Correspondence: Tamás Papp, Department of Microbiology, Faculty of Science and Informatics, University of Szeged, H-6726 Szeged, Közép fasor 52, Hungary. Tel.: +36 62 544 516; fax: +36 62 544 823; e-mail: firstname.lastname@example.org
Ophiobolins are sesterterpene-type phytotoxins produced by fungi belonging mainly to the genus Bipolaris. In this study, the antifungal effect of ophiobolins A and B on different zygomycetes has been examined. Depending on the zygomycete tested, MIC values of 3.175–50 μg mL–1 were found for ophiobolin A and 25–50 μg mL–1 for ophiobolin B. Ophiobolin A inhibited sporangiospore germination of Mucor circinelloides and caused morphological changes; the fungus formed degenerated, thick or swollen cells with septa. Cytoplasm effusions from the damaged cells were also observed. Fluorescence microscopy after annexin and propidium iodide staining of the treated cells suggested that the drug induced an apoptosis-like cell death process in the fungus.
There are only a few reports on the antifungal effect of ophiobolins. In an earlier study, ophiobolin A was found to inhibit the growth of Gloeosporium, Glomerella, Corticium, Macrosporium and Trichophyton species (Nakamura & Ishibashi, 1958). It also showed a potent inhibitory effect against Aspergillus flavus, Candida albicans, Torulopsis cremoris and Torulopsis petrophilum (Li et al., 1995). Similarly, both ophiobolins A and B exerted strong activity against Trichophyton mentagrophytes in an agar-well diffusion assay (Au et al., 2000a).
Apart from these studies, the activity of these compounds against species representing other fungal groups, such as the class Zygomycetes, has never been studied. Zygomycetes are important as postharvest pathogens of agricultural products; Rhizopus, Mucor and Gilbertella species are among the most frequently isolated causative agents of rots in fruits and vegetables (Csernetics et al., 2005). Rhizopus, Rhizomucor and some other species are also known as opportunistic pathogens of humans and animals (Papp et al., 2008). These fungi have a substantial intrinsic resistance to the most widely used antifungal drugs. In this study, the effect of ophiobolins A and B on zygomycetes was investigated.
Materials and methods
Strains and media
The tested fungal strains are listed in Table 1. Growth inhibition tests were performed in a yeast extract–peptone–glucose medium (SPEC; 0.1% yeast extract, 0.05% peptone, 2.0% glucose). Investigations of the fungistatic–fungicidic effect of the drugs and cultivation for microscopy were performed on a solid or in a liquid yeast extract–glucose medium (YEG, 0.5% yeast extract, 1% glucose, 1.5% agar).
Table 1. MIC90 values of ophiobolins A and B (μg mL–1) determined for different zygomycetes fungi
ATCC, American Type Culture Collection (Rockville, MD); CBS, Centraalbureau voor Schimmelcultures (Baarn, the Netherlands); ETH, Swiss Federal Institute of Technology Culture Collection (Zurich, Switzerland); NRRL, Agricultural Research Service C.C. (Peoria, IL); SZMC, Szeged Microbial Collection (Hungary).
Letters c and s indicate fungicide or a fungistatic effect, respectively.
Ophiobolin A was purchased from Sigma, while ophiobolin B was purified on TLC after a diethyl ether extraction of the culture supernatant of a Bipolaris sp. strain. Briefly, culture supernatants were extracted with an equal volume of diethyl ether and the organic phase was dried under a nitrogen gas stream; the dried extract was resuspended in ethyl acetate and placed on silica gel F256 (Merck), which was developed with toluene-ethyl acetate-formic acid (5 : 4 : 1). The appropriate band was extracted and dried again. Stock solutions were performed from both compounds, diluting them in 10% methanol.
Antifungal activity assay
The in vitro antifungal activity of ophiobolins was determined in a 96-well microtiter plate bioassay by measuring the absorbance of the fungal cultures at 620 nm. The wells contained SPEC medium supplemented with ophiobolin A or B and inoculated with the appropriate sporangiospore suspension (105 spores mL–1). The drug concentrations applied were 100, 50, 25, 12.5, 6.25, 3.175 and 1.5875 μg mL–1, respectively. The plates were incubated for 72 h at 24 or 37 °C depending on the culturing requirements of the strains. Absorbances were measured using an ASYS Jupiter HD microplate reader (ASYS Hitech) every 24 h. Each test plate contained a sterile control (containing medium alone), a growth control (containing inoculated medium without the drugs) and a drug-free control (containing inoculated medium and methanol in the appropriate dilution without the ophiobolins). The uninoculated medium was used as the background for the calibration of the spectrophotometry. Absorbance of the untreated control cultures was referred to 100% of growth in each case. To decide whether the antifungal effect was fungistatic or fungicidic, 10 μL of each suspension in the microdilution plates was dropped onto YEG plates. After incubation for 24 h, the plates were checked visually. If colony formation was observed, the antifungal effect was considered to be fungistatic; otherwise, it was fungicidic. Each experiment was repeated three times.
