Embryonal life histories: Desiccation plasticity and diapause in the Argentinean pearlfish Austrolebias bellottii

2.1 Housing and data collection

Animal care and handling protocols were approved by the Animal Welfare committee of the Leiden Faculty of Sciences and Medicine. The twenty-eight individuals of Austrolebias bellottii used in this experiment (14 males and 14 females from different family lines) were descendants from a population of 50 individuals collected in Ingeniero Maschwitz, Argentina, in 2004 and hatched from eggs collected in 2008. There were two groups of pairs with an age difference of about 4 months (hatched 2 or 6 months before the start of egg collection, on April 20, 2010). Pairs were kept in 20-L aquaria in a climate room at 19°C with a 12-hr light/dark cycle. The aquaria contained some plants (e.g. Elodea densa, Vesicularia dubyana, Lemna minor) and a black, plastic 2-L container with boiled coco peat where the fish could hide and lay eggs.

We collected eggs on different days, with intervals of 3–7 days between them. The day before collecting eggs, a clean container with 300 ml glass beads and 300 ml peat granules (both sieved and boiled) replaced the other container (Moshgani & Van Dooren, 2011). Twenty-four hours later, eggs were collected and placed individually in wells of a 24-well plate with 1 ml of UV-sterilized tap water (DUNEA Leiden, the Netherlands). Each clutch was divided over two plates placed in different climate rooms with a small temperature difference between them (19.5°C vs. 20.5°C). Temperature differences between eggs and with adults were unplanned and due to within- and between room variability. Next to the control treatment, which consisted of eggs continually incubated in water, we exposed the eggs to different “desiccation” treatments in the range where expected survival of most stages would allow developmental responses to be observed (Podrabsky, Carpenter, & Hand, 2001). We set up desiccators with demineralized water and saturated salt solutions of KNO₃, NH₄H₂PO₄ and KCl that would generate air humidities from 100% relative humidity (Water), 94%–93%, 93% and 86%, respectively (O'Brien, 1948). The humidity levels were verified with dataloggers (Gemini Tinytag PlusII, https://www.geminidataloggers.com) which could only confirm that the relative humidity was below 100% for KCl (relative humidity 85%). Loggers might have failed to capture humidities reliably due to wear on the probe or limited precision at high humidities, or some solutions were insufficiently saturated and did not produce the expected regime (i.e., remained above 93%). We therefore decided to analyze differences between the different desiccation regimes with categorical variables and not by means of regressions. Developmental states were determined at least once per week for each individual embryo, except for a single longer interval in summer (Figure 2). Eggs were transferred to the desiccation treatments at ages between 17 and 50 days. Desiccation treatments lasted between 51 and 94 days. Treatments and control were terminated by replacing the water in the wells with a mix of tap water, demineralized water and peat extract to promote hatching (Varela-Lasheras & Van Dooren, 2014). Embryos were checked for hatching responses over successive days, and hatchlings were removed from their well on the same day. Approximately 30 days later, the water was replaced with sterilized water again for the eggs that had not hatched. There were no further observations made during a period of 61 days. The hatching procedure was then repeated (112–119 days after the first rewetting). In all, embryos experienced environmental sequences with up to four changes and were followed for up to 244 days.

Based on previous studies (Varela-Lasheras & Van Dooren, 2014; Wourms, 1972a), individual state was scored as "dead" or as being alive in one of five developmental stages that can be distinguished well (Figure 2). Embryos were assigned to stage 1 when collected. A neural keel and the first somites appearing delineate the start of stage 2, the presence of optic cups the transition to stage 3 and the pigmentation of the eyes the transition to stage 4. Finally, stage 5 starts when the embryo completely surrounds the yolk sac and ends when the fry hatches or dies. For embryos in stages four and five, we noted at each observation whether the tail coiled over the left or right side of the head. Diapause I occurs in the first stage, diapause II near the end of stage 2 or early in stage 3, diapause III in stage 5.

