Ecological and evolutionary factors of intraspecific variation in inducible defenses: Insights gained from Daphnia experiments

Abstract Phenotypic variation among individuals and species is a fundamental principle of natural selection. In this review, we focus on numerous experiments involving the model species Daphnia (Crustacea) and categorize the factors, especially secondary ones, affecting intraspecific variations in inducible defense. Primary factors, such as predator type and density, determine the degree to which inducible defense expresses and increases or decreases. Secondary factors, on the other hand, act together with primary factors to inducible defense or without primary factors on inducible defense. The secondary factors increase intraspecies variation in inducible defense, and thus, the level of adaptation of organisms varies within species. Future research will explore the potential for new secondary factors, as well as the relative importance between factors needs to be clarified.


| INTRODUC TI ON
Organisms can change their phenotypic traits (morphology, behavior, and physiology) and adapt to environmental variations. The ability of a single genome to produce a range of phenotypes in response to environmental conditions is called phenotypic plasticity (Agrawal, 2001;Fordyce, 2006). In general, the degree of phenotypic plasticity has a direct effect on fitness and therefore represents an important feature of the organism's adaptation.
The change in traits observed in phenotypic plasticity may not be binary (high and low) or represented by an on/off reaction but rather a continuous process in individuals (Auld, Agrawal, & Relyea, 2010;Forsman, 2015). Owing to this variation, individual organisms differ in cost and/or adaptive status relative to that of the optimal phenotype in a giving environment. Costs of inducible phenotypes are a central component of the evolution of plasticity (Auld et al., 2010;DeWitt, Sih, & Wilson, 1998) but have proven difficult to measure empirically. Variation in phenotypic plasticity can produce several adaptive states (i.e., adaptive, maladaptive, or neutral); therefore, studies of phenotypic plasticity tend to focus on cost detection and adaptation status (Auld et al., 2010;Murren et al., 2015). Because even trait variation of phenotypic plasticity is linked to evolution (Bolnick et al., 2011), it is important to clarify why variance in plasticity traits occurs and is maintained in the environment.
Predation is an important factor driving natural selection, and defensive traits are expressed against predators in a plastic or constitutive manner. Daphnia (Arthropoda Crustacea) is an excellent model system for studying predator-induced plasticity (Lass & Spaak, 2003;Tollrian & Dodson, 1999), with alterations in their phenotype against predators including changes in body size, head shape, tail length, number of eggs, reproduction status, and distribution depth (Lass & Spaak, 2003). To express predator-induced plasticity, Daphnia need | 8555 NAGANO ANd dOI to perceive predatory kairomone (chemical substance) and/or other factors besides predators; the former is called primary factor and the latter secondary factor (Riessen & Gilbert, 2019). Riessen and Gilbert (2019) suggested in a review that secondary factors are related to increases or decreases in the degree of plasticity. This suggests that predator-induced plasticity displays different trait values among individuals owing to the interaction between primary and secondary factors. Therefore, a wide range of factors can induce predator-induced plasticity. Considering variations in predator-induced plasticity, it is important to consider how secondary factors as well as the essential triggers work. There are numerous studies focusing on the predator-induced plasticity of Daphnia, making it potentially feasible to target and synthesize the various secondary factors affecting variations in this plasticity. Daphnia are tractable in various experimental settings and can be analyzed with modern genomic tools (Miner, De Meester, Pfrender, Lampert, & Hairston, 2012) and large-scale gene expression technology (Colbourne et al., 2011). Specifically, Daphnia pulex is the first crustacean to have its whole genome sequenced (Colbourne et al., 2011). Moreover, multiple studies of Daphnia have identified the neural mechanisms associated with predator-induced defenses (Miyakawa, Sugimoto, Kohyama, Iguchi, & Miura, 2015;. It can also be argued that, based on the predator-prey system, the elucidate secondary factors regulating variations in Daphnia plasticity could lead to a deeper understanding of phenotypic plasticity. The goal of this review is to clarify variations in predator-induced plasticity in Daphnia and summarize the secondary factors influencing those variations. We begin with a brief overview of variations of inducible defenses in Daphnia and then examine the relationship between plasticity variation and the various secondary factors involved. Recent theoretical works indicate that intraspecific trait (nonplasticity) variation can have significant ecological effect (Bolnick et al., 2011) and the variation of degree of expression in inducible defense might have likewise significant relationship ecological and evolutionary context. Exploring such variations associated with inducible defense is a critical step in clarifying how changes in traits occur and are maintained according to the environment.

