Monday, 10 November 2014

Diabolical Parasites: Ophiocordyceps variabilis and Relatives

Ophiocordyceps variables growing from larva in rotting log

My fungi friend Ulli and I were crashing through the bush on our way back to the car after a day of searching for oddities near Dorset, Ontario, when she pointed out these little beauties on a large rotting log. "Good eye," I said, since whatever they were, they were barely a centimetre tall. 

Having no reading glasses with me, nor time to inspect them with my loupe, and thinking the orange blurs were simply shrivelled jelly or coral fungi (it hadn't rained in a while), I roughly carved them out of the log and stuck them in a container in my basket.  
parasitic Ophiocordyceps variabilis with larval host
Ophiocordyceps variabilis with larval host (visible at right)
It wasn't until I had a chance to look at them more closely the next day that I realized they were something good, something really good—a Cordyceps of some kind, an ascomycete that parasitizes insects. Not only that, but by collecting them along with a chunk of the rotting log, which I confess I'd done for no other reason than to not lose them among other small finds of the day, I'd conveniently also gathered the larval host, a larva that, because of its bright orange-brown colour, I assumed was a youthful Coleoptera, or beetle.

Xylophagus fly larva, host of Ophiocrodyceps variabilis
Xylophagus fly larva mimicking a beetle larva—host of Ophiocrodyceps variabilis

Silly me. 

After a bit of micro work and some research, I got a name, Ophiocordyceps variabilis, a species that actually only attacks wood-inhabiting Diptera, or fly, larvae. The hosts of mine have been identified by Ziggy, an entomologist friend (Ulli's daughter, in fact!), as belonging to the genus Xylophagus—which is exactly the genus cited in a paper about O. variabilis Ziggy fished up for me. It helps to have friends in buggy places! 

Ophiocordyceps variabilis orange stromata and perithecial ostioles
The fertile portion of Ophiocordyceps variabilis stromata are
strongly textured by protruding perithecial ostioles.
Though O. variabilis, like many Ophiocordycipitaceae, is not well studied, it probably invades an unsuspecting host in a similar way to more studied species: after adhering to the exoskeleton of a larva, a spore germinates, grows a specialized structure that, in combination with pressure and the secretion of lipase and protease enzymes, allows it to penetrate the exoskeleton. Once past this barrier, the fungus grows and divides and, in some species at least, causes aberrant behaviour in the host before the host dies. Post death, the fungus consumes the innards, filling the intact exoskeleton with tightly packed, often peanut-shaped, hyphal bodies. During this process, the larval victim of O. variabilis changes colour, darkening from cream to burnt sienna, which makes it easy to mistake it for the larva of a beetle, so easy, in fact, that beetle larvae have regularly been erroneously named as the hosts of O. variabilis in the original description and elsewhere. 

Cordyceps variabilis and larva
A second, surprise, Ophiocordyceps variabilis I unknowingly collected
the same day as the ones pictured above that 
I found in a
container in the fridge 
a month later.
Eventually, when conditions are appropriate, O. variabilis sprouts its usually single fruiting body—a cylindrical yellowish orange stroma, ranging from 2-24 mm tall, with a terminus that is ornamented with variably shaped cushions holding orange-tipped perithecia. These perithecia contain characteristically thick-capped asci with multiseptate filiform ascospores that, once released, fragment into part-spores. 

And the life cycle goes on. 

Which is all fine and good, but I have a question. If the spores are dispersed by air current, as is assumed, how do they get inside the rotting logs where the eggs that produce these Diptera larvae are deposited? Ziggy has watched another wood-inhabiting Diptera, a cranefly, inserting its ovipositor deep into a rotting log and suggests it is perhaps via this apparatus that spores could be introduced into the wood. But she says, too, that since Xylophagidae larvae prey on other wood-inhabiting larvae, these other larvae could be carrying spores on their cuticles or inside their guts, thus infecting their assassins as a kind of post-mortem revenge.  

Ophiocordyceps variabilis asci and fragmented spores
The asci of Ophiocordyceps variabilis wear "hats" and their multiseptate
filiform spores fragment into "part-spores" when they're released.
Almost all Ophiocordyceps and their close relatives, Cordyceps, are entomopathogenic: they  parasitize insects, and usually kill them as well. Though there are a number of temperate-zone species, the greatest diversity is distributed throughout the tropics, where several hundred have been identified. Though most of these fungi are restricted to a single host species, or closely related ones, as a group they are known to victimize a wide range of arthropods from at least ten different orders, ranging from flies to Lepidoptera to grasshoppers to spiders. Some species specifically attack adults, while others infect larval, nymph, or pupal stages. 

