Microbes engage in diverse interactions to which social evolution theory developed for animals can be usefully applied. In turn, studies on microbes offer insight into social evolution theory as it applies to larger organisms. Here we review key evolutionary concepts as applied to microbial interactions, then describe some prominent examples of how microbes interact to obtain resources, communicate, move, attack and defend themselves from competitors, prey, or predators, and influence multicellular hosts.

The Dictyostelia present a splendid opportunity for studies of sociality and genetic conflicts of interest. These eukaryotic amoebae aggregate upon starvation to form a multicellular individual in which some formerly independent individuals die to form a stalk that lifts the others to a better place for sporulation and dispersal. Dictyostelia vary in their social organization and can be cultured from soil samples in nearly all parts of the world. Genome sequences are available for increasing numbers of species; many molecular pathways are known; experimental evolution is feasible. Who lives, who dies, and how sociality is structured are great questions that are easily addressed in this group.

Predation is likely to influence the function of bacterial communities and the evolution of bacterial pathogens, because characteristics that permit escape from predators often overlap with traits used for biocontrol of plant pathogens, virulence, or even bioremediation. Soil bacteria are preyed upon by a variety of microorganisms, including the amoeba Dictyostelium discoideum , which has led some strains to evolve resistance. We identified genes required for three Pseudomonas species associated with D. discoideum to evade predation by screening more than 6,000 transposon mutants for loss of resistance. One species required a variety of genes including toxins and secondary metabolism genes, but the other two appear to have functionally redundant mechanisms of resistance, since disruption of genes with pleiotropic effects was required to render them susceptible. We determined that GacA, which positively regulates secondary metabolism, is required for resistance in all three species. Predation resistance also appears to be a social trait based on enrichment of cooperative genes in one species and rescue of mutants by wild type in another. Many genes required for resistance are conserved among both resistant and susceptible species, but several are found in few genomes and some of these have homologs in distantly related species. Gain and loss of resistance appears to be a dynamic process in which regulatory and structural genes are well conserved across species, the specific toxins they regulate may be lost in the absence of predators, and new toxins may be acquired through horizontal gene transfer. ### Competing Interest Statement The authors have declared no competing interest.

Generalist predators often live in environments that vary in the type and abundance of prey species. The prey species in turn vary in their susceptibility and suitability to predation. How do generalists navigate this variation in prey abundance and profitability and seek out their preferred prey. We investigated this in the soil protist Dictyostelium discoideum, a generalist predator of many species of bacteria. Despite their generalist diet, amoebas experience considerable variation in growth rate on their prey bacteria. We tested if amoebas innately prefer the more profitable prey bacteria. We found that the levels of chemoattraction of naive amoebas towards prey bacteria increased with increasing growth rate of the amoebas on those bacteria. This suggests that D. discoideum amoebas have an innate prey preference that is adaptive. Next, we tested how experience with prey bacteria affects chemoattraction in amoebas. Given the number of bacterial species in soil, learning from experience should be advantageous. However, we found no evidence for learning; instead amoeba preference in learning experiments was partially explained by innate attractiveness of the prey bacteria. Our results suggest that generalist amoebas are innately attracted to the more profitable prey bacteria and this innate attraction cannot be overridden by recent experience. ### Competing Interest Statement The authors have declared no competing interest.

The evolution of symbiotic interactions may be affected by unpredictable conditions. However, a link between prevalence of these conditions and symbiosis has not been widely demonstrated. We test for these associations using Dictyostelium discoideum social amoebae and their bacterial endosymbionts. D. discoideum commonly hosts endosymbiotic bacteria from three taxa: Paraburkholderia, Amoebophilus and Chlamydiae. Three species of facultative Paraburkholderia endosymbionts are the best studied and give hosts the ability to carry prey bacteria through the dispersal stage to new environments. Amoebophilus and Chlamydiae are obligate endosymbiont lineages with no measurable impact on host fitness. We tested whether the frequency of both single infections and coinfections of these symbionts were associated with the unpredictability of their soil environments by using symbiont presence-absence data from D. discoideum isolates from 21 locations across the eastern United States. We found that symbiosis across all infection types, symbiosis with Amoebophilus and Chlamydiae obligate endosymbionts, and symbiosis involving coinfections were not associated with any of our measures. However, unpredictable precipitation was associated with symbiosis in two species of Paraburkholderia, suggesting a link between unpredictable conditions and symbiosis.

