C. elegans outside the Petri dish

1The free-living nematode Caenorhabditis elegans is a major model species that is used in a range of biological research. After initial work by Emile Maupas and Victor Nigon on its mode of reproduction, meiosis and development, experiments by Sydney Brenner and collaborators in the 1960s and 1970s raised C. elegans to the status of a premier model organism. Besides its genome, the first to be sequenced for a multicellular organism, an extensive body of knowledge is now available on the molecular, cellular, developmental and behavioral biology of this organism. A number of key discoveries have been made by studying C. elegans, including the molecular mechanisms of apoptosis and gene silencing by small RNAs.

2The C. elegans reference strain N2 is a laboratory animal. This strain, originally isolated in Bristol, England, was cultured in the laboratory for many years before it was first frozen. Its laboratory environment consisted of agar plates seeded with Escherichia coli as a food source. E. coli was used because, as another model organism, it was already available in many laboratories—not because it was originally associated with wild C. elegans. An independent subculture from the same wild Bristol isolate was maintained in liquid axenic culture, and by comparing both strains, we now know of several mutations that appeared and were fixed in the N2 lineage, some of which appear adaptive in the agar plate environment. These mutations pleiotropically affect many traits, such as behavior, reproduction, susceptibility to pathogens, body size and entry into the dauer stage (where C. elegans undergo developmental arrest at the third larval stage). Like other model organisms, C. elegans N2 has thus been modified by domestication.

3For a century, the only information available on the natural history of C. elegans was that it could be found in compost heaps and in rich humus. From 2000 onward, this began to change when researchers embarked on an extensive sampling of wild C. elegans populations. Their efforts were motivated by wanting to develop C. elegans into a model organism for evolutionary biology and ecology research, which could also leverage the tools and knowledge already acquired for this species. In turn, exploring the natural history of C. elegans provides a context for, and also informs, basic biological research conducted with this organism, such as studies of its genome, development, behavior and immune system. In this article, we summarize our current knowledge of the natural history of this C. elegans, show how the isolation of natural pathogens of C. elegans informed basic biological research, and discuss a number of open questions.

4C. elegans is found worldwide, predominantly in humid temperate areas (Figures 1 and 2A-D). This species was originally isolated in rich soil or compost, where it is mostly found in a non-feeding stage called the dauer. More recently, feeding and reproducing stages of C. elegans have been found in decomposing plant material, such as fruits and thick herbaceous stems (Figure 2E–G). These rotting substrates in their late stages of decomposition provide abundant bacterial food for the nematode. Like other model organisms, C. elegans is thus partially associated with human activity (cultivated fruits and stems, compost), but the species is also commonly found on stems and fruits in wilder settings, such as woods. New types of habitat and geographical locations may still be discovered.

5Two alternative life cycles have been described in the laboratory for C. elegans, depending on environmental conditions. If well fed, newly hatched individuals pass through four larval stages (L1, L2, L3, L4) and reach the adult stage after 3 days. Under stressful conditions (such as crowding, limited food supply, and heat stress), individuals can shift during the L1 stage to an alternative developmental route and enter a predauer stage (L2d), followed by the non-feeding diapause stage called dauer (an alternative L3 stage) (Figure 3). Dauer larvae are resistant to various stresses and can survive for several months without food. Upon their return to more favorable conditions, dauer larvae feed again and resume development.

6Population demographic surveys at the local scale in orchards and woods indicate that C. elegans has a boom-and-bust lifestyle. C. elegans metapopulations evolve in a fluctuating environment where optimal habitats are randomly distributed in space and time (Figure 4A). A cycle of colonization of a food source likely begins when one to several dauer larvae discover a fruit or stem, exit the dauer stage and seed a growing population of up to 104 feeding nematodes at different life-cycle stages (Figure 3). Some moderate-sized populations found in rotting fruits and stems do not contain any dauer larvae, but larger ones always include only adults, L1, L2d and dauers. As a food source runs low, dauers may leave it to explore the neighboring environment for new islands of resources. Most of them will fail.

