- Email: [email protected]
Toxicity testing of destruxins and crude extracts from the insect-pathogenic fungus Metarhizium anisopliae
FEMS Microbiology Letters 251 (2005) 23–28 www.fems-microbiology.org
Toxicity testing of destruxins and crude extracts from the insect-pathogenic fungus Metarhizium anisopliae Anke Skrobek, Tariq M. Butt
*
School of Biological Sciences, University of Wales Swansea, Singleton Park, Swansea SA2 8PP, UK Received 21 April 2005; accepted 19 July 2005 First published online 8 August 2005 Edited by M.J. Bidochka
Abstract Increasing sensitivity towards secondary metabolites from fungal biological control agents (BCAs) has prompted the toxicological risk assessment of metabolites produced by the insect pathogenic fungus Metarhizium anisopliae. Viability studies on one human and one insect cell line were used to compare the two approaches of testing individual metabolites (destruxins A, B and E) or the complete crude extract from liquid cultures. Furthermore, crude extract was separated into fractions, which did not contain the main destruxins A, B and E. Evaluation of the cytotoxic activity of these different compounds suggested that a wide range of metabolites with synergistic or adverse effects are present in the crude extract. The results indicate that identification and toxicological assessment of each individual metabolite produced by a BCA is not only time and cost-intensive, but also does not convey the whole picture. Testing of the crude extract offers an alternative approach and is recommended when assessing the risks of metabolites for registration purposes. Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Metarhizium anisopliae; Destruxins; Fungal metabolites; Risk assessment; Toxicity; Cell lines
1. Introduction Increasing public sensitivity to environmental pollution and problems of pest resistance to chemical pesticides has provided the impetus for the development of alternate, benign strategies for pest control. Insect pathogenic fungi like Metarhizium anisopliae show considerable promise for use in integrated pest management (IPM) programmes [1]. However, consumer concerns regarding mycotoxins entering the food chain has prompted closer scrutiny of the secondary metabolites * Corresponding author. Tel.: +44 1792 295374; fax: +44 1792 295447. E-mail addresses: [email protected] (A. Skrobek), t.butt@swansea. ac.uk (T.M. Butt).
of all fungal biocontrol agents (BCAs). Regulatory authorities often require detailed information on ‘‘relevant metabolites’’. It is not clear what constitutes a relevant metabolite when most fungi secrete a disparate array of bioactive compounds [2]. Production of these metabolites may vary between species or strains and is also influenced by cultural conditions (e.g. [3–6]). Most often the metabolites are secreted in extremely small quantities even under optimal production systems [2]. Evaluation of the toxicological risks of each secondary metabolite could prove onerous and highly expensive partly because methods and tools have not been developed for the risk assessment of metabolites of fungal BCAs. Furthermore, no guidelines or simulation models exist to evaluate the fate of secreted fungal metabolites in the environment. Little is known about the full range
0378-1097/$22.00 Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsle.2005.07.029
24
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
the food chain and pose a risk to human and animal health.
of metabolites produced and whether they enter the food chain so posing a risk to human and animal health. Without doubt the high cost and time required for risk assessment of each metabolite will discourage many bio-manufacturers from developing fungal BCAs and will result in growers being trapped between the diminishing number of chemical pesticides and the lack of safe alternatives [7,8]. Clearly, sensible strategies that are simple and cost-effective are needed for the risk assessment of metabolites of fungal BCAs. One plausible approach is to evaluate the risk of crude extracts from culture filtrates that would contain the widest range of metabolites the fungus could ever produce in vitro. This approach has several highly commendable attributes; it is indifferent to variations in culture conditions and isolates and would take into account any additive or synergistic interactions between different compounds. Furthermore, it represents the ‘‘worse case scenario’’ since the metabolites are more concentrated than they would be in natural systems. The aim of this study was to determine the toxicity of crude extract preparations of M. anisopliae culture filtrates and compare these with purified destruxins (dtxs), the major metabolites secreted by this fungus both in vivo and in vitro (e.g. [9,10]). Toxicological (viability) assays were conducted using one human and one insect cell line. In vitro test systems with cell lines are increasingly gaining acceptance as tools to determine the toxicological effects of pesticides instead of whole animal testing [11,12]. Mammalian and insect cell lines have already shown considerable promise in the evaluation of the toxicological properties of some fungal metabolites [13–18]. This study forms part of a major programme to assess whether metabolites from fungal BCAs enter
2,000
2. Materials and methods 2.1. Test compounds Cells were exposed to the three main dtxs A, B and E and to their mixtures in order to study synergistic and adverse effects. Dtx A was obtained from Sigma (UK). Dtxs B and E were a kind gift from Dr. Alain Vey (INRA, France). As an alternative approach, cells were also treated with three different batches of crude extract from M. anisopliae V275 culture filtrate isolated in three independent experiments, which was obtained with organic solvents with a wide polarity range and analysed by HPLC as described by [5]. Additionally, the crude extract batches were separated into four fractions with decreasing polarity excluding dtxs with a Foxy Junior Fraction Collector (Dionex, UK) as indicated in Fig. 1. Cells were exposed to these individual fractions in order to eliminate the effect of dtxs in the crude extract. Reagents, media components and solvents were purchased from Sigma (UK) if not indicated otherwise. 2.2. Cell line assays The human leukemic cell line HL60 (Caucasian, promyelotic) was a kind gift from Dr. Luigi Macchia (Bari, Italy). Cells were maintained in RPMI1640 medium supplemented with 10% foetal calf serum and 1% glutamine–penicillin–streptomycin solution at 37 °C
WVL:215 nm
mAU
dtx e dtx a 1,500
1,000
Fraction 2 500
Fraction 1
dtx b Fraction 3
Fraction 4
min
-200 0.0
5.0
10.0
15.0
20.0
25.0
33.0
Fig. 1. Different HPLC fractions of crude extracts from culture filtrates of M. anisopliae V275 used for cytotoxicity testing.
