A large close relative of C. elegans is slow-developing but not long-lived
Variation in body size is thought to be a major driver of a wide variety of ecological and evolutionary patterns, including changes in development, reproduction, and longevity. Additionally, drastic changes in natural context often have profound effects on multiple fitness-related traits. Caenorhabditis inopinata is a recently-discovered fig-associated nematode that is unusually large relative to other members of the genus, including the closely related model system C. elegans. Here we test whether the dramatic increase in body size and shift in ecological context has led to correlated changes in key life history and developmental parameters within this species.
Using four developmental milestones, C. inopinata was found to have a slower rate of development than C. elegans across a range of temperatures. Despite this, C. inopinata did not reveal any differences in adult lifespan from C. elegans after accounting for differences in developmental timing and reproductive mode. C. inopinata fecundity was generally lower than that of C. elegans, but fitness improved under continuous-mating, consistent with sperm-limitation under gonochoristic (male/female) reproduction. C. inopinata also revealed greater fecundity and viability at higher temperatures.
Consistent with observations in other ectotherms, slower growth in C. inopinata indicates a potential trade-off between body size and developmental timing, whereas its unchanged lifespan suggests that longevity is largely uncoupled from its increase in body size. Additionally, temperature-dependent patterns of fitness in C. inopinata are consistent with its geographic origins in subtropical Okinawa. Overall, these results underscore the extent to which changes in ecological context and body size can shape life history traits.
KeywordsBody size Life history C. elegans Lifespan Heterochrony
Feminization of germline
Generalized linear model
Fourth larval stage
Likelihood ratio test
Trade-offs dominate life history evolution. Organisms have access to limited energy resources, and these must be allocated in a balance between self-maintenance and reproductive output. In keeping with the expectation that different distributions of life history traits (such as age of maturity, reproductive duration, and age-specific fecundity, among others) should be sensitive to different distributions of selective pressures on those traits, a huge diversity of patterns among life history traits has emerged across the broad scope of animal diversity [1, 2, 3, 4, 5]. As a consequence, many organisms exhibit well-documented correlations among traits such as fecundity and survival [6, 7, 8], fecundity and developmental rate [1, 9, 10, 11], and reproductive quantity and quality [12, 13].
Body size is a particularly potent component of life history syndromes. Body size is usually correlated with a multitude of fitness-related traits including developmental rate, offspring number, offspring size, gamete size, and lifespan [14, 15, 16, 17]. Body size is also known to covary with physiological traits, such as metabolic rate, thought to underlie trade-offs among life history traits [15, 17]. These factors in turn generate allometric relationships that appear to explain scale-based trends for a wide variety of traits across many orders of magnitude . Indeed, body size appears to be a central component of broad macroevolutionary trends among lineages over geological timescales . But which is cause and which is effect? To what extent does change in body size due to selection on body size per se lead to collected changes in such a wide array of life history traits and to what extent does body size change because of selection acting directly on these traits?
Life history theory suggests that selection for increased body size can be balanced against the benefits of faster reproduction and the costs of lower offspring viability and lower initial fecundity , weighed against a backdrop of differential allocation of physiological and metabolic resources to each of these processes and to growth itself [17, 19]. At the same time, selection on body size itself must be mediated via environmental factors such as resource availability and/or predation . Although these various causes are not mutually exclusive and likely overlap, the proximate and ultimate causes of body size change—particularly the relationship between these two—remain largely unresolved.
However, body size is not the only factor known to influence life history traits—environmental and ecological change is also expected to promote life history evolution. Indeed, whenever a change in environment impacts the optimal survival and fecundity of different age classes, then life histories will evolve in response . Changes in both abiotic (temperature, salinity, humidity, etc.) and biotic (predation pressure, resource abundance, pollinator density, etc.) environments can impact life history strategies: for instance, differential predation on larval stages in guppies due to spatial differences in predator abundance promoted changes in reproductive effort across populations [21, 22]. Furthermore, the extent of regularity in temporal environments (such as season length) is thought to underlie bet-hedging strategies (such as the germination time decision), and life histories are expected to evolve in response to changes in the timing of environmental cycles as well . Thus, environmental and ecological context plays a critical role in the evolution of life histories. How do changes in ecological context interact with the constraints imposed by body size to promote the evolution of life history traits?
The nematode Caenorhabditis elegans has for decades been an important model for genetics, development, and biology in general . However, the degree and extent of trade-offs between body size and other life history traits in systems like C. elegans remain largely unknown and/or have generated somewhat ambiguous or contradictory results [24, 25, 26, 27, 28, 29, 30, 31, 32]. Further, because nearly all known members of this genus share a common natural ecological niche of rotting plant material , it has not been possible to use a comparative approach to investigate how change in ecological circumstances might drive changes in the relationship between body size and life history . Here, we address this question by taking advantage of a highly phenotypically and ecologically divergent close relative of C. elegans: the recently discovered fig-associated nematode C. inopinata.
