, such as for hepatitis (28) and fungal infections (3, 4, 22, 29) have been released. The above information indicates that some steroidal saponins do have low toxicity and may become therapeutic agents. This prompted us to explore the potential of this class of compounds as antifungal leads for the development of systemic antifungal drugs.
![]() View larger version (19K): [in a new window] |
FIG. 1. Structures of compounds 1 to 28.
|
|
|
|---|
-spirost-3ß-hydroxy-3-O-
-L-rhamnopyranosyl-(1-3)-ß-D-glucopyranosyl-(1-2)-[ß-D-xylopyranosyl-(1-3)]- ß-D-glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 1) (10), agamenoside G (compound 2) (10), degalactotigonin (compound 3) (15), (25R)-5
-spirost-3ß-hydroxy-3-O-ß-D-xylopyranosyl-(1-3)-ß-D-glucopyranosyl-(1-2)-[ß-D-xylopyrano-syl- (1-3)]-ß-D-glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 4) (15), agamenoside C (compound 5) (9), cantalasaponin 1 (compound 6) (9), agamenoside A (compound 7) (12), agavesides A (compound 8) (11) and B (compound 9) (11), (25R)-5
-spirost-3ß-hydroxy-12-oxo-3-O-ß-D-glucopyranosyl-(1-2)-ß-D-glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 10) (11), (25R)-5
-spirost- 3ß-hydroxy-12-oxo-3-O-ß-D-glucopyranosyl-(1-2)-[ß-D-xylopyranosyl-(1-3)]-ß-D- glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 11) (11), agamenosides D (compound 12) (11) and E (compound 13) (11), (25R)-5
-spirost-3ß-hydroxy-12-oxo-3-O-ß-D-xylopyranosyl-(1-3)-ß-D-glucopyranosl-(1-2)-[ß-D-xylopyranosyl-(1-3)]-ß-D-glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 14) (15), agamenoside F (compound 15) (11, 15), (25R)-5
-spirost-3ß,6
-diol-6-O-ß-D-glucopyranoside (compound 16) (14), (25R)-5
-spirost-3ß,6
-diol-3-O-ß-D-xylopyranosyl-(1-3)-ß-D-glucopyranosyl-(1-2)-[ß-D-xylopyranosyl-(1-3)]-ß-D-glucopyranosyl-(1-4)-ß-D-galactopyranoside (compound 17) (15), prosapogenin A of dioscin (compound 18) (16), deltonin (compound 19), dioscin (compound 20) (13), collettiside I (compound 21), polygonatoside A (compound 22) (16), tigogenin (compound 23) (10), agavegenin A (compound 24) (8), hongguanggenin (compound 25) (9), hecogenin (compound 26) (11), chlorogenin (compound 27) (14), and 9(11)-dehydroxyhecogenin (compound 28) (14). AMB and doxorubicin, used as control drugs for antifungal and cytotoxicity testing, respectively, were purchased from ICN Biomedicals, Ohio. FLU, used as a control drug for antifungal testing, was obtained from Pfizer, Morris Plains, N.J.
Assay for antifungal activity.
Reference strains were obtained from the American Type Culture Collection (ATCC) (Manassas, VA) and included Candida albicans ATCC 90028, Candida glabrata ATCC 90030, Candida krusei ATCC 6258, Cryptococcus neoformans ATCC 90113, and Aspergillus fumigatus ATCC 90906. Susceptibility testing was performed using a modified version of the CLSI (formerly NCCLS) methods (25, 26). Samples (dissolved in dimethyl sulfoxide) were serially diluted using 20% dimethyl sulfoxide in 0.9% saline and transferred in duplicate to 96-well flat-bottom microplates. Candida spp. and C. neoformans inocula were prepared by picking 1 to 3 colonies from agar plates and resuspending in
5 ml 0.9% sterile saline. The optical density at 630 nm of the saline suspensions was compared to the 0.5 McFarland standard. The microorganisms were diluted in broth (RPMI 1640-2% dextrose-0.165 M morpholinepropanesulfonic acid [MOPS] at pH 4.5 [Cellgro] for Candida spp. and Sabouraud dextrose [Difco] for C. neoformans) to afford final target inocula of 5.0 x 103 for Candida spp. and 5.0 x 105 for C. neoformans. The A. fumigatus inoculum was prepared by gently removing spores from an agar slant, transferring to
5 ml 0.9% saline, and filtering through Miracloth (Calbiochem, La Jolla, CA). The filtrate was diluted appropriately (via comparison to a standard curve) in 5% Alamar blue (BioSource International, Camarillo, CA)-RPMI 1640 broth (2% dextrose buffered with 0.165 M MOPS at pH 7.3) to afford a final target inoculum of 4.0 x104 CFU/ml. The fungal inocula were added to the samples to achieve a final volume of 200 µl and final sample concentrations starting with 20 µg/ml. Growth (saline only), solvent, and blank (media only) controls were included on each test plate. Fungicidal AMB and fungistatic FLU were included as positive controls. All organisms were read at 630 nm using the EL-340 