For morphological examinations, the Mucor circinelloides strain ATCC 1216b was cultured on a solid and in a liquid YEG medium containing different concentrations of the drug (1.6, 3.2, 6.25 or 12.5 μg mL–1) at 24 °C. If the fungus was cultured on a solid medium, microscopy was performed after incubation for 24 h. In the case of the liquid cultures, ophiobolin A was added to the fungus either at the time of spore inoculation (0 h) or 4 h postinoculation, and cells were examined microscopically 5 h after the addition of the inhibitor (5 or 9 h postinoculation, respectively). Treated cells were stained using the annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (Sigma) according to the manufacturer's instructions. For nuclear staining, cells were resuspended in 1 mL of 0.1 μg mL–1 4′-6-diamidino-2-phenylindole (DAPI) staining solution and were allowed to stain for 30 min at room temperature. Stained spores were collected, washed twice with distilled water (dw), and resuspended in 50 μL dw. Microscopic examinations were performed with a Zeiss Jenalumar fluorescence microscope using an excitation filter U 205 g, a barrier filter G-244 and a 510 nm dichroic splitter.
Results and discussion
Sensitivity to ophiobolins A and B
The susceptibility to ophiobolins A and B of 17 fungal isolates representing six different genera (Micromucor, Mortierella, Mucor, Rhizomucor, Rhizopus and Gilbertella) was tested and their MIC values were determined (Table 1). With the exception of Mortierella wolfii, ophiobolin A showed higher antifungal activity against the tested isolates than ophiobolin B. This fungus proved to be the least sensitive to ophiobolin A, which inhibited the germination of its sporangiospores only at a concentration of 50 μg mL–1. Ophiobolin A proved to be highly active against the other tested strains: MIC90 values were found in a range 3.2–12.5 μg mL–1. For comparison, in the case of the opportunistic human pathogen Rhizopus oryzae, MIC values with complete blockade of growth were found in the ranges of 2–4, 2–4 and 0.5–2 μg mL–1 for amphotericin B, miconazole and itraconazole, respectively, whereas nystatin, griseofulvin and fluconazole exerted only a minimal inhibition effect on the fungus (Nyilasi et al., 2010; I. Nyilasi, unpublished data). In another study, MICs of ophiobolin A against A. flavus and C. albicans were found to be 25 and 12.5 μg mL–1, respectively (Li et al., 1995).
Effect of ophiobolin A on the sporangiospore germination and morphology of M. circinelloides
To study the effect of ophiobolin A on the development of a zygomycete, an M. circinelloides strain was cultured on a solid and in a liquid medium containing different concentrations of the drug and the cells that were formed were then examined microscopically. On the solid ophiobolin A-containing medium, the fungus formed degenerated, thick or swollen cells with septa instead of the normal coenocytic hyphae; cytoplasm effusions at the apical part of the germ tubes were often observed (Fig. 2a and b). If the concentration of the inhibitor was low (e.g. 1.6 μg mL–1), cells finally overcame the effect of the drug and hypha formation normalized in time (Fig. 2c and d). In the liquid medium, the effect of ophiobolin A was more pronounced. When the drug was added to the medium simultaneously with the spore inoculation (0 h), it blocked the germination of the sporangiospores in a concentration-dependent manner (Fig. 3c, g and m). If the drug was added to the culture during the formation of the germ tubes (e.g. at 4 h postinoculation), cytoplasm effusions at the hyphal tips (Fig. 3e), hyphal growth retardation and germ tube destruction (Fig. 3i and k) could be detected. After a 5-h incubation of the precultured cells in the presence of a high concentration of ophiobolin A (e.g. 6.25 μg mL–1 or higher), germ tubes almost completely disintegrated and a large amount of hyphal fragments appeared in the medium (Fig. 3o).