C. inopinata (formerly known as C. sp. 34) is remarkable in that it displays tremendous ecological and phenotypic differences compared to its close relatives [34, 35]. Compared to other Caenorhabditis, C. inopinata is huge: it can grow to be nearly twice as long as other members in the genus [34, 35]. C. inopinata also develops nearly half as quickly, has sperm three times the size, and embryos 20% longer than C. elegans . Furthermore, in contrast to the rotting-plant material ecological niche of C. elegans and other Caenorhabditis species , it thrives in the fresh, intact Okinawan figs of Ficus septica [34, 35, 37]. C. inopinata thus appears to have experienced a radically different selective environment that has led to its highly divergent suite of life history traits. And, as C. inopinata is much larger in size and develops much more slowly than its close relatives, it can therefore be used as a natural system to test the predictions of life history theory using a comparative approach. Here, we do just this by describing the developmental timing, lifespan, fecundity, and viability of C. inopinata and C. elegans at multiple temperatures.
C. inopinata develops more slowly yet does not differ from C. elegans in lifespan and reproductive duration
C. inopinata is sperm-limited and reveals higher fitness at higher temperatures
When examining the relationship between developmental rate and fecundity, the intrinsic rate of increase (r) is likely a better measure of fitness than total fecundity (R0) [1, 45]. Under this approach, fitness is a function of age-specific fecundity and viability, and the age of first reproduction can highly influence the population growth rate . So although C. inopinata and C. elegans have comparable brood sizes under continuous mating conditions, they likely differ in fitness because of their different developmental rates. Indeed, despite their comparable brood sizes, C. elegans has a rate of increase (r=1.54, 95% confidence interval (CI)=1.26-1.72) that is over twice as high as C. inopinata (r=0.66, 95% CI=0.54-0.74). This difference in fitness is even greater in mating conditions with just overnight access to males (C. elegans r=2.09, 95% CI=1.88-2.24; C. inopinata r=0.63, 95% CI=0.55-0.69). Thus continuous access to males is not sufficient to overcome the detriment to fitness due to slow development in C. inopinata.
Most C. elegans genes with life history phenotypes conventionally associated with large body size intersect with only one phenotype
Possibly because it is both obvious and easy to measure, body size variation has been studied extensively for centuries. The range in body size across the tree of life is so immense as to demand explanation (21 orders of magnitude [16, 50]), and this incredible diversity has spawned a vast and rich literature attempting to comprehend its origins and maintenance. One major conclusion from this research program is that body size is correlated with nearly every trait, such that long-established relationships between body size and growth, reproduction, and lifespan underscore a prominent role for body size in the evolution of life histories [14, 15, 50]. Here, we found that an exceptionally large and ecologically divergent close relative of C. elegans exhibits slow growth and low fecundity across a range of temperatures yet is not long lived. Together with the extensive C. elegans literature and the foundations of life history theory, these observations can contribute to our understanding of the causes and consequences of large-scale changes in body size and ecological divergence.
The impact of ecological divergence on life history syndromes
Changes in ecological context are expected to impact life history traits. Here, we described the life history traits of C. inopinata, an organism that occupies an exceptional ecological niche when compared with its close relatives. Most Caenorhabditis species, including C. elegans, thrive in rotting plant material . C. inopinata proliferates in fresh F. septica figs, living in close association with its pollinating fig wasps [34, 37]. How might this dramatic shift in ecological context explain the patterns of life history traits observed here?
C. inopinata grows at nearly half the rate as C. elegans (Fig. 1). One possible explanation for the divergence in developmental rate could be its novel natural context. Most Caenorhabditis proliferate in the ephemeral environments of rotting vegetation ; it is thought that the rapid turnover and spatial patchiness of its bacterial food resources has driven its rapid development, high fecundity, and its resource-dependent decision to enter the dispersal dauer larval stage . C. inopinata, conversely, grows in the presumably more stable environment of the fig lumen and obligately disperses on pollinating fig wasps [34, 37]. Figs generally take weeks to develop , although it is unclear how many generations of worms occur within a single fig. It is then possible that the extreme divergence in developmental rate might be connected to differences in the transience of resource availability between these environments. Furthermore, as C. inopinata animals disperse to new figs via pollinating wasps , their life cycle is necessarily closely tied to patterns of wasp development and emergence, consistent with correlations among Ceratosolen fig wasp and C. inopinata developmental stages that have been found in previous field studies . This contrasts with C. elegans and its other close relatives, who disperse on a plethora of invertebrate carriers (including isopods, myriapods, and gastropods) . Future longitudinal field studies of single fig trees at finer temporal resolution will be required to determine the relative paces of fig, fig wasp, and nematode development in nature and to test hypotheses regarding the ecological drivers of heterochrony.