Symbiotic interactions may change depending on third parties like predators or prey. Third-party interactions with prey bacteria are central to the symbiosis betweenDictyostelium discoideumsocial amoeba hosts andParaburkholderiabacterial symbionts. Symbiosis with inedibleParaburkholderiaallows hostD. discoideumto carry prey bacteria through the dispersal stage where hosts aggregate and develop into fruiting bodies that disperse spores. Carrying prey bacteria benefits hosts when prey are scarce but harms hosts when prey bacteria are plentiful, possibly because hosts leave some prey bacteria behind while carrying. Thus, understanding benefits and costs in this symbiosis requires measuring how many prey bacteria are eaten, carried and left behind by infected hosts. We found thatParaburkholderiainfection makes hosts leave behind both symbionts and prey bacteria. However, the number of prey bacteria left uneaten was too small to explain why infected hosts produced fewer spores than uninfected hosts. Turning to carried bacteria, we found that hosts carry prey bacteria more often after developing in prey-poor environments than in prey-rich ones. This suggests that carriage is actively modified to ensure hosts have prey in the harshest conditions. Our results show that multi-faceted interactions with third parties shape the evolution of symbioses in complex ways.

Cooperation is widespread across life, but its existence can be threatened by exploitation. The rise of obligate social cheaters that are incapable of contributing to a necessary cooperative function can lead to the loss of that function. In the social amoeba Dictyostelium discoideum, obligate social cheaters cannot form dead stalk cells and in chimeras instead form living spore cells. This gives them a competitive advantage within chimeras. However, obligate cheaters of this kind have thus far not been found in nature, probably because they are often enough in clonal populations that they need to retain the ability to produce stalks. In this study we discovered an additional cost to obligate cheaters. Even when there are wild-type cells to parasitize, the chimeric fruiting bodies that result have shorter stalks and these are disadvantaged in spore dispersal. The inability of obligate cheaters to form fruiting bodies when they are on their own combined with the lower functionality of fruiting bodies when they are not represent limits on obligate social cheating as a strategy.

Chlamydiae represent a diverse group of obligate intracellular bacteria with elusive hosts in environmental settings. This study used one of the largest collections of wild amoebae (Dictyostelium discoideum and D. giganteum, 106 clones) collected over the past two decades to screen for novel environmental chlamydiae. We found that novel environmental chlamydiae are prevalent in two wild Dictyostelium species and assembled 42 novel chlamydiae metagenome-assembled genomes (MAGs). The MAGs represent three chlamydiae species previously only reported using 16S sequencing. Their genomes are divergent enough from other species to warrant placing them in two new genera (tentatively called Ca. Dictychlamydia sp. LF1, Ca. Dictychlamydia sp. LF2, and Ca. Feichlamydia sp. LF3). In addition, these chlamydiae species show strong host specificity with two Dictyostelium amoeba hosts, except one amoeba sample. Ca. Dictychlamydia sp. LF1 and Ca. Feichlamydia sp. LF3 was exclusively observed in D. discoideum, while Ca. Dictychlamydia sp. LF2 was found only in D. giganteum. Phylogenetic and comparative genomic analyses suggest that all three chlamydiae are close to arthropod-associated chlamydiae and likely have some intermediate characteristics between previously reported amoeba-associated and vertebrate-associated chlamydiae. This study significantly broadens our understanding of the chlamydial host range and underscores the role of amoebae as vital hosts for environmental chlamydiae. ### Competing Interest Statement The authors have declared no competing interest.

AbstractPredation is likely to influence the function of bacterial communities and the evolution of bacterial pathogens, because characteristics that permit escape from predators often overlap with traits used for biocontrol of plant pathogens, virulence, or even bioremediation. Soil bacteria are preyed upon by a variety of microorganisms, including the amoebaDictyostelium discoideum,which has led some strains to evolve resistance. We identified genes required for threePseudomonasspecies associated withD. discoideumto evade predation by screening more than 6,000 transposon mutants for loss of resistance. One species required a variety of genes including toxins and secondary metabolism genes, but the other two appear to have functionally redundant mechanisms of resistance, since disruption of genes with pleiotropic effects was required to render them susceptible. We determined that GacA, which positively regulates secondary metabolism, is required for resistance in all three species. Predation resistance also appears to be a social trait based on enrichment of cooperative genes in one species and rescue of mutants by wild type in another. Many genes required for resistance are conserved among both resistant and susceptible species, but several are found in few genomes and some of these have homologs in distantly related species. Gain and loss of resistance appears to be a dynamic process in which regulatory and structural genes are well conserved across species, the specific toxins they regulate may be lost in the absence of predators, and new toxins may be acquired through horizontal gene transfer.

