The Evolutionary Story of Cordyceps and its Allies

© Thomas S. Jenkinson
Original publication: Mycena News, January 2010

For many field mycologists, one of the most thrilling mushroom finds in the wild must be the Cordyceps mushroom. The first time I observed one of these fungi in nature was not too long ago. I fondly remember the discovery, foraging with several mentors in the sweltering, deep-summer woods outside of Baton Rouge, Louisiana. I had seen plenty of textbook photos of these amazing mushrooms, pored over their life cycles, and read their descriptions. None of those experiences however, could prepare me for the thrill of seeing this organism in hand. Finally, there it was: Cordyceps militaris, a single bright orange, club-shaped mushroom, erupting from the head of a mummified insect pupa. Since that day, my fascination with these fungi has only been bolstered by the newest scientific research on the evolution of these organisms.


The insect-attacking Cordyceps species exhibit a fascinating life cycle exemplified by the type species, C. militaris. To complete the sexual phase of its life cycle, this fungus has evolved to invade and take over the body of an insect host. Spores of the fungus adhere to the host and germinate, producing enzymes that allow it to breach the insect’s exoskeleton. Once inside the host, the fungus initially grows in a yeast-like stage, circulating toxins that eventually result in the death of the host. Fungal hyphae then feed on the unfortunate insect, growing throughout and consuming all internal organs. In the end, the internal tissue of the insect is replaced with a mass of mycelium, and all that remains of the host is a ghostly exoskeleton supporting the stroma (mushroom). The mushroom is often found bursting from the head of its victim’s remains.

Increasingly, popular attention has turned to this genus of mushrooms in part due to the recent BBC television production, Planet Earth. An episode in the series featured jaw-dropping time-lapse footage of a Cordyceps stroma fruiting from an infected ant. This footage has done wonders to pique the attention of the general public. In addition to the Planet Earth series, several recently published popular and scientific articles have cultivated a collective fascination in the genus.

Without exception, the most attention has been paid to the economically significant species Ophiocordyceps sinensis (= Cordyceps sinensis). Endemic to the Tibetan Plateau region, the “Chinese Caterpillar Fungus” fruits from the corpse of a parasitized ghost moth larva. There its harvest and sale serves as the main source of income for many rural Tibetans (Winkler 2008). The harvested caterpillar fungus is sold mainly in costal Chinese cities where it is referred to in Mandarin as “winter worm, summer grass.” In China, the fungus and its associated host caterpillar are sold at exorbitant prices as an herbal panacea, most notably as an aphrodisiac. The harvest of this cash “crop” has alreadycaused significant distortion to local economies, and has prompted local governments to take measures to limit the environmental impacts of harvesting and ensure a sustainable resource for future generations (Cannon et al. 2009).

I would like to emphasize that the intrigue and biodiversity of the genus Cordyceps does not end with these few e t h n omy c o l o g i c a l l y significant and welldocumented species. In fact, new species of Cordyceps continue to be described from sampling locations worldwide. This increase in our knowledge of Cordyceps diversity is coupled with exciting new conclusions we are able to draw about the evolutionary history of these unique fungi. Modern techniques to extract and sequence DNA, and the computing power necessary to analyze sequence data, have finally put previously unresolved hypotheses about evolution and speciation in a testable framework.

First, I must clarify that not all species of Cordyceps parasitize animals. There are also members of this genus that specialize in attacking underground truffles of the genus Elaphomyces. All species of Cordyceps do, however, belong to the Ascomycete order Hypocreales, (I’ll return to talk more about these relatives in a moment). One of the most fascinating conclusions we have been able to draw from DNA sequence evidence is that all members of the genus Cordyceps, in the old sense, were not each other’s closest relatives, as previously thought. The old genus Cordyceps was an assemblage of distantly related species that had evolved the similar characteristic of insect parasitism. The distantly related Cordyceps lineages have now been formally transferred to new genera (Ophiocordyceps, Metacordyceps, etc.). With apologies to the taxonomists, in this article I will continue to use the generic name Cordyceps in the old sense for the convenience of the reader. If you are curious about the updated taxonomy please refer to Sung et al. (2007).

