Symbiosis

Abstract

Microorganisms arose and diversified before the appearance of large multicellular organisms. The bodies of the latter provided new potential habitats for microbes — habitats that were persistent, stable (from a microbial perspective) and nutrient-rich. As a result, large organisms, from oaks to humans, have been continuously enmeshed in complex interactions with microorganisms during their evolution. Biologists have paid most attention to associations with microbes that are pathogenic; typically, microbial infection has been viewed as deleterious, or at best irrelevant, to vigor and reproduction. But the last 15 years have witnessed increased appreciation of interactions that benefit the host as well as the microbe. These interactions are loosely grouped under the term ‘symbiosis’ and the microbial partners called ‘symbionts’. Although the term ‘symbiont’ is typically applied to mutualistic microorganisms, it is often used to include associates for which the full spectrum of effects on hosts is not known. Symbiosis is ubiquitous in terrestrial, freshwater and marine communities. It has played a key role in the emergence of major life forms on Earth and in the generation of biological diversity. Numerous authors have pointed out that Darwin did not emphasize symbiosis as an evolutionary mechanism; its role is, of course, not inconsistent with Darwin's main ideas, but appreciation of symbiosis as a source of evolutionary novelty has developed relatively recently. Its importance is perhaps most strikingly illustrated by the symbiotic events that led to the evolution of eukaryotic organelles — plastids and mitochondria — from cyanobacterial and alpha-Proteobacterial ancestors, respectively. For plants, associations with fungi and bacteria were key innovations in the colonization of land and of specific habitats. Eukaryote-associated microbes act as metabolic partners for accessing limiting nutrients and also as protectors, producing toxins that ward off herbivores or pathogens. Similar associations have arisen with animals, allowing colonization of diverse niches, such as the specialized feeding on plant or animal tissues, and the use of deep ocean hydrothermal vent habitats. Often, the associations are persistent for the hosts, frequently being transmitted vertically across generations, from mother to progeny. The organisms involved in a symbiosis may be sufficiently fused that they cannot live apart or be recognized as distinct entities without close scrutiny. Before biological research became focused on genetic model organisms, greater attention was paid to the study of symbiosis. Microscopy was used extensively to discover and elucidate symbioses before the middle of the 20th century. The reasons for a period of relative neglect, until about 1990, are complex. One contributing factor has been the unsuitability of most symbiotic partners as models for laboratory studies of genetics and development. The organisms that are easiest to grow and study in the laboratory, such as Escherichia coli, Arabidopsis thaliana, Drosophila melanogaster and Mus domesticus, are weedy species adapted to show fast growth in temporary niches. These species have lifestyles relatively free of complex interdependencies with other species. But most microorganisms in natural communities are likely to have obligate dependencies on other species (often other microbes), explaining why 99% of microorganisms are difficult or impossible to culture. Similarly, most symbionts of plants and animals cannot be readily cultured independently of hosts, precluding most conventional microbiological analyses. The recent proliferation of research on symbionts is part of a broad development in biology which is seeing the extension of the tools of molecular biology to the study of biological diversity, evolutionary history and the metabolic functioning of non-model organisms. Genomic approaches have enabled massive progress in understanding the metabolism of natural microbial communities in general and of symbiotic systems in particular. These new approaches have yielded extensive genomic and evolutionary data for a broad sampling of symbiont types, allowing new insight into evolutionary innovations and constraints imposed by symbiotic lifestyles, for both bacterial symbionts and their hosts. Symbiotic microorganisms form intimate associations with host species, but these associations take myriad forms. Symbioses vary in the mechanisms for achieving and maintaining the associations, in the role of the microbe in the host biology, in the evolutionary history of the relationship, and in the genomic features of the microorganism. Some symbioses are highly specialized and constant, and appear to constrain the evolution of novelty in host lineages. Others are dynamic, with symbiont genomes acting as portals for the import of new genes that affect the host ecology. Some of the main variables in symbioses are considered here. In mutualistic associations, both host and symbiont often evolve to accommodate one another. Prime examples are the vertically transmitted symbionts that live in insect hosts, such as Buchnera aphidicola in aphids. The insect host has evolved to produce specialized cells or organs to house the symbionts and to facilitate passage of the symbiont from mother to offspring before birth or oviposition (Figure 1). The symbiont has evolved to overproduce amino acids or other nutrients required by the host and, presumably, to regulate its replication so as to avoid damage to host functioning. These situations can be described as co-evolved in the sense that both parties have evolved so as to sustain the relationship. The term ‘codiversification’ refers to a