Cordycepin, a metabolite of Cordyceps militaris, reduces immune-related gene expression in insects

Hypocrealean entomopathogenic fungi (EPF) (Sordariomycetes, Ascomycota) are natural regulators of insect populations in terrestrial environments. Their obligately-killing life-cycle means that there is likely to be strong selection pressure for traits that allow them to evade the effects of the host immune system. In this study, we quantified the effects of cordycepin (3′-deoxyadenosine), a secondary metabolite produced by Cordyceps militaris (Hypocreales, Cordycipitaceae), on insect susceptibility to EPF infection and on insect immune gene expression. Application of the immune stimulant curdlan (20 µg ml(−1), linear beta-1,3-glucan, a constituent of fungal cell walls) to Drosophila melanogaster S2r+ cells resulted in a significant increase in the expression of the immune effector gene metchnikowin compared to a DMSO-only control, but there was no significant increase when curdlan was co-applied with 25 µg ml(−1) cordycepin dissolved in DMSO. Injection of cordycepin into larvae of Galleria mellonella (Lepidoptera: Pyralidae) resulted in dose-dependent mortality (LC(50) of cordycepin = 2.1 mg per insect 6 days after treatment). Incubating conidia of C. militaris and Beauveria bassiana (Hypocreales, Cordycipitaceae; an EPF that does not synthesize cordycepin) with 3.0 mg ml(−1) cordycepin had no effect on the numbers of conidia germinating in vitro. Co-injection of G. mellonella with a low concentration of cordycepin (3.0 mg ml(−1)) plus 10 or 100 conidia per insect of C. militaris or B. bassiana caused a significant decrease in insect median survival time compared to injection with the EPF on their own. Analysis of predicted vs. observed mortalities indicated a synergistic interaction between cordycepin and the EPF. The injection of C. militaris and B. bassiana into G. mellonella resulted in increased expression of the insect immune effector genes lysozyme, IMPI and gallerimycin at 72 h post injection, but this did not occur when the EPF were co-injected with 3.0 mg ml(−1) cordycepin. In addition, we observed increased expression of IMPI and lysozyme at 48 h after injection with C. militaris, B. bassiana and sham injection (indicating a wounding response), but this was also prevented by application of cordycepin. These results suggest that cordycepin has potential to act as a suppressor of the immune response during fungal infection of insect hosts.