Multiple working hypotheses for hyperallometric reproduction in fishes under metabolic theory
Introduction
Hyperallometric reproduction, defined as a more than proportional increase of the fecundity relative to the body mass of individuals within a species, has a wide range of consequences for our understanding of population dynamics, our predictions in a changing environment and ultimately our conservation policies (Marshall et al., 2019). Such a reproductive scaling pattern has recently been described for a wide range of fish species (Barneche et al., 2018) and implies that large females contribute more to the renewal of the population than do small females, relatively to their body mass. Large females therefore have a higher conservation value and their protection should be prioritized over small females to increase the sustainability of the population (Birkeland and Dayton, 2005).
Ideally, population dynamics models should be driven by an underlying model of bioenergetic processes that can explicitly capture the observed scaling of reproduction. A number of ‘growth models’ now exist, including some derived from general theories of how metabolism works, i.e. ‘metabolic theories’ (van der Meer, 2006). It was recently argued that the observed hyperallometric scaling of reproduction could not be explained using current bioenergetic models (Marshall and White, 2019). If this were true, and hyperallometric reproduction is indeed widespread, it would call into question the generality of existing theory and limit our capacity to develop robust models of the life history and population dynamics for many species.
However, a pattern of hyperallometric reproduction observed in wild populations is challenging to interpret when the age, environmental life history and genetic background of the individuals displaying this relationship are unknown (Kearney, 2019). This is the case for the data underlying hyperallometric reproduction in fish which derive almost entirely (>98%) from observations of wild individuals (Barneche et al., 2018). Investigating a fecundity to body mass scaling naturally results from the comparison of different sized individuals. But what is driving this variability? Why are large individuals larger? Multiple biological explanations are possible.
First, in nature, there is no reason to believe that all animals encounter the same environmental conditions over their life history. This is particularly true for species with wide geographical and habitat ranges (Riede, 2004; Wheeler, 1975), including the Atlantic cod and European sea bass. Comparing the weight and reproductive output of wild-harvested individuals of such species necessarily involves confounding effects related to environmental history. These effects are amplified when investigating the variability in weight without evaluating the age, since large animals are more likely to have encountered better environments compared to small individuals of the same age.
Second, whether it is the consequence of predation, of sudden environmental challenges or of human activities, selective pressures in the wild are particularly numerous and diversified. These multiple sources of selection are considered as the primary mechanisms of polymorphism in nature (Orr and Smith, 1998). Metabolic capacities driving growth, maturation, reproduction and ultimately fitness are central phenotypic targets of selective pressures (Pettersen et al., 2018). Within a given population, one can expect genetically based variability in metabolic capacities resulting in intrinsic individual differences in assimilation and growth capacity (Besson et al., 2019). Thus, the life history patterns obtained from wild-harvested individuals may also reflect the action of selection.
The environmental and genetic variation inherent in samples from wild populations means that one should be cautious in attributing a pattern of reproductive hyperallometry to metabolic mechanisms involved at the individual level. Rather than using such an empirical observation to dismiss existing models of individual growth and metabolism, one can instead ask under what environmental and genetic circumstances the model would produce the empirical pattern in question. This then leads to clear and testable predictions about what might be occurring in nature. In parallel, detailed laboratory experiments are critical to control and evaluate selective pressures and environmental variabilities and thereby evaluate theoretical expectations.
A common point of contention and misunderstanding in modelling growth centers around the distinction between phenomenological and mechanistic models (White and Marshall, 2019). A phenomenological model provides a quantitative description of a process using a simple function thought to capture the essence of an underlying process; often these are allometric functions in the case of growth models. In contrast, mechanistic models are derived through the explicit representation of processes occurring at a lower level to the phenomenon in question. In growth models, these processes are the chemical transformations from food to biomass, which are modelled on the basis of physicochemical principles including energy and mass conservation. Successful mechanistic models can predict dynamics under complex sequences of environmental conditions with the same variables and parameters – something phenomenological models cannot do without adding parameters or changing their functional form. In this sense, mechanistic modeling can help reconstruct environmental history when they are inverted to fit growth data from the wild (Lavaud et al., 2019; Pecquerie et al., 2009).
In this paper, we explore the potential for reproductive hyperallometry under the framework of Dynamic Energy Budget (DEB) theory; a general, mechanistic metabolic theory that captures the environmental (food, temperature) and internal (chemical transformations and allocations) constraints on development, growth and reproduction from first principles (Kooijman, 2010, 1986). In particular, we consider three working hypotheses (Fig. 1) and illustrate them using the European sea bass, Dicentrarchus labrax, a species shown to exhibit hyperallometric reproduction in the wild (Mayer et al., 1990):
- 1)
the ‘ontogenetic hypothesis’ that hyperallometric reproduction can emerge for an individual simulated under DEB theory across its ontogeny;
- 2)
the ‘environmental variation hypothesis’ that hyperallometric reproduction can be produced by realistic variation in feeding rate and temperature environments among individuals with the same metabolic capacities;
- 3)
the ‘metabolic capacities variation hypothesis’ that inter-individual variation in DEB parameter values can lead to hyperallometric reproduction.
