Detection and Identification of Amanitins in the Wood-Rotting Fungi Galerina fasciculata and Galerina helvoliceps

Abstract

More than 600 strains of wood-rotting fungi were screened for the detection of amanitins. Three strains of Galerina fasciculata and 18 strains of Galerina helvoliceps contained amanitins. These strains contained mainly α- and β-amanitins in the native fruit bodies, while α- and γ-amanitins were found in liquid-cultured mycelia. Purified amanitins were confirmed by their chromatographic profiles, spectra (UV, Fourier transform infrared, and atmospheric ionization mass), cytotoxicity for mammalian cell lines (3T3 and SiHa), and inhibitory effects on RNA polymerase II. The results revealed that the purified amanitin fractions from these species are identical to authentic amanitins and suggest that these two species must be handled as poisonous mushrooms.

Amanitins belong to a family of bicyclic octapeptide mycotoxins that bind tightly to and inhibit eukaryotic DNA-dependent RNA polymerase II (4, 8, 9). Four types of amanitin (α, β, γ, and ɛ) are characterized by differences in their side groups (16–18) and are used as inhibitors in eukaryotic gene transcription analysis. These compounds are extracted from poisonous mushrooms, especially Amanita phalloides (15). Many attempts to obtain mycelial subcultures of Amanita species in artificial media have been unsuccessful because of their mycotrophy. For some species in the genera Lepiota and Galerina, identification of amanitins in fruit bodies has been reported (3, 7), and preliminary work on the production of amanitins in liquid medium with a strain of Galerina marginata has been reported by Benedict et al. (2). By screening culturable strains possessing the ability to produce amanitins, two species were found and identified as Galerina fasciculata and Galerina helvoliceps. In this paper, we report the isolation and identification of amanitins from natural mushrooms and cultured mycelia of these species.

MATERIALS AND METHODS

Chemicals.

Authentic α- and β-amanitins were purchased from Sigma Chemical Co. (St. Louis, Mo.). γ-Amanitin was purified from a mycelial extract of G. fasciculata as described below and used as a standard based on its molar extinction coefficient (15). Columns for high-pressure liquid chromatography (HPLC) were as follows: Superdex Peptide HR 10/30 (Amersham Pharmacia Biotech, Tokyo, Japan), Zorbax octyldecyl silane (ODS) (4.6 mm by 25 cm; Shimadzu-Dupont, Kyoto, Japan), Wakosil II 5C₁₈ types RS, HG, and AR (4.6 mm by 25 cm; Wako Pure Chemical, Osaka, Japan), μBondasphere 5 μ C₁₈-100Å (19 mm by 15 cm; Waters, Tokyo, Japan), and Discovery RP-Amide C16 5μm (4.6 mm by 25 cm; Sigma). HeLa cell nuclear extract for in vitro transcription was prepared as described by Manley et al. (10). [α-³²P]UTP (400 Ci/mmol, 6 mCi/ml) was from Amersham Pharmacia Biotech. All other reagents were of the highest grade available from commercial sources.

Collection and isolation of amanitin-containing basidiomycetes.

Fruiting bodies of wood-rotting fungi were collected between July 1995 and October 1997 from decayed wood in forests in the middle latitudes of Japan. The mushrooms were identified by referring to studies by Ammirati et al. (1), Imazeki and Hongo (6), and Singer (11). After HPLC analysis as described by Enjalbert et al. (5), the strains found to contain amanitins were subcultured. The liquid medium (HSV) for amanitin production contained the following, per liter: 1 g of yeast extract (Difco, Detroit, Mich.), 2 g of glucose, 0.1 g of NH₄Cl, 0.1 g of KCl, 0.1 g of CaSO₄ · 1/2H₂O, 1 mg of thiamine · HCl, and 0.1 mg of biotin (medium pH with HCl, 5.2). Agar medium (HSVA) for subculture contained 2% agar in HSV. Vegetative mycelial stocks were prepared by culturing aseptic fragments of fruiting bodies on HSVA plates. Fungal colonies were transferred and reisolated until pure cultures were obtained. The stocks were subcultured every 6 months and deposited at The Mushroom Research Institute of Japan (Kiryu, Gunma, Japan).

Screening of amanitin-producing strains.

Isolated strains were cultured in 30 ml of HSV (in a 100-ml Erlenmeyer flask by rotary shaking at 150 rpm) at 25°C for 30 days in the dark. The mycelia were collected by centrifugation at 3,000 × g for 20 min, washed twice with distilled water, lyophilized, and weighed. The extraction and determination of amanitin content were carried out according to the methods described by Enjalbert et al. (5). For the analytical reversed-phase HPLC, a 4.6- by 250-mm Zorbax ODS column was used.