7Developmental regulation and the behavior of dauer larvae are central to the C. elegans lifestyle. Dauer larvae display active locomotion and a specific behavior called nictation, where they stand on their tail and wave their body in the air. Remarkably, dauers may also congregate to form a column and nictate as a group (see the video; http://www.wormatlas.org/dauer/behavior/Images/DBehaviorVID4.mov). These behaviors are thought to help dauers to find passing invertebrate hosts that they can use for their dispersal, such as isopods, snails and slugs. Together, dauer physiology and behavior suggest that this developmental stage plays a key role in C. elegans' stress resistance, long-distance dispersal, and possibly its overwintering capacity.

8Over the year, in surveys performed in France and Germany, C. elegans populations in rotting fruits typically peak in the fall, with proliferation possible in spring through to early winter. This seasonal dynamic is consistent with that observed in a semi-natural habitat, such as a compost heap, where C. elegans population size varies over the year, from a dense population in autumn to one that declines and undergoes bottlenecks in winter (Figure 4B). These bottlenecks can lead to local extinctions. Where this happens, a new genotype may colonize the compost heap when favorable conditions return (Figure 4C).

9The spatio-temporal distribution of C. elegans genetic diversity reflects this boom-and-bust lifecycle coupled to active migration. At the global scale, C. elegans genetic diversity is low and displays little geographical structure, which must be due to long-distance migration, perhaps aided by larger hosts, such as birds, rodents or humans (see Box 1). In contrast, at a local scale, founder effects have a huge impact, with low diversity in a given sample contrasting with high molecular diversity but weak haplotype diversity at the scale of a square kilometer.

10C. elegans has a very peculiar mode of reproduction called androdioecy. It can reproduce either by self-fertilizing (selfing) hermaphrodites or by hermaphrodites (XX) breeding with males (X0). Hermaphrodites develop a female soma, and early in adulthood produce sperm in limited quantities (ca. 200–300 sperm cells). Males occur by non-disjunction of the X chromosomes at meiosis or in the progeny of male-hermaphrodite crosses. The non-disjunction of X chromosomes occurs overall at a low frequency (e.g., 0.1%) in the laboratory, and this frequency varies with genotype and the environment. In natural populations, outcrossing does occur but the overall proportion of males is low, generally similar to the estimated rate of chromosomal non-disjunction measured in the lab. In rare instances, outcrossing is evident from a high proportion of males or of heterozygotes in a given sample.

11This combination of metapopulation dynamics, a very low outcrossing rate and selection has a major impact on the C. elegans genome. These three factors reduce the effective population size, and thus genetic diversity, and can wipe out diversity in a large genomic region. Indeed, positively selected alleles carry with them a large part of the genome by linkage, at the scale of megabases, particularly in the center of chromosomes where recombination is low. For example, most C. elegans isolates carry a recently swept genotype spanning a large part of chromosome V; several other chromosomes have also experienced detectable sweeps. Because of high linkage disequilibrium due to selfing, selective sweeps may cover large fractions of chromosomes and have erased much genetic diversity. Background selection (the loss of non-deleterious variants due to selection against linked deleterious alleles) may further reduce linked genetic diversity in this selfing species. As a consequence, genome-wide pairwise genetic diversity is low (ca. 10−3/bp), especially compared to outcrossing relatives. Overall, this means that much of the C. elegans genome has been shaped by selection at linked sites, not directly by selection at the locus of interest. These extreme sweeps also mean that recent events in historical times, perhaps mediated by humans, may have had a significant impact on C. elegans.

12The selfing lifestyle of C. elegans also has an impact on phenomena such as inbreeding and outbreeding depression. Most animal species that reproduce by outcrossing suffer from inbreeding depression through the accumulation of deleterious recessive alleles. In a predominantly self-fertilizing organism, such as C. elegans, alleles are most often in a homozygous state due to repeated selfing, which allows deleterious recessive alleles to be purged by natural selection. Instead of inbreeding depression, wild C. elegans suffer from outbreeding depression. This phenomenon is best explained by the fact that selfing creates and maintains beneficial gene combinations, while outcrossing favors their disruption. Thus, the progeny of outcrossing are less fit than are the progeny of selfing. However, at least under certain circumstances of experimental evolution, outbreeding may contribute positively to C. elegans adaptation and inbreeding depression may reappear.