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
25
and 5% CO2. The SF9 insect cell line (derived from Spodoptera frugiperda ovaries) was purchased from the European Culture Collection (ECACC) and maintained in TNMFH insect medium supplemented with 10% foetal calf serum and 1% glutamine–penicillin–streptomycin solution at 26 °C. Tests were carried out in 96 well flat bottom microtitre plates (Nunc, UK), which contained 105 cells/well. Tests consisted of 50 ll (for 4 h incubation) and 100 ll (for 24 h incubation) of diluted test compound in order to study acute toxic effects as well as effects after longer exposure. The solvent concentration (methanol:acetonitrile, 1:1, v/v) did not exceed 0.5%. Controls consisted of 50 ll phosphate buffered saline (PBS, 0.2 g/l KCl, 0.2 g/l KH2PO4, 8 g/l NaCl, 1.15 g/l Na2HPO4 Æ 7H2O in deionised water, pH 7) for 4 h incubation or 100 ll culture medium for 24 h incubation as well as of 0.5% pure solvent in PBS or culture medium. After incubation for 4 or 24 h, 50 ll calcein AM (0.5 lM in PBS, Molecular Probes, NL) were added and cytotoxicity was determined by reading the absorption in a BMG labsystems ‘‘Polar Galaxy’’ microtitre plate reader after 30 min incubation at optimum conditions for the respective cell line. The following filter combination was used: 485– 15 nm excitation and 538 nm emission. Cell viability was calculated by integrating absorbance values with a pre-determined calibration function. The percentage of living cells in comparison with the untreated control indicated toxic effects after 4 or after 24 h of exposure. For better comparison LC50 values (=lethal concentration that caused 50% mortality) were calculated.
licates per treatment, were analysed to determine means, normal distribution and equal variance (p 6 0.05). Data that were not distributed normally were analysed with the H-test (Kruskal–Wallis ANOVA on ranks, p 6 0.05). Significant differences (denoted by different letters) were then determined using the Dunns-test. Data with normal distribution were further analysed using the Tukey-test. LC50 values and their lower and upper fiducial limits (=confidence limits 95%) were calculated by PriProbit 1.63 (1996–2000 Masayuki Sakuma).
2.3. Statistical analysis
Three different batches of crude extract from M. anisopliae V275, extracted with an equal mixture of polar and non-polar solvents, were cytotoxic to both cell lines. The effects on HL60 cells with LC50 values of 221 and 193 ppm after 4 and 24 h, respectively, were stronger
The results were analysed statistically with the program SigmaStat 2.03 for Windows (SPSS Inc., US). Results of three independent assessments, with four rep-
3. Results 3.1. Effects of pure dtxs and their mixtures on cell viability Pure dtxs A, B and E exhibited no effect on the cell viability of human leukemic HL60 cells up to 500 ppm (Table 1). There was no decrease in the number of live cells either after 4 or after 24 h (data not shown). Likewise, mixtures of two or three dtxs did not show any effect on HL60 cells (Table 1). Pure dtxs or their mixtures had no acute toxic effect on SF9 cells up to 500 ppm but dtx A and its mixtures exhibited toxic activity after 24 h of exposure with LC50 values ranging between 5 and 12 ppm (Table 1). Dtxs B and E or their mixture had no effect on cell viability after 24 h incubation. 3.2. Effects of crude extract and its fractions on cell viability
Table 1 Toxicity of different polar HPLC fractions, purified destruxins (dtxs), combination of dtxs and crude extracts of M. anisopliae V275 to human (HL60) and insect (SF9) cell lines (LC50 values with lower and upper fiducial limits (95%) as calculated by PriProbit 1.63 (1996–2000 Masayuki Sakuma)) Test compound
LC50 (ppm) HL60
dtx A dtx B dtx E dtx A + dtx B dtx A + dtx E dtx B + dtx E dtx A + dtx B + dtx E Crude extract Fraction 1 Fraction 2 Fraction 3 Fraction 4
SF9
4 h incubation
24 h incubation
4 h incubation
24 h incubation
n.e. n.e. n.e. n.e. n.e. n.e. n.e. 221 (162; 331) 252 (225; 424) >MCT >MCT 80 (10; 1100)
n.e. n.e. n.e. n.e. n.e. n.e. n.e. 193 (120; 345) >MCT >MCT >MCT 370 (164; 453)
n.e. n.e. n.e. n.e. n.e. n.e. n.e. >MCT 320 (260; 410) 200 (n.c.) 97 (49; 210) 92 (45; 198)
5 (4; 16) n.e. n.e. 12 (8; 32) 9 (7; 33) n.e. 8 (6; 24) >MCT 120 (92; 170) 240 (160; 440) >MCT 110 (79; 150)
n.e., no effect up to 500 ppm; n.c., fiducial limits not reliable; MCT, max concentration tested.
26
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
than those on SF9 cells with LC50 values above the highest concentration tested (500 ppm, Table 1). HPLC analyses established that there was a high variation in the dtx content of the three crude extract batches: in 100 ppm of crude extract there were 18 ± 11 ppm dtx A, 8 ± 6 ppm dtx B and 30 ± 19 ppm dtx E. Crude extract fractions, which excluded dtxs, exhibited cytotoxic effects after 4 h incubation with HL60 and SF9 cells (Figs. 2 and 3, Table 1). While fractions 1 and 4 showed stronger effects on the human cell line, fractions 2 and 3 exhibited higher activity against the
insect cells. The least polar fraction 4 expressed the strongest activity with 97% insect cell mortality and 100% human cell mortality at 500 ppm (Figs. 2 and 3). Toxic effects after 24 h exposure could be observed for all fractions and for both cell lines (Table 1). Fraction 1, which was more active against HL60 after 4 h than after 24 h incubation, exhibited stronger toxic effects against SF9 after 24 h with LC50 values of 252 and 320 ppm for 4 h and >500 and 120 ppm for 24 h, respectively. Fraction 2 exhibited stronger effects on the insect cell line at 4 and 24 h incubation. Fraction 3 was more active against SF9 cells after 4 h (97 vs. >500 ppm). Fraction 4 had a lower LC50 value for HL60 cells than for SF9 cells after 4 h of incubation (80 vs. 92 ppm) but the LC50 value for SF9 cells was lower for 24 h of incubation with 110 vs. 370 ppm. Higher LC50 values for 24 h incubation compared to 4 h incubation indicated that HL60 cells and SF9 cells recovered from the acute toxic effects of fraction 1 and fraction 3, respectively. In contrast, fraction 1 was more toxic to the insect cells after 24 h incubation than after 4 h.
4. Discussion
Fig. 2. Effect of different fractions of crude extracts from M. anisopliae V275 on the viability of human leukemic cells (HL60) after 4 h incubation. (Values represent the means of three independent assessments with four replicates each, significant differences between controls and different concentrations of individual fractions are indicated by different letters, DUNNS, p 6 0.05.)
Fig. 3. Effect of different fractions of crude extract from M. anisopliae V275 on the viability of insect cells (SF9) after 4 h incubation. (Values represent the means of three independent assessments with four replicates each, significant differences between controls and different concentrations of individual fractions are indicated by different letters, TUKEY, p 6 0.05.)
