Host-pathogen infections and possible effects on co-evolutionary patterns depend on the genotypes of both host and pathogen. Obligate fungal pathogens of plants are often characterized by host-pathogen genotype-by-genotype (GxG) interactions, but whether these patterns exist in obligate insect fungal pathogens is unclear. We take advantage of the obligate insect pathogenic fungus Entomophthora muscae, where individual isolates are specific to different dipteran host species in nature but can cross-infect multiple fly species in the laboratory. We collected three new isolates of E. muscae from Drosophila species. Phylogenetic analysis showed that Drosophila-isolated E. muscae represents a distinct geographically widespread Drosophila lineage compared to the house fly (Musca domestica) or Delia species-isolated E. muscae. We used the three new E. muscae isolates from Drosophila spp. together with a genetically distinct E. muscae isolate from house flies and assessed their virulence in a cross-infection experiment using one house fly, three Drosophila suzukii, and two D. melanogaster genotypes as hosts. All fungal isolates successfully infected hosts, induced behavioural manipulation, sporulated in all fly hosts, and differed in virulence between host genotypes, revealing GxG interactions. While house flies were most susceptible to fungal infection with 99% mortality, we found a lower virulence of 49% and 25% mortality in D. melanogaster and D. suzukii genotypes, respectively. Furthermore, all isolates harboured a specific mycovirus (family Iflaviridae), but co-phylogenetic branching patterns did not support fungus-virus co-speciation. We show that the genetic makeup of both fungal pathogen and fly host influence E. muscae infectivity, confirming GxG interactions in obligate fly fungal pathogens.

Many entomopathogenic fungi cause infections that kill their insect host. Little is understood about changes in the reproductive investment that occurs during an infection by a lethal disease over the waning life of an insect. Life-history theory suggests the host will respond by investing resources into fighting the disease or increasing reproduction. Here, we investigate how the reproductive life of adult house flies, Musca domestica, is impacted by its host-specific fungal pathogen, Entomophthora muscae. Specifically, we test how the week-long infection alters the mating behavior of virgin adult male house flies. We find that the pathogen significantly decreases male libido, an effect which grows stronger over the course of the infection. Furthermore, females were significantly less likely to choose an infected male, reducing male mating success. Additionally, we assessed sperm viability to understand the reproductive costs for monandrous females to mate with infected males. Analyses revealed that sperm quality decreases as early as 3 days post-infection. These results show that E. muscae, which can have a prevalence near 100% in wild populations, causes severe lifetime reproductive costs to male house flies. Understanding how host-pathogen interactions affect host life history is crucial for elucidating all the negative effects pathogen virulence exerts on hosts. In nature, some diseased animals face a decision: to mate or to get better. The house fly has a fungal disease that enters its body, grows over a week, and just before death turns it into a "zombie," before exiting through its soft body tissue. This lethal fungus makes the males lose their desire to mate and "kills" their sperm, and makes healthy females refuse sex with them, leaving little choice for infected male flies.

Host manipulation by pathogens and parasites is a widespread phenomenon, but the molecular mechanisms are poorly understood. We investigated the summiting disease caused by the fungus Entomophthora muscae in houseflies, where infected flies climb to elevated positions and die, releasing infectious conidia. We performed dual-RNA sequencing of fly heads at different time points and identified candidate genes from both the host and the pathogen that may be involved in this summiting phenotype. Surprisingly, we also detected an extremely high abundance of a novel positive sense single-stranded (+ss) RNA Iflavirus in infected fly heads. We show that the virus load increases over time and shows signs of accumulation in fly heads and thoraces. We also reveal predicted interactions between fungal secreted proteins and insect host proteins related to neurological and immune functions, suggesting a possible role of these proteins in host manipulation. Furthermore, we find that E. muscae encodes a homologue of ecdysteroid UDP-glucosyltransferase (egt), a gene that has been implicated in host manipulation by other pathogens. Our study reveals a complex interplay between a fungus, a virus and a fly, and suggests that convergent evolution of egt for host manipulation mechanisms may occur in different pathogens. ### Competing Interest Statement The authors have declared no competing interest.

Insect pathogenic fungi (IPF) and insects have ubiquitous interactions in nature. The extent of these interkingdom host-pathogen interactions are both complex and diverse. Some IPF, notably of the order Entomophthorales, manipulate their species-specific host before death. The fungus-induced altered insect behaviours are sequential and can accurately be repeatedly characterised temporally, making them a valuable model for understanding the molecular and chemical underpinnings of behaviour and host-pathogen co-evolutionary biology. Here, we present methods for the isolation and laboratory culturing of the emerging behaviourally manipulating model IPF Entomophthora muscae for experimentation.•E. muscae isolation and culturing in vitro.•Establishing and maintaining an E. muscae culture in vivo in houseflies (Musca domestica).•Controlled E. muscae infections for virulence experiments and quantification of conidia discharge per cadaver.

Some insect-pathogenic fungi have evolved the ability to behaviorally manipulate their insect hosts. This has required the fungi to develop intricate mechanisms of infection, proliferation, and behavioral hijacking, which has led to speculation that behaviorally manipulating fungi must only infect a narrow range of hosts. One well-known example is the insect-pathogenic fungus Entomophthora muscae, which infects dipterans. Here, we present the different stages of the life cycle of E. muscae, focusing on the unique adaptations that allows the fungus to enter and proliferate inside its hosts, the possible ways it manipulates behavior, how the fungus exits the killed host to seek new susceptible hosts, and the ecological implications of these adaptations for determining the host range and intra-specific variation of E. muscae. We address the biology of E. muscae from an evolutionary ecology perspective and discuss the capacity of the fungus for behavioral manipulation within an extended phenotype framework. We highlight areas where further research is needed to fully develop E. muscae as a model system for host-pathogen research, for example to address questions relating to fitness consequences of an infection.

Conventional monitoring methods for disease vectors, pollinators or agricultural pests require time-consuming trapping and identification of individual insects. Automated optical sensors that detect backscattered near-infrared modulations created by flying insects are increasingly used to identify and count live insects, but do not inform about the health status of individual insects. Here we show that deep learning in trained convolutional neural networks in conjunction with sensors is a promising emerging method to detect infected insects. Health status was correctly determined in 85.6% of cases as early as two days post infection with a fungal pathogen. The ability to monitor insect health in real-time potentially has wide-reaching implications for preserving pollinator biodiversity and the rapid assessment of disease carrying individuals in vector populations. One sentence summary Automated optical sensors distinguish between fungus-infected and healthy insects.

Termites are widely used as a food resource, particularly in Africa and Asia. Markets for insects as food are also expanding worldwide. To inform the development of insect-based foods, we analysed selected minerals (Fe-Mn-Zn-Cu-Mg) in wild-harvested and commercially available termites. Mineral values were compared to selected commercially available insects. Alate termites, of the genera Macrotermes and Odontotermes , showed remarkably high manganese (Mn) content (292–515 mg/100 gdw), roughly 50–100 times the concentrations detected in other insects. Other mineral elements occur at moderate concentrations in all insects examined. On further examination, the Mn is located primarily in the abdomens of the Macrotermes subhyalinus ; with scanning electron microscopy revealing small spherical structures highly enriched for Mn. We identify the fungus comb, of Macrotermes subhyanus , as a potential biological source of the high Mn concentrations. Consuming even small quantities of termite alates could exceed current upper recommended intakes for Mn in both adults and children. Given the widespread use of termites as food, a better understanding the sources, distribution and bio-availability of these high Mn concentrations in termite alates is needed.