Originally written for the 2011 McGraw Hill Yearbook of Science & Technology, republished with permission.
Leaf-cutting (attine) ants have an ancient mutualistic relationship with basidiomycete fungi, which they cultivate for food and defend from parasites with bacterially generated antibiotics (Fig. 1). The bacteria involved in these multipartite relationships are an ancient group called actinobacteria, notably Pseudonocardia species, which most attine ants shelter and feed in specialized structures on their cuticles called crypts.
Modern attine ant gardens (Fig. 2) are a complex microbial mix of fungi, yeasts, and bacteria—a living biomass dominated by the basidiomycete monoculture that attracts a diversity of transient and easily removed nonmutualistic “weed” fungi, including Trichoderma and Fusarium. However, the most dangerous and best-studied garden invader is the filamentous, ascomycetous fungi of the genus Escovopsis, which kill the food fungi, eat the dead remains, and leave the ants to starve. This is a catastrophe for ant communities, some of which house up to five million individuals in huge subterranean nests.
Escovopsis is a shape-changing fungus that appears to earn its entire living stealing food from leaf-cutter ants. Found only in ant gardens and associated refuse dumps, Escovopsis fulfill the postulates of pathogenicity elucidated by Robert Koch in the late 1800s, that is, when isolated from diseased gardens and reapplied to healthy gardens, they cause the same disease.
Scientists think Escovopsis is vectored into ant colonies by neighboring invertebrates because founder queens carefully select disease-free garden patches following their mating flight, and this particular fungus does not produce airborne spores. Regardless of how it arrives into the nests, Escovopsis seems to be an integral member of the attine ant-microbe community and probably evolved into an obligate pathogen during its long association with the ants and their cultivar.
Actinobacteria are a ubiquitous, diverse, and very successful group of bacteria occupying a large number of different niches and taking many different forms, although all are Gram-positive (having tough outer coats that absorb the blue dye in Gram staining), which indicates that they were among the earliest inhabitants of Earth.
Pseudonocardia, which resemble fungi in their filamentous form, are soil dwellers, and soil is a busy, competitive place—a perfect laboratory for the creation of small molecules, some of which have been adapted by humans, along with ants, for use as antibiotics. Indeed, four of the best-known human antibiotics —actinomycin, neomycin, streptomycin, and vancomycin— were derived from actinobacteria.
One of the Pseudonocardia-manufactured chemical weapons that attine ants use to protect their fungal gardens was described in 2009. Called “dentigerumycin” in honor of Apterostigma dentigerum, the ant that helped choreograph its evolution, this small molecule (molecular formula: C40H67N9O13) actively inhibits Escovopsis, but largely spares the ant’s garden fungi.
Dentigerumycin, a cyclic depsipeptide (a peptide with one or more ester bonds in addition to the amide bonds) containing unusual amino acids, also slows the growth of a multidrug-resistant strain of the human pathogen Candida albicans and may provide humans with an antifungal alternative to existing drugs.
Escovopsis, like Candida, can develop antibiotic resistance, but Pseudonocardia have millions of years of experience evolving new small molecules and usually outpace the fungus. In the event that this does not happen fast enough, the ants simply acquire new strains of actinobacteria, possibly from other symbiotic stock or the soil surrounding their nests.
Researchers, however, are not yet able to fully explain how this is done. But, it appears that leaf-cutting ants tamed free-living strains of Pseudonocardia many times over during the course of their long evolutionary history together and that the acquisition is very selective; actinobacteria are the only clearly established symbionts of fungusgrowing ants.
Understanding the chemistry underlying all that Pseudonocardia do for the ants is a major research issue and will ultimately yield important insights about how they and bacteria in general deal with the world. For example, biosynthetic pathways called “orphan pathways” are turned on in special and completely unknown circumstances and it is important to know how and why this happens. Moreover, it is unlikely that Pseudonocardia developed the small molecules that the ants use as antibiotics for the express purpose of warding off the ants’ garden parasites; they probably have more peaceful uses in the wild.
Bacteria tend to be gregarious, gathering in high-density populations, probably reflecting a safety in numbers strategy. Furthermore, they “chatter” continuously and their words are chemical. Thus, one important use for the actinobacterial small molecules is probably “quorum sensing,” a bacterial monitoring language that tells them how many and what type of microbes are gathered in their neighborhood.
Genome sequencing is needed to define the small molecule-generating potential of Pseudonocardia; analytical chemistry is needed to isolate and characterize the molecules themselves; and functional assays are necessary to determine what these molecules actually do. The sequencing began in earnest in 2009, with three ant genomes and 14 ant-associated fungal and bacterial genomes under active investigation. Finalizing the genome sequences and annotation will take years, but results are being made available to the scientific community as they progress.
Black yeast prey on Pseudonocardia
Adding to the complexity of leaf-cutting ant communities, actinobacteria were discovered to have their own specialized fungal predators—a black yeast called Phialophora, which grows on and around the crypts housing the ants’ antibiotic-producing bacteria.