Cordyceps locustiphila Susanne Sourell
Cordyceps locustiphila (Susanne Sourell, Mushroom Observer)
The morphology of their spore-producing structures is as varied as the host species they attack. Some form clusters of stromata, while others sprout one or two long, serpentine appendages with mace-like tips. Some of these erupt willy-nilly from their victims, while others always grow from a specific part of the anatomy, such as between the head and the thorax. There is a fabulous book out there, by Japanese mycologists D. Shimizu and K. Kobayasi, featuring exquisite illustrations, a book I'd love to get my hands on if I ever have a few hundred dollars to spare. You can also click here to see Daniel Winkler's spectacular photos of numerous species. 

illustration of Cordyceps discoideocapitata from Illustrated Vegetable Wasps and Plant Worms in Colour
Cordyceps discoideocapitata from the Japanese book,  冬虫夏草図鑑―カラー版,
(Illustrated Vegetable Wasps and Plant Worms in Colour) ISBN-10: 4-259-53866-7.
Cordyceps sp. from the same book.

Beyond the aesthetics of their fabulously sculptural stromata, Ophiocordyceps and Cordyceps species are attracting attention from researchers worldwide for the abundance of biologically active compounds they produce. O. sinensis, an Asian species endemic to the Tibetan plateau, which has been used in traditional Asian medicine since at least the 15th century, and C. militaris, which occurs in North America, both produce a biometabolite, Cordycepin, which has been shown to be a broad-spectrum antibiotic and polyadenylation inhibitor, and is currently being investigated for tumour suppression properties. Anamorphs of several other species are already being used as insect biological control agents, or pesticides. 

But there's another really interesting "control" aspect of at least some species of Ophiocordyceps and Cordyceps: they brainwash their victims, causing them to behave in ways that are advantageous for the pathogen. 

yellow Ophiocordyceps
(David Hughes)
In the case of Ophiocordyceps unilateralis and other closely related species that infect mostly tropical, tree-dwelling carpenter ants, the process works something like this: a spore that has adhered to the exoskeleton of an ant grows an appressorium that produces hypha that penetrate the ant's protective armour using a combination of mechanical pressure and enzymes. Once inside, the parasite grows as free-living yeast cells, eventually producing nerve toxins that alter the ant's behaviour. The ant staggers away from the colony, descending from the dry canopy to the humidity of the ground where, between convulsions, it ascends a small sapling at high noon. The ant clamps down on the main vein of a leaf with its mandibles at a height of about 25 cm. above the soil surface. The ant soon dies, but the life of the fungus continues. Hyphae emerge from orifices and joints, welding the ant to the leaf, while at the same time saprophytic mycelia run rampant through the corpse. Eventually, the fruiting stalk bursts forth from the rear of the head, its fertile surface knobbly with perithecia that produce the asci, which forcibly eject spores into the forest (watch BBC video here).     

zombie fungus Ophiocordyceps unilateralis
Ophiocordyceps unilateralis growing from a carpenter ant. (David Hughes)
David Hughes of Penn State University is at the forefront of research into these fungi and the chemical mechanisms behind their creation of zombie ants. Controlling a host's behaviour is an example of an extended phenotype, or, as Hughes writes: "While the manipulated individual may look like an ant, it represents a fungal genome expressing fungal behaviour through the body of an ant." 

Ophiocordyceps camponoti-balzan zombie ant fungus
Ophiocordyceps camponoti-balzani, a recently discovered species,
growing out of a "zombie" ant's head. (David Hughes)
Hughes and his colleagues have shown that an infected ant's jaw muscles atrophy, keeping the mandibles locked in place long after the ant's demise, a "death grip" that is clearly advantageous to the fungus since it keeps the host in the optimum location long enough for spores to be produced. With the help of gene-sequencing and metabolomics (the analysis of bioactive chemicals produced by a particular genome), Hughes and his team have now identified the likely chemical compound that causes the atrophy. Other colleagues, working at Penn State's Genomics Institute, have found molecules that play a key role in the "mind control" of the so-called zombie ants.  

These compounds and others that will can expect to be isolated from various Ophiocordyceps and Cordyceps species will likely have wide-ranging uses in agriculture, pharmacology, and other fields—yet another reason to preserve, and fund studies in, our planet's diverse ecosystems.

But back to my O. variabilis: I was curious about whether or not this fungus is known to manipulate the behaviour of its fly larva hosts and posed the question to Hughes. His reply was that, though there are anecdotal suggestions of other insects (flies, crickets, moths) being manipulated, i.e. being found in unusual locations, no one has looked at any of these species in detail. At least not yet. 

Resources & References:

Codyceps.US - an Electronic Monograph of Cordyceps and Related Fungi

Daniel Winkler's Cordyceps blog

David Pacchioli’s article about David Hughes & his work, “Getting to the bottom of the zombie ant phenomenon” on Penn State news

Hughes, D.P., Andersen, S.* Hywel-Jones, N.L. , Himaman, W., Bilen, J and J.J. Boomsma. Behavioral mechanisms and morphological symptoms of zombie ants dying from fungal infection BMC Ecology 2011, 11:13doi:10.1186/1472-6785-11-1

Hughes, D.P. Parasites and the Superogranism (2012). In Host Manipulation by Parasites Edited by David P. Hughes, Jacques Brodeur, and Frédéric Thomas 

Hodge, K.T.; Humber, R.A.; Wozniak, C.A. 1998. Cordyceps variabilis and the genus Syngliocladium. Mycologia. 90:743-753

Fantastic BBC video showing time-lapse of Cordyceps growing from ant's head!