Overall, C. inopinata reveals a lower viability (Fig. 5) and fecundity (Fig. 4) than C. elegans in laboratory conditions, although lifetime access to males greatly improves C. inopinata fecundity (Fig. 4b). How might its unique ecology underlie these patterns? A particularly interesting avenue to pursue is based on the observation that wild bacteria associated with Caenorhabditis can have both positive or negative influences on fecundity and growth [54, 55] and that different species of Caenorhabditis are associated with different microbes in nature . Thus the nutritional environment can have a profound effect on fitness and life history traits. The natural microbial food of C. inopinata is currently unknown. As C. inopinata exhibits reduced gonads in laboratory culture , it may be experiencing nutritional deficiencies. The reduced fecundity of C. inopinata may then reflect a plastic response to an adverse environment as opposed to a trade-off with increased body size. The potential influence of natural microbial associates of Ficus septica figs on C. inopinata fitness affords an exciting opportunity for future research.
The large body size of C. inopinata could also be more directly connected to its novel environmental context. Body size is broadly correlated with fecundity across nematode phylogeny [56, 57, 58], and the fig microcosm may represent an environment with less selective pressure on body size than rotting vegetation. Hence C. inopinata may be large because its ecological context reveals relaxed selection on body size compared to its close relatives. Such pressures may entail predation or pathogens—various mites, fungi, bacteria, and viruses are known to prey on or infect Caenorhabditis . Figs maintain many defenses against antagonistic invertebrates and microorganisms [59, 60, 61], and it is likely that this environment harbors a less diverse community than rotting vegetation. In this case any trade-offs among predation or infection avoidance and body size would be lifted, facilitating body size change resulting from selection on increased fecundity. However, within C. elegans, no evidence for correlations between body size and fecundity was found after artificial selection for large body size . Thus it is entirely possible that such trade-offs are not driving body size change in this case and that other factors (adaptive or not) are responsible upon shifting to the fig environment. Further studies explicitly addressing the possible ecological determinants of body size change will be needed to delineate these possibilities.
Developmental timing and body size
It makes intuitive sense that larger organisms should develop more slowly. Being more massive, presumably more cell divisions and/or biosynthetic reactions must take place for their construction and it therefore follows that their development should take longer than smaller organisms. And this intuition bears out across vast phylogenetic distances: from bacteria to sequoias, body size covaries with generation time . Here, we found that in all temperatures, C. inopinata grows nearly half as quickly as C. elegans, consistent with previous observations (Fig. 1; [34, 35]). Indeed, C. inopinata needs to be grown at 30°C to approach a rate of development comparable to that of C. elegans when grown at 20°C. Thus, the observation that this very large species also develops much more slowly than its close relatives is in line with decades of allometric studies. Further, as cell size is coordinated with cell division decisions in multiple organisms [62, 63], body size change could occur even in the absence of cell number change through the modification of cell cycle timing. This may explain the case of C. inopinata, as previous observations observed no change in cell number despite its large size and slow development .
However, there are reasons to suspect slow development may not underlie large body size in this case. It has been argued that the allometric trends observed in birds and mammals cannot be easily extended to poikilotherms because of difficulties in comparing physiological time due to rapid change in metabolic rates . More notable is the common observation that developmental timing can be decoupled from body size in C. elegans. Most mutations in C. elegans that extend body length do not also slow the rate of growth: only 29% of the genes in the C. elegans genome known to control body length also promote slower development (Fig. 6). Furthermore, experimental evolution and mutation accumulation studies in C. elegans and C. briggsae have not generally reported correlated changes in body size and developmental timing [25, 27, 28, 64]. Thus, it appears that body size and rate of growth need not be strongly coupled in Caenorhabditis and that the relationship between these traits observed in C. inopinata may not necessarily be causative. Furthermore, as mentioned above, the slow growth of C. inopinata may also be better understood with respect to its natural ecological context (or could also be due to nutritional deficiencies in a laboratory context); framing the causes of heterochronic change with respect to this context is an exciting avenue for future studies.