Finally, we provide empirical data on the allometric scaling of reproduction under controlled environmental conditions and limited selection pressure. We discuss our results in the light of DEB predictions under such environmental conditions. Since the DEB model is generic between fish species (only parameters vary), the results of our study are transposable to most other fish species.
Section snippets
The DEB model for D. labrax
The DEB parameters for D. labrax used in our simulations are available online (Lika et al., 2018) and are provided in Supplementary tables 1, 2 and 3.They were inversely estimated from observations on development times (at multiple temperatures), lengths and weights at birth/hatch, metamorphosis, maturation and ultimate size as well as growth curves in length and weight, reproduction and feeding rate vs. weight and ammonia production vs. temperature (Stavrakidis-Zachou et al., 2019). The
Results
Fig. 3 summarizes the outcomes of DEB simulations leading to hypo-, iso- or hyper-allometric reproduction, as a result of our three working hypotheses.
Discussion
Collectively, our results demonstrate that multiple biological explanations integrated in DEB simulations can lead to the hyperallometric reproduction previously observed for a wide range of fish species in the wild (Barneche et al., 2018). DEB models have been used to capture the life cycle of more than 2000 animal species across all major phyla; our results are therefore likely applicable to many other species (Kooijman et al., 2020). We thus show that it is premature to conclude that
Data availability statement
The data that support the findings of this study are available as supplementary material.
CRediT authorship contribution statement
Bastien Sadoul: Conceptualization, Methodology, Writing - original draft, Software, Formal analysis, Visualization. Benjamin Geffroy: Data curation, Resources, Writing - review & editing. Stephane Lallement: Data curation, Resources, Writing - review & editing. Michael Kearney: Conceptualization, Methodology, Writing - original draft, Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (38)
- et al.
Chemical and physical differences in milkfish (Chanos chanos) eggs from natural and hormonally induced spawns
Aquaculture
(1994) - et al.
The importance in fishery management of leaving the big ones
Trends Ecol. Evol.
(2005) - et al.
Short- and long-term dietary effects on female sea bass (Dicentrarchus labrax): seasonal changes in plasma profiles of lipids and sex steroids in relation to reproduction.
Comp. Biochem. Physiol. Part C
(1995) Reproductive hyperallometry does not challenge mechanistic growth models
Trends Ecol. Evol.
(2019)- et al.
Estimating inter-individual variability of dynamic energy budget model parameters for the copepod Nitocra spinipes from existing life history data
Ecol. Modell.
(2020) Energy budgets can explain body size relations
J. Theor. Biol.
(1986)Metabolic acceleration in animal ontogeny: an evolutionary perspective
J. Sea Res., Dyn. Energy Budget Theory
(2014)- et al.
The energetic basis of population growth in animal kingdom
Ecol. Modell.
(2020) - et al.
Scenarios for acceleration in fish development and the role of metamorphosis
J. Sea Res., The AquaDEB project (phase II)
(2011) - et al.
Reconstructing physiological history from growth, a method to invert DEB models
J. Sea Res., Ecosyst. Based Manag. Biosph.
(2019)
Have we outgrown the existing models of growth?
Trends Ecol. Evol.
Ecology and speciation
Trends Ecol. Evol.
Modeling fish growth and reproduction in the context of the Dynamic Energy Budget theory to predict environmental impact on anchovy spawning duration
J. Sea Res., Metab. Organ.
Prediction of long-term variation in offspring metabolism due to BPA in eggs in rainbow trout using the DEB model
J. Sea Res., Ecosyst. Based Manag. Biosph.
A DEB model for European sea bass (Dicentrarchus labrax): parameterisation and application in aquaculture.
J. Sea Res., Ecosyst. Based Manag. Biosph.
Metabolic theories in ecology
Trends Ecol. Evol.
Should we care if models are phenomenological or mechanistic?
Trends Ecol. Evol.
A clockwork fish: age prediction using DNA methylation-based biomarkers in the European seabass
Mol. Ecol. Resour.
Physiological modes of action across species and toxicants: the key to predictive ecotoxicology
Environ. Sci.
Cited by (10)
Bioenergetics
2022, Fish NutritionGlobal assessment of the response to chronic stress in European sea bass
2021, AquacultureCitation Excerpt :For the 4 organs, the function selected the same two reference genes: l13 and fau. The DEB model has previously been parametrized for European sea bass (Lika et al., 2018) and used to explain metabolic trends both in nature (Sadoul et al., 2020) and in controlled conditions (Stavrakidis-Zachou et al., 2019). Based on longitudinal data of body mass and length, we used the DEB model (equations provided in Table S1) to estimate the amount of assimilated energy and test whether chronic stress has significant effects on this variable.
Size scaling of sea turtle reproduction may reconcile fundamental ecology and conservation strategies at the global scale
2022, Global Ecology and Biogeography