shared history of vertical descent, and this is also exemplified by the aphid–Buchnera association and other insect nutritional symbioses. In these cases, phylogenetic studies have shown that diversification of both host and symbiont have occurred in parallel, through a history of constant association over millions of years. It is important to keep in mind the distinction between coevolution and codiversification; either can occur without the other. For example, the signaling between Hawaiian bobtail squid and their Vibrio fischeri symbionts indicates a history of coevolution, even though there is no strong case for long codiversification of the two lineages. Many symbionts provide clear-cut cases of mutualism, providing their hosts with nutrients or defenses. But other cases are less readily categorized. Any microbe that forms chronic infections in an individual host or in a host lineage may evolve to conserve or even to benefit its host, as this will help to maintain its immediate ecological resource. Many host-associated microorganisms are thus expected to combine mutualistic and pathogenic properties, in the sense that they can invade novel hosts but do not kill them immediately. Examples include the Mollicutes, the chlamydiae, the rickettsia and other bacteria that persist in hosts for long periods. Symbiosis may be obligate for the host, for the symbiont, for both or for neither. The most intensively studied cases involve highly specialized associations in which both partners can live only with one another. Examples include the bacterial symbionts of insects that make nutrients for their hosts, such as Buchnera aphidicola and aphids. In other cases, the association is obligate for the symbiont but not for the host; examples include a large number of insect associates such as Wolbachia pipientis, Spiroplasma species, and facultative symbionts of aphids and relatives. Some symbioses, such as the association between Vibrio fischeri and certain squid and fish species, appear to be largely obligate for both parties, but also entail a life cycle stage in which the symbiont replicates outside of the host. As considered further below, these distinctions make a very big difference for the genome evolution of the symbiont and, in turn, for the biology of the host. Many symbionts are transmitted vertically as their major or sole means of colonizing hosts. In these cases, transmission is usually through the maternal line. In certain cases, maternal transmission is usual, but occasionally horizontal transmission — that is, transmission from an outside source — occurs, permitting colonization of naïve hosts. Finally, in some cases the host must be colonized anew every generation or even repeatedly during an individual's lifespan. In these cases, symbionts may be less clearly distinguishable from pathogens. Of particular interest are the symbioses that are transmitted maternally along genetic lineages of hosts. Mitochondria and plastids, each descended from free-living bacteria that invaded eukaryotic ancestors more than 1 billion years ago, can be considered as extreme examples of this: they are effectively fused with the host, and form part of the essential components of each host cell. But although genomic data have revealed that many genes have moved from the organelle genome to the nuclear genome of the eukaryotic host, the part of the genome that remains separate enforces important constraints and generates complexity in eukaryotic evolution. In the cases of organelles, as well as some other maternally transmitted microorganisms, the symbiont genome is clonal or near-clonal, and has different mutational processes and a different population structure from the host's nuclear genome (Figure 2). This added genomic complexity is a recurring theme in symbiotic systems and potentially subjects the host lineage to evolutionary forces that would not exist if the genomes were unified. Many symbionts live in the cytosol of the host cells, sometimes in direct contact and sometimes in a vacuole surrounded by a host-derived membrane. The latter may differ little from those living in close proximity to host cells. The same variety of cellular locations is seen in pathogenic bacteria and the mechanisms by which symbionts interact with host cells are often similar to those found in pathogens. For example, many cases are now known, for both pathogens and symbionts, in which the molecular syringes known as type III secretion systems are used for delivering substances to host cells and for entering the cytosol. Similarly, toxins used by pathogens in exploiting hosts are also encoded in symbiont genomes and appear to sometimes function in providing protection against natural enemies. Certain symbionts exhibit the extremes of evolutionary stasis in genome arrangement and content. The very small genomes of Buchnera and Blochmannia (symbionts of carpenter ants) show no rearrangements or gene uptake over many millions of years of divergence, and lack any bacteriophage or mobile genetic elements. In contrast, symbionts with higher rates of horizontal transmission have more dynamic genome content. For example, in facultative symbionts of insects, bacteriophage are associated with novel gene acquisition, and repetitive sequences may facilitate frequent recombination among strains that encounter one another in co-infected hosts. Symbionts face the evolutionary challenge of spreading themselves, that is, increasing the frequency of infected individuals in a host population. Many of the strategies that have evolved in a wide variety of microbe lineages infecting animal, plant and other hosts depend on mutualistic effects that enhance the fitness of the host and consequently — because infected hosts persist longer and reproduce more — of the symbiont. Symbiosis plays many of the same roles in both plant and animal hosts, contributing to nutrition and defense. Symbionts can also spread via exploitation of hosts. Three main categories of symbiont-imposed effects emerge repeatedly. These can be correlated with other variables; for example, genome stasis and long-term codiversification have been most frequently associated with nutritional benefits to hosts. Symbionts often enhance host ability to acquire nutrients from the environment or provide the pathways for synthesis of needed organic compounds or for catabolism of molecules available in the environment. Among the most important and widespread symbioses are the mycorrhizal associations which enable terrestrial plants to obtain inorganic phosphate from soil. Similarly, many vascular plants associate with bacterial symbionts that boost their nitrogen supplies by fixating atmospheric nitrogen. Photosynthetic capacity itself has evolved repeatedly in unicellular eukaryotic lineages through repeated events of symbiont acquisition. For animals, the genes encoding enzymes for biosynthesis of many required organic compounds were lost early in evolution. Thus, animals are unable to make many of the compounds required for cellular function and are limited in their abilities to use different energy sources. The required organic compounds are commonly provided by bacterial or fungal symbionts. Acquisition of a symbiont has often been a key step in colonizing specialized niches in which dietary availability of these compounds would otherwise be inadequate. In simple cases, symbionts provide essential amino acids (those required in animal diets) or vitamins (mainly cofactors required for functioning of central metabolic enzymes). Some nutritional symbioses are maternally transmitted, and these symbionts typically have highly reduced and static genomes, having lost all but the essential components of cell metabolism plus the pathways required for host provisioning. Examples include Buchnera aphidicola in aphids feeding on phloem sap, Wigglesworthia glossinidia in tsetse feeding on vertebrate blood, and the paired symbionts Sulcia muelleri and Baumannia cicadellinicola in sharpshooters feeding on xylem sap. In some cases, such as the bacterial symbionts of gutless marine oligochaetes and tube worms, symbionts dwell within specialized organs and enable hosts to exploit unusual energy sources such as sulphur oxidation. Nutritional contributions are also provided by the less-specialized symbionts living in guts, which not only produce vitamins but also help to catabolize molecules in food that would otherwise not be available. Microorganisms have capacity for synthesis of a great variety of bioactive compounds that can be used as defensive weapons by hosts. Plants associate with a diverse set of endophytic fungi and other microbes that have been shown to provide protection from herbivores and pathogens. These symbionts may be vertically transmitted, within seeds, as in the case of some grass–fungi associations, or they may be acquired during the lifespan of individual plants or even leaves, as in the endophytes of tropical trees. Animals are generally limited in capacity to produce a variety of bioactive compounds; a far greater diversity of antibiotic molecules is produced by bacteria, plants and fungi. In many animals, associations with microbes supply defensive compounds that ward off predators, parasites and pathogenic microorganisms, or that help to subdue prey. Clear examples of chemical defenses provided by symbionts have been reported for bryozoans, marine crustacea, and insects. Nutritional interactions between symbionts and hosts tend to be relatively static in a particular environment; but where a host depends on symbionts for protection, this often needs to be dynamic, changing in response to new natural enemies or newly evolved countermeasures to existing defenses. Protective symbioses can involve a diversity of microbes in the same host, and may involve symbionts that themselves have dynamic gene content. For example, in the Paederus beetles protected from spider predation by a Pseudomonas symbiont, the protective compounds are polyketides which, in turn, are underpinned by a recently acquired ‘genomic island’ within the bacterial chromosome. Similarly, the defensive symbiont of aphids, Hamiltonella defensa, contains variable bacteriophage that carry genes encoding eukaryote-targeted toxins; these may play a role in the ability of the symbionts to subdue parasitic wasp larvae. In some defensive associations, the symbionts are less conspicuous and may not be obligate. Sometimes they consist of a diverse community of microorganisms, for which the categorization as symbionts may not be clear-cut. For example, sponges are typically associated with a large variety of microorganisms; some are simply food for these filter feeders, while others produce bioactive compounds and likely function as defensive symbionts. Some cases of protective symbiosis involve mechanisms other than toxins; an example is the light organ symbioses in which bacteria bioluminescence enables fish and squid to make themselves less apparent to potential predators. Toxins of invertebrates have repeatedly turned out to be synthesized by bacterial symbionts, rather than by the animals themselves. Examples include bryostatins from bryozoans and antibiotic compounds used by solitary hymenopterans to protect their developing brood. Some of these compounds are of potential medical use as novel anti-tumor agents or antibiotics. These defensive substances are likely to be underpinned by dynamic elements of