Biokinetics reader (Bio-Tek Instruments, Vermont) prior to and after incubation (Candida spp. at 37°C for 18 to 24 h; C. neoformans and A. fumigatus at 30°C for 72 h), except A. fumigatus, which was read at 544-nm excitation/590-nm emission using the Polarstar Galaxy plate reader (BMG LabTechnologies, Germany). Percent growth was calculated and plotted versus test concentration to afford the IC50 (sample concentration that affords 50% growth of the organism). The MIC was defined as the lowest test concentration that allowed no detectable growth (or no more than 20% growth for FLU; for A. fumigatus, no color change from blue to pink). Minimum fungicidal concentrations (MFCs) were determined by removing 5 µl from each clear (or blue) well, transferring to agar, and incubating as previously mentioned. The MFC was defined as the lowest test concentration that allows no growth of the organism on agar.
Assay for in vitro cytotoxicity. A panel of mammalian cells obtained from ATCC (Manassas, VA) included five human cancer cell lines (SK-MEL [melanoma], KB [epidermal carcinoma, oral], BT-549 [ductal carcinoma, breast], SK-OV-3 [ovary carcinoma], and HepG2 [human hepatic carcinoma]) and one noncancerous Vero cell line (African green monkey kidney fibroblast). The cells were cultured in 75-cm2 culture flasks in RPMI 1640 medium (Gibco, Invitrogen Corp., CA) supplemented with bovine calf serum (10%) and amikacin (60 mg/liter), at 37°C in an atmosphere of 95% humidity, 5% CO2. The assay was performed in 96-well microplates. Cells were seeded to the wells of the plate at a density of 25,000 cells/well and grown for 24 h at 37°C. Samples, diluted appropriately in RPMI 1640 medium, were added to the cells and again incubated for 48 h. The number of viable cells was determined by neutral red procedure (2). Briefly, after incubation with the samples, the cells were washed with saline and incubated for 90 min with the medium containing neutral red (166 µg/ml). After washing the plate to remove extracellular dye, a solution of acidified isopropanol (0.33% HCl) was then added to lyse the viable cells. The absorbance was read at 540 nm. The IC50 (the concentration of the test compound that caused a growth inhibition of 50% after 48 h exposure of the cells) was calculated from the dose curves generated by plotting percent growth versus the test concentration on a logarithmic scale using Microsoft Excel. Doxorubicin was used as a positive control in the cytotoxicity assay.
|
|
|---|
The antifungal testing results indicated that steroidal saponins 1 to 4, 6, 11, 14, 17, and 19 to 20 were active against Candida albicans, C. glabrata, C. krusei, Cryptococcus neoformans, and Aspergillus fumigatus, while the remaining steroidal saponins (compounds 5, 7 to 10, 12, 13, 15, 16, 18, and 21 to 22) and all the steroidal sapogenins (compounds 23 to 28) were inactive at the highest test concentration of 20 µg/ml. The antifungal activities of the active saponins are shown in Table 1. The lack of antifungal activity of the steroidal sapogenins is consistent with the previous report that similar compounds show little activity against agricultural pathogens P. oryzae and Hansenula anomala (7). Among the 10 active saponins that exhibited various degrees of activity against the five human pathogenic fungi, tigogenin saponins 1 to 4 with a sugar moiety of four or five monosaccharide units displayed remarkable activity against C. neoformans with MFCs comparable to the positive control AMB (MFC, 1.25 µg/ml). Compounds 1, 3, and 4 also exhibited potent activity against A. fumigatus with MICs of 2.5 to 5 µg/ml. Two diosgenin saponins, compounds 19 and 20, showed activity against C. albicans and C. glabrata similar to the four tigogenin saponins but were not active in C. neoformans and A. fumigatus at the highest test concentration of 20 µg/ml. To our knowledge, the in vitro antifungal activity of compounds 1, 3, and 4 against C. neoformans is the most potent within the steroidal saponin class that has so far been reported, while their antifungal potency against A. fumigatus is comparable to that of diuranthoside A, a neotigogenin saponin with the same sugar moiety as in compound 3 (37). In view of limited antifungal drugs for cryptococcosis and aspergillosis (17, 36), it is of particular significance for this discovery.