2.2 Statistical analysis

Developmental life histories consist of the age intervals embryos spend in each developmental stage before either moving into the next stage, hatching, or dying (Figure 2). Events (deaths, developmental transitions) are assumed to have occurred just before they are observed. The data were analyzed using Cox proportional hazard models with random effects, using the libraries survival and coxme in R (Therneau, 2015; Therneau & Grambsch, 2000). We modeled all transition rates per stage separately, that is, the mortality rates and developmental transitions to the next stage (Varela-Lasheras & Van Dooren, 2014). The models per stage in the main text combine all durations in the different environmental conditions, so that effects of environmental conditions can be tested by removing them from a model. In the supplement, results are presented for models where the durations were analyzed per environment separately (Control/Desiccation Treatments/Rewetted) or censored before the long interobservation summer interval. In some environment/stage combinations, we had very few individuals at risk and therefore either did not fit any models to the data, or Cox models with fixed effects. We analyzed zero-one responses or proportions of states present at specific time points using logistic regressions.

3.1 Survival

Transformed cumulative hazard functions per stage allow a comparison of relative magnitudes of mortalities between stages and ages (Figure 3a-e). These curves can be interpreted as survivorship curves, assuming that competing developmental transitions can be treated as censors. For ages where these curves are steep, mortality rates are high. Where they become flat, mortality becomes negligible. Figure three shows that, by the end of the observation period, the cumulative hazard is lowest for prehatching embryos (Figure 3e) and highest for embryos that remained in the earliest developmental stage (Figure 3a). In stage 1, most deaths occur in the first month. After that, remaining in this stage does not impose an immediate disadvantage in terms of further mortality relative to the other stages. The death rate decreases with age to become zero in most stages and treatment groups. By the age where all surviving embryos are in the rewetting phase, there are hardly any new deaths occurring. Table 2 summarizes which variables were retained in the models for survival per stage. Table S1 does that for a dataset restricted to the observations made before the start of the long interval (days 96–133, Figure 2). There are significant fixed effects for stages two and five, and other effects suggested by Figure 3 were not significant. Embryos that are desiccated in stage 2 have an increased death rate (confidence interval [0.11, 1.49]) and embryos which have previously spent a longer duration in stage 1 have a decreased death rate (slope [−0.12, −0.02]). In stage 5, there is negligible mortality in the wet phase (Figure 3 and Supporting Information Figure S5). Death rates in the prehatching stage are increased by any of the desiccation regimes, even when eggs were in air with 100% relative humidity (increase in mortality rate for H₂O [0.33, 3.00]). Confidence intervals for mortality rates of KNO₃ ([2.45, 4.81]) and NH₄H₂PO₄ ([2.51, 4.87]) desiccation regimes overlap with H₂O, for KCl it does not ([3.68, 6.31]). The order of mortality rates matches the intended strengths of desiccation. In the prehatching stage, mortality increases with temperature ([0.89, 1.81]). The supplementary material contains Table S3 and figures of survivorship curves per stage and per period (Wet/Dry/Rewetted). These results suggest that the lower survival in the driest desiccator was caused by a briefly elevated mortality at the start of the rewetting phase. There are no significant effects of parental age (Table 2). However, in stage 5 and when each period is analyzed separately (Table S3), we recover parental age effects not found when the data are jointly analyzed across periods. Prehatching embryos with older parents have lower mortality in the dry period [parameter estimate −1.49 (SE = 0.42)] but a higher mortality in the following rewetted period (1.43 (0.75)) resulting in no net overall effect.

When we inspect the results on the random effects, variances between parental pairs and between plates are significant in stage 1 only, while random effects of collection date are present in three out of five stages (one, three and four). A linear collection date effect is not different from zero for stage 1 (0.004 (0.004)). Mortality rates decrease with collection date in stages three (−0.052 (0.017)) and four (−0.032 (0.015)), respectively. However, these linear effects are insufficient to explain all variation between collection dates (AIC differences larger than ten).