Reproduction and body size
The relationship between body size and reproduction varies both within and between taxa. In birds and mammals, larger species tend to have lower fecundities than smaller species . Conversely, body size appears to be positively correlated with fecundity in insects  and nematodes [56, 57, 58]. C. inopinata was generally found to have lower brood sizes than C. elegans across a range of temperatures (Fig. 4a), although continuous mating greatly improves fecundity in C. inopinata (Fig. 4b). The relatively low fecundity of C. inopinata is then incongruent with patterns of fecundity and body size that have been previously reported among nematodes [56, 57, 58]. C. inopinata’s gonochoristic mode of development cannot explain its low brood size, as multiple male/female species of Caenorhabditis have been reported to have higher brood sizes [42, 43, 66, 67, 68, 69]. However, the sperm-limited fecundity of C. inopinata (Fig. 4b) is consistent with previous observations with the gonochoristic C. remanei [42, 43]. It is possible that the evolution of extreme body size in the case of C. inopinata reveals a trade-off with reproductive output, wherein resources usually allocated to progeny have instead been shifted to increase self-maintenance and growth. Yet most genes known to regulate body length in the C. elegans genome do not appear to have a pleiotropic role in brood size (only 28% do; Fig. 6). This is also consistent with experimental evolution studies in Caenorhabditis , wherein fecundity and body size do not necessarily trade-off. So again, the precise causal relationship here bears further study. Additionally, as mentioned above, patterns of fecundity in C. inopinata may also be better understood with respect to its natural environment. It is possible that C. inopinata may be experiencing nutritional deficiencies in the laboratory context, and future studies with natural associated microbes will help to inform this possibility.
Lifespan and body size
Lifespan is often positively correlated with body size, and from an allometric perspective is usually thought to be regulated by variation in developmental and metabolic rates [15, 17]. And although the age of maturity is sensitive to selection under a range of trait distributions in life history theory , from an evolutionary perspective it is thought that late-life traits are generally not subject to selection as its strength falls to zero once reproduction ends . Despite its large size and slow development, C. inopinata was found to have only a marginally longer lifespan than C. elegans (Fig. 2). And, when differences in developmental timing and reproductive mode are taken into account, C. inopinata adult lifespan is not significantly different from that of C. elegans (Fig. 2b). The lack of lifespan change in this system is consistent with the view that lifespan is under weak selection, as C. inopinata has experienced dramatic change in many other traits under its novel ecological context [34, 35, 37]. Indeed, most lifespan-extending mutations identified in C. elegans have not been associated with pleiotropic effects on body size (Fig. 6). Similarly, experimental evolution studies in C. elegans show no correlated responses in lifespan upon artificial selection on early fecundity  and body size . Additionally, no relationships between lifespan and fecundity have been found in mutation-accumulation lines  or among wild isolates . These observations are inconsistent with the antagonistic pleiotropy explanation of aging, which posits that the greater fitness contribution of early life survival and reproduction leads to late life deterioration because of negative genetic correlations of these traits . Rather, lifespan appears to be possibly largely uncoupled from fitness-related traits in this group, consistent with the unchanged longevity observed in C. inopinata. However, the nutritional caveats in this system noted in the above interpretation of observed patterns of fecundity also apply here. It is possible that C. inopinata will be longer-lived under different rearing conditions, and measurements of lifespan of C. inopinata raised on bacterial food originating from its natural context need to be performed.
Variation, inbreeding, and fitness-related traits
The observations reported here constitute comparisons of two species with each represented by one genetic strain. How does this fact impact the implications of this work discussed here? From a comparative phylogenetics perspective, this sample size is simply insufficient to make broad generalizations about patterns of life history trait covariation in the Caenorhabditis genus. It remains possible that C. inopinata represents an outlier that defies meaningful biological trends that we would otherwise not capture because of our limited phylogenetic sample. Furthermore, as we have only interrogated one strain of C. inopinata, this strain may also not be representative of this species as such. Ultimately, broad sampling and measures across Caenorhabditis phylogeny will be needed to make solid claims along these lines. However, can previous observations of variation in fitness-related traits in Caenorhabditis help to overcome this limitation or better inform these results?
C. inopinata is an exceptional Caenorhabditis species with respect to its large body size [34, 35], and most members of the Elegans group are difficult to distinguish morphologically [71, 72]. But, as mentioned above, there is variation both within and between Caenorhabditis species in fecundity (Additional file 1: Figures S5-S6; Additional file 2) and developmental rate (Additional file 1: Figures S7-S8; Additional file 3). Among four recent studies that measured fecundity in Caenorhabditis (including this study) [38, 42, 73] , which includes 24 strains among four species, our estimate of C. inopinata fecundity at 20°C is the lowest (Additional file 1: Figures S5-S6). But as C. inopinata and other gonochoristic Caenorhabditis are sperm-limited (Fig. 4) [42, 43], this is likely an underestimate of its reproductive capacity. However, even limited access to males can maintain brood sizes in C. remanei wild isolates that exceed those observed in many strains of selfing species (Additional file 1: Figures S5-S6) , including C. elegans N2, which is considered a domesticated laboratory strain . With respect to previous studies of developmental timing, C. inopinata is even more extreme in its divergence from its close relatives, developing at about half the rate as the next slowest strain (C. tropicalis JU1630) among those considered in two previous publications (Additional file 1: Figures S7-S8) [38, 75]. Thus a broader phylogenetic context with more Caenorhabditis species and strains also suggests that this C. inopinata strain harbors an exceptionally slow developmental rate and a low (but sperm-limited) fecundity in addition to its large body size.