|
View this table: [in a new window] |
TABLE 1. Antifungal activity of steroidal saponins 1 to 4, 6, 11, 14, 17, and 19 to 20 (IC50/MIC/MFC,a µg/ml)
|
-hydroxy at C-6 and a ketone functionality at C-12 of the tigogenin skeleton, respectively, only compounds 11, 14, and 17 show moderate/marginal antifungal activity, although they have the same sugar chains as in the potent tigogenin saponin 3 or 4. Also, two antifungal diosgenin saponins (compounds 19 and 20) that have an aglycone with a double bond between C-5 and C-6 show a different antifungal profile than the four antifungal tigogenin saponins (compounds 1 to 4). The sugar chain in each saponin is necessary and plays a key role for its antifungal activity. For example, the antifungal activity of the hecogenin saponins (compounds 8 to 15) is largely dependent on the composition of sugar moiety. No activity is detected in the saponin when its sugar moiety is less than four monosaccharide units. Pentaglycoside 14 is more active than tetraglycoside 11 and shows extended antifungal spectrum against A. fumigatus. In the diosgenin saponin series (compounds 18 to 21), only triglycosides 19 and 20 are active against C. albicans and C. glabrata, while the monoglycoside 21 and diglycoside 18 do not show any activity. Similarly, within the group of the four most potent tigogenin saponins (compounds 1 to 4), their antifungal capacity is slightly influenced by the composition of the sugar moiety. The replacement of a glucosyl unit with a xylosyl unit in compound 1 from compound 2 shows enhanced activity against A. fumigatus, while an extended xylosyl unit in compound 4 from compound 3 reduces antifungal activity in Candida species.
The in vitro cytotoxicity of the 10 antifungal saponins 1 to 4, 6, 11, 14, 17, and 19 to 20 on the human cancer cell lines SK-MEL, KB, BT-549, SK-OV-3, and HepG2 as well as the noncancerous Vero cells was evaluated (Table 2) and compared with the positive control doxorubicin. Diosgenin saponins 19 and 20 were more cytotoxic than other saponins. Their IC50s for cytotoxicity to the mammalian cells (1.9 to 6.8 µg/ml) were comparable to the IC50s in the fungal cells (1.5 to 15 µg/ml), indicating a general cytotoxic effect of these compounds. However, selectivity was found in the four most potent antifungal tigogenin saponins (compounds 1 to 4). Saponins 1, 2, and 4 did not show any cytotoxicity to cancer cells up to the highest test concentration of 20 µg/ml (except compound 2 showed an IC50 of 6.9 µg/ml in HepG2 cells), although mild to moderate cytotoxicity was seen in the Vero cells (IC50s, 15, 3.7, and 7.5 µg/ml, respectively). Based on this comparison, saponin 1 appears to have the most favorable activity profile with selectivity indices (IC50 of Vero cells/IC50 of fungal cell) of 4.6, 5.7, 2.7, 37.5, and 10 for C. albicans, C. glabrata, C. krusei, C. neoformans, and A. fumigatus, respectively. Although higher selectivity indices are observed in C. neoformans and A. fumigatus for saponin 1, other antifungal saponins in this study remain less selective due to their cytotoxicity to mammalian cells. It is interesting that saponin 1 was much less toxic to Vero cells than compound 2 (IC50 of 15 µg/ml versus 3.7 µg/ml), whereas its antifungal activity, particularly for A. fumigatus, is more potent than that of compound 2 (MIC of 2.5 µg/ml versus 20 µg/ml). These observations indicate that the antifungal activity of these saponins is not necessarily correlated with their cytotoxicity to mammalian cells. The minor structural difference, for example, between compounds 1 and 2 or between compounds 1 and 4 (with the difference of one monosaccharide residue), imposes a significant difference on their cytotoxicity. The selectivity and specificity of a particular saponin is apparently associated with the drug target at the cellular or molecular level, which is still unknown at this point. Therefore, it is worthwhile to further study these antifungal compounds and, in a broad sense, the SAR of more steroidal saponins to explore the therapeutic potential of this important class of natural products as antifungal leads for drug discovery.
|
View this table: [in a new window] |
TABLE 2. Cytotoxicity of antifungal saponins 1 to 4, 6, 11, 14, 17, and 19 to 20
|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»