Dtxs, cyclic peptides, are the main metabolites secreted by M. anisopliae [2,9]. They are known to have cytotoxic and cytostatic effects on mammalian and insect cells with dtx A and dtx E being the most toxic [19–24]. Odier et al. [20] reported acute cytotoxic and cytostatic activities of dtx E on mouse lymphocytes from 0.001 and 0.1 ppm on for cell lines P388 and L1210, respectively. In a later study, they showed a total cytotoxic effect of dtx E on blood and thymic lymphocytes with doses of 16 and 3.3 ppm after 48 h incubation as well as an inhibition of P388 proliferation at 0.33 ppm for dtx E, 16 ppm for dtx A and 3.33 ppm for dtx B [19]. This is the first study to compare the toxicity of crude extracts with purified dtxs A, B and E of M. anisopliae on human and insect cell lines. All three dtxs at 500 ppm had no effect on human leukemic cells up to 24 h incubation whereas dtx A caused mortality in SF9 insect cells, suggesting huge differences in the sensitivity of different cell lines. Furthermore, HL60 human leukemic cells were more sensitive than SF9 cells to the crude extract of M. anisopliae than to the pure dtxs. These observations suggest that several different cell lines should be chosen for toxicological assessments. Strains of M. anisopliae differ considerably in their metabolite profiles [3–5]. This is also true for other insect-pathogenic fungi including Beauveria bassiana and Tolypocladium sp. [25,26]. Evaluating the effect of crude extracts instead of individual compounds takes such differences into account since extracts of culture filtrates
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
contain the widest range of metabolites the fungus can produce in vitro. In addition, metabolites are far more concentrated than they would be in the environment [6]. For example, Amiri-Besheli et al. [3] showed that the dtx content of one Galleria mellonella larva infected with M. anisopliae V245 was 0.44 lg dtx A and 0.5 lg dtx B on the day of death. This means that 1000 g Galleria larvae (ca. 7 g bodyweight per larva) would contain 0.063 mg dtx A and 0.071 mg dtx B. In contrast, one litre (equals 1000 g) of liquid culture of the same strain contains maximally 12 mg dtx A, 4 mg dtx B and 17 mg dtx E [5]. Thus, when assessing the toxicity of BCA metabolites it is better to look at the crude extract than at individual compounds. Toxicological tests conducted using crude extracts and its fractions were variable compared with pure dtxs. This may be due to variations in the levels of dtxs, and possibly other metabolites, in the different batches of crude extract, since the same strain may not always produce the same quantity of metabolites even when grown under the same cultural conditions. Thus, crude extracts also take into account variations in metabolite profiles between fermentation batches. Crude extracts amplify the signal for the main metabolites but may also take into account interactions between compounds. Genthner et al. [27] reported that dtxs alone were not responsible for the toxicity of M. anisopliae crude extracts to mysids. Our studies also show that compounds other than dtxs are cytotoxic to HL60 and SF9 cells after 4 h incubation. Metarhizium secretes an array of bioactive metabolites including cytotoxic compounds like swainsonine and cytochalasins C and D (e.g. [28–32]), which potentially could account for the toxicity of crude extracts. Synergy and antagonism between compounds could also influence the overall toxicity of crude extracts. Dtx A, of all the compounds tested, was the most toxic to SF9 insect cells but in crude extracts it was less effective presumably due to antagonism from other compounds. For example, 5 ppm pure dtx A caused 50% mortality in SF9 cells while 500 ppm crude extract containing 90 ppm dtx A did not cause such a high mortality. When looking at the different fractions, some were weakly cytotoxic after a short incubation period but increased in toxicity over time. Some fractions were comparatively more toxic after only a brief exposure. High concentrations of the complete crude extract had to be used to cause 50% death of HL60 cells after 24 h incubation. Complete crude extract was less toxic to SF9 cells than the separate fractions. These findings suggest that the cytotoxic activity of crude extracts from BCAs cannot be attributed to one compound alone and that crude extracts represent the worse case scenario. Risk assessment of crude extracts requires less time and resources and subsequently would be more cost-effective for companies wishing to register fungal BCAs.
27
Acknowledgements The authors thank Dr. Vey (F) for kindly providing standards for dtxs B and E and Dr. Macchia (I) for the human leukemic cell line. This work was supported by the European Commission, Quality of Life and Management of Living Resources Programme (QoL), Key Action 1 on Food, Nutrition and Health QLK1-200101391 (RAFBCA).
References [1] Butt, T.M., Jackson, C.W. and Magan, N. (2001) Fungi as Biocontrol Agents: Potential, Progress and Problems. CAB International, Wallingford, 416 pp. [2] Vey, A., Hoagland, R. and Butt, T.M. (2001) Toxic metabolites of fungal biocontrol agents In: Fungi as Biocontrol Agents: Potential, Progress and Problems (Butt, T.M., Jackson, C.W. and Magan, N., Eds.), pp. 311–345. CAB International, Wallingford. [3] Amiri-Besheli, B., Khambay, B., Cameron, S., Deadman, M.L. and Butt, T.M. (2000) Inter- and intra-specific variation in destruxin production by insect pathogenic Metarhizium spp., and its significance to pathogenesis. Mycol. Res. 104, 447– 452. [4] Kershaw, M.J., Moorhouse, E.R., Bateman, R.P., Reynolds, S.E. and Charnley, A.K. (1999) The role of destruxins in the pathogenicity of Metarhizium anisopliae for three species of insect. J. Invertebr. Pathol. 74, 213–223. [5] Wang, C.S., Skrobek, A. and Butt, T.M. (2004) Investigations on the destruxin production of the entomopathogenic fungus Metarhizium anisopliae in liquid and solid media. J. Invertebr. Pathol. 85, 168–174. [6] Strasser, H., Vey, A. and Butt, T.M. (2000) Are there any risks in using entomopathogenic fungi for pest control, with particular reference to the bioactive metabolites of Metarhizium, Tolypocladium and Beauveria species? Biocontrol Sci. Technol. 10, 717–735. [7] Zimmermann, G., Lu¨th, P., Bode, E., Hommes, M., Selzsam, B., Guske, S. and Huber, J. (2004) Bericht u¨ber das dritte Fachgespra¨ch zum biologischen Pflanzenschutz ‘‘Kommerzialisierung mikrobiologischer Pflanzenschutzmittel’’. Nachrichtenbl. Deut. Pflanzenschutzd. 56, 131–136. [8] Blum, B., Ehlers, R.L., Haukeland-Salinas, S., Hokkanen, H., Jung, K., Kuhlmann, U., Menzler-Hokkanen, I., Ravensberg, W.J., Strasser, H., Warrior, P. and Wilson, M. (2003) Letter to the editors – Biological control agents: safety and regulatory policy. BioControl 48, 474–487. [9] Roberts, D.W. (1966) Toxins from the entomogenous fungus Metarhizium anisopliae. II. Symptoms and detection in moribund hosts. J. Invertebr. Pathol. 8, 222–227. [10] Samuels, R.I., Charnley, A.K. and Reynolds, S.E. (1988) The role of destruxins in the pathogenicity of 3 strains of Metarhizium anisopliae for the Tobacco Hornworm Manduca sexta. Mycopathologia 104, 51–58. [11] Camatini, M., Colombo, A., Bonfanti, P., Doldi, M., Urani, C., Dibisceglia, M. and Nagelkerke, J.F. (1996) In vitro biological systems as models to evaluate the toxicity of pesticides. Int. J. Environ. Anal. Chem. 65, 153–167. [12] Malich, G., Markovic, B. and Winder, C. (1997) The sensitivity and specificity of the MTS tetrazolium assay for detecting the in vitro cytotoxicity of 20 chemicals using human cell lines. Toxicology 124, 179–192.