Gonochoristic Caenorhabditis species are susceptible to inbreeding depression [76, 77]. Could C. inopinata strain NKZ2 have experienced inbreeding depression during laboratory culture that impacts estimates of fitness-related traits? This possibility cannot be definitively ruled out, and an important caveat of these results is that inbreeding depression may influence the patterns of life history traits observed here. However, there are reasons to suspect inbreeding depression may not be a major influencing factor in this case. Primarily, C. inopinata NKZ2 is a wild isolate that has not been deliberately inbred, and it is a strain derived from multiple founding individuals . If inbreeding depression were the only cause of declines in fitness-related traits, then C. inopinata NKZ2 should be expected to have higher fecundity than isofemale wild isolates and inbred lines of other gonochoristic species as it is derived from more than one founding individual. As this is not the case (Additional file 1: Figures S5-S6), this suggests that inbreeding depression alone may not be driving these patterns. Furthermore, the ecology of C. inopinata may render it less susceptible to inbreeding depression through its obligate dispersal on fig wasps . Despite this, the embryo-to-adult viability of C. inopinata is low across multiple temperatures (Fig. 5), which is itself suggestive of inbreeding depression. As we did not measure larval survival directly, it is unclear at what stage animals are undergoing developmental arrest. Additionally, it is possible that behavior could also be driving these results—larvae that crawl off plates and do not mature will artificially deflate viability measurements. However, as dead embryos have been anecdotally observed in these cultures, embryonic lethality is likely driving at least some of these patterns. It also remains possible that the culture conditions designed for C. elegans are insufficient for the reliable development of this species, and native fig, wasp, or microbial factors may be needed for robust C. inopinata viability. Regardless, future studies that rear this species in more ecologically-relevant culture conditions, use C. inopinata inbred lines and wild isolates, and implement population genomic approaches in natural populations will be needed to disentangle these possibilities.
Pleiotropy and life history syndromes
Here we note relationships among life history traits across two species of Caenorhabditis. What role does pleiotropy play in the patterns observed here, and are life history syndromes the result of indirect selection, direct selection on multiple characters, or both? Because this study does not directly interrogate the genetic basis of these traits, the underlying genetic causes of these relationships are undetermined. However, the vast background information associated with the C. elegans model system can provide context to generate hypotheses regarding the evolution of life history strategies. How many genes have pleiotropic effects on multiple life history traits in C. elegans? As discussed above, most genes associated with one of four life history WormBase phenotypes  that might be associated with large body size (“slow growth”, “reduced brood size,” “extended life span,” and “long”) do not intersect with one another (Fig. 6). That is, most genes with any of these phenotypes are associated with only one of the four (74%; Fig. 6), and only a fraction of them reveal evidence of pleiotropic effects (26%; Fig. 6). Thus these life history traits appear to be largely genetically decoupled in this group, and pleiotropy need not underlie the correlated evolution of these traits. However, as some genes do influence multiple phenotypes (Fig. 6), pleiotropy may still contribute to the evolution of life history syndromes in this case. Furthermore, as many C. elegans studies are largely concerned with only a few phenotypic traits of interest, these results are likely to underestimate the extent of pleiotropy among C. elegans genes. Regardless, future work investigating the genetic bases of these traits in C. inopinata will be needed to understand the role of pleiotropy in shaping life history syndromes.