| RE VIS ITING THE IMP ORTAN CE OF VARIATI ON S IN INDUCIB LE DEFEN S E
Ecologist have long recognized intraspecific variation in inducible defense; here we explore the factors involved in intraspecific variation in the inducible defense of Daphnia and synthesize the findings reported by empirical studies. Phenotypic changes show both qualitative (the presence or absence of spines) and quantitative (body size, spine length, and/or migration behavior) traits. Moreover, Daphnia express a combination of several unique, species-specific defensive traits in response to chemical cues (self-induced defense; a primary factor) initiated by predators, such as fish and invertebrates (Boeing, Ramcharan, & Riessen, 2006a, 2006bBoersma, Spaak, & De Meester, 1998). Although predator-induced plasticity in Daphnia includes a broad range of traits and shows complicated expression patterns, studies might underestimate or overestimate the variation based on evaluation of only average values for a single trait. Stoks, Govaert, Pauwels, Jansen, and De Meester (2016) used univariate and multivariate analyses of phenotypic plasticity to identify a natural Daphnia magna population capable of rapidly tracking changes in fish predation. This integrated, multi-trait approach improved our understanding of the evolution of phenotypic plasticity. The combined value of all the variation capacities of an individual (growth stage and multiple traits) in phenotypic plasticity would be measured as a potential capacity for adaptation.
Even if a change in one trait appears to be adaptive, other traits may appear to be maladaptive. This discrepancy is referred to as "trait compensation" (DeWitt, Sih, & Hucko, 1999) and suggests that the adaptability of an individual cannot be measured using only one trait. Specific traits complement one another, and inducible defenses can show both progression and regression of multiple traits in an individual (Boeing et al., 2006b;Boersma et al., 1998). In fact, these can occur simultaneously, which warrants the simultaneous observation of multiple traits. From a cost-benefit perspective, Daphnia might develop only a few inducible defense characteristics (Boersma et al., 1998), indicating that the expression of multiple defensive traits is associated with a certain cost in the forms of maintenance, production, and information acquisition. If a single trait is sufficient as an inducible defense against multiple predators, it could be unnecessary to develop multiple defensive traits. For example, development of only an elongated spine can make it more difficult for Daphnia to be captured by several predators (Caramujo & Boavida, 2000), which lowers the cost of acquiring this characteristic (Laforsch & Tollrian, 2004b). In this situation, the costs remain the same, but the benefits increase if it helps against multiple predators at once.
The primary factor is the most important aspect of variation in inducible defense in Daphnia. The factors of predators can be separated into "predator species/type," "predatory kairomone," and "kairomone concentration" as main or primary factors. First, Daphnia must contend with predators that are size-selective regarding to their prey (Dodson, 1974). The predation type for invertebrates is generally gape-limited predation that shows preference for small zooplankters, whereas vertebrate predators, such as fish, tend to be large zooplankters (Brooks & Dodson, 1965). Therefore, Daphnia will know exactly what kinds of predators are existing there and will express a moderate degree of defense accordingly. In a meta-analysis, Riessen (1999) showed that the life history responses of Daphnia to Chaoborus larvae differ substantially from those to Notonecta and fish. In the presence of small-size-selective predation by Chaoborus larvae, Daphnia mature later and show a larger size at that time. By contrast, under large-size-selective predation by fish, Daphnia reproduce early and are small at maturity (Riessen, 1999). Daphnia sizes vary among species (Gliwicz, 1990); body size is an important factor in terms of inducible defense traits.
The essential trigger includes predatory kairomone or kairomone concentration. Several studies report strong evidence for dose dependence where inducible defense is concerned (Dennis, Carter, Hentley, & Beckerman, 2010;Hammill, Rogers, & Beckerman, 2008;Parejko & Dodson, 1990), and the degree of defense expression tends to vary directly with predator abundance or kairomone concentration. However, studies show that the degree of dose-specific plasticity does not increase indefinitely as kairomone concentration increases, but reach a saturation point beyond which no additional changes in plasticity occur (Hammill et al., 2008;Reede, 1995;Weetman & Atkinson, 2002). This suggests that plasticity expression is constrained by what is not predatory kairomone. Given the avoidance of mismatching phenotypes, secondary factor may help the control, accelerate, and limit of the expression of defensive plasticity, in addition to ensuring the reliability of primary factors.