3.2 Development

The transformed cumulative hazard functions for rates of development in Figure 3f–J represent the proportions of embryos that have made a developmental transition as a function of the time already spent in each stage (assuming deaths can be treated as censors). For embryos in stages one and five, the cumulative hazards clearly stop increasing at a certain age within-stage before all individuals have made a developmental transition (Figure 3f,J), showing that diapause I and III occur irrespective of mild desiccation. There is a small fraction of abortive hatchings from stage 5. Diapause II is not observed as a clear levelling off in the fraction that made a transition (Figure 3g,h), even with prolonged but mild desiccation. For stage 2, the cumulative hazard does seem to stop increasing but at a value implying that the fraction remaining in diapause in this stage would be very small, below 5%. We also see a similar amount of individuals that do not make a developmental transition from stage 4 into 5, where no diapause is expected. What we clearly observe is that the curves for development in stage 2, three, and four, all show a period of about twenty days between days 20 and 50 with fewer developmental transitions. When data are censored before the single long time interval between observations, there is no such period. We conclude that there are no slow/fast alternative developmental trajectories in these stages. For the shortened dataset and relative to the control, embryos in saturated air develop significantly more slowly (−0.37 (0.17)) from stage 1 into 2. However, the effect is not clearly visible in graphs, there are no significant effects for the other desiccation levels and the effect is not detected in the complete data or when the dry period is separately analyzed. We therefore denote it as spurious. Desiccation changed the developmental rate from stage 2 into stage 3 relative to wet conditions (Table 3, parameter estimates H₂O 0.27 (0.23); KNO₃ –0.21 (0.25); NH₄H₂PO₄ –0.57 (0.26)); KCl 1.51 (1.06)). This pattern of estimates suggests a nonlinear effect of desiccation intensity, but that strongly depends on the estimate for KCl which has large sampling variance and is the least reliable. The pairwise difference from the control group is only significant for the NH₄H₂PO₄ treatment. These differences are not discernible in Figure 3g. However, they are in Supporting Information Figure S7. Individuals that spend more time in stage 1 have a slower development rate from stage 2 into 3 (Table 3, slope −0.030 (0.006)). Supporting Information Figure S9 in the supplement suggests that for developmental rates from stage 4 into 5, KCl and the control groups have a lower rate of development than other treatments. The mixed Cox models confirm this, with development into stage 4 significantly increased relative to control in a desiccator with H₂O (1.73 (0.41)) KNO₃ (1.49 (0.43)) and NH₄H₂PO₄ (1.18 (0.46)). The rate of development in stage 4 tends to be accelerated for embryos that have spent more time in the preceding stage (slope 0.044 (0.024), Table 3, Tables S2, S4). The rate of spontaneous hatching from stage 5 is significantly smaller for embryos that are in the desiccators (parameter estimate −2.24 (0.72)), and there are no abortive hatchings in the rewetting phase (Supporting Information Figure S10).

We find significant variation between pairs for the developmental rates in first and fourth stage (Table 3). Plate variance effects occur in slightly more developmental stages than for survival rates (Tables 3 and Tables S2, S4). The parental pair effect on the rate of development from stage 1 into stage 2 is not correlated with that for development in stage 4, nor with the pair effect on mortality in stage 1. The pair effect on the mortality rate in stage 1 is significantly and negatively correlated with the development rate in stage 4 (Spearman's rank correlation r_s = −0.63, p = 0.03). At the level of pair variation, low mortality in stage 1 is associated with faster development from stage 4 into the prehatching stage.

When we fit linear collection date effects in the mixed effect Cox models, the model AIC always increases while estimated effects differ from zero. In stages one and two development slows down with collection date (−0.013 (0.003); −0.019 (0.005)), in stage 4 it accelerates (0.06 (0.02)).

3.3 Multistate overview

Figure 4 presents an overview of fractions of embryos in different stages in dependence on age, with different panels for the different regimes. The panels again show the decreases in rates when embryos have been a while in a treatment—the boundaries between states become more horizontal. The amount of events is overall small in the rewetting period. Among the individuals that did not hatch, stage 5 embryos are the most numerous at the end of the experiment, suggesting that in the environments we investigated, diapausing individuals are mainly of diapause type III after some time. Note also that among the rewetted embryos, there is a small fraction of individuals in stage 3. This could not be seen in Figure 3h. The data suggest that after some time, there are small numbers of individuals remaining in stages two to four, and larger ones in the first and last stages. When we plot the dependency of the total number of surviving embryos and the fraction in diapause I on the rates of survival and development (Figure 5), it becomes clear that changes in nearly all rates have effects on these two fitness components but with different strengths. The percentile confidence intervals for the bootstrapped slope estimates did not include zero except for mortality in stage 4. Effects of rates on number surviving and fraction in diapause I are of opposing sign except for the death rate in stage 1. If individuals want to increase the fraction ready for an extra generation within a season, they should increase all developmental rates and decrease mortality rates in stages two, three and five for earlier collection dates and in younger parents. We found such date effects in the rates of development from stages one and two and the opposite effect in the mortality during stage 3.