Temperature-dependent patterns of fitness-related traits in C. Inopinata
Notably, C. inopinata was more fit at higher than lower temperatures (Fig.4a, Fig. 5). Temperature-dependent plasticity of fitness-related traits varies both within and between species in Caenorhabditis, and these patterns often coincide with ecological context. Within C. briggsae, there are definable clades that are genetically structured by latitude [78, 79], and these wild isolates reveal temperature-dependent patterns of fecundity that are consistent with their geographical origin . Additionally, the tropical species C. nigoni [66, 81] and C. tropicalis  have higher fitness at warmer temperatures. As C. inopinata has only been found in the subtropical islands of Okinawa [34, 35], its temperature-dependent patterns of fitness are consistent with these previous observations. And further, the temperatures where C. inopinata has shown the highest fitness here are comparable to natural Ficus septica fig temperatures measured in nature . Additionally, C. inopinata needs to be grown at 30°C to approach a rate of development comparable to that of C. elegans when grown at 20°C. Could the slow growth of C. inopinata more appropriately be interpreted as an adjustment of optimal developmental timing imposed by its subtropical environment? This explanation is appealing as thermal plasticity in growth is widespread in ectotherms, and wild F. septica fig interiors were found to harbor temperatures of 29°C on average . However, if temperature were a major driver of a universal, optimal developmental rate in Caenorhabditis, then we would expect to see much slower development in tropical strains and species than has been reported (Additional file 1: Figure S8). Indeed, there are no detectable developmental timing differences between tropical and temperate strains of C. briggsae (Additional file 1: Figure S8) , which nonetheless do reveal clade-specific, temperature-dependent differences in fecundity . And although C. tropicalis, which is typically found in warmer climates than C. inopinata [51, 83], harbors a slower developmental rate than C. elegans (Additional file 1: Figure S7-S8) , it remains far faster than that of C. inopinata (Additional file 1: Figure S7-S8). Thus, it seems more likely that the slow growth of C. inopinata is connected to its novel ecological context in F. septica figs, its exceptionally large body size, or both, rather than its subtropical locality alone. Regardless, as a close relative of C. elegans, this species is well positioned for uncovering the genomic bases of temperature adaptation.
Body size and ecological divergence are major drivers of evolutionary change in multiple taxa, and such changes often co-occur with widespread change in life history traits. Here, we examined the life history traits of a large, ecologically-divergent close relative of C. elegans. We found that C. inopinata develops at nearly half the rate as C. elegans, revealing a likely trade-off between growth and body size. Conversely, longevity does not evolve as part of correlated response to selection on body size in this system, consistent with previous studies and indicative of genetic decoupling of longevity from other life-history traits. Future studies that situate these systems within their natural ecological contexts will be needed to fully disentangle matters of cause and effect among the traits that constitute life history strategies. Taken together, these observations reveal that drastic change in ecological context and body size do not necessarily have an all-encompassing impact on life history syndromes.
Strains and maintenance
Animals were maintained on Nematode Growth Media (with 3.2% agar to discourage burrowing) supplemented with Escherichia coli strain OP50-1 for food. The C. inopinata wild isolate strain NKZ2  was utilized for all observations in this report. C. elegans N2 and the obligate outcrossing C. elegans fog-2(q71) JK574  mutant strain were also used for most comparisons. Notably, C. elegans is hermaphroditic, while C. inopinata is male/female or gonochoristic. This makes interspecific comparisons problematic. Thus the fog-2(q71) mutation, which prevents spermatogenesis only in hermaphrodites but promotes no obvious somatic defects in either sex , was used to control for differences in reproductive mode in various comparisons of life history traits.
The timing of four developmental milestones (hatching, fourth larval stage (L4), adult stage/young adulthood, and the onset of reproduction/reproductive adulthood) was measured at four temperatures: 15°C, 20°C, 25°C, and 30°C. For synchronization, mid-stage embryos (blastula to 1.5 fold stage) were picked from plates cultured at 25°C to new plates and then shifted to the given rearing temperature. Plates were then monitored hourly (for hatching) and then daily (for L4, young adulthood, and reproductive adulthood) for the onset of developmental milestones. Male tail and female/hermaphrodite vulva morphologies were used to define L4 and young adult stages. The onset of reproduction was scored only among females and hermaphrodites by the presence of embryos in the uterus. Plates were assayed until the number of individuals at or older than a given milestone did not increase for two hours or days. Animals who failed to reach a given milestone were not used for subsequent analysis. For analysis, animals were plotted by their developmental status (“0” = yet to reach milestone; “1” = reached milestone) over time and logistic regression was used to estimate the median time to a given event via the “glm” function (using a binomial distribution) in the R statistical language. This models approach was used for hypothesis testing and for calculating 95% confidence intervals (see Additional file 4; data are available in Additional files 5 and 6).
