Life-History Traits of the Model Organism Pristionchus pacificus Recorded Using the Hanging Drop Method: Comparison with Caenorhabditis elegans

The nematode Pristionchus pacificus is of growing interest as a model organism in evolutionary biology. However, despite multiple studies of its genetics, developmental cues, and ecology, the basic life-history traits (LHTs) of P. pacificus remain unknown. In this study, we used the hanging drop method to follow P. pacificus at the individual level and thereby quantify its LHTs. This approach allowed direct comparisons with the LHTs of Caenorhabditis elegans recently determined using this method. When provided with 5×109 Escherichia coli cells ml–1 at 20°C, the intrinsic rate of natural increase of P. pacificus was 1.125 (individually, per day); mean net production was 115 juveniles produced during the life-time of each individual, and each nematode laid an average of 270 eggs (both fertile and unfertile). The mean age of P. pacificus individuals at first reproduction was 65 h, and the average life span was 22 days. The life cycle of P. pacificus is therefore slightly longer than that of C. elegans, with a longer average life span and hatching time and the production of fewer progeny.


Introduction
Model organisms with short generation times and able to tolerate cultivation under laboratory conditions have long captured the attention of biologists, who have used them to test mechanistic hypotheses at the lowest experimental cost and temporal scale possible [1,2]. For instance, the nematode Caenorhabditis elegans Maupas 1900 has for decades served as an animal model for researchers throughout the world [3,4,5]. In the wake of the success of C. elegans, other Nematoda species have been used to investigate a broader range of traits [6,7].
One example is the nematode Pristionchus pacificus Sommer et al. 1996, which holds several important advantages as a model organism such as hermaphroditism, short generation time, and easy cultivation using Escherichia coli as a standardized food source [8]. Only in the past five years, 107 publications have focused on P. pacificus [9], thus revealing its growing use as a model organism.
Providing with an ecological context of model organisms is a crucial point in our understanding of evolutionary biology [10]. Pristionchus pacificus has an interesting ecology in its highly host-specific necromenic behavior, a form of phoresy, in which the nematode enters a resting stage once inside its beetle host and awakens upon host death to feed on the 'soup' of microbial decomposers growing within the beetle carcass [11,12]. This association is of such intricate nature that horizontal gene transfer from insect to nematode has been reported [13]. Another remarkable ecological trait of P. pacificus is its ability to switch from a bacterivorous to a predacious phenotype according to physiological and sociological cues [14,15,16]. The unique traits and phenotypic plasticity of P. pacificus make this nematode a relevant evolutionary model that is likely to yield important molecular, developmental, and ecological insights (e.g. [8,12,14,17,18]).
However, to date, the basic features of P. pacificus life history, such as its population growth rate and reproductive traits, have not been measured. While genetics determine life-history traits (LHTs) such as fitness and survival, both are inclusively necessary for a broader understanding of animal evolution and population dynamics [19,20]. Thus, for the LHTs of model species, accurate data from empirical measurements are required to establish a robust knowledge of the physiology and evolutionary biology of that species [21]. For instance, the number of offspring and the life span of wild strains recorded under optimal conditions are commonly used as benchmarks to assess divergence in mutant strains or endpoints in ecotoxicological tests [22,23,24].
Multiple nematode culture methods can be used to record LHTs (e.g. [25]). The 'hanging drop method' proposed by Muschiol & Traunspurger [26] combines the advantages conferred by culture methods using solid and liquid media in terms of accurately observing, describing, and recording nematode LHTs. This method is based on the use of semi-solid droplets of culture medium that hang from the lid of plastic culture multi-well plates. This set up allows nematodes to be monitored individually over time at any degree of precision. By eliminating intraspecific interferences in batch cultures and the uneven distribution of food items in solid medium, it avoids several sources of error and artefacts.
In this study, we accurately define the major characteristics of the life cycle and LHTs of P. pacificus strain PS312 under standard conditions. Our methodological approach is based on the 'hanging drop method', which is optimal for such purposes. Additionally, we compared the life cycle and LHT of P. pacificus with those recently obtained from C. elegans using this same method [27].

