On conflict and resolution in the termite-fungus symbiosis

Life is organized in a hierarchical fashion; smaller replicating entities cooperate to make more complex organizational forms. From an evolutionary perspective there is a tension between lower-level selection and higher-level organization; natural selection at the lower level can oppose the higher-level organization, if reproductive interests between the two levels are not aligned. In this thesis I explored this tension and the stabilizing mechanisms that align the interests of different organizational levels at different levels of selection in the termite-fungus symbiosis.

The symbiosis between fungus-growing termites (Macrotermitinae) and Termitomyces fungi (Basidiomycota) evolved once, approximately 30 million years ago, without subsequent reversals to non-symbiotic states. The symbiosis has often been described as a farming system in which the termite farmers cultivate their domesticated fungus. Over time both the termites and their fungi have become mutually and obligately dependent on each other, even though in most cases the termites and fungi have retained independent reproduction and dispersal. Independent reproduction implies that the reproductive interests of the termites and their symbionts are not completely aligned, leaving room for conflict between the partners. Since the symbiosis has remained stable over evolutionary time, it is likely that there are mechanisms that have stabilized this level of organization.   

One of the major questions in the termite-fungus symbiosis is how sexual reproduction in the partners is correlated in time. Even though the termites and their fungal symbionts reproduce and disperse independently to establish new colonies, the fungal symbiont typically forms mushrooms a few weeks after the colony has produced reproductive termites. It has been hypothesized that this timing of mushroom formation is due to a trade-off between alate and worker production by the queen of the termite colony. Under the assumption of a maximal rate of termite reproduction, investment in the production of alates leads to a reduction in the production of workers. Because workers consume the fungus, reduced numbers of workers will allow mushrooms to ‘escape’ from the host colony. In chapter 2 we tested a specific version of this hypothesis, viz. that the typical asexual structures found in all species of Termitomyces – nodules – are immature stages of mushrooms that are normally harvested in a primordial stage, except when there are too few workers. We refuted this version of the hypothesis by showing that nodules and mushrooms are completely different structures from the earliest developmental stages that we could sample.

Due to the independent reproduction and dispersal of the termites and their fungi, the interaction between host and symbiont needs to be re-established at the start of each termite colony. It is known that there is a certain interaction specificity between termites and Termitomyces fungi, but it is unknown what factors contribute to the observed combinations of termite and fungus. It has been hypothesized that substrate provisioning by termite farmers could explain the observed interaction specificity. In chapter 3 we explored whether differences in nutrient requirement between fungi from different termite species can be found.

In the termite-fungus symbiosis, horizontal symbiont transmission is also associated with sexual reproduction of the fungus. The dispersing fungal spores are sexual spores produced in the mushrooms. It has been shown that for inhabitant symbionts, like Termitomyces, those that undergo little genetic change should be selected as they live in a stable biotic environment to which they have become adapted. Following from this observation, there should have been selection for a low recombination rate in Termitomyces fungi. In chapter 4 we constructed a new, more contiguous reference assembly of the Termitomyces symbiont of M. natalensis that allows for the study of recombinational landscapes. Also, we isolated a full-sibling mapping population of this Termitomyces species and used these to create the first linkage map of a Termitomyces fungus using a Genotyping-by-Sequencing approach. Finally, we performed an initial study into the recombination landscape of this Termitomyces species and showed that its recombination rate varies substantially across the genome. To be able to answer whether Termitomyces fungi indeed have evolved a low recombination rate, the recombination landscapes of more Termitomyces species as well as those of its close, free-living relatives should be studied.

In chapter 5 we zoomed in on the basidiomycete life cycle and explored how the peculiarities of basidiomycete life cycle open possibilities for lower-level selection that conflicts with the higher-level organization (the fungal mycelium). The first difference between basidiomycetes and the vast majority of sexual life cycles is that after gamete fusion, the nuclei remain separate for almost the whole life cycle. The second difference is that the nuclei of two fusing gametes can move through the whole body – the whole mycelium - of their mating partner. We show that by remaining separate, the fates of these two separate nuclei are not fully aligned, which means that selection can act on the individual nuclei at the cost of the dikaryon. Also, we show that these life cycle peculiarities could enhance the conflict of interest between nuclei and mitochondria, possibly leading to reduced fitness of the dikaryon.

Although the stability of the termite-fungus symbiosis has attracted the interest of many evolutionary biologists, the interest in the termite-fungus symbiosis is not for fundamental questions only. All mushrooms of Termitomyces fungi are edible and considered delicacies in the areas where they are found. The work described in this thesis concerning Termitomyces will also aid the search for Termitomyces mushroom cultivation methods. The work in chapter 2 brings us closer to find the factors that promote mushroom formation. The work in chapter 3 will aid the optimization of Termitomyces growth substrate. Finally, the work in chapter 4 could in the future help for breeding and analysing desirable traits for the cultivation of Termitomyces mushrooms, so that in future we will be able to re-domesticate the fungus that was domesticated by termites 30 million years ago.