Large-scale cultivation.

To confirm the amanitins, mainly two strains, G. fasciculata GF-060 and G. helvoliceps GH-343, were used for large-scale cultivation. The mycelia cultivated for 10 days in 30 ml of HSV medium (in a 100-ml Erlenmeyer flask) as described above were dispersed aseptically with a homogenizer (Biomixer SBM-1; Nihonseiki, Tokyo, Japan) for 10 s at 30,000 rpm. The dispersed mycelia (30 ml) were mixed in the same medium (400 ml) in a 1-liter Erlenmeyer flask, and the mixture was further cultivated under the same conditions for 30 days.

Purification of amanitins. (i) Step 1: preparation of cell extract.

Washed and lyophilized mycelia from 430 ml of culture broth were suspended in 500 ml of methanol containing 0.083% (vol/vol) HCl. The suspension was treated with a homogenizer (at 30,000 rpm for 5 min at 4°C). The supernatant fluid was obtained by centrifugation at 15,000 × g for 15 min at 4°C, and the precipitates were extracted three times under the same conditions. The combined supernatant fluids were concentrated and dried in a rotary evaporator at 35°C. The cell extract was dissolved in 10 ml of distilled water and defatted with an equal volume of diethyl ether.

(ii) Step 2: solid-phase extraction by C₁₈ cartridge.

The extract from step 1 was diluted to 10 ml with distilled water and applied to a Sep-Pak Vac tC₁₈ cartridge (200 mg; Waters) previously equilibrated with distilled water. The cartridge was washed with 10 ml of distilled water, and then the amanitins were eluted with 40% (vol/vol) methanol. The amanitin fractions in 10 ml of 40% (vol/vol) methanol were dried in vacuo at 35°C.

(iii) Step 3: size exclusion HPLC.

The extract from step 2 was dissolved in water, passed through a membrane filter (pore size, 0.45 μm; Millipore, Tokyo, Japan), and applied to size exclusion HPLC (LC-10AT; Shimadzu) under the following conditions: column, Superdex Peptide HR 10/30; elution buffer, 100 mM ammonium acetate containing 35% (vol/vol) acetonitrile (pH 6.5 with acetic acid); flow rate, 1.0 ml/min at 25°C; and monitoring with a photodiode array detector (MD-910; Jasco Corp., Tokyo, Japan) from 210 to 500 nm. The fraction corresponding to the amanitins (eluted at 14.6 to 16.3 min) was collected and concentrated in vacuo at 35°C.

(iv) Step 4: preparative reversed-phase HPLC.

Purification of each amanitin from the step 3 fraction was carried out by binary gradient HPLC (LC-6A; Shimadzu): column, μBondasphere 5μ C₁₈-100 Å; solvent A, 20 mM ammonium acetate containing 5% (vol/vol) methanol (pH 5.0 with acetic acid); solvent B, 20 mM ammonium acetate containing 80% (vol/vol) methanol (pH 5.85 with acetic acid). Separation was carried out at a flow rate of 10 ml/min at 35°C with the following three eluents: 20% solvent B for 15 min, a linear gradient of 20 to 99% solvent B for 15 min, and 99% solvent B for 10 min. Detection and UV spectral analysis of the eluents were performed with a photodiode array detector at 210 to 500 nm. The fractions, corresponding to β-amanitin eluted at 16.5 min, α-amanitin eluted at 18.5 min, and γ-amanitin eluted at 25.2 min, were collected separately and lyophilized. The purified fractions from the Galerina species (defined as the β-, α-, and γ-amanitin fractions) were used in the following experiments.

Spectrometric analyses.

Fourier transform infrared (FT-IR) spectra were obtained with a KBr pellet on an FT-IR spectrophotometer (model 1600; Perkin-Elmer, Yokohama, Japan). UV spectra in water were obtained with a spectrophotometer (U-3210; Hitachi, Tokyo, Japan). β-, α-, and γ-amanitin fractions were dissolved in solvent (2.5 mM ammonium acetate in 50% [vol/vol] methanol), and their mass spectra were analyzed on a Perkin-Elmer API-100 by atmospheric ionization (API) in positive-ion mode.

Determination of melting point.

The melting point was determined in a glass capillary with a melting point apparatus (MFB595; Gallenkamp, Sussex, United Kingdom).