the symbiont genome, as in the case of the Paederus beetles mentioned above. A microbe associated with a multicellular eukaryote can evolve to benefit the host under a range of circumstances, since by improving the survival and reproduction of the host it may improve the longevity or extent of its current habitat. But it can also evolve to affect hosts in ways that are not mutualistic but that have interesting consequences for long-term patterns of host evolution and diversification. The pre-eminent example of a symbiont group that furthers its own replication at the expense of host fitness is Wolbachia pipientis, infecting arthropods. Several other symbionts from different bacterial phyla have been shown to have similar effects on insect hosts; these include spiroplasmas and the Bacteroidetes symbiont Cardinium hertigii. These bacteria impose a variety of effects, including reproductive incompatibility among host strains from the same species, parthenogenesis of otherwise sexually reproducing host females, and son-killing and other mechanisms of sex ratio biases favoring females. All of these effects on hosts reflect adaptations to maternal transmission: whereas hosts pass their genes equally through sons and daughters, symbionts are passed only through females, leading to the evolution of symbiont features that favor spread of the infected matriline, at the expense of other matrilines. The incompatibility of male and female hosts with different infection status can potentially lead to reproductive isolation and speciation of hosts. Thus, symbionts may play a role in the generation of biological diversity through direct effects on speciation, in addition to their capacity for expanding ecological niches. Several model systems for microbe–animal symbiosis have emerged that illustrate the diversity of symbiotic lifestyles. Full genome sequences have been obtained for many of the symbionts. Together, these give a picture of the various modes of symbiosis, the role of symbiosis in eukaryote ecology and evolution, and the genomic processes that are important in symbiont adaptation. These cases represent the tip of the iceberg, as most symbioses remain unrecognized or unstudied. Almost all aphids contain the obligate symbiont Buchnera aphidicola. This symbiont has cospeciated with aphids throughout their evolution, through strict maternal transmission. Several Buchnera genomes have been sequenced, and these exhibit a set of features that is typical of long-term symbionts and pathogens replicating only within hosts. They are among the smallest of bacterial genomes, in the range of 0.5–0.65 megabase pairs with strong AT bias and very few regulatory genes. Despite their small size, the genomes retain genes encoding enzymes for the biosynthesis of essential amino acids that are lacking in phloem sap diets of hosts. Buchnera genomes have accumulated many mutations that destabilize gene products, and they overexpress proteins typically involved in the response to heat shock. Many aphids also have facultative relationships with a variety of other symbionts. Although inherited vertically in the laboratory, they are horizontally transmitted among host matrilines in the field. Several are closely related to human pathogens, such as Salmonella enterica and Yersinia pestis in the Enterobacteriaceae. These facultative symbionts have been shown to provide hosts with defense against parasitoids and pathogenic fungi, as well as to increase tolerance to heat stress. Partial sequencing of the genome of one such symbiont, Hamiltonella defensa, indicates the presence of phage and a variety of toxin-encoding genes, which vary among strains of the symbiont and may play a role in the ability of this symbiont to kill parasitoid wasp larvae within the host aphid body. African tsetse flies (genus Glossinidia) feed exclusively on vertebrate blood and are major vectors of human disease. These insects contain two intracellular symbionts, Wigglesworthia species and Sodalis glossinidius, the latter being one of the few obligate symbionts cultured in isolation from the host. Genomes of both these symbionts have been sequenced. Wigglesworthia, which resembles Buchnera in having undergone long-term cospeciation with hosts and in its tiny genome (∼0.7 megabase pairs), contrasts with Buchnera in its biosynthetic capabilities: in keeping with the nutritional deficiencies of its host's blood diet, Wigglesworthia has many pathways for the biosynthesis of vitamins but lacks those for biosynthesis of amino acids present in Buchnera. Sodalis is believed to be a more recently derived symbiont and has a much larger (4.1 megabase pairs) and more dynamic genome, containing many phage genes and mobile genetic elements. Many Sodalis genes seem to have undergone recent mutational inactivation and consequently its genome has a low coding density. It retains three type III secretion systems, which are used in colonization of host cells, paralleling the mechanisms used by pathogens such as Salmonella enterica. Like the facultative symbionts in aphids, Sodalis is maternally transmitted, but phylogenetic analyses indicate some incidence of horizontal transfer among host lineages. This host, Euprymna scolopes, uses counterlighting to reduce its conspicuousness to predators, and the light is provided by luminescent Vibrio fischeri living in specialized light organs. In contrast to most of the insect symbioses, these symbionts must infect juvenile hosts after they emerge from eggs, with symbionts colonizing from the surrounding water. The interaction is nonetheless species-specific and obligate, and the host has a variety of developmental and biochemical adaptations for recruitment and selection of the