13Contributing also to a reduction in effective outcrossing, C. elegans possesses a widespread genetic incompatibility system, whereby a sperm poison (PEEL-1) arrests the development of embryos that do not synthesize the antidote (ZEEL-1). The zeel-1 and peel-1 genes are closely linked on chromosome I. The two allelic combinations, with both genes or without both genes, co-exist in the same geographical location, and the rest of the genome shows evidence of mixing. Yet this system, and other incompatibilities, may further contribute to lowering the effective outcrossing rate.

14C. elegans has evolved selfing from an ancestor that reproduced through females and males (see Box 2). Male-specific genes are overall conserved yet appear to be subject to relaxed selection. Although C. elegans males still sense female pheromones from closely related species, hermaphrodites no longer produce pheromones to attract males, suggesting a partial degeneration of hermaphrodite outcrossing. On the male side, some aspects of mating behavior also appear to have degenerated, such as the ability to deposit a copulatory plug. The effect, and the future, of males in C. elegans natural populations remains an intriguing question.

15C. elegans shares its microhabitat with arthropods and with other microorganisms (bacteria and fungi) and invertebrates (Table 1), including other nematodes such as Oscheius, Pristionchus, Panagrellus and other Caenorhabditis species, such as Caenorhabditis briggsae. If not with E. coli, it is noteworthy that C. elegans shares its rotting fruit habitat with two other top model organisms, Drosophila melanogaster and Saccharomyces cerevisiae. A specific association is actually found between another Caenorhabditis species and another Drosophila species: this nematode species, C. drosophilae, feeds on rotting cactus in desert areas and its dauer juveniles use a local Drosophila species as a vector to move between cacti.

16The C. elegans diet mainly consists of bacteria and small eukaryotes, but has not been further characterized. The trophic interaction between the nematode and its prey is not univocal: as C. elegans adults grow old, they can serve as food to the same microorganisms.

17Predators of C. elegans are also little studied. From co-occurrence in the wild and from laboratory experiments, possible natural predators include small arthropods, such as mites or springtails (Figure 2H), other nematodes, such as Pristionchus spp., and trapping nematophagous fungi (Table 1).

18C. elegans also constantly interacts with a variety of obligate and non-obligate parasites, such as fungi, microsporidia, bacteria and viruses (Table 1). These parasites infect their host via the two most exposed parts of the nematode, the cuticle and the intestine. Some non-invasive bacteria form a biofilm along the nematode's cuticle or directly stick to it. Other bacteria proliferate in the nematode gut, which may induce constipation and likely impairs nutrient uptake. The most intrusive parasites enter and proliferate inside the nematode body. Some pierce the cuticle (e.g., Drechmeria coniospora, Figure 2J), while others enter intestinal cells via the apical membrane (e.g., microsporidia and Orsay virus).

19Numerous characteristics of C. elegans biology arise from its natural history and explain why this animal has become such an emblematic species for scientists. While some of these characteristics were clear for the early pioneers of C. elegans research, others were only discovered later.

20First, like other model organisms, C. elegans rapidly proliferates in the wild upon encountering a food source, and can remain in a resting stage for long periods of time. In the lab, these features translate into organisms with short generation times (for their group) and good storage ability. Indeed, C. elegans displays a short generation time compared with many other nematodes and can simply be maintained on cholesterol-supplemented agar plates and fed E. coli. The natural ability of dauer larvae to live for several months offers a convenient option for short-term storage. C. elegans survives freezing, which enables their long-term storage at −80°C or in liquid nitrogen—an extremely useful feature for genotype storage and experimental design.

21Second, the natural mode of reproduction of C. elegans makes this nematode a dream come true for geneticists. Selfing permits the generation of homozygous stocks, without inbreeding depression. The ability to induce males makes genetic crosses easy.

22Third, C. elegans egg shells are naturally resistant to many environmental stresses, allowing the use of bleach. Bleaching is widely employed in the laboratory: in a population, only embryos will survive bleaching, thus allowing synchronization and removal of microbial contaminants.