Cytotoxicity test.

Two mammalian cell lines (3T3 [laboratory stock] and SiHa [gift of T. Kanda, National Institute of Health, Tokyo, Japan]) (12) were used to assay cytotoxicity. Cells were grown in 96-well plates (no. 3860; Iwaki Glass, Chiba, Japan) on Dulbecco’s modified Eagle’s medium (Iwaki Glass) supplemented with 10% fetal bovine serum (Iwaki Glass) at 37°C in 5% CO₂, with the medium changed every 3 days. Authentic α-amanitin and the α-amanitin fraction were diluted to the appropriate concentration with Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum. After incubation for 7 days with or without α-amanitin, cell growth was measured with an MTT (3-[4,5-dimethyl-2-thiazolyl]-2,5-diphenyl-2H-tetrazolium bromide) assay kit (Sigma) at A₅₇₀ to A₆₂₀. After the calculation, each 50% lethal dose (LD₅₀) was estimated from the dose-response curve.

Inhibitory effect on RNA polymerases II and III.

The specific inhibition of RNA polymerase II activity was measured by transcribing specific templates with class II and class III promoters. Two DNAs, plasmid pSmaF, containing the SmaI-F fragment of adenovirus type 2 major late promoter (template for RNA polymerase II), and pAdSalIC, containing the VA1 RNA gene of adenovirus type 5 (for RNA polymerase III), provided by H. Handa and I. Saito (The Institute of Medical Science, University of Tokyo), respectively, were used as the templates (14) for in vitro transcription in a reconstituted system of HeLa cell components (10). Authentic α-amanitin or the α-amanitin fraction was added to each reaction mixture at the appropriate concentration. The transcription reactions contained 0.5 μl of [α-³²P]UTP (400 Ci/mmol, 6 mCi/ml); reactions were carried out according to methods described by Manley et al. (10). Transcripts were analyzed in denaturing 4% polyacrylamide gel containing 7 M urea. Autoradiographs of the transcripts were obtained by exposing the gel to an X-ray film with an intensifying screen overnight at −70°C.

RESULTS AND DISCUSSION

Screening of amanitin-containing basidiomycetes.

Strains containing amanitin in their fruiting bodies are listed in Table 1. They were identified as G. helvoliceps and G. fasciculata. In general, these two species share similar habitats but are distinguished by fruiting body size (Fig. 1) and cystidia shape. Amanitin content in the fruiting body of each strain is also shown in Table 1. The reasons for the various amanitin content levels might be geographical differences, seasonal conditions, moisture content, or the diversity of inheritance. According to the LD₅₀ of α-amanitin in humans (13), the ingestion of 5 to 10 g of these fresh mushrooms would be fatal for an adult. When these strains were cultured in HSV medium, α-amanitin and small amounts of γ-amanitin accumulated intracellularly, while little β-amanitin was observed (Fig. 2C). On the other hand, in the fruiting bodies or mycelia from cultivation on solid HSVA, mainly α- and β-amanitins were observed, with only trace amounts of γ-amanitin detected (Fig. 2A and B). Trace amounts of extracellular amanitins were detected at every stage of the culture period (data not shown). These results suggest that the production patterns of amanitins are affected by culture conditions, such as liquid or solid medium.

TABLE 1.

Galerina strains containing amanitins in Japan

Strain	Date of collection (mo/day/yr)	Location	Amanitin content (μg/g of fresh wt)a
			α	β
G. helvoliceps
GH-323	09/08/1996	Azegata, Nikko, Tochigi	138.41	97.51
GH-324	09/08/1996	Azegata, Nikko, Tochigi	125.98	81.78
GH-326	09/08/1996	Azegata, Nikko, Tochigi	89.21	38.49
GH-327	09/08/1996	Azegata, Nikko, Tochigi	145.21	99.68
GH-340	09/15/1996	Koutoku, Nikko, Tochigi	127.91	81.97
GH-343	09/15/1996	Azegata, Nikko, Tochigi	391.06	165.16
GH-344	09/15/1996	Azegata, Nikko, Tochigi	122.34	120.24
GH-345	09/15/1996	Azegata, Nikko, Tochigi	52.92	17.71
GH-349	09/15/1996	Azegata, Nikko, Tochigi	64.29	28.80
GH-353	09/15/1996	Kuriyama, Tochigi	89.47	37.65
GH-355	09/15/1996	Kuriyama, Tochigi	233.66	97.81
GH-364	09/23/1996	Mizorogi, Akagi, Gunma	243.05	87.11
GH-368	09/23/1996	Mizorogi, Akagi, Gunma	210.17	74.93
GH-380	09/29/1996	Tanbara, Numata, Gunma	172.72	55.04
GH-386	10/04/1996	Ashiwada, Yamanashi	128.45	112.14
GH-403	10/20/1996	Nagaoka, Niigata	41.73	10.21
GH-408	09/21/1997	Azegata, Nikko, Tochigi	38.48	9.33
GH-409	10/10/1997	Hotaka, Tone, Gunma	41.53	13.45
G. fasciculata
GF-060	09/27/1995	Kawamata, Tochigi	112.84	147.12
GF-329	09/08/1996	Azegata, Nikko, Tochigi	100.29	122.04
GF-405	11/10/1996	Shitada, Niigata	255.59	Tr