3.4 Distributions across stages

At first rewetting, 42 individuals were in stage 1, 6 in stage 2, 19 in stage 3, 7 in stage 4, and 525 in stage 5. In 21 out of 67 clutches with more than one embryo alive at this moment, embryos in stages one and five co-occur, in agreement with expectations for diversified bet-hedging. We only analyzed fractions in stages one and three relative to stage 5. For the fraction of embryos in stage 1 relative to stage 5 at rewetting, we found significant differences between parental pairs (Figure 6, χ² (13) = 53.19, p < 0.001). The fraction in stage 1 at rewetting goes from zero to 24% depending on the parental pair. Parental pair random effects on the fraction in stage 1 at rewetting correlate negatively with the mortality random effect in stage 1 (Spearman's rank r_s = −0.61, p = 0.021, Figure 7). There is a negative correlation (r_s = −0.55, p = 0.041) with the speed of development from stage 1 into 2. These outcomes are in agreement with expectations based on Figure 5. When we correlate the fraction of embryos alive at rewetting per pair with the significant parental pair random effects, we find that it correlates negatively with the mortality effect in stage 1 (r_s = −0.97, p < 0.001, Figure 7).

The fraction of embryos that are in stage 3 at rewetting differs between parental pairs. It is smaller for older parents (effect estimate −2.06 (SE = 0.53); χ² (1) = 17.8, p < 0.001). For the distributions over stages at the end of the observation period, we find the same effects. These results suggest different speeds within treatments at which the individuals settle in diapauses I and III, but no consistent treatment differences in the eventual fractions. When we represent the distribution over stages at first rewetting in a graph and label parents according age class (Figure 6), it can be seen that the older parents contributed more individuals overall and in particular in the prehatching stage, as they produced larger clutch sizes sooner. There are no further effects of collection date. At the end of the period where we observed individual state regularly, 139 of the stage 5 individuals had hatched or died.

3.5 Hatching

In 37 out of 65 clutches with more than one embryo in stage 5, there were embryos that hatched and others that did not. On average, the hatching probability of the control group where embryos were in water at all times is 12%. For the H₂O group, it is 27%, in the KNO₃ desiccator the hatching probability is 13%, from NH₄H₂PO₄ 12% and there was no hatching in the KCl desiccators. Mixed models for hatching probabilities did not converge well, and we refitted them as glm's. For the first rewetting, we found significant effects of collection date (χ² (11) = 62.90, p < 0.001), a desiccation treatment effect (χ² (5) = 60.04, p < 0.001), and an effect of hatching date (χ² (3) = 29.99, p < 0.001). In the H₂O treatment, the intercept term is increased relative to the control (H₂O 2.92 (0.46)). There were no significant correlations of parental pair random effects for survival or development with hatching probability per female. When the categorical collection date effect was replaced by a linear effect of collection date, this did not suffice to capture all variation between dates. Moreover, hatching probability increases with collection date (0.037 ; SE = 0.011): eggs from later clutches hatch with a larger probability.

When we inspected embryos before the second rewetting (Table 4), a fraction of the eggs had developed further starting from stages one to four. Other embryos had remained in the stages where they were two months before (stages 1, 2, 3, and 5). Survival did not differ significantly between stages. Correlations between parental effects on developmental rate and mortality in stage 1 and the fraction of individuals in stage 1 were slightly weaker (respectively r_s = −0.42 and r_s = −0.45) and nonsignificant. Overall, survival was 83% across this period. When peat water was added, 48% of the stage 5 individuals hatched. We found significant effects of storage temperature (χ² (1) = 4.05, p = 0.04) and of the date where the embryos were rewetted before (χ² (3) = 26.84 p < 0.0001). For embryos stored at 20.5 ºC, the hatching probability was reduced (parameter estimate effect −0.53 (0.27)). Two first rewetting dates had significantly reduced hatching probabilities at the second rewetting (parameter differences approx. −1.5). There were again no pair effects on hatching probability.