Consumers range from specialists that feed on few resources to generalists that feed on many. Generalism has the clear advantage of having more resources to exploit, but the costs that limit generalism are less clear. We explore two understudied costs of generalism in a generalist amoeba predator, Dictyostelium discoideum , feeding on naturally co-occurring bacterial prey. Both involve costs of combining prey that are suitable on their own. First, amoebas exhibit a reduction in growth rate when they switched to one species of prey bacteria from another compared to controls that experience only the second prey. The effect was consistent across all six tested species of bacteria. These switching costs typically disappear within a day, indicating adjustment to new prey bacteria. This suggests that these costs are physiological. Second, amoebas usually grow more slowly on mixtures of prey bacteria compared to the expectation based on their growth on single prey. There were clear mixing costs in three of the six tested prey mixtures, and none showed significant mixing benefits. These results support the idea that, although amoebas can consume a variety of prey, they must use partially different methods and thus must pay costs to handle multiple prey, either sequentially or simultaneously.

Most living things that we can see with our naked eyes develop from a single fertilized egg cell. Even dinosaurs hatched from eggs. Why are eggs so successful over the evolutionary history of life? One theory is eggs result in a multicellular body where the daughter cells are genetically identical, and therefore have nothing to gain from conflict with other cells. Another theory is that when individual cells acquire changes independently of other cells, they may become incompatible with each other. Such cells can be thought of as speaking different genetic languages, such that they lose the ability to communicate during a crucial process like the development of the body. Here we have tested this latter theory, which we call the "coordination hypothesis" in a species of social amoeba. Of several traits tested, one showed strong evidence of poorer coordination in mixtures, supporting the coordination hypothesis.

Selection for cooperation or conflict in multicellular organisms that develop from a mixture of cells has been widely appreciated, but unselected effects of mixing could also have important fitness consequences to the resulting chimeras. We formalized this idea as the ‘developmental incompatibility hypothesis’ and empirically tested it in the social amoeba Dictyostelium discoideum . We mixed D. discoideum that evolved in isolation for generations and accumulated mutations that have not been tested against each other by selection. To quantify the effect of developmental incompatibilities in experimental lines that evolved in the presumed absence of kin recognition, we measured the developmental traits of slug migration and spore production. Our results show no evidence for incompatibilities in coordinated movement towards light in the social amoeba. We found support for developmental incompatibilities from our spore production result. The effect was especially clear for a small proportion of fruiting bodies that developed after migration. For slugs that did not migrate, mixing could be harmful if they are smaller or if, the stalks of such fruiting bodies are shorter which would hamper dispersal. Our results indicate that the detrimental fitness consequences of developmental incompatibilities may be context-dependent. ### Competing Interest Statement The authors have declared no competing interest.

Consumers lie on a continuum between diet specialization on few resources to being generalist feeders on many resources. Generalism has the clear advantage of having more resources to exploit, but the costs that limit generalism are less clear. We explore two understudied costs of generalism in a super-generalist amoeba predator, Dictyostelium discoideum , feeding on naturally co-occurring bacterial prey. Both involve costs of combining different prey. First, amoebas exhibit a reduction in growth rate when they switch from one species of prey bacteria to another, something we call resource-switching costs. These switching costs typically disappear within a day, indicating adjustment to new prey bacteria. Second, amoebas usually divide more slowly on mixtures of bacteria compared to on single bacteria, something we call resource-mixing costs. Both results support that idea that, although amoebas can consume a huge variety of prey, they must use partially different methods and thus must pay costs to handle multiple prey, either sequentially or simultaneously. Significance Statement Perhaps the most fundamental conflict in nature occurs when one organism consumes another. Diet generalists benefit from the advantage of eating many prey but then must deal with many prey defences. We explore costs associated with a broad diet in a protist microbial predator, Dictyostelium discoideum . These predators of bacteria show a delay in growth when switched from one bacteria to another, supporting the hypothesis that they must deploy different strategies. They also experience costs when grown on many bacteria at once, suggesting that the alternative strategies for consuming different prey are partly incompatible with each other. Our findings shed light on the nature of diet generalism and highlight the complexity of predation in the microbial world.