28
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
[13] Visconti, A., Minervini, F., Lucivero, G. and Gambatesa, V. (1991) Cytotoxic and immunotoxic effects of Fusarium mycotoxins using a rapid colorimetric bioassay. Mycopathologia 113, 181– 186. [14] Calo, L., Fornelli, F., Ramires, R., Nenna, S., Tursi, A., Caiaffa, M.F. and Macchia, L. (2004) Cytotoxic effects of the mycotoxin beauvericin to human cell lines of myeloid origin. Pharmacol. Res. 49, 73–77. [15] Calo, L., Fornelli, F., Nenna, S., Marrone, D., Tursi, A., Caiaffa, M.F. and Macchia, L. (2003) Beauvericin cytotoxicity to the invertebrate cell line SF-9. J. Appl. Genet. 44, 515–520. [16] Fornelli, F., Minervini, F. and Logrieco, A. (2004) Cytotoxicity of fungal metabolites to lepidopteran (Spodoptera frugiperda) cell line (SF-9). J. Invertebr. Pathol. 85, 74–79. [17] Cai, P., Smith, D., Cunningham, B., Brown-Shimer, S., Katz, B., Pearce, C., Venables, D. and Houck, D. (1998) Epolones: novel sesquiterpene-tropolones from fungus OS-F69284 that induce erythropoietin in human cells. J. Nat. Prod. (Lloydia) 61, 791– 795. [18] Dumas, C., Ravallec, M., Matha, V. and Vey, A. (1996) Comparative study of the cytological aspects of the mode of action of destruxins and other peptidic fungal metabolites on target epithelial cells. J. Invertebr. Pathol. 67, 137–146. [19] Odier, F., Vey, A. and Bureau, J.P. (1992) In vitro effect of fungal cyclodepsipeptides on leukemic cells: study of destruxins A, B and E. Biol. Cell 74, 267–271. [20] Odier, F., Vago, P., Quiot, J.M., Devauchelle, G. and Bureau, J.P. (1987) Cytometric study of the effects of destruxin E on leukemic cells in mice. C.R. Acad. Sci. III 305, 575–578. [21] Morel, E., Pais, M., Turpin, M. and Guyot, M. (1983) Cytotoxicity of cyclodepsipeptides on murine lymphocytes and on L 1210 leukemia cells. Biomed. Pharmacother. 37, 184–185. [22] Quiot, J.M., Vey, A. and Vago, C. (1984) Effects of mycotoxins on invertebrate cells in vitro. In Vitro Cell Dev. B 20, 242. [23] Dumas, C., Robert, P.H., Pais, M., Vey, A. and Quiot, J.M. (1994) Insecticidal and cytotoxic effects of natural and hemisynthetic destruxins. Comp. Biochem. Physiol. C 108, 195–203.
[24] Vey, A. and Quiot, J.-M. (1989) Etude in vitro et chez linsecte hoˆte des destruxins toxines cyclodepsipeptidiques produites par le champignon entomopathoge`ne Metarhizium anisopliae. Can. J. Microbiol. 35, 1000–1008. [25] Quesada-Moraga, E. and Vey, A. (2003) Intra-specific variation in virulence and in vitro production of macromolecular toxins active against locust among Beauveria bassiana strains and effects of in vivo and in vitro passage on these factors. Biocontrol Sci. Technol. 13, 323–340. [26] Bandani, A.R., Khambay, B.P.S., Faull, J.L., Newton, R., Deadman, M. and Butt, T.M. (2000) Production of efrapeptins by Tolypocladium species and evaluation of their insecticidal and antimicrobial properties. Mycol. Res. 104, 537–544. [27] Genthner, F.J., Chancy, C.A., Couch, J.A., Foss, S.S., Middaugh, D.P., George, S.E., Warren, M.A. and Bantle, J.A. (1998) Toxicity and pathogenicity testing of the insect pest control fungus Metarhizium anisopliae. Arch. Environ. Contam. Toxicol. 35, 317–324. [28] Hino, M., Nakayama, O., Tsurumi, Y., Adachi, K., Shibata, T., Terano, H., Kohsaka, M., Aoki, H. and Imanaka, H. (1985) Studies of an immunomodulator, swainsonine. I. Enhancement of immune response by swainsonine in vitro. J. Antibiot. 38, 926– 935. [29] Patrick, M., Adlard, M.W. and Keshavarz, T. (1993) Production of an indolizidine alkaloid, swainsonine by the filamentous fungus Metarhizium anisopliae. Biotechnol. Lett. 15, 997–1000. [30] Roberts, D.W. (1981) Toxins of entomopathogenic fungi In: Microbial Control of Pests and Plant Diseases 1970/1980 (Burges, H.W., Ed.), pp. 441–464. Academic Press, New York. [31] Pedras, M.S.C., Zaharia, I.L. and Ward, D.E. (2002) The destruxins: synthesis, biosynthesis, biotransformation, and biological activity. Phytochemistry 59, 579–596. [32] Oredipe, O.A., Furbert-Harris, P.M., Laniyan, I., Green, W.R., White, S.L., Olden, K., Parish-Gause, D., Vaughn, T., Griffin, W.M. and Sridhar, R. (2003) Enhanced proliferation of functionally competent bone marrow cells in different strains of mice treated with swainsonine. Int. Immunopharmacol. 3, 445–455.
