Til projektoversigt

How zombie-flies can inform us about pathogen adaptation

Postdoc-stipendium i Danmark | 24/11/2016

Why do some pathogens infect a new host species virtually on first encounter, whereas other pathogens never manage to infect new host species despite repeated exposure?

Fungus mind-control

The fungus Entomophthora muscae (in Danish: “Flueskimmel”) is an insect-pathogen that causes disease and eventually kills its insect hosts. This fungus is highly host-specific, meaning that it naturally only infects very few host species, most often the common house fly Musca domestica (in Danish: “Almindelig stueflue”). If you have seen a dead fly attached to your window, surrounded by a white halo of spores – that’s the one.

Figure 1: Dead “zombie-fly” killed by the fungus Entomophthora muscae. The fungus can be seen as the white patches growing out between the cuticular segments of the abdomen.

When spores of this insect-pathogenic fungus land on a fly, they are able to germinate and penetrate the hard outer layer made up by the insect’s cuticle. Once inside the body cavity of the fly, E. muscae grows and gradually eats the fly alive. Towards the end of the infection, something unique happens. E. muscae takes over and control the behaviour of the infected fly – performing a type of mind-control. The still alive, but doomed, fly is forced to seek elevated positions, bite onto the substrate and spreads its wings away from the body in an unnatural position. The fly finally dies, and the fungus grows out from the abdomen and releases millions of infective spores that can infect new flies. By controlling when and where the fly dies, and by forcing the wings away from the abdomen, the fungus increases the chance of spores reaching new susceptible flies.

Figure 2: Close-up of the abdomen with white patches of fungi growing out between the cuticular segments, which are covered by hair.

The analogy to zombie-flies – or the living-dead – is thus straightforward, and I can’t help but feel some sympathy for what must be a horrible way to die for these flies. In my work, I study the evolution of pathogen host adaptation, and the objective of this Carlsberg Foundation funded project was to discover the underlying genetic components of pathogen host adaptation. The study of host adaptation is central for understanding what determines pathogen host ranges, which potentially has great implications for managing plant, animal and human health, and predicting disease emergence.

Why zombie-flies?

The fly-infecting pathogen E. muscae belongs to a special group of insect-pathogenic fungi in the family Entomophthoraceae.

Pathogen host adaptation

The choice of a host is similar to the problem of diet or habitat choice in free-living organisms, although the specifics are different. It is rare that organisms can specialize in one resource or host without lowering their efficiency on another. Therefore, pathogen specialization in one host often results in reduced infectivity on other hosts.

This family primarily contain species that are host-specific and obligate insect pathogens [1]. These fungi infect a range of insects, such as aphids, beetles, hemipterans, flies, cicadas etc. The specific fungus E. muscae is actually composed of several species in a so-called species complex, where individual species infect different host species, all within the order of flies (Diptera). Because of the high host-specificity, and because closely related pathogens have different host-ranges, this group of fungi is ideal for studying the genetic and evolutionary factors of host adaptation.

A comparative approach

A particular species in the E. muscae species complex infects cabbage flies (Delia radicum, in Danish: “Lille kålflue”), but looks exactly like the house fly infecting E. muscae. These two E. muscae fungi are very closely related, but will most likely, due to the non-overlapping habitats of house flies and cabbage flies, only rarely come into contact with each other.

Figure 3: Catching infected flies in a cow stable on Zealand.

Because of their close relatedness, the amount of random mutations that have accumulated in these two fungi since their most recent common ancestor is thus minimal. It is therefore reasonable to assume that most of the genetic differences between these two fly-infecting fungi is due to host-specific natural selection.

In a thorough comparison of the complete set of genes between these two fly-infecting E. muscae fungi, it was evident that there are genetic differences [2]. It was much trickier to make sense of these differences, because the physiological function of many of the variable genes is unknown. In some cases, however, it was clear that fungal genes involved in using lipids inside the hosts contained genetic markers of being specifically selected for. This indicates that differences in lipid molecules or lipid content between house flies and cabbage flies have resulted in specific adaptations of their respective fungal pathogens.

Among insect-pathogenic fungi, classical virulence factors, such as the production of toxins or proteases and other cuticle-degrading enzymes, have traditionally been thought to be important for host specificity. Instead, the obtained results add to a growing body of evidence, which indicates that utilization of nutritional resources may be a significant driver of pathogen variation and specialization. So, even though these fungi have evolved advanced manipulative strategies to increase transmission via insect mind-control, it is the basic need for nutrients and energy that appears to determine their host range.

To infect – or not to infect? Ecological factors determine the likelihood of pathogen encounters with new, potential hosts, for example by overlapping geographical distributions or shared seasonal presence. Physiological factors, in turn, determine whether the parasite can actually infect, grow within the host, and be transmitted to new hosts [3].


Knowing the range of hosts which a given pathogen can infect, and how it does so, is important for predicting disease emergence in new species. For example, many agricultural crops suffer from host-specific and highly adapted fungal diseases that significantly reduce the harvest. On a global scale, this threatens our capacity to feed the growing human population. Despite selective planting of resistant varieties, such fungal diseases occasionally shift to new host plants – being able to predict such events would be extremely useful. Similarly, when insect-pathogenic fungi are used in biological control of pest insects, unintended changes of target are of real concern, because beneficial insects, such as pollinators, should not be affected.

What it means to receive funding from the Carlsberg Foundation

“With the support from the Carlsberg Foundation, I was given the opportunity to return to Denmark after having spent two years as a postdoc in Sweden. I joined one of the world’s leading groups on insect-pathogenic fungi in the section for Organismal Biology at the Department of Plant and Environmental Sciences, University of Copenhagen. The grant allowed me to start an independent research program and initiate new collaborations in Denmark and internationally. The funding from the Carlsberg Foundation has therefore been immensely important for my career”, says Henrik H. de Fine Licht.


[1] De Fine Licht HH, Hajek AE, Eilenberg J, Jensen AB. Utilizing Genomics to Study Entomopathogenicity in the Fungal Phylum Entomophthoromycota: A Review of Current Genetic Resources. Adv Genet 2016;94:41–65. doi:10.1016/bs.adgen.2016.01.003.

[2] De Fine Licht HH, Jensen AB, Eilenberg J. Comparative transcriptomics reveal host-specific nucleotide variation in entomophthoralean fungi. Mol Ecol 2016. doi:10.1111/mec.13863.

[3] Schmid-Hempel P. Evolutionary parasitology: The integrated study of infections, immunology, ecology and genetics. Oxford: Oxdord University Press; 2011.