ABSTRACT Many microbes interact with one another, but the difficulty of directly observing these interactions in nature makes interpreting their adaptive value complicated. The social amoeba Dictyostelium discoideum forms aggregates wherein some cells are sacrificed for the benefit of others. Within chimeric aggregates containing multiple unrelated lineages, cheaters can gain an advantage by undercontributing, but the extent to which wild D. discoideum has adapted to cheat is not fully clear. In this study, we experimentally evolved D. discoideum in an environment where there were no selective pressures to cheat or resist cheating in chimeras. D. discoideum lines grown in this environment evolved reduced competitiveness within chimeric aggregates and reduced ability to migrate during the slug stage. By contrast, we did not observe a reduction in cell number, a trait for which selection was not relaxed. The observed loss of traits that our laboratory conditions had made irrelevant suggests that these traits were adaptations driven and maintained by selective pressures D. discoideum faces in its natural environment. Our results suggest that D. discoideum faces social conflict in nature, and illustrate a general approach that could be applied to searching for social or non-social adaptations in other microbes. SIGNIFICANCE STATEMENT Microbes interact in diverse and important ways, but the difficulty of directly observing microbes in nature can make it challenging to understand the adaptive significance of these interactions. In this study, we present an experimental evolution approach to infer the selective pressures behind an apparently social trait in the microbe Dictyostelium discoideum. We take advantage of the observation that organisms ‘use it or lose it’ – when selective pressures are relaxed, adaptations that evolved in response to those pressures tend to be lost. Our work helps resolve debate over the importance of cheating in D. discoideum, and demonstrates a general approach that could be applied to the study of other microbial traits that are difficult to observe in nature.

AbstractThe soil amoebaDictyostelium discoideumacts as both a predator and potential host for diverse bacteria. We tested fifteenPseudomonasstrains that were isolated from transiently infected wildD. discoideumfor ability to escape predation and infectD. discoideumfruiting bodies. Three predation-resistant strains frequently caused extracellular infections of fruiting bodies but were not found within spores. Furthermore, infection by one of these species induces secondary infections and suppresses predation of otherwise edible bacteria. Another strain can persist inside of amoebae after being phagocytosed but is rarely ingested. We sequenced isolate genomes and discovered that predation-resistant isolates are not monophyletic. ManyPseudomonasisolates encode secretion systems and toxins known to improve resistance to phagocytosis in other species, as well as diverse secondary metabolite biosynthetic gene clusters that may contribute to predation resistance. However, the distribution of these genes alone cannot explain why some strains are edible and others are not. Each lineage may employ a unique mechanism for resistance.

Many microbes interact with one another, but the difficulty of directly observing these interactions in nature makes interpreting their adaptive value complicated. The social amoeba Dictyostelium discoideum forms aggregates wherein some cells are sacrificed for the benefit of others. Within chimaeric aggregates containing multiple unrelated lineages, cheaters can gain an advantage by undercontributing, but the extent to which wild D. discoideum has adapted to cheat is not fully clear. In this study, we experimentally evolved D. discoideum in an environment where there were no selective pressures to cheat or resist cheating in chimaeras. Dictyostelium discoideum lines grown in this environment evolved reduced competitiveness within chimaeric aggregates and reduced ability to migrate during the slug stage. By contrast, we did not observe a reduction in cell number, a trait for which selection was not relaxed. The observed loss of traits that our laboratory conditions had made irrelevant suggests that these traits were adaptations driven and maintained by selective pressures D. discoideum faces in its natural environment. Our results suggest that D. discoideum faces social conflict in nature, and illustrate a general approach that could be applied to searching for social or non-social adaptations in other microbes.