Toxicity testing of destruxins and crude extracts from the insect-pathogenic fungus Metarhizium anisopliae Anke Skrobek, Tariq M. Butt
*
School of Biological Sciences, University of Wales Swansea, Singleton Park, Swansea SA2 8PP, UK Received 21 April 2005; accepted 19 July 2005 First published online 8 August 2005 Edited by M.J. Bidochka
Abstract Increasing sensitivity towards secondary metabolites from fungal biological control agents (BCAs) has prompted the toxicological risk assessment of metabolites produced by the insect pathogenic fungus Metarhizium anisopliae. Viability studies on one human and one insect cell line were used to compare the two approaches of testing individual metabolites (destruxins A, B and E) or the complete crude extract from liquid cultures. Furthermore, crude extract was separated into fractions, which did not contain the main destruxins A, B and E. Evaluation of the cytotoxic activity of these different compounds suggested that a wide range of metabolites with synergistic or adverse effects are present in the crude extract. The results indicate that identification and toxicological assessment of each individual metabolite produced by a BCA is not only time and cost-intensive, but also does not convey the whole picture. Testing of the crude extract offers an alternative approach and is recommended when assessing the risks of metabolites for registration purposes. Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Metarhizium anisopliae; Destruxins; Fungal metabolites; Risk assessment; Toxicity; Cell lines
1. Introduction Increasing public sensitivity to environmental pollution and problems of pest resistance to chemical pesticides has provided the impetus for the development of alternate, benign strategies for pest control. Insect pathogenic fungi like Metarhizium anisopliae show considerable promise for use in integrated pest management (IPM) programmes [1]. However, consumer concerns regarding mycotoxins entering the food chain has prompted closer scrutiny of the secondary metabolites * Corresponding author. Tel.: +44 1792 295374; fax: +44 1792 295447. E-mail addresses: [email protected] (A. Skrobek), t.butt@swansea. ac.uk (T.M. Butt).
of all fungal biocontrol agents (BCAs). Regulatory authorities often require detailed information on ‘‘relevant metabolites’’. It is not clear what constitutes a relevant metabolite when most fungi secrete a disparate array of bioactive compounds [2]. Production of these metabolites may vary between species or strains and is also influenced by cultural conditions (e.g. [3–6]). Most often the metabolites are secreted in extremely small quantities even under optimal production systems [2]. Evaluation of the toxicological risks of each secondary metabolite could prove onerous and highly expensive partly because methods and tools have not been developed for the risk assessment of metabolites of fungal BCAs. Furthermore, no guidelines or simulation models exist to evaluate the fate of secreted fungal metabolites in the environment. Little is known about the full range
0378-1097/$22.00 Ó 2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsle.2005.07.029
24
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
the food chain and pose a risk to human and animal health.
of metabolites produced and whether they enter the food chain so posing a risk to human and animal health. Without doubt the high cost and time required for risk assessment of each metabolite will discourage many bio-manufacturers from developing fungal BCAs and will result in growers being trapped between the diminishing number of chemical pesticides and the lack of safe alternatives [7,8]. Clearly, sensible strategies that are simple and cost-effective are needed for the risk assessment of metabolites of fungal BCAs. One plausible approach is to evaluate the risk of crude extracts from culture filtrates that would contain the widest range of metabolites the fungus could ever produce in vitro. This approach has several highly commendable attributes; it is indifferent to variations in culture conditions and isolates and would take into account any additive or synergistic interactions between different compounds. Furthermore, it represents the ‘‘worse case scenario’’ since the metabolites are more concentrated than they would be in natural systems. The aim of this study was to determine the toxicity of crude extract preparations of M. anisopliae culture filtrates and compare these with purified destruxins (dtxs), the major metabolites secreted by this fungus both in vivo and in vitro (e.g. [9,10]). Toxicological (viability) assays were conducted using one human and one insect cell line. In vitro test systems with cell lines are increasingly gaining acceptance as tools to determine the toxicological effects of pesticides instead of whole animal testing [11,12]. Mammalian and insect cell lines have already shown considerable promise in the evaluation of the toxicological properties of some fungal metabolites [13–18]. This study forms part of a major programme to assess whether metabolites from fungal BCAs enter
2,000
2. Materials and methods 2.1. Test compounds Cells were exposed to the three main dtxs A, B and E and to their mixtures in order to study synergistic and adverse effects. Dtx A was obtained from Sigma (UK). Dtxs B and E were a kind gift from Dr. Alain Vey (INRA, France). As an alternative approach, cells were also treated with three different batches of crude extract from M. anisopliae V275 culture filtrate isolated in three independent experiments, which was obtained with organic solvents with a wide polarity range and analysed by HPLC as described by [5]. Additionally, the crude extract batches were separated into four fractions with decreasing polarity excluding dtxs with a Foxy Junior Fraction Collector (Dionex, UK) as indicated in Fig. 1. Cells were exposed to these individual fractions in order to eliminate the effect of dtxs in the crude extract. Reagents, media components and solvents were purchased from Sigma (UK) if not indicated otherwise. 2.2. Cell line assays The human leukemic cell line HL60 (Caucasian, promyelotic) was a kind gift from Dr. Luigi Macchia (Bari, Italy). Cells were maintained in RPMI1640 medium supplemented with 10% foetal calf serum and 1% glutamine–penicillin–streptomycin solution at 37 °C
WVL:215 nm
mAU
dtx e dtx a 1,500
1,000
Fraction 2 500
Fraction 1
dtx b Fraction 3
Fraction 4
min
-200 0.0
5.0
10.0
15.0
20.0
25.0
33.0
Fig. 1. Different HPLC fractions of crude extracts from culture filtrates of M. anisopliae V275 used for cytotoxicity testing.