symbionts. The symbionts must undergo a series of winnowing events to gain entry into the host organ. This system provides excellent examples of how homologous mechanisms can be used in both pathogenesis and mutualism, with different outcomes. For example, surface molecules implicated in virulence in other bacteria, such as those causing whooping cough and gonorrhea, are used by the squid to recognize and recruit V. fischeri. In addition, the genome sequence of V. fischeri reveals a large number of infection mechanisms that are homologous with those known from pathogenic Vibrio species. Wolbachia pipientis is an ancient clade of alpha-Proteobacteria that exhibits two major lifestyles. In arthropod hosts, in which it was first characterized, Wolbachia is maternally transmitted with high efficiency but also undergoes extensive horizontal transfer in field populations. This horizontal transfer is sometimes between distantly related hosts, through mechanisms that are largely unidentified. It infects up to 70% of insect species and is likely to have played a large role in the evolutionary history of insects and other terrestrial arthropods. The mutualistic effects of Wolbachia on its hosts appear to be limited. Wolbachia invades host populations by manipulating their reproductive systems in several ways. Most commonly, mating with infected males, which cannot transmit the symbionts to progeny, reduces the fecundity of females that are uninfected or infected with a different Wolbachia strain (reproductive incompatibility). In other cases, infected females can be transformed to undergo parthenogenetic, rather than sexual, reproduction, producing only female progeny each of which can pass on the infection. In some hosts, genetic males, if infected, are feminized so that they become capable of transmitting Wolbachia via eggs. Wolbachia is present in a large number of pests of agricultural crops and of disease vectors including mosquitoes; since it can potentially cause sweeps of genes within host populations, it has been considered a possible tool in strategies aimed at manipulating pest insect populations. In contrast, Wolbachia living in filarial nematodes appear to be obligate mutualists of their hosts, which in turn are serious parasites of humans and livestock. They have relatively small genomes and show a history of cospeciation with hosts, similar to Buchnera and other nutritional symbionts of insects, and in contrast to the Wolbachia living in arthropods. The finding that these pathogenic nematodes require bacterial symbionts for reproduction has led to new strategies for the prevention of African river blindness, which is caused by the nematode Onchocerca volvulus. Species of Xenorhabdus and Photorhabdus, related genera in the Enterobacteriaceae, are obligate associates of nematodes that kill insects and then consume their cadavers, reproducing within them. Thus, these symbionts combine life-cycle phases that are mutualistic (to the nematodes) as well as virulent (to the insects) and competitive (with other microorganisms). These different needs are reflected in the production of a correspondingly high diversity of bioactive compounds, including molecules that ensure the death of the insect prey, enzymes that participate in conversion of the insect body for use as nematode food, and antibiotic compounds, including polyketides and nonribosomal peptides, that prevent the invasion of other bacteria or fungi while the nematodes feed and reproduce. The symbionts then reinfect the young nematodes, which proceed to forage for additional insect hosts. The diversity of products and the complexity of the life cycles are reflected in the genomes of these symbionts which are relatively large (>5 megabase pairs), about an order of magnitude larger than those of nutritional symbionts, such as Buchnera, and relatively dynamic, with evidence for gene acquisition and loss. The Photorhabdus luminescens genome has genes for the production of a high diversity of bioactive compounds, toxic to insects as well as microorganisms; these are of interest as a potential source of compounds useful for antibiotic and insecticidal applications. Some toxin-encoding genes overlap with those of pathogens. P. luminescens also has a large number of genes involved in regulation and environmental sensing. The digestive tracts of animals contain a wide variety of microorganisms that are a mixture of transient opportunists, food-borne commensals and mutualists that are typical for a particular host and interact in specialized ways with the host, affecting development, immune system functioning, and processing of food. Some members of the intestinal fauna play critical roles in the biology of vertebrates, which have evolved to depend on interactions for normal development. For example, Bacteroidetes thetaiotamicron induces vascularization and other aspects of the development of the mammalian gut. The human gut fauna includes hundreds of species of microorganisms and the extent to which these include specific or obligate interactions that benefit hosts is not yet clear. Symbionts persist by virtue of major effects on their hosts. As a result, though near-invisible and often completely overlooked, symbionts can be critical factors in determining the outcome of ecological and evolutionary processes. It is now clear that symbioses have been crucial in adaptive radiation, lineage evolution, and ecological diversification. It is likely that they also play a role in evolutionary constraint and in extinction of host lineages. Finally, due to the potency of symbionts as agents affecting host health and reproduction, their study can reveal new means for controlling populations and diseases.