3.6 Coiling

We estimated rates of change of the tail coiling orientation with a glm as in Harney et al. (2013). In stage 4, we found a significant treatment × time in stage interaction (χ² (4) = 908.84, p < 0.0001) and differences between parental pairs (χ² (6) = 13.59, p = 0.035), but none of the parameter estimates were significantly different from zero. When simplifying the model to a single effect of time within age, we found that it was nonsignificant.

In stage 5, there is a significant treatment × age within stage interaction (χ² (3) = 9.072, p = 0.0284), but again no separate parameters could be shown to be significantly different from zero. For the group desiccated with NH₄H₂PO₄, the rate tends to be lower (−1.48 (0.82), p = 0.07). Simplifying to a single covariate—age within stage 5—we find that the probability of changing direction decreases with age in stage 5 (parameter estimate (−0.092 (0.038)). According to this simple model, the probability to change coiling direction at age zero within stage 5 is 5% per day and at day forty 0.1%. For stage 4, the probability to change coiling direction is 7% per day.

4.1 Adaptive development

We can assess three aspects of adaptation in the development of annual killifish using results from this experiment. First, we can check whether patterns of individual variation are consistent with bet-hedging variation. Second, we can inspect parental age and collection date effects to check whether development might be adapted to exploit opportunities to complete an additional generation within the same year. Third, we can check whether desiccation could be used as a cue for seasonal timing, not only announcing the arrival of the dry season, but also the rewetting when it ends.

We found co-occurring fractions of individuals in diapause I and III in many clutches. Many parental pairs and many clutches have fractions of embryos in different developmental stages at rewetting, consistent with diversified bet-hedging. Patterns of exit from diapause characterize germ banking strategies (Evans & Dennehy, 2005). Here, hatching is probabilistic, also at the within-clutch level. The fraction of individuals in diapause I at rewetting, a key aspect of the delaying strategy, is not plastic and depends on parental pair variation. This leaves the possibility open for maternal determination of this strategy, as in Nothobranchius (Polačik et al., 2017). We could not find any coupling of rates of development to hatching probabilities. Processes of diapause entry and exit seem different.

Evidence for adaptive responses to exploit opportunities for an extra generation is ambiguous and relatively weak. Younger parents are not producing faster-developing embryos. Development in late developmental stages accelerates with collection date. Hatching probabilities increased substantially at the second rewetting. These effects are unexpected according to this hypothesis. Development in early stages decelerates with collection date, in agreement with this hypothesis. At occasions very late in the season preference should be given to remain in the egg bank longer. The positive collection date effect on hatching probability is in agreement with embryos timing seasonal progress using their own individual age. Younger embryos then assume that they are early in the year.

Desiccation seems to announce the dry season relatively well: In agreement with this hypothesis, late development is accelerated in the presence of mild desiccation and so is the hatching response. The collection date effect on late development could also be explained as a response for a cue that seasons progress and that a dry period will be followed by rewetting. The collection date effect on hatching disagrees with this hypothesis.

We have found stage-specific effects of collection date (stages three and four) and of the desiccation treatment (prehatching stage) on mortality. If such effects occur in the field, then they will affect which decisions taken by individuals are optimal. Altogether, responses to age, date and desiccation seem to indicate that a tool such as dynamical programming will be needed to assess adaptation in development and hatching well and that we need to move beyond hypotheses that we can assume and test without detailed modeling and ignoring stage-specific survival. Similarly, a solid demonstration that the distribution over diapauses I and III and probabilistic hatching represent bet-hedging will require a demographic model and fitness calculations of the entire life cycle (Childs et al., 2010; de Jong, Haccou, & Kuipers, 2011).

Important is that we could demonstrate by means of simulations and our data that random pair variation found in rates during development is "visible" for selection: rates of development and survival affect components of fitness. If adapted, they are likely kept at their present values because of trade-offs between rates and fitness components or because selection on the life history components we investigated is weak.