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
25
and 5% CO2. The SF9 insect cell line (derived from Spodoptera frugiperda ovaries) was purchased from the European Culture Collection (ECACC) and maintained in TNMFH insect medium supplemented with 10% foetal calf serum and 1% glutamine–penicillin–streptomycin solution at 26 °C. Tests were carried out in 96 well flat bottom microtitre plates (Nunc, UK), which contained 105 cells/well. Tests consisted of 50 ll (for 4 h incubation) and 100 ll (for 24 h incubation) of diluted test compound in order to study acute toxic effects as well as effects after longer exposure. The solvent concentration (methanol:acetonitrile, 1:1, v/v) did not exceed 0.5%. Controls consisted of 50 ll phosphate buffered saline (PBS, 0.2 g/l KCl, 0.2 g/l KH2PO4, 8 g/l NaCl, 1.15 g/l Na2HPO4 Æ 7H2O in deionised water, pH 7) for 4 h incubation or 100 ll culture medium for 24 h incubation as well as of 0.5% pure solvent in PBS or culture medium. After incubation for 4 or 24 h, 50 ll calcein AM (0.5 lM in PBS, Molecular Probes, NL) were added and cytotoxicity was determined by reading the absorption in a BMG labsystems ‘‘Polar Galaxy’’ microtitre plate reader after 30 min incubation at optimum conditions for the respective cell line. The following filter combination was used: 485– 15 nm excitation and 538 nm emission. Cell viability was calculated by integrating absorbance values with a pre-determined calibration function. The percentage of living cells in comparison with the untreated control indicated toxic effects after 4 or after 24 h of exposure. For better comparison LC50 values (=lethal concentration that caused 50% mortality) were calculated.
licates per treatment, were analysed to determine means, normal distribution and equal variance (p 6 0.05). Data that were not distributed normally were analysed with the H-test (Kruskal–Wallis ANOVA on ranks, p 6 0.05). Significant differences (denoted by different letters) were then determined using the Dunns-test. Data with normal distribution were further analysed using the Tukey-test. LC50 values and their lower and upper fiducial limits (=confidence limits 95%) were calculated by PriProbit 1.63 (1996–2000 Masayuki Sakuma).
2.3. Statistical analysis
Three different batches of crude extract from M. anisopliae V275, extracted with an equal mixture of polar and non-polar solvents, were cytotoxic to both cell lines. The effects on HL60 cells with LC50 values of 221 and 193 ppm after 4 and 24 h, respectively, were stronger
The results were analysed statistically with the program SigmaStat 2.03 for Windows (SPSS Inc., US). Results of three independent assessments, with four rep-
3. Results 3.1. Effects of pure dtxs and their mixtures on cell viability Pure dtxs A, B and E exhibited no effect on the cell viability of human leukemic HL60 cells up to 500 ppm (Table 1). There was no decrease in the number of live cells either after 4 or after 24 h (data not shown). Likewise, mixtures of two or three dtxs did not show any effect on HL60 cells (Table 1). Pure dtxs or their mixtures had no acute toxic effect on SF9 cells up to 500 ppm but dtx A and its mixtures exhibited toxic activity after 24 h of exposure with LC50 values ranging between 5 and 12 ppm (Table 1). Dtxs B and E or their mixture had no effect on cell viability after 24 h incubation. 3.2. Effects of crude extract and its fractions on cell viability
Table 1 Toxicity of different polar HPLC fractions, purified destruxins (dtxs), combination of dtxs and crude extracts of M. anisopliae V275 to human (HL60) and insect (SF9) cell lines (LC50 values with lower and upper fiducial limits (95%) as calculated by PriProbit 1.63 (1996–2000 Masayuki Sakuma)) Test compound
LC50 (ppm) HL60
dtx A dtx B dtx E dtx A + dtx B dtx A + dtx E dtx B + dtx E dtx A + dtx B + dtx E Crude extract Fraction 1 Fraction 2 Fraction 3 Fraction 4
SF9
4 h incubation
24 h incubation
4 h incubation
24 h incubation
n.e. n.e. n.e. n.e. n.e. n.e. n.e. 221 (162; 331) 252 (225; 424) >MCT >MCT 80 (10; 1100)
n.e. n.e. n.e. n.e. n.e. n.e. n.e. 193 (120; 345) >MCT >MCT >MCT 370 (164; 453)
n.e. n.e. n.e. n.e. n.e. n.e. n.e. >MCT 320 (260; 410) 200 (n.c.) 97 (49; 210) 92 (45; 198)
5 (4; 16) n.e. n.e. 12 (8; 32) 9 (7; 33) n.e. 8 (6; 24) >MCT 120 (92; 170) 240 (160; 440) >MCT 110 (79; 150)
n.e., no effect up to 500 ppm; n.c., fiducial limits not reliable; MCT, max concentration tested.
26
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
than those on SF9 cells with LC50 values above the highest concentration tested (500 ppm, Table 1). HPLC analyses established that there was a high variation in the dtx content of the three crude extract batches: in 100 ppm of crude extract there were 18 ± 11 ppm dtx A, 8 ± 6 ppm dtx B and 30 ± 19 ppm dtx E. Crude extract fractions, which excluded dtxs, exhibited cytotoxic effects after 4 h incubation with HL60 and SF9 cells (Figs. 2 and 3, Table 1). While fractions 1 and 4 showed stronger effects on the human cell line, fractions 2 and 3 exhibited higher activity against the
insect cells. The least polar fraction 4 expressed the strongest activity with 97% insect cell mortality and 100% human cell mortality at 500 ppm (Figs. 2 and 3). Toxic effects after 24 h exposure could be observed for all fractions and for both cell lines (Table 1). Fraction 1, which was more active against HL60 after 4 h than after 24 h incubation, exhibited stronger toxic effects against SF9 after 24 h with LC50 values of 252 and 320 ppm for 4 h and >500 and 120 ppm for 24 h, respectively. Fraction 2 exhibited stronger effects on the insect cell line at 4 and 24 h incubation. Fraction 3 was more active against SF9 cells after 4 h (97 vs. >500 ppm). Fraction 4 had a lower LC50 value for HL60 cells than for SF9 cells after 4 h of incubation (80 vs. 92 ppm) but the LC50 value for SF9 cells was lower for 24 h of incubation with 110 vs. 370 ppm. Higher LC50 values for 24 h incubation compared to 4 h incubation indicated that HL60 cells and SF9 cells recovered from the acute toxic effects of fraction 1 and fraction 3, respectively. In contrast, fraction 1 was more toxic to the insect cells after 24 h incubation than after 4 h.
4. Discussion
Fig. 2. Effect of different fractions of crude extracts from M. anisopliae V275 on the viability of human leukemic cells (HL60) after 4 h incubation. (Values represent the means of three independent assessments with four replicates each, significant differences between controls and different concentrations of individual fractions are indicated by different letters, DUNNS, p 6 0.05.)
Fig. 3. Effect of different fractions of crude extract from M. anisopliae V275 on the viability of insect cells (SF9) after 4 h incubation. (Values represent the means of three independent assessments with four replicates each, significant differences between controls and different concentrations of individual fractions are indicated by different letters, TUKEY, p 6 0.05.)