Experimental set-up
Escherichia coli (strain OP50) served as the sole food resource for P. pacificus. Frozen E. coli were inoculated on LB medium and allowed to grow overnight at 37°C. This inoculum was then used to prepare agar plates for the maintenance of stock cultures on nematode growth medium (NGM). All experimental steps were performed in a sterile environment under a laminar airflow fume hood.
For the experimental set-up, standard E. coli concentrations were prepared based on the optimal requirements of C. elegans and Caenorhabditis briggsae, which are between 10 9 and 10 10 bacterial cells ml -1 [28,29]. We therefore used E. coli concentration of 5×10 9 cells ml -1 as described by Muschiol et al. [27] for C. elegans, which allowed a direct comparison of its LHTs with those of P. pacificus determined in this study.
To obtain the desired E. coli concentration, liquid cultures were centrifuged and the resulting pellets were suspended in K-medium to determine E. coli density based on the absorbance-to-cell density correlation (OD 600; [26]). To facilitate storage, E. coli suspensions of known concentrations were centrifuged and the pellets were stored in the dark at 8°C for up to 20 days. The pellets were then re-suspended as needed in sterilized semi-liquid nematode growth gelrite medium (NGG), which is analogous to the standard NGM but bacto agar is replaced by a bacterial exopolysacharidic gellam gum (Gelrite, Merck & Co., Kelco Division). This substitution produces a semi-solid medium, enabling nematodes to be individually placed in hanging drops. NGG is similar in appearance and consistency to 0.4% bacto-agar, and can be poured into Petri dishes for the maintenance of cultures (see further details in [26]). Prepared NGG was stored in the dark at 8°C for up to 4 days.
Drops of prepared NGG (8 μl) were placed on the upper lids of sterile 12-well plates. To avoid desiccation of the drops, cellulose gauze pads soaked in distilled water were placed beneath each drop in the corresponding wells. The plates were sealed with Parafilm and incubated in the dark at 20°C. The semi-fluid consistency of the medium allowed the drops to hang on the lids, with evaporation prevented by replacing the wet gauze pads. Individual nematodes could thus be observed uninvasively under a stereomicroscope (×9-90) as required [27].

Nematode culture maintenance and acclimatization
Cultures of a wild isolate of P. pacificus (strain PS312) were obtained from the Caenorhabditis Genetics Center (University of Minnesota, Minneapolis, MN, USA). Nematodes were delivered embedded on agar plates containing NGM. Agar plates containing living worms were stored in the dark at 20°C. Approximately every 6 days, the nematodes were transferred to new agar plates previously inoculated with prepared E. coli suspension. All manipulations of stock cultures were done at 20°C and under sterile conditions. Prior to the experiments, the nematodes were transferred from NGM to NGG and allowed to acclimatize for five days, to reduce potential influences due to acclimatization stress, maternal effects, and contamination.

Life stage synchronization and experimental run
To obtain accurate LHT measures, nematode life stages were synchronized at inoculation. Ten young adult nematodes with already developed gonads, but non-gravid, were transferred in drops (10 μl) of NGG containing 5×10 9 E. coli cells ml -1 . The nematodes were then left to lay eggs during 30-50 h of incubation in the dark at 20°C. The drops were checked every 3 h. Additional nutritive medium (5 μl) was added daily to provide fresh food and to limit dehydration of the drops. Once the eggs had hatched, juveniles hatched within a 4 h interval (3 h plus 1 h of handling time) were selected for further study. This method ensured that the LHT measures were as temporally accurate as possible.
Synchronized juveniles (n = 33) were gently picked up with a 'nematode picker' (one eyebrow hair glued to the tip of a Pasteur pipette) and then individually positioned in their respective drops. Juveniles were checked for life parameters every 24 h, at which time they were transferred to a fresh drop of food medium. When they had reached sexual maturity, egg-laying hermaphrodites were transferred to a fresh drop every 6 h and the number of eggs laid in the drops was recorded. After the adults had been transferred, the drops were left undisturbed for 48 h (originally set to 24 h, which was shown to be insufficient), which allowed the number of hatching vs. non-hatching eggs to be recorded. After 126 h, the maximum fecundity was expected to decline and the worms were transferred to new drops first every 12 h, and then, after 294 h, every 24 h. The experiment was run until the death of the last two remaining adults (after 764 h:~32 days).