4.2 Annual versus nonannual rivulids

In comparison to nonannual rivulids where initially over 10% of individuals changed coiling per day (Varela-Lasheras & Van Dooren, 2014), the activity of prehatching annual fish embryos seems overall lower from the start. Overall, the effects of desiccation are limited and different from the pattern observed among nonannual rivulids. The desiccation regimes in this study are much prolonged in comparison to Varela-Lasheras and Van Dooren (2014), where the median duration was eight days, with visible effects on the state of the eggs after one hour. There, delays in response to brief desiccation were early in development and desiccation provoked an increased incidence of delayed hatching (similar or equal to diapause III) in stage 5 in some species. Regarding early development in the first stage of the Argentinean pearlfish, we only found a significant desiccation effect in a reduced dataset and for the weakest desiccation level, casting doubt on its existence. The speed of development near head formation was slowed after rewetting for one desiccation level and showed a positive response to desiccation late in development. In nonannuals, desiccation significantly increased mortality in stages three and four and decreased in Cynodonichthys magdalenae again after return to water. In the annual species, developing embryos are quite resistant to the desiccation levels we applied as survival effects are limited to the prehatching stage.

Slowing down during early development in nonannuals in response to desiccation might be mostly an immediate response, therefore quiescence, even while there are some lagged effects (however, see Furness et al., 2018). If genetic assimilation of diapause has occurred in killifish via the evolution of plasticity in response to desiccation, then there has clearly been further plasticity evolution after the emergence of diapause-like characteristics. The plasticity now involves an accelerating effect.

4.3 The three diapauses

According to Wourms (1972c), diapause I is common in South American Austrolebias killifish, whereas diapause III is obligatory and diapause II facultative, as in Nothobranchius (Furness, Lee, et al., 2015; Wourms, 1972c). In agreement with that, diapause I and III occurred in benign conditions, consistent with a requirement for diapause that can occur in the absence of an environmental stressor. Note that there were no escape embryos in the sense of Polačik et al. (2017) that reach the prehatching stage within a month.

If diapause II occurred in this experiment, it did so in below 5% of the individuals that stayed alive in stage 2 and after more than a month in that stage. Similar proportions of individuals remained in the fourth developmental stage, where no diapause has ever been inferred. We also observed embryos that spent two months in stages two or three between the first and second rewetting.

Recently, diapause II was detected in Austrolebias nigrofasciatus (da Fonseca et al., 2018). Its incidence differed depending on incubation conditions. Lowered oxygen levels might be an environmental cue occurring in the closed vials with batches of eggs used by da Fonseca et al. (2018), or when organic matter present decomposes. Diapause II might have more characteristics of a quiescence than a diapause, occurring mostly in conditions of environmental stress. It might occur rarely in Austrolebias in comparison to the other diapauses, but data for most species are currently lacking. Diapause II is not constitutive for this genus and our results on the stage where diapause II occurs can be more parsimoniously explained by the general decrease of rates in all stages.

Several papers recently brought new data and ideas on the diapauses in annual fish. Furness (2016) believes diapause I to have a short duration in benign conditions and to be mainly brought about by harsh conditions. Diapause II, after the phylotypic stage, would contribute most to variability in developmental duration (Furness, 2016). Our results contradict this. Podrabsky et al. (2001) found that Austrofundulus embryos resisted desiccation best in diapause II. We found that most developmental stages suffer no survival effects of prolonged mild desiccation in Austrolebias. Given the results here, diapause II is certainly not the most prominent stage of arrest in all annual fish as proposed by Furness, Reznick, et al. (2015). Being more facultative than diapause III at least in Austrolebias, it cannot be equated to the annualism syndrome.

It has not been clear which criteria were previously followed to assign embryos to diapause or direct-developing pathways (Furness, Lee, et al., 2015; Furness, Reznick, et al., 2015) and there has been some circularity in the data analysis, with developmental outcomes used as an explanatory variable of the same process. Data on Austrolebias nigripinnis in Furness, Reznick, et al. (2015) suggest that developmental trajectories are not very dichotomous between embryos with direct development to a prehatching stage or not, whereas direct developers in Nothobranchius have significantly larger heads. This suggests that next to developmental delays, morphological embryonal life history traits have evolved differently in the different annual groups.

4.4 Parental effects

We found a very limited number of significant effects of parental age on the survival and development of embryos and hatching. However, we did find variation between parental pairs for survival and developmental rates, which could encompass genotypic variation. Several of the pair effects on rates could be correlated to fitness components, and two random effects were correlated as well. We did not find parental pair effects on hatching and therefore on the exit from diapause. The potential for evolutionary responses in this system seem larger with respect to the composition of the germ bank than in the exit from "germ banking".