ABSTRACT Microbes adapt to the presence of other species, but the fitness consequences of specific interactions are difficult to study in their natural context. We experimentally evolved symbiotic microbes in an artificial environment without access to the partners with whom they interact in nature. As organisms will tend to lose adaptations that they do not need due to drift or pleiotropic tradeoffs, we expect normally symbiotic microbes evolved in isolation to lose adaptations to help or harm their natural partners. The direction and magnitude of such changes can suggest whether the microbes had historically been selected to help or harm one another. We apply this method to the symbiosis between the social amoeba Dictyostelium discoideum and three intracellular bacterial endosymbionts, Paraburkholderia agricolaris, P. hayleyella, and P. bonniea. A minority of strains of Paraburkholderia and D. discoideum evolved differences in their effects on one another’s fitnesses, implying the existence of adaptations to one another that were lost when no longer relevant. Our results suggest that the degree to which D. discoideum and Paraburkholderia have adapted to help or harm one another can differ substantially between strains within each species, with some strains appearing to have a historically adversarial relationship, some strains a more benign relationship, and many strains no clear adaptations to one another at all. Our results underscore the complexity of microbial interactions in nature and suggest experimental evolution under relaxed selection is a potentially useful approach for studying adaptation in microbes.

The social amoebaDictyostelium discoideumengages in a complex relationship with bacterial endosymbionts in the genusParaburkholderia, which can benefit their host by imbuing it with the ability to carry prey bacteria throughout its life cycle. The relationship betweenD. discoideumandParaburkholderiahas been shown to take place across many strains and a large geographical area, but little is known aboutParaburkholderia's potential interaction with other dictyostelid species. We explore the ability of threeParaburkholderiaspecies to stably infect and induce bacterial carriage in other dictyostelid hosts. We found that all threeParaburkholderiaspecies successfully infected and induced carriage in seven species ofDictyosteliumhosts. While the overall behaviour was qualitatively similar to that previously observed in infections ofD. discoideum, differences in the outcomes of different host/symbiont combinations suggest a degree of specialization between partners.Paraburkholderiawas unable to maintain a stable association with the more distantly related hostPolysphondylium violaceum. Our results suggest that the mechanisms and evolutionary history ofParaburkholderia's symbiotic relationships may be general withinDictyosteliumhosts, but not so general that it can associate with hosts of other genera. Our work further develops an emerging model system for the study of symbiosis in microbes.

Some endosymbionts living within a host must modulate their hosts' immune systems in order to infect and persist. We studied the effect of a bacterial endosymbiont on a facultatively multicellular social amoeba host. Aggregates of the amoeba Dictyostelium discoideum contain a subpopulation of sentinel cells that function akin to the immune systems of more conventional multicellular organisms. Sentinel cells sequester and discard toxins from D. discoideum aggregates and may play a central role in defence against pathogens. We measured the number and functionality of sentinel cells in aggregates of D. discoideum infected by bacterial endosymbionts in the genus Paraburkholderia. Infected D. discoideum produced fewer and less functional sentinel cells, suggesting that Paraburkholderia may interfere with its host's immune system. Despite impaired sentinel cells, however, infected D. discoideum were less sensitive to ethidium bromide toxicity, suggesting that Paraburkholderia may also have a protective effect on its host. By contrast, D. discoideum infected by Paraburkholderia did not show differences in their sensitivity to two non-symbiotic pathogens. Our results expand previous work on yet another aspect of the complicated relationship between D. discoideum and Paraburkholderia, which has considerable potential as a model for the study of symbiosis.

The social amoeba Dictyostelium discoideum engages in a complex relationship with bacterial endosymbionts in the genus Paraburkholderia, which can benefit their host by imbuing it with the ability to carry prey bacteria throughout its life cycle. The relationship between D. discoideum and Paraburkholderia has been shown to take place across many strains and a large geographical area, but little is known about Paraburkholderia's potential interaction with other dictyostelid species. We explore the ability of three Paraburkholderia species to stably infect and induce bacterial carriage in other dictyostelid hosts. We found that all three Paraburkholderia species successfully infected and induced carriage in seven species of Dictyostelium hosts. While the overall behaviour was qualitatively similar to that previously observed in infections of D. discoideum, differences in the outcomes of different host/symbiont combinations suggest a degree of specialization between partners. Paraburkholderia was unable to maintain a stable association with the more distantly related host Polysphondylium violaceum. Our results suggest that the mechanisms and evolutionary history of Paraburkholderia's symbiotic relationships may be general within Dictyostelium hosts, but not so general that it can associate with hosts of other genera. Our work further develops an emerging model system for the study of symbiosis in microbes.