When the black yeast was put together with actinobacteria in a petri dish, Phialophora ate the bacteria, thereby robbing the ants of an important anti-Escovopsis defense. However, parasitism in this multipartite system is not considered a detrimental thing; it helps to keep the cooperators honest.
For example, making antifungal agents is expensive, and selfish actinobacteria would be tempted to reproduce more and protect the ants less without the black yeast around. Thus, with the yeast, antibiotic production is also in the best interest of the threatened bacteria.
Moreover, the black yeast undoubtedly adds value to the relationship because, like the cultivar, Escovopsis, and actinomycetes, the yeast has probably been part of the leaf-cutting ants’ microbial balancing act since they first began farming about 50 million years ago.
Cordyceps infection of ants
Ants spend a great deal of time foraging about in the soil, and it should come as no surprise that they, as well as their fungal food gardens, are vulnerable to infection with pathogenic fungi. Principal among these are Cordyceps, which comprise at least 400 species and infect a wide range of insects and spiders in addition to ants.
Ant queens are at particular risk of parasitic infections during the colony-founding stage because of the many energetic demands that must be met simultaneously. In addition, although it appears that vigorous general defenses against food crop fungi indirectly protect the ants from a range of parasitic fungi, an ant-associated genus of Cordyceps, called Ophiocordyceps stilbelliformis, has been found parasitizing leaf-cutter ants in Panama.
Fortunately for farming ants, this is a rare event because Cordyceps are a particularly nasty group of fungi that keep the infected ants alive just long enough to distance themselves from the nest and then become spore-making factories.
Attine ant-microbe evolution
The ancestors of fungus-growing ants probably transitioned from hunter-gatherers of arthropod prey, nectar, and plant juices to the farming life fortuitously, by taming the wild fungi growing on the walls of their nests in leaf litter or from a system of myrmecochory (spore or seed dispersal by ants) where specialized fungi used the ants for their own dispersal.
Ants routinely ingest fungal spores and hyphal material and such infrabuccal contents are eventually expelled as pellets on nest middens (refuse dumps) and elsewhere, providing the fungi with a way of dispersing their spores and hyphae. Thus, the fungi were probably not passive symbionts that happened to come under ant control, but rather played a proactive role in the ant evolution from hunter-gatherer to fungus farmer.
Recent work using culture-independent genomic sequencing technologies is lending further weight to the long-standing scientific consensus that the ant-actinomycete association evolved for mutual benefit rather than as a coevolutionary arms race between antibiotic-producing Pseudonocardia and Escovopsis parasites. However, these new technologies must be used properly, statistically valid conclusions must be drawn, and wild as well as laboratory ant colonies must be tested in these endeavors.
Importantly, the new tools and new data will attract further investigations by the growing community of researchers fascinated by the attine tribe of sophisticated farming ants.
See also: Antibiotic; Bacteria; Chemical ecology; Ecological communities; Ecology; Food web; Fungal ecology; Fungi; Hymenoptera; Mutualism; Social insects; Soil ecology; Trophic ecology
l C. R. Currie et al., Coevolved crypts and exocrine glands support mutualistic bacteria in fungus-growing ants, Science, 311:81–83, 2006 DOI:10.1126/science.1119744
l C. R. Currie et al., Fungus-growing ants use antibiotic-producing bacteria to control garden parasites, Nature, 398:701–704, 1999 DOI:10.1038/19519
l O. Dong-Chan et al., Dentigerumycin: A bacterial mediator of an ant-fungus symbiosis, Nat. Chem. Biol., 5:391–393, 2009 DOI:10.1038/nchembio.159
l D. P. Hughes et al., Novel fungal disease in complex leaf-cutting ant societies, Ecol. Entomol., 34:214–220, 2009 DOI:10.1111/j.1365-2311.2008.01066.x
l G. Yim, H. Huimi Wang, and J. Davies, Antibiotics as signalling molecules, Philos. T. Roy. Soc. B., 362:1195–1200, 2007 DOI:10.1098/rstb.2007.2044
l J. J. Boomsma and D. K. Aanen, Rethinking crop-disease management in fungus-growing ants, Proc. Natl. Acad. Sci. USA, 106(42):17611–17612, 2009 DOI:10.1073/pnas.0910004106
l H. Fernández-Marín et al., Reduced biological control and enhanced chemical pest management in the evolution of fungus farming in ants, P. R. Soc. B., 276:2263–2269, 2009 DOI:10.1098/rspb.2009.0184
l M. Poulsen and C. R. Currie, Symbiont interactions in a tripartite mutualism: Exploring the presence and impact of antagonism between two fungus-growing ant mutualists, PLoS ONE, 5(1):1–13, 2010
l H. T. Reynolds and C. R. Currie, Pathogenicity of Escovopsis weberi: The parasite of the attine ant-microbe symbiosis directly consumes the ant-cultivated fungus, Mycologia, 96(5):955–959, 2004 DOI:10.2307/3762079
Decoding Science. One article at a time.