Non-target effects of Metarhizium brunneum (BIPESCO 5/F 52) in soil show that this fungus varies between being compatible with, or moderately harmful to, four predatory arthropods

Research output: Contribution to journalJournal articleResearchpeer-review


Biological control with entomopathogenic fungi is a feasible option for regulation of pest insect populations. However, possible effects on beneficial arthropods must be considered. We assessed the non-target effects of the microbial biological control agent Metarhizium brunneum (isolate BIPESCO 5/F 52) applied in soil on four different predatory arthropods: the predatory mite Gaeolaelaps aculeifer (Canestrini), the predatory bug Orius majusculus (Reuter), the rove beetle Dalotia coriaria (Kraatz) and the gall midge Aphidoletes aphidimyza Rondani. All are widespread and naturally occurring in Europe, they represent different classes of arthropods and different insect orders; furthermore, their life cycles involve different levels of contact with the soil. Adult G. aculeifer, O. majusculus, and D. coriaria, and last instar A. aphidimyza larvae were exposed to natural soil (control) or natural soil inoculated with M. brunneum at a concentration of 5 × 106 conidia/g of soil; this represents a worst-case scenario. Mortality, longevity, fecundity and Metarhizium outgrowth on dead individuals were assessed for the first three species; for A. aphidimyza, only mortality (non-emergence rate) and fecundity of emerged females were assessed. The fungal treatment resulted in a significantly higher mortality of O. majusculus and D. coriaria, 96%, and 7.3% respectively, compared with 19%, and 2% for their respective controls. Mortality of G. aculeifer was not significantly affected by exposure to the fungus in the soil. Longevity of O. majusculus and D. coriaria was significantly reduced following exposure to the fungus in the soil (log-rank test: p < 0.0001, Wilcoxon test p < 0.0001 and log-rank test: p = 0.029, Wilcoxon test: p = 0.027, respectively), while G. aculeifer longevity was not affected. Fecundity of O. majusculus and D. coriaria was negatively affected following exposure to the fungus in the soil, which reduced their oviposition by 20% and 4%, respectively, compared with the control, while G. aculeifer fecundity was not affected. Aphidoletes aphidimyza larval mortality was higher following exposure to the fungus in the soil (60% dead) than in the control (40% dead) but its fecundity was not statistically significantly affected by treatment. In conclusion, the predatory arthropods studied demonstrated a range of fitness responses to M. brunneum exposure in the soil, from no response (G. aculeifer), to intermediate (D. coriaria and A. aphidimyza) and high response (O. majusculus). This study demonstrates the relevance of using several fitness parameters and different arthropod species to determine whether a biological control agent should be considered a low-risk substance with respect to non-target effects.

Original languageEnglish
JournalBiological Control
Pages (from-to)18-24
Number of pages7
Publication statusPublished - 1 Apr 2019

    Research areas

  • Biological control, Entomopathogenic fungus, Gall midges, Predatory bugs, Predatory mites, Rove beetles

Number of downloads are based on statistics from Google Scholar and

No data available

ID: 213712148