Data analysis
To concisely summarize LHT measures, the data were presented in the form of life tables. The following parameters were calculated: hatching time, net production rate, total fertility rate, life span, generation time, population doubling time, and rate of natural increase. Life span, age at first egg deposition, and maximum rate of egg-laying were expressed as means with their standard deviations. To facilitate interpretation of the results, a glossary of the terms used herein is provided in Fig 1. Hatching time, net production rate, fecundity, rate of natural increase, and alternative measures of generation time were calculated using methods described in the literature and from empirical measurements, as described in detail below: Hatching time. Hatching time was preferred over egg-laying time, as egg-laying can be delayed such that, in extreme cases, intra-uterine hatching may occur (also known as matricidal hatching or endotokia matricida [30,31]). Hence, because the experiment started with juveniles and not eggs, a measure of hatching time was needed to accurately assess the entire P. pacificus life cycle. Our approach was based on the protocol described for C. elegans by Muschiol et al [27], which was modified to account for the longer developmental period of P. pacificus. Briefly, 100 adults and stage-four juveniles of P. pacificus were randomly picked from an exponential stage culture raised on NGG. The worms were dispatched in ten 20-μl drops of NGG (10 adults per drop) for 32 h, after which the drops were fixed with 4% formaldehyde and stained with Rose Bengal for observation under a stereomicroscope (×90). Hatching time was calculated as: where: H t = estimated hatching time N eggs = number of eggs laid during the experimental time N juv = number of juveniles hatched during the experimental time T h = experimental time (in hours) Net production rate. The net production rate was calculated as: where: R 0 = net production rate l x = age-specific survival probability m x = age-specific fecundity The value of R 0 can be interpreted as the average number of offspring produced by an individual of the study population during its entire life [20].
Fecundity. The fecundity was calculated individually for all sexually mature, alive individuals and then as an arithmetic mean that was converted into a percentage. This allowed an easier interpretation of the resulting graphs. Note that individual deaths due to natural causes were excluded from the calculation of fecundity as an individual mean; they were, however included in the calculations of survivorship and the rate of natural increase of the population.
Intrinsic rate of natural increase. The intrinsic rate of natural increase (r m ) was proposed by Lotka [32] to measure the growth potential of a population. It is defined as the rate of increase per individual in the absence of adverse conditions [33] and is calculated based on Euler's equation: r m = intrinsic rate of natural increase x = time [d] l x = age-specific survival probability m x = age-specific fecundity This equation, also known as the 'Lotka equation' [33], is solved by substitutions of the r m value until the equation equality is met (see [20] for calculation details).
Population doubling time. In the nematode LHT literature, the term "population doubling time" (PDT) is very broadly employed but with little explanation. The PDT is the time needed for a growing population to duplicate [34]. As such, in a synchronized population, the PDT is the time needed for the population to double after sexual maturity has been achieved; that is, once reproduction has begun. Hence, in an assessment of the entire life cycle of a population, as was our aim with P. pacificus, the PDT of a synchronized population requires the prior determination of the mean age at first reproduction. PDT is calculated as: PDT ¼ lnð2Þ=r m where: PDT = population doubling time ln = natural logarithm r m = rate of natural increase Alternative measures of generation time. Generation time is sometimes regarded as a subjective concept due to its dependence on population structure and temporal resolution, hence alternative measurements have been proposed [35]. Following these indications, we measured: G 0 or cohort generation time; G 1 , which is the time necessary for the increasing population to grow by a factor of R 0 ; and G h , the mean parental age at which a new generation is produced [35]. To calculate G 0 , G 1 , and G h , the following formulas were used:  Fig 2). These two juveniles became comparatively smaller in the first few days and died after 62 and 89 h. In the remainder of the population, the maximum life span was 32 days, when the last two surviving individuals died. The average length of the population life span was 22.5 ± 7.3 days (Table 1).