| Abiotic factors
Organisms may remember more accurate and reliable cues in order to predict and to know the presence of predators, although reliable cue selection mechanisms are unknown. If the emergence of predators is seasonal/temporal, Daphnia may be able to detect and respond to abiotic seasonal factors. Abiotic factors, including water temperature (Bernot, Dodds, Quist, & Guy, 2008;Hanazato, 1991;Lass & Spaak, 2003;Sakwinska, 1998;Weetman & Atkinson, 2002;Yurista, 2000), turbulence ( Laforsch & Tollrian, 2004b;2006), light (Boeing, Leech, Williamson, Cooke, & Torres,2004 ;Rhode, Pawlowski, & Tollrian, 2001;Rose, Williamson, Fischer, Connelly, Olson, & Noe, 2012 ), and copper and other minerals (Hunter & Pyle,2004 ;Mirza & Pyle, 2009), can affect the degree of predator-induced plasticity, but there is no fixed trend. These factors may work together with the primary factors, or they may work on their own. These abiotic factors may change the chemical composition of the predatory kairomone and thus reduce their effect on the organism. Temperature manipulation has shown that the degree of plasticity varies with differences in temperature alone, regardless of kairomone concentration (Sakwinska, 1998), and that other crustaceans have spines that elongate in the absence of kairomone but only at high temperatures (Miehles, McAdam, Bourdeau, & Peacor, 2013). Since these abiotic factors strongly influence the survival and life history traits of daphniids in the first place, abiotic factors may often limit expression plasticity even when the primary factors are detected.
The degree of expressed plasticity is thought to be both enhanced and suppressed in such environments and may be enhanced when Daphnia links periodic changes (i.e., seasons) in predator presence to physical stimuli and may be suppressed in the absence of relationships with cycles (Riessen & Gilbert, 2019 type of factor a "proxy cue" (Miehles et al., 2013). These factors are associated with local predator regimes and thereby cause intraspecific variation between populations. If primary factors are not reliable cues of predation risk, the abiotic factors would be accurate and useful factors. Moreover, abiotic factors that correlate with selective agents work similarly to primary factors and alone can cause an inducible defense on their own (Miehles et al., 2013). The phenomenon of inducible defense without primary factors is well known, although there is a lack of experimental support for identifying these factors. This factor may be the most reliable cue of the emergence, presence, and predation cycle of predators that is closest to Daphnia itself.

| Ecological and evolutionary traps
Organisms can incorrectly express phenotypes owing to artificial

| Food
Food level is not only a basic element of growth, but also a critical factor in modifying inducible defenses (e.g., depth-selective behavior [Loose & Dawidowicz, 1994]; morphological defenses [Tollrian,1995b]  For instance, inducible defense under low food level is expressed, but to a lesser extent (Barry, 1995). The degree of expressed plasticity has been found to be greater at high food levels and lower at low food levels, with other clones responding in the opposite direction (Jeyasingh & Weider, 2005

| Clones/genotypes
The degree of expression plasticity commonly varies between clones (morphological defense, Boeing et al., 2006;Declerck & Weber, 2003;Ferrari, Müller, Karaaijeveld, & Godfray, 2001;Havel, 1985;Hammill et al., 2008;Jeyasingh & Weider, 2005;Lively, Hazel, Schellenberger, & Michelson, 2000;Miyakawa et al., 2015;Rabus & Laforsch, 2011;Spitze, 1992;Weider, 1985;Wiąckowski, Fyda, Pajdak-Stós, & Adamus, 2003, life history traits ;Weider &Pijanowska, 1993, andbehavioral traits Michels, Amsinck, Jeppesen, &De Meester, 2007). Interclonal variation in the expression of inducible defenses originates from habitats with different predation regimes (Boeing et al., 2006a;Boersma et al., 1998Boersma et al., , 1999Dennis et al., 2010). The interclonal variations in the type and degree of inducible defense of Daphnia hyalina result from seasonal variations in the clonal composition of field populations (1985Stibor & Lampert, 2000. Moreover, this might partly account for the seasonally different occurrence of defended and undefended morphs in the field, caused by changing predator regimes (Havel, 1985). The variation in degree of expression inducible defense is predicted might be greater between species than between clones, although F I G U R E 2 Conceptual diagram outlining the factors of intraspecific variations in predator-induced plasticity no comparisons have been made. However, clonal variations are not negligible or small enough to be ignored. If the variation in the degree of plasticity is greater for clonal variation than for interspecific variation, then natural selection might be working strongly within the species.

| Instars
Although it is unclear how Daphnia itself perceives own body size, the body size is an important factor in determining the extent to which inducible defense should be expressed (Hart & Bychek, 2011;Tollrian, 1995a,). This is because predation sensitivity changes with age/instar changes in body size. It is important to be able to identify the type of predator, that is, gape-limited or visual predator, by primary factors at first. Chaoborus larvae prefer a narrow range of small-sized prey (Pastorok, 1981;Swift & Fedorenko, 1975), whereas fish prefer larger-sized prey, because they are readily visible (Brooks & Dodson, 1965;Nunn, Tewson, & Cowx, 2012). Hence, inducible defense varies among instars. For example, neckteeth induction is stronger at the 2nd and 3rd instars of Daphnia than at other stages (Tollrian, 1993;Tollrian, 1995aTollrian, , 1995bImai, Naraki, Tochinai, & Miura, 2009), because the former are the most vulnerable to Chaoborus larva predation. Therefore, depending on the trait, the degree of expression plasticity can be varied large within instar. The presence of fish chemicals decreases Daphnia body size (Boersma et al.,1999Brett, 1992Carter, Silva-Flores, Oyanedel, & Ramos-Jiliberto, 2013;Fisk, Latta, Knapp, & Pfrender, 2007;Weber & Declerck, 1997). Daphnia expresses inducible defense throughout its entire lifespan in the presence of predators capable of ingesting prey of any size (Laforsch & Tollrian, 2004b;Rabus & Laforsch, 2011).