Perhaps at this point you are wondering if the truffle-attacking Cordyceps belong in their own evolutionary lineage, while all of the insect-attacking species share a common ancestor. Actually, DNA sequence data reveals that this assumption is not so. The truffleattacking Cordyceps are more closely related to the medicinal caterpillar fungus, C. sinensis, than either lineage is to the insect-attacking type species of Cordyceps, C. militaris. These findings are very interesting from an evolutionary viewpoint, because they indicate a dramatic shift in the host organism of these Cordyceps. At some point in their evolutionary history, an ancestral cicada-pathogenic Cordyceps species is thought to have made an underground host-jump to the mycorrhizal truffle then underwent a radiation of species (Nikoh & Fukatsu 2000).

Host specificity of parasitic organisms is a likely stable outcome of evolution, but, sometimes, parasitic species adapt to attack a novel host, thereby opening a new ecological niche to occupy, leading to the potential evolution of new species. Biologically, the host jump to attacking truffles is quite significant. It has been shown that closely related Cordyceps species have made dramatic host shifts between arthropod orders. These host shifts between arthropod orders pale in comparison to the jump from an animal host to a fungal host, a jump between separate biological kingdoms.

So if these morphologically and ecologically similar insect pathogen species are not each other’s closest relatives, then who are? The answers may be revealed by looking at a wider sampling of the Hypocreales, and in particular, the family Clavicipitaceae. A recent study provides genetic sequence evidence showing that the closest relatives to the lineage containing the truffle-attacking Cordyceps and C. sinensis is a lineage of fungi composed in part by plant symbionts (Spatafora et al. 2007). Ethnomycologically important species are present in these Cordyceps relatives as well, for example the causative fungus of rye ergot, Claviceps purpurea. The published evidence shows that the lineage containing the medicinal caterpillar Cordyceps is more closely related to ergot of rye and other grass symbionts than either lineage is to the type species of the genus Cordyceps. These fascinating developments show that the evolution of distantly related Cordyceps lineages, and their relatives in the Clavicipitaceae has been characterized by several shifts in nutritional mode from animalbased to plant and fungal-based nutrition. We can make a confident conclusion that the evolutionary ancestor of these Clavicipitaceae, including rye ergot, was a Cordyceps- like pathogen of animals.

These latest evolutionary conclusions about the flexibility of host associations within Cordyceps and their relatives also raise further questions about what makes this family of organisms so adept at breaking down host defenses. Members of this family are obviously well equipped with a chemical arsenal of secondary metabolites with which to alter host organisms. We have observed that an array of natural chemicals from both Cordyceps and others of the family Clavicipitaceae have the ability to affect human physiology and perception. I begin to wonder what additional compounds exist in Cordyceps and their relatives that remain to be discovered?

New questions like those above are a fundamental reason that research on the diversity and evolution of these species is vital to humans and our sustainability on Earth. Only when we know what we have can we make better decisions on how to protect it. I do suspect that the average Tibetan or Bhutanese caterpillar fungus harvester might not care too much that his/her harvest has been formally transferred to a new genus, better reflective of its evolutionary history. I’m sure, however, that all of us could find common ground in the imperative that populations of these natural wonders (and their allies) must be preserved, so that they may be passed on to next generations as intact as we found them. Until then, I will continue to be on the lookout for those wild Cordyceps stroma that tell the tales of their unlucky host below.

Literature Cited:

Thomas Jenkinson is an M.S. candidate studying systematic mycology with the Desjardin lab at San Francisco State University. He has assisted in teaching introductory biology laboratory and spring fungi courses at San Francsico State. Thomas has worked as a collaborator in scientific field surveys of fungal biodiversity, and has published contributions to the Assembling the Fungal Tree of Life project.