- Email: [email protected]
Developmental plasticity and the evolution of parental effects
Review
Developmental plasticity and the evolution of parental effects Tobias Uller Edward Grey Institute, Department of Zoology, University of Oxford, Oxford OX1 3PS, UK
One of the outstanding challenges for evolutionary biologists is to understand how developmental plasticity can influence the evolutionary process. Developmental plasticity frequently involves parental effects, which might enable adaptive and context-dependent transgenerational transmission of phenotypic strategies. However, parent–offspring conflict will frequently result in parental effects that are suboptimal for parents, offspring or both. The fitness consequences of parental effects at evolutionary equilibrium will depend on how conflicts can be resolved by modifications of developmental processes, suggesting that proximate studies of development can inform ultimate questions. Furthermore, recent studies of plants and animals show how studies of parental effects in an ecological context provide important insights into the origin and evolution of adaptation under variable environmental conditions. Developmental plasticity and parental effects There is a growing awareness that evolutionary biologists need to redirect their focus away from a narrow genecentered view and toward developmental aspects of phenotypic evolution, to fully understand the evolution of organismal form and function [1–6]. This change in focus requires us first to extend our view of heredity to include epigenetic inheritance [1–3] and second to address the role of developmental plasticity in initiating and directing adaptive evolution [4,5]. Together, the transmission of non-genetic developmental factors and developmental plasticity result in an effect of the parental phenotype on offspring phenotype that cannot solely be ascribed to inherited genes. Thus, parental effects [7] represent a form of plasticity that spans generations. Although parental effects initially received scant treatment by ecologists and evolutionary biologists, there has been a rapid accumulation of studies since a landmark publication 10 years ago [7]. This book has been instrumental for the general acceptance of parental (in particular maternal) effects as an important factor in evolution. However, many limitations to our understanding of evolution of parental effects still remain. For example, whereas theoretical work has focused on the consequences of parental effects for the short-term response of traits to selection [8–11], empirical studies tend mainly to be concerned with whether or not parental effects are adaptive [7,12–14]. Thus, it remains unclear under what circumstances parental effects are selected for, Corresponding author: Uller, T. ([email protected])
432
and the relative contribution of evolution of parental and offspring strategies [14,15]. This review attempts to take a step toward filling this gap by outlining a framework for the evolution of parental effects. It does so by integrating selection, parent–offspring conflict [16,17], developmental plasticity and proximate mechanisms of interactions between parents, offspring and their environments. Such an approach helps to clarify the circumstances that will select for the parental and offspring strategies that comprise parental effects, and to estimate their fitness consequences. Importantly, the approach also shows that understanding the evolution of parental effects requires knowledge both of the selection pressures involved and of the ways in which conflicts and constraints can be resolved. Evolution when everyone agrees: adaptive transgenerational phenotypic plasticity Parental effects represent a form of phenotypic plasticity spanning generations. Models of within-generation phenotypic plasticity generally predict that plasticity is favored when (i) there is environmental heterogeneity (spatial or temporal), (ii) there are cues that reliably predict future environmental conditions and, hence, selective regimes and (iii) the cost of plasticity is low [18,19]. Applied to the context of parental effects, this suggests that parentally induced plasticity is favored when (i) there is temporal or spatial fluctuation in environments across generations, (ii) offspring environmental conditions are predictable from the parental environmental conditions or the parental phenotype and (iii) the cost of obtaining information and responding appropriately is low for both parents and offspring. Furthermore, the level of parent– offspring conflict should be low, because otherwise the optimal level of plasticity will differ for parents and Glossary Development: phenotypic change during the lifetime of an individual. Developmental plasticity: the ability of a genotype to give rise to multiple phenotypes in different environments. Epigenetic inheritance: according to the broad definition, epigenetic inheritance means transgenerational transmission of form by other means than DNA. According to the narrow definition it implies heritable variation in DNA expression or activity, for example via persistent DNA methylation. Parent–offspring conflict: evolutionary conflict arising from different fitness optima in parents and offspring. Parental effect: an effect of the parental phenotype on offspring phenotype that cannot solely be ascribed to the parental or offspring genome, non-parental components of the environment or the interaction between these. Phenotypic plasticity: synonymous with developmental plasticity.
0169-5347/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2008.04.005 Available online 27 June 2008
Review
Trends in Ecology and Evolution
Vol.23 No.8
offspring and fitness might not be maximized for any of the individuals (see below). The above criteria seem to apply to many taxa, and parental effects that increase the fitness of both offspring and parents have indeed been documented in both plants and animals (Boxes 1,2). However, across-generation heterogeneity and selection for developmental plasticity are not sufficient for parental effects to be favored. Cues to what constitutes optimal phenotypic development can be obtained by the offspring either through the parents or through other aspects of offspring environment. Thus, whether the parental phenotype will affect offspring
development should also depend on the reliability of information provided by parents versus that obtained through the non-parental environment. Furthermore, parental effects will only be favored when there is a high likelihood that parental induction will allow an appropriate timing of the adaptive response in offspring (Box 1). For example, in plants and insects, lack of offspring sensory organs that can detect reliable cues early in development might explain why germination, diapause and dispersal strategies are commonly under maternal control, rather than being affected by environmental conditions experienced by the offspring [20,21].
Box 1. Predictability and the evolution of adaptive parental effects
Evolution of parental effects under parent–offspring conflict Parental effects represent the outcome of interactions between parents and offspring. Thus, their evolution will be affected by selection pressures acting on both generations, and parent–offspring conflicts will occur whenever there are different optimal strategies for parents and offspring [16,17]. This means that selection on parents might favor parental effects that reduce offspring fitness [14,22]. By contrast, offspring are under selection to maximize their own fitness, not that of the parents [16]. Thus, the evolutionary dynamics of parent–offspring conflict will depend on the relative costs and benefits to the interacting individuals, the strength of selection and the potential for control. The fitness consequences of parental effects for parents and offspring reflect the evolutionary outcome of this conflict (i.e. who wins, and at what cost?) [16,17,23].
Adaptive parental effects rely upon cues that predict offspring environment. Three scenarios are particularly likely to fulfill this criterion, because