Synchronized animals were generated by allowing gravid females/hermaphrodites (20 C. elegans hermaphrodites or C. elegans fog-2(q71) pseudo-females per plate; about 100 C. inopinata females per plate) to lay for 2-3 hours. After a few days, synchronized L4 virgin females/hermaphrodites were moved to new plates, with about 30 nematodes per plate. All animals were transferred every day for the first 4-5 days of adulthood as hermaphrodites reproduced. Subsequently, animals were scored every 1-3 days as either living or dead up until the point that all animals had died. All measurements were performed at 25°C. The number of days alive after egg-laying was taken as the measure of total lifespan. Lifespan and longevity studies in C. elegans are often concerned with the basis of aging, which is generally thought to largely occur in adulthood after developmental growth . Thus we report here both total lifespan (starting at embryogenesis) and adult lifespan (starting at the onset of maturation). As C. inopinata and C. elegans display different rates of developmental growth, this also allows a comparison of the rate of aging in adults that accounts for this difference. Adult lifespan was taken as the total lifespan minus two (C. elegans) or four (C. inopinata) days, as C. inopinata develops at about half the rate as C. elegans. Statistical analyses were performed as in , with the survival package for the R statistical language being used to generate survivorship curves and the coxme package being used to generate Cox proportional hazard models and perform hypothesis tests (see Additional file 4; data are available in Additional file 7).
Daily offspring production was measured following overnight mating and under continuous exposure to males. For all observations, L4 C. inopinata NKZ2 and C. elegans fog-2(q71) animals raised at 25°C were isolated and raised for one (C. elegans) or two (C. inopinata) days to adulthood (see above). For overnight mating, single adult females/pseudo-females were shifted to the given experimental rearing temperature and mated with six males overnight. Brood sizes were measured at 15°C, 20°C, 25°C, and 30°C. The next day males were removed. Every day, embryos and larvae were counted, and egg-laying females were moved to new plates. New progeny were scored until females stopped laying for at least one (C. elegans) or two (C. inopinata) consecutive days. Continuous mating conditions were similar, except that single females were always in the presence of six males. Males that crawled up the side of the plate or otherwise died before the female stopped laying embryos were replaced with young adult males. The continuous mating observations were performed at 25°C. The instantaneous rate of natural increase  was calculated in Python as in  using life tables for C. elegans and C. inopinata constructed from the viability, fecundity, and lifespan data developed here (see Additional file 8; data are available in Additional files 9, 10, 11 and 12).
Nematode embryos were synchronized by allowing gravid females/hermaphrodites (20 C. elegans hermaphrodites or C. elegans fog-2(q71) pseudo-females per plate; about 100 C. inopinata females per plate) to lay for 2-3 hours. After the parents were removed, the number of embryos per plate were counted, and the plates were shifted to their respective rearing temperatures (15°C, 20°C, 25°C, or 30°C). L4 and adult worms were counted 4-5 days later. This fraction of mature worms/initial worm embryos was reported as the viability. Data are available in Additional file 13.
Intersection of WormBase phenotypes related to life history traits among C. elegans protein-coding genes
Functional annotations for all C. elegans protein-coding genes were retrieved using the simplemine tool in WormBase (http://www.wormbase.org/tools/mine/simplemine.cgi, link labeled “query all C. elegans”; Additional file 14). Genes with mutant or RNAi phenotypes “long,” “slow growth,” “extended life span,” and “reduced brood size” were extracted, and a spreadsheet denoting the intersection of these four phenotypes for every gene that included at least one of these phenotypes was created with Linux (see Additional file 15 for software) and Perl (https://github.com/religa/stats/blob/master/merge) tools (see Additional files 14 and 15 for data). The UpSetR package  was used to make Fig. 6 with this data (Additional file 4).
We thank two anonymous reviewers whose helpful comments improved this manuscript.
This work was supported by funding from the National Institutes of Health to GCW (5F32GM115209-03) and PCP (R01 GM-102511; R01 AG049396). This funding body had no role in the design of the study; the collection, analysis, and interpretation of data; and in the drafting and revising of the manuscript.