Dtxs, cyclic peptides, are the main metabolites secreted by M. anisopliae [2,9]. They are known to have cytotoxic and cytostatic effects on mammalian and insect cells with dtx A and dtx E being the most toxic [19–24]. Odier et al. [20] reported acute cytotoxic and cytostatic activities of dtx E on mouse lymphocytes from 0.001 and 0.1 ppm on for cell lines P388 and L1210, respectively. In a later study, they showed a total cytotoxic effect of dtx E on blood and thymic lymphocytes with doses of 16 and 3.3 ppm after 48 h incubation as well as an inhibition of P388 proliferation at 0.33 ppm for dtx E, 16 ppm for dtx A and 3.33 ppm for dtx B [19]. This is the first study to compare the toxicity of crude extracts with purified dtxs A, B and E of M. anisopliae on human and insect cell lines. All three dtxs at 500 ppm had no effect on human leukemic cells up to 24 h incubation whereas dtx A caused mortality in SF9 insect cells, suggesting huge differences in the sensitivity of different cell lines. Furthermore, HL60 human leukemic cells were more sensitive than SF9 cells to the crude extract of M. anisopliae than to the pure dtxs. These observations suggest that several different cell lines should be chosen for toxicological assessments. Strains of M. anisopliae differ considerably in their metabolite profiles [3–5]. This is also true for other insect-pathogenic fungi including Beauveria bassiana and Tolypocladium sp. [25,26]. Evaluating the effect of crude extracts instead of individual compounds takes such differences into account since extracts of culture filtrates
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
contain the widest range of metabolites the fungus can produce in vitro. In addition, metabolites are far more concentrated than they would be in the environment [6]. For example, Amiri-Besheli et al. [3] showed that the dtx content of one Galleria mellonella larva infected with M. anisopliae V245 was 0.44 lg dtx A and 0.5 lg dtx B on the day of death. This means that 1000 g Galleria larvae (ca. 7 g bodyweight per larva) would contain 0.063 mg dtx A and 0.071 mg dtx B. In contrast, one litre (equals 1000 g) of liquid culture of the same strain contains maximally 12 mg dtx A, 4 mg dtx B and 17 mg dtx E [5]. Thus, when assessing the toxicity of BCA metabolites it is better to look at the crude extract than at individual compounds. Toxicological tests conducted using crude extracts and its fractions were variable compared with pure dtxs. This may be due to variations in the levels of dtxs, and possibly other metabolites, in the different batches of crude extract, since the same strain may not always produce the same quantity of metabolites even when grown under the same cultural conditions. Thus, crude extracts also take into account variations in metabolite profiles between fermentation batches. Crude extracts amplify the signal for the main metabolites but may also take into account interactions between compounds. Genthner et al. [27] reported that dtxs alone were not responsible for the toxicity of M. anisopliae crude extracts to mysids. Our studies also show that compounds other than dtxs are cytotoxic to HL60 and SF9 cells after 4 h incubation. Metarhizium secretes an array of bioactive metabolites including cytotoxic compounds like swainsonine and cytochalasins C and D (e.g. [28–32]), which potentially could account for the toxicity of crude extracts. Synergy and antagonism between compounds could also influence the overall toxicity of crude extracts. Dtx A, of all the compounds tested, was the most toxic to SF9 insect cells but in crude extracts it was less effective presumably due to antagonism from other compounds. For example, 5 ppm pure dtx A caused 50% mortality in SF9 cells while 500 ppm crude extract containing 90 ppm dtx A did not cause such a high mortality. When looking at the different fractions, some were weakly cytotoxic after a short incubation period but increased in toxicity over time. Some fractions were comparatively more toxic after only a brief exposure. High concentrations of the complete crude extract had to be used to cause 50% death of HL60 cells after 24 h incubation. Complete crude extract was less toxic to SF9 cells than the separate fractions. These findings suggest that the cytotoxic activity of crude extracts from BCAs cannot be attributed to one compound alone and that crude extracts represent the worse case scenario. Risk assessment of crude extracts requires less time and resources and subsequently would be more cost-effective for companies wishing to register fungal BCAs.
27
Acknowledgements The authors thank Dr. Vey (F) for kindly providing standards for dtxs B and E and Dr. Macchia (I) for the human leukemic cell line. This work was supported by the European Commission, Quality of Life and Management of Living Resources Programme (QoL), Key Action 1 on Food, Nutrition and Health QLK1-200101391 (RAFBCA).
References [1] Butt, T.M., Jackson, C.W. and Magan, N. (2001) Fungi as Biocontrol Agents: Potential, Progress and Problems. CAB International, Wallingford, 416 pp. [2] Vey, A., Hoagland, R. and Butt, T.M. (2001) Toxic metabolites of fungal biocontrol agents In: Fungi as Biocontrol Agents: Potential, Progress and Problems (Butt, T.M., Jackson, C.W. and Magan, N., Eds.), pp. 311–345. CAB International, Wallingford. [3] Amiri-Besheli, B., Khambay, B., Cameron, S., Deadman, M.L. and Butt, T.M. (2000) Inter- and intra-specific variation in destruxin production by insect pathogenic Metarhizium spp., and its significance to pathogenesis. Mycol. Res. 104, 447– 452. [4] Kershaw, M.J., Moorhouse, E.R., Bateman, R.P., Reynolds, S.E. and Charnley, A.K. (1999) The role of destruxins in the pathogenicity of Metarhizium anisopliae for three species of insect. J. Invertebr. Pathol. 74, 213–223. [5] Wang, C.S., Skrobek, A. and Butt, T.M. (2004) Investigations on the destruxin production of the entomopathogenic fungus Metarhizium anisopliae in liquid and solid media. J. Invertebr. Pathol. 85, 168–174. [6] Strasser, H., Vey, A. and Butt, T.M. (2000) Are there any risks in using entomopathogenic fungi for pest control, with particular reference to the bioactive metabolites of Metarhizium, Tolypocladium and Beauveria species? Biocontrol Sci. Technol. 10, 717–735. [7] Zimmermann, G., Lu¨th, P., Bode, E., Hommes, M., Selzsam, B., Guske, S. and Huber, J. (2004) Bericht u¨ber das dritte Fachgespra¨ch zum biologischen Pflanzenschutz ‘‘Kommerzialisierung mikrobiologischer Pflanzenschutzmittel’’. Nachrichtenbl. Deut. Pflanzenschutzd. 56, 131–136. [8] Blum, B., Ehlers, R.L., Haukeland-Salinas, S., Hokkanen, H., Jung, K., Kuhlmann, U., Menzler-Hokkanen, I., Ravensberg, W.J., Strasser, H., Warrior, P. and Wilson, M. (2003) Letter to the editors – Biological control agents: safety and regulatory policy. BioControl 48, 474–487. [9] Roberts, D.W. (1966) Toxins from the entomogenous fungus Metarhizium anisopliae. II. Symptoms and detection in moribund hosts. J. Invertebr. Pathol. 8, 222–227. [10] Samuels, R.I., Charnley, A.K. and Reynolds, S.E. (1988) The role of destruxins in the pathogenicity of 3 strains of Metarhizium anisopliae for the Tobacco Hornworm Manduca sexta. Mycopathologia 104, 51–58. [11] Camatini, M., Colombo, A., Bonfanti, P., Doldi, M., Urani, C., Dibisceglia, M. and Nagelkerke, J.F. (1996) In vitro biological systems as models to evaluate the toxicity of pesticides. Int. J. Environ. Anal. Chem. 65, 153–167. [12] Malich, G., Markovic, B. and Winder, C. (1997) The sensitivity and specificity of the MTS tetrazolium assay for detecting the in vitro cytotoxicity of 20 chemicals using human cell lines. Toxicology 124, 179–192.