Reproduction started after an average of 65.1 h (65.1 ± 8.4, Table 1) and lasted until day 13 (Fig 2). During its life span, each nematode laid an average of 270 eggs (269.6 ± 73.4), which resulted in the mean net production of 112 juveniles by all nematodes that reached sexual maturity (112.1 ± 21.1). This corresponded to a mean net production of 109 juveniles per nematode when considering nematode survivorship (R 0 = 108.95).
The rate of egg-laying peaked twice: between 5 and 7 days, when fecundity was nearly maximum (96%), and at the end of the reproductive period, after 13-14 days, when fecundity had ended (Fig 2). As time resolution was higher during maximum fecundity and early reproduction, the mean number of eggs laid daily was greater in the first peak (25 vs. 18 eggs laid in 24 h, Fig 2). Survival, fecundity and egg production of Pristionchus pacificus (strain PS312). Nematodes were cultured at 20°C by means of the hanging drop method. The survival data fit a type I attenuation curve, which is typical in species with a low level of juvenile mortality [36]. Values are reported as the means (N = 33, ±SD). From the beginning of reproduction, fecundity showed a general positive trend during the first 6 days, corresponding to a maximum fertility rate of 97.4% after 6.1 days. After that, fecundity decreased dramatically such that at day 13 it was negligible, with only one juvenile emerging out of the 257 eggs laid by all living individuals. This corresponded to a fecundity of 0.21% after 12.8 days and null (0%) thereafter.
For P. pacificus, the TFR (the maximum number of offspring produced by an individual with a maximum life span) was 115 ( Table 2). The intrinsic rate of natural increase of the P. pacificus population was 1.125. The PDT after the start of reproduction averaged 14.8 h (Table 1).

Discussion
Our results provide the first assessment of the standard life cycle and LHTs of the model nematode Pristionchus pacificus. This species is being increasingly used in evolutionary and developmental biology as a 'satellite' model organism of C. elegans [8]. However, until now, an accurate assessment of the standard life parameters of P. pacificus that allowed a direct comparison with C. elegans was lacking. The basic measures of standard population metrics reported herein and the use of optimal procedures to culture and observe this nematode provide a robust benchmark for further practical laboratory investigations using P. pacificus models. The hanging drop method [26,27] used in this study has several advantages over cultures in solid or liquid medium. First, it allows the accurate calculation and even distribution of bacterial food, because the semi-solid consistency of the NGG drops avoids sinking of the bacteria and worms. Since fresh food is provided periodically to individual worms, optimal conditions and absence of metabolites accumulation and poor oxygenation are assured. The follow up to the individual level allowed the accurate calculation of life parameters. Moreover, the hanging drop method enables observations of individuals through the upper plastic lid of multi-well plates or Petri dishes and thus avoids disturbance of the nematodes. Hence, many additional traits can be recorded (e.g. pumping rate, locomotion), leading to a high-throughput of behavioural data if this method is coupled with automated-image tracking procedures [37,38]. As it can be adjusted to any desired degree of temporal resolution, we were able to obtain accurate data of fertility and egg production at the time of maximum fecundity, when high numbers of eggs and juveniles may interfere during counting. Given its practicability, since first proposed by Muschiol and Traunspurger [27] this method has been successfully used in several studies of nematode LHTs (listed in Table 3).
In agreement with previous findings, the complete reproductive cycle of P. pacificus was <4 days [41]. However, this estimate includes the hatching time of juveniles. In fact, age at first reproduction was lower in P. pacificus than in C. elegans, reported to be a fast reproducer [41]. Nevertheless, the earlier start of P. pacificus reproduction was compensated for by a much longer hatching time ( Table 4). The apparent variation in the literature regarding hatching time can be explained considering constraints, such as food depletion [31], which can delay hatching to the point of matricidal hatching. In old individuals, matricidal hatching may simply reflect malfunction of the egg-laying machinery [42]. This mechanism is not known for P. pacificus and in this study we observed it only once, in an old hermaphrodite (9.1 days). An egg laid by another old hermaphrodite (9.6 days) hatched less than 12 h after its deposition. Hence, the delayed egg-laying and matricidal hatching observed here in P. pacificus appear to be consequences of the aging process and thus independent of the food conditions. The hatching time of 25 h measured in this study agrees with the embryonic development period of approximately 24 h previously reported for P. pacificus [44]. The relatively longer development time has been attributed to the complexity of the buccal structure of P. pacificus [45,46]. Additionally, as a member of the Diplogastromorpha, P. pacificus hatches directly as a second stage juvenile, with the first moult taking place inside the egg [46]. Thus, despite its younger age at first reproduction, P. pacificus has a slightly longer total life cycle than C. elegans (3.5 days [47]).