GCW and PCP designed the experiments; GCW and EJ performed the experiments; GCW analyzed the results; GCW and PCP wrote the paper. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.Stearns SC. The evolution of life histories, vol. 249. Oxford: Oxford University Press; 1992.Google Scholar
- 2.Roff DA. Life history evolution; 2002.Google Scholar
- 4.Emlen JM. Age specificity and ecological theory. Ecology. 1970;51(4):588–601.Google Scholar
- 5.Charlesworth B. Evolution in age-structured populations. Cambridge: Cambridge University Press; 1980.Google Scholar
- 6.Clutton-Brock TH, Guinness FE, Albon SD. Red deer: behavior and ecology of two sexes. Chicago: University of Chicago press; 1982.Google Scholar
- 7.Reznick D. Costs of reproduction: an evaluation of the empirical evidence. Oikos. 1985;44:257–67.Google Scholar
- 9.Reznick D. The structure of guppy life histories: the tradeoff between growth and reproduction. Ecology. 1983;64(4):862–73.Google Scholar
- 10.Robinson B, Doyle R. Trade-off between male reproduction (amplexus) and growth in the amphipod Gammarus lawrencianus. Biol Bull. 1985;168(3):482–8.Google Scholar
- 12.Fleming IA, Gross MR. Latitudinal clines: a trade-off between egg number and size in Pacific salmon. Ecology. 1990;71(1):1–11.Google Scholar
- 13.Sadras VO. Evolutionary aspects of the trade-off between seed size and number in crops. Field Crop Res. 2007;100(2):125–38.Google Scholar
- 14.McMahon TA, Bonner JT. On size and life. New York: Scientific American Library; 1983.Google Scholar
- 15.Calder WA. Size, function, and life history. Cambridge: Harvard University Press; 1984.Google Scholar
- 16.Schmidt-Nielsen K. Scaling: why is animal size so important? Cambridge: Cambrige University Press; 1984.Google Scholar
- 17.Peters RH. The ecological implications of body size, vol. 2. Cambridge: Cambridge University Press; 1986.Google Scholar
- 19.Sibly RM, Calow P. Physiological ecology of animals. 19. Oxford: Blackwell Scientific Publications; 1986.Google Scholar
- 34.Kanzaki N, Tsai IJ, Tanaka R, Hunt VL, Tsuyama K, Liu D, et al. Biology and genome of a newly discovered sibling species of Caenorhabditis elegans. Nat Commun. 2018;9:1–12. https://www.nature.com/articles/s41467-018-05712-5
- 35.Woodruff GC, Willis JH, Phillips PC. Dramatic evolution of body length due to post-embryonic changes in cell size in a newly discovered close relative of C. elegans. Evolution Letters. 2018;2:427–41.Google Scholar
- 37.Woodruff GC, Phillips PC. Field studies reveal a close relative of C. elegans thrives in the fresh figs of Ficus septica and disperses on its Ceratosolen pollinating wasps. BMC Ecol. 2018;18:1–16. https://bmcecol.biomedcentral.com/articles/10.1186/s12898-018-0182-z
- 43.Diaz SA, Haydon DT, LindstrÖM J. Sperm-limited fecundity and polyandry-induced mortality in female nematodes Caenorhabditis remanei. Biol J Linn Soc. 2010;99(2):362–9.Google Scholar
- 45.Charlesworth B, Charlesworth D. Elements of evolutionary genetics, vol. 42. Greenwood Village, CO: Roberts and Company Publishers; 2010.Google Scholar
- 50.Bonner JT. Size and cycle. Princeton: Princeton University Pres; 1965.Google Scholar
- 52.Janzen DH. How to be a fig. Annu Rev Ecol Syst. 1979;10(1):13–51.Google Scholar
- 54.Dirksen P, Marsh SA, Braker I, Heitland N, Wagner S, Nakad R, et al. The native microbiome of the nematode Caenorhabditis elegans: gateway to a new host-microbiome model. BMC Biol. 2016;14(1):1.Google Scholar
- 57.Morand S. Life-history traits in parasitic nematodes: a comparative approach for the search of invariants. Funct Ecol. 1996;10:210–8.Google Scholar
- 58.Skorping A, Read AF, Keymer AE. Life history covariation in intestinal nematodes of mammals. Oikos. 1991;60:365–72.Google Scholar
- 59.Herre EA, Jandér KC, Machado CA. Evolutionary ecology of figs and their associates: recent progress and outstanding puzzles. Annu Rev Ecol Evol Syst. 2008;39:439–58.Google Scholar
- 61.Proffit M, Schatz B, Bessière J-M, Chen C, Soler C, Hossaert-McKey M. Signalling receptivity: comparison of the emission of volatile compounds by figs of Ficus hispida before, during and after the phase of receptivity to pollinators. Symbiosis (Rehovot). 2008;45(1):15.Google Scholar
- 65.Honěk A. Intraspecific variation in body size and fecundity in insects: a general relationship. Oikos. 1993;66:483–92.Google Scholar
- 70.Williams GC. Pleiotropy, natural selection, and the evolution of senescence. Evolution. 1957;11:398–411.Google Scholar
- 71.Sudhaus W, Kiontke K. Comparison of the cryptic nematode species Caenorhabditis brenneri sp. n. and C. remanei (Nematoda: Rhabditidae) with the stem species pattern of the Caenorhabditis Elegans group. Zootaxa. 2007;1456:45–62.Google Scholar
- 77.Barrière A, Yang S-P, Pekarek E, Thomas CG, Haag ES, Ruvinsky I. Detecting heterozygosity in shotgun genome assemblies: lessons from obligately outcrossing nematodes. Genome Res. 2009;19:470–80.Google Scholar
- 85.Hill C. Learning scientific programming with python. Cambridge: Cambridge University Press; 2016.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.