23Finally, a convenient feature of the reference C. elegans strain, N2, that is found in some but not all C. elegans wild isolates or Caenorhabditis species is an active RNA interference pathway. This pathway was perhaps selected in response to encounters with viruses and transposons.

24The exploration of the natural habitat of C. elegans opens new doors to study its biology, for example, its development, physiology and behavior. Strikingly, for a top model organism such as C. elegans, a large proportion of its genes are, as yet, not associated with any mutant phenotype. Our exploration of its natural history advances biological research in two key respects. First, the quest for wild isolates of C. elegans has yielded deletion mutants for many genes. Second, our improved knowledge of its wild habitat gives us new environmental parameters to test for mutant phenotypes.

25As an example, an obvious ecological parameter is the interaction between C. elegans and its natural parasites, which have likely had a considerable impact on its recent evolution. From genomic analyses, expanded gene families, such as ubiquitin-adaptor genes, were hypothesized to have evolved under diversifying selection provided by pathogens or toxins. Recently, a ubiquitin-mediated response was indeed shown to take part in the C. elegans defense against two wild parasites, Nematocida parisii (microsporidia) and the Orsay virus, as well as in behavioral avoidance of a bacterial pathogen.

26The isolation of the Orsay virus, an RNA virus, was also instrumental in demonstrating in C. elegans the antiviral role of genes involved in RNA interference (so far studied in response to artificially administered double-stranded RNAs). Moreover, an exploration of natural variation in C. elegans identified, by a genome-wide association study, a locus called drh-1. This locus was shown to carry in many C. elegans isolates a deletion that renders them permissive for Orsay virus replication. The drh-1 gene is homologous to the RIG-I family of genes that encode mammalian viral sensors. In C. elegans, this likely upstream sensor of viral RNAs was shown to activate an antiviral small RNA pathway. Further molecular immunity pathways will likely be unraveled by studying the interaction of C. elegans with its natural associates and its corresponding natural phenotypic variation.

27How C. elegans development and behavior are modulated by the environment will also be greatly enriched by a better knowledge of its natural history. A key component of its population cycles that remains to be explored is the regulation of dauer entry and exit by relevant environmental factors, besides crowding and high temperature. Behaviorally, the dauer larva has been little studied; for example, we still know little about its olfactory responses. Recent work has demonstrated the importance of a specific set of neurons (called IL2 cells) and of cholinergic transmission in controlling dauer nictation. When present at high density, the dauers can nictate in large groups. The mechanistic basis and environmental regulation of this host-seeking behavior are still poorly understood.

28Another striking resource provided by the exploration of C. elegans natural history is the phenotypic diversity at the intra- and inter-specific level (see Box 2). As an example, the mode of reproduction has been modified multiple times in the Caenorhabditis genus. While most species reproduce through males and females, C. elegans, C. briggsae and Caenorhabditis tropicalis are exceptions by having evolved spermatogenesis in a female soma. Phylogenetic studies and molecular dissection of germ line sex determination argue for a convergent evolution of androdioecy, three times in the Caenorhabditis genus. Another opening provided by the huge rise in Caenorhabditis species and strain collections is the ability to study reproductive isolation and speciation mechanisms in this group.

29Despite intense laboratory work on C. elegans, many important features of this species in the wild remain mysterious (Box 1). We still know very little concerning this nematode's feeding, dauer entry and exit cues, migration, overwintering, and we have only just started to glimpse organisms associated with it, be they food or pathogens. A greater understanding of its natural history could greatly inform our studies of its physiology, behavior and immunity. Comparative studies between Caenorhabditis species also promise to provide a better understanding of the role and the evolutionary history of specific gene families and other orphan genes.

30In addition, improving our understanding of C. elegans' natural history will help establish it as a model species for evolutionary and ecological studies. The diversity of modes of reproduction in the Caenorhabditis genus makes them attractive models for studies of the causes and consequences of such variation. Its boom-and-bust lifestyle is favorable for studies of C. elegans metapopulation dynamics in different environments.

31To conclude, knowing more about C. elegans' life outside of the Petri dish is opening new avenues to laboratory research. As for other model organisms, it also helps us to realize how idiosyncratic some features of the C. elegans N2 reference strain are—only a snapshot in laboratory evolution.