^aAmanitins in natural fruiting bodies were measured by reversed-phase HPLC analysis. 

FIG. 1.

Fruiting bodies of amanitin-containing mushrooms G. helvoliceps (A) and G. fasciculata (B). Bar = 1 cm.

FIG. 2.

Typical chromatograms of intracellular amanitins of G. helvoliceps GH-343, obtained by analytical reversed-phase HPLC. (A) Naturally occurring fruiting body; (B) mycelia cultured on solid (HSVA) medium; (C) mycelia cultured on liquid (HSV) medium. Separation conditions are described in Materials and Methods.

Purification and confirmation of amanitins.

The purification and confirmation of the amanitins were carried out as described in Materials and Methods. The purity of each amanitin fraction was confirmed, and each produced a single peak on analytical reversed-phase HPLC with various columns and elution conditions. The purities of the final products of both the α- and β-amanitin fractions, calculated from the peak areas in analytical reversed-phase HPLC, were more than 99% (data not shown). Elution profiles of the amanitin fractions from Galerina strains were also compared in six types of reversed-phase chromatography (Zorbax ODS; Wakosil II 5C₁₈ types RS, HG, and AR; μBondasphere 5μ C₁₈-100 Å and Discovery RP-Amide C16 5μm). All profiles of α- and β-amanitin fractions obtained in these chromatographies were identical to those of authentic α- and β-amanitin, respectively (data not shown). Physical and biological properties were also compared. Estimation of the molecular weight by API mass spectrometry (919), determination of the melting point (decomposition at 254°C), and calculation of the LD₅₀s for 3T3 (0.35 μg ml⁻¹) and SiHa (0.32 μg ml⁻¹) cells showed no differences between the authentic α-amanitin and the α-amanitin fraction. The UV and FT-IR spectra of the amanitin fractions were identical to those of the authentic amanitins (data not shown). α- and β-amanitin fractions from mycelial extracts from other Galerina strains, as listed in Table 1, grown on HSVA or HSV medium were also confirmed by the same spectrometric analyses. All γ-amanitin fractions from the listed strains were confirmed by reference to the various spectrometric analyses and chromatographic studies described by Wieland (15). Because amanitins are well known as inhibitors of mammalian RNA polymerase II, the inhibitory activities of authentic α-amanitin and the α-amanitin fraction were tested (Fig. 3). Both α-amanitins inhibited the transcription of RNA polymerase II (major product from pSmaF, 536 bases) at a low concentration (0.4 μg/ml), while neither α-amanitin preparation inhibited the transcription of RNA polymerase III (major product from pAdSalIC, 156 bases), even at a high concentration (1.6 to 2.0 μg/ml). These observations illustrate one important characteristic of amanitins.

FIG. 3.

Inhibitory effects of authentic α-amanitin and the α-amanitin fraction from G. fasciculata GF-060 on RNA polymerases II and III in an in vitro transcription system. Two DNAs, pSmaF for polymerase II and pAdSalIC for polymerase III, were used as templates for the reaction. The conditions for in vitro transcription analysis are described in Materials and Methods.

In conclusion, it was found that G. fasciculata and G. helvoliceps produce α-, β-, and γ-amanitins in cultured mycelia, in addition to in their fruit bodies. In the past, these Galerina species were considered to be poisonous mushrooms if ingested; however, the toxic components were not clarified. In this report, endogenous amanitins in these species were identified. Therefore, these two species must be handled as poisonous mushrooms. The availability of two fermentation styles (solid and liquid medium culture) for these strains makes it possible to obtain sufficient amounts of each amanitin for further studies. Moreover, these amanitin-producing strains will contribute to the elucidation of the amanitin biosynthesis pathways.