The mode of the antifungal action of ophiobolin A remains to be clarified. An earlier study reported that it could induce hyphal malformation in Phytophthora capsici, a pathogenic oomycete on green pepper; this effect was supposed to be due to the inhibition of β-1,3 glucan synthetase (Fukushima et al., 1993). However, the biological actions of ophiobolins are diverse and only their phytotoxic activities have been studied in detail. Early studies suggested that ophiobolins might act on the plasma membrane of the plants, inhibiting proton extrusion and impairing different transport processes (Cocucci et al., 1983; Reissig & Kinney, 1983). It was later demonstrated that ophiobolin A interacts with the maize calmodulin (Leung et al., 1984). Today, calmodulin, a central signal transducer subunit in a number of signaling complexes, is regarded as the main target for the toxin (Au et al., 2000a). Lysines 75, 77 and 148 of the calmodulin molecule have been shown to serve as binding sites for ophiobolin A, with lysine 75 as the primary inhibitory site (Au & Leung, 1998; Au et al., 2000a). In filamentous fungi, calcium signaling involving calmodulin plays a critical role in several processes of development and morphogenesis including cell cycle, formation and germination of spores, growth of hyphal tips as well as orientation and branching of the hyphae (Osherov & May, 2001; Zelter et al., 2004).
Annexin and propidium iodide staining of M. circinelloides cells
Ophiobolin A was described as a potent apoptosis-inducing agent in mammalian cells (Fujiwara et al., 2000). Moreover, there is evidence suggesting that the calcium/calmodulin signaling affects the fungal death response (Ramsdale, 2006). Therefore, we examined whether ophiobolin A would induce apoptosis-like cell death in zygomycetes and cells treated with ophiobolin A in liquid cultures stained with annexin V-FITC and propidium iodide using an apoptosis detection kit. The fluorescent probe annexin V-FITC binds to phosphatidylserine in the membrane and detects phosphatidylserine externalization in cells in the early stage of the apoptotic process. Propidium iodide binds to the DNA in the cytoplasm of cells, in which the membranes have been disorganized. Intact living cells are not stained either by the propidium iodide or by the annexin V-FITC. Accordingly, these reagents did not stain the untreated control (Fig. 3b). Cells treated with 1.6 μg mL–1 ophiobolin A formed germ tubes and hyphae with a morphology more or less similar to those of the untreated control, but these cells proved to be annexin- and propidium iodide positive, suggesting an apoptosis-like cell death process (Fig. 3d and f). At 3.2 μg mL–1 ophiobolin A concentration, spore germination was blocked and only spherical growth was observed. The homogeneous propidium iodide staining indicated that the inner membrane structures of the cells were totally disorganized (Fig. 3h). Cells treated with the same concentration of the inhibitor at 4 h postinoculation were also stained with both reagents (Fig. 3j and l). In the presence of higher drug concentrations, the totally disintegrated spores and hyphae showed intensive propidium iodide staining (Fig. 3n and p).
DAPI staining of ophiobolin A-treated M. circinelloides and Rhizopus stolonifer sporangiospores displayed the typical tubular and degenerated nuclear images corresponding to chromatin fragments (Fig. 4), whereas the untreated cells exhibited the normal bright, round-shaped nuclei.
During the past decade, there has been evidence of programmed cell death (PCD) in fungi (Ramsdale, 2006). PCD involves several different biological processes, among which, externalization of phosphatidylserine and the accumulation of DNA-strand breaks are characteristic of apoptosis, a specific type of PCD (Hamann et al., 2008). Although apoptotic processes have been described in a number of yeasts and filamentous fungi, zygomycetes have remained poorly characterized in this respect. There has only been one report on the apoptosis-like cell death process in zygomycetes (Roze & Linz, 1998), where the apoptotic process was triggered by the HMG-CoA reductase inhibitor, lovastatin, in Mucor racemosus. The described changes in the sporangiospore germination and hyphae formation were similar to those observed in our experiments. In that study, DNA fragmentation, with laddering, associated with the apoptosis-like process was also observed. This feature could be detected only when the treated cells were incubated at pH 7.45; the usual incubation pH (generally at pH 4.5) prevented the activation of the DNA fragmentation response. In our experiments, DNA laddering was detected neither at pH 4.5 nor at pH 7.45 (result not shown). However, it is worth mentioning that DNA laddering associated with PCD has rarely been observed in fungi and that this phenomenon is also not an absolute feature of apoptosis in mammalian cells (Ramsdale, 2006).
Currently, further experiments are in progress to elucidate the molecular background of the antifungal effect of ophiobolins and their possible interaction with fungal calmodulins. Our results suggest that these compounds may offer a promising tool to examine the death-related signaling pathways in fungi.
This work was supported by a grant from the Hungarian Scientific Research Fund and the National Office for Research and Technology (CK 80188).