28
A. Skrobek, T.M. Butt / FEMS Microbiology Letters 251 (2005) 23–28
[13] Visconti, A., Minervini, F., Lucivero, G. and Gambatesa, V. (1991) Cytotoxic and immunotoxic effects of Fusarium mycotoxins using a rapid colorimetric bioassay. Mycopathologia 113, 181– 186. [14] Calo, L., Fornelli, F., Ramires, R., Nenna, S., Tursi, A., Caiaffa, M.F. and Macchia, L. (2004) Cytotoxic effects of the mycotoxin beauvericin to human cell lines of myeloid origin. Pharmacol. Res. 49, 73–77. [15] Calo, L., Fornelli, F., Nenna, S., Marrone, D., Tursi, A., Caiaffa, M.F. and Macchia, L. (2003) Beauvericin cytotoxicity to the invertebrate cell line SF-9. J. Appl. Genet. 44, 515–520. [16] Fornelli, F., Minervini, F. and Logrieco, A. (2004) Cytotoxicity of fungal metabolites to lepidopteran (Spodoptera frugiperda) cell line (SF-9). J. Invertebr. Pathol. 85, 74–79. [17] Cai, P., Smith, D., Cunningham, B., Brown-Shimer, S., Katz, B., Pearce, C., Venables, D. and Houck, D. (1998) Epolones: novel sesquiterpene-tropolones from fungus OS-F69284 that induce erythropoietin in human cells. J. Nat. Prod. (Lloydia) 61, 791– 795. [18] Dumas, C., Ravallec, M., Matha, V. and Vey, A. (1996) Comparative study of the cytological aspects of the mode of action of destruxins and other peptidic fungal metabolites on target epithelial cells. J. Invertebr. Pathol. 67, 137–146. [19] Odier, F., Vey, A. and Bureau, J.P. (1992) In vitro effect of fungal cyclodepsipeptides on leukemic cells: study of destruxins A, B and E. Biol. Cell 74, 267–271. [20] Odier, F., Vago, P., Quiot, J.M., Devauchelle, G. and Bureau, J.P. (1987) Cytometric study of the effects of destruxin E on leukemic cells in mice. C.R. Acad. Sci. III 305, 575–578. [21] Morel, E., Pais, M., Turpin, M. and Guyot, M. (1983) Cytotoxicity of cyclodepsipeptides on murine lymphocytes and on L 1210 leukemia cells. Biomed. Pharmacother. 37, 184–185. [22] Quiot, J.M., Vey, A. and Vago, C. (1984) Effects of mycotoxins on invertebrate cells in vitro. In Vitro Cell Dev. B 20, 242. [23] Dumas, C., Robert, P.H., Pais, M., Vey, A. and Quiot, J.M. (1994) Insecticidal and cytotoxic effects of natural and hemisynthetic destruxins. Comp. Biochem. Physiol. C 108, 195–203.
[24] Vey, A. and Quiot, J.-M. (1989) Etude in vitro et chez linsecte hoˆte des destruxins toxines cyclodepsipeptidiques produites par le champignon entomopathoge`ne Metarhizium anisopliae. Can. J. Microbiol. 35, 1000–1008. [25] Quesada-Moraga, E. and Vey, A. (2003) Intra-specific variation in virulence and in vitro production of macromolecular toxins active against locust among Beauveria bassiana strains and effects of in vivo and in vitro passage on these factors. Biocontrol Sci. Technol. 13, 323–340. [26] Bandani, A.R., Khambay, B.P.S., Faull, J.L., Newton, R., Deadman, M. and Butt, T.M. (2000) Production of efrapeptins by Tolypocladium species and evaluation of their insecticidal and antimicrobial properties. Mycol. Res. 104, 537–544. [27] Genthner, F.J., Chancy, C.A., Couch, J.A., Foss, S.S., Middaugh, D.P., George, S.E., Warren, M.A. and Bantle, J.A. (1998) Toxicity and pathogenicity testing of the insect pest control fungus Metarhizium anisopliae. Arch. Environ. Contam. Toxicol. 35, 317–324. [28] Hino, M., Nakayama, O., Tsurumi, Y., Adachi, K., Shibata, T., Terano, H., Kohsaka, M., Aoki, H. and Imanaka, H. (1985) Studies of an immunomodulator, swainsonine. I. Enhancement of immune response by swainsonine in vitro. J. Antibiot. 38, 926– 935. [29] Patrick, M., Adlard, M.W. and Keshavarz, T. (1993) Production of an indolizidine alkaloid, swainsonine by the filamentous fungus Metarhizium anisopliae. Biotechnol. Lett. 15, 997–1000. [30] Roberts, D.W. (1981) Toxins of entomopathogenic fungi In: Microbial Control of Pests and Plant Diseases 1970/1980 (Burges, H.W., Ed.), pp. 441–464. Academic Press, New York. [31] Pedras, M.S.C., Zaharia, I.L. and Ward, D.E. (2002) The destruxins: synthesis, biosynthesis, biotransformation, and biological activity. Phytochemistry 59, 579–596. [32] Oredipe, O.A., Furbert-Harris, P.M., Laniyan, I., Green, W.R., White, S.L., Olden, K., Parish-Gause, D., Vaughn, T., Griffin, W.M. and Sridhar, R. (2003) Enhanced proliferation of functionally competent bone marrow cells in different strains of mice treated with swainsonine. Int. Immunopharmacol. 3, 445–455.