| Maternal effect
Inducible defense can be transmitted to the next generation as a history of predation. The degree of defensive traits in the daughter generation of Daphnia cucullata depends on the extent to which the maternal line was exposed to predation by Chaoborus larvae (Agrawal, Laforsch, & Tollrian, 1999). The exposure of kairomone during embryonic and postembryonic development of D. pulex is required to allow adequate extension of head length. (Dennis, LeBlanc, & Beckerman, 2014;Miyakawa et al., 2010). However, not all plasticity traits are dependent on maternal effects (Mikulski & Pijanowska, 2017), and it is adaptive because the next generation can express the defensive trait without the cost of perceiving primary factors.

| CON CLUS ION
The variation of degree in inducible defense of Daphnia among conspecific individuals has long been recognized in experimental and field work. Despite a fast-growing study on the variation in inducible defense, we lack a general framework for understanding the variation by which factors influences to express. Then, we classified seven secondary factors related to evolutionary and ecology in predator-induced plasticity. The secondary factors can be distinguished by their relative relationship to primary factors, that is, presence of predator and/or predatory kairomone. Abiotic factors, food, clone/ genotype, and instars are promoted or inhibited the degree of expression in inducible defense by working with primary factors. And while abiotic factors, ecological traps and alarm substance, and maternal effect may work alone, but the degree of expression by them may be equivalent, smaller, or larger compared with the degree of variation from the primary factors. Variation of inducible defense is associated with vulnerability of predator. Therefore, it will be important to clarify the factors and the degree of variation in the future.

| FUTURE D IREC TI ON S
Research into inducible defenses in field populations is informative; however, recent studies were often based on laboratory experiments. In the laboratory, predatory kairomones are prepared based on a "kairomone recipe" that is generally established at a much higher concentration than that in nature. It is believed that Daphnia will react sufficiently in the presence of appropriate stimuli; therefore, preparation of a "kairomone recipe" does not assume the same response in any population of any species. Additionally, the degree of expression of inducible defense inducible defenses differ among populations of the same species owing to local adaptation (Boersma, De Meester, & Spaak, 1999;Boeing et al., 2006a;Reger, Lind, Robinson, & Beckerman, 2018 (Luecke & Litt, 1987;Nagano & Doi, 2018). We will attempt to elucidate the reasons for the discrepancy between experimental and field specimens in terms of their comparative degrees of inducible defense expression.
A major goal of evolutionary biology is to understand the mechanisms involved in creating biodiversity. Recent data concerning variations in phenotypic plasticity have promoted ecological speciation but with little empirical evidence . Although speciation involves several processes , phenotypic plasticity is thought to be helpful in the early stages of speciation (Forsman, 2015;Snell-Rood, 2013 Because water temperature is a major secondary factor, research into the phenotypic plasticity of living organisms in response to climate change will become increasingly significant in the future (Crispo et al., 2010;. Future studies should still consider not only the response of physiological activity against climate change, but the effect on predator-prey dynamics. Animal personality remains constant, regardless of environmental variation (Dingemanse et al., 2009;Sih, Bell, Johnson, & Ziemba, 2004;Wolf & Weissing, 2012), and inducible defenses can vary because of personality differences (e.g., bold and shy) regardless of the presence of predators (crucian carp; Hulthén, Chapman, Nilsson, Hollander, & Brönmark, 2014). This study showed that bold individuals undergo more substantial morphological changes than shy individuals. In contrast, shy individuals vary considerably in terms of evasion behavior. Therefore, personality-induced variation in inducible defense may be seen as both an adaptive and a maladaptive response. Under various environments and situations within the same species, bold individuals will have wide activity ranges, whereas shy individuals will have a narrow range. As Daphnia seem to have a personality (Heuschele, Ekvall, Bianco, Hylander, & Hansson, 2017), this species merits further investigation of personality as a factor contributing to variations in inducible defense.
Depending on personality, the degree of expression in plasticity is expected to vary, as in the case of the crucian carp.

ACK N OWLED G M ENTS
Thanks to Dr. Sakamoto for his valuable comments on an earlier version of the manuscript. We would like to thank anonymous reviewers for insightful comments on this manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.