As suggested by Charlesworth, we used alternative measures of the generation time [35]. Surprisingly, there was a large difference between G 0 (137 h) and G h (80 h) ( Table 1). The results of age at first reproduction of 65 hours (65.1 ± 8.4) and PDT of 15 hours (14.8 h) are in concordance with the production of a new generation (G h ) in 80 hours.
The LHTs of P. pacificus were compared with those of other rhabditid nematodes assessed using the hanging drop method (Table 3). During its reproductive period, P. pacificus produced fewer offspring than C. elegans (109 vs. 290 average offspring per lifespan). As a result, the rate of the natural population increase was lower for P. pacificus than for C. elegans (1.125 vs. 1.375; [27]). Nevertheless, the population increases of other rhabditid nematodes, as determined in other studies, were much smaller: For instance, Panagrolaimus sp. (strain NFS-24) had a population increase of 0.53 at 21°C [39]. Panagrolaimus sp. and Poikilolaimus sp., both isolated from a cave ecosystem, had population increases of 0.309 and 0.165, respectively [26]. Only Steinernema riobrave (strain Sr 7/12), an entomopathogenic nematode, had a similar population increase of 1.13 at 25°C [40]. Indeed, temperature might affect the dynamics of metabolic processes in poikilothermic small invertebrates such as nematodes. Nevertheless, considering the unique host-specific necromenic ecology of P. pacificus, which has been proposed as an intermediate step towards parasitism [48], the population increase determined for P. pacificus is very similar to the values reported for the necromenic C. elegans and the entomopathogenic nematode S. riobrave.
Despite its relatively high rate of natural increase, offspring production by P. pacificus was similar to that of slower reproducers such as Panagrolaimus spp. [26,39]. A possible explanation is the relatively large number of unfertile eggs produced by P. pacificus hermaphrodites (270 eggs over a life-time, 56 of which were laid after the overall fecundity was null). Since P. pacificus is a self-fertilizing hermaphrodite with a limited amount of sperm, a larger number of progeny would be expected in the presence of males [49], such as occurs in wild populations or batch cultures.
Juvenile mortality in P. pacificus was minor; the few juveniles that did not reach sexual maturity were comparatively smaller than those able to mature. Observations of the second generation of juveniles were beyond the scope of this study; however, we did note that old adults (>8 days) with a low fecundity produced smaller offspring and in some cases seemingly aberrant nematodes. The food source in this study was OP50 E. coli, as in laboratory protocols in which P. pacificus served as the model organism [50]. In the wild, P. pacificus shows an omnivorous, necromenic life style [8]. This facultative switch to a scarab beetle host brings complexity into the life cycle of this nematode, which can behave both as a bacterial feeder and as an antagonistic entomophile [51]. Since entomopathogenic nematodes are highly dependent on temperature [52] and achieve higher progeny and growth rates with increasing food concentrations [40], reproductive values for P. pacificus higher than those determined in this study may be possible at higher food concentrations and temperature. Further studies are needed to shed more light into the complex ecology of this model organism.

Conclusions
This was the first study to measure the LHTs of the model organism P. pacificus based on observations during culture of this nematode using the hanging drop method. Compared to C. elegans, P. pacificus has a slightly longer life cycle, a longer life span, and longer hatching times and produces fewer progeny under standard conditions of food and temperature (5×10 9 E. coli cells ml -1 ; 20°C). Although the singular ecology of P. pacificus prevents the extrapolation of our results to the fitness and performance of this species in natura, we believe that our LHT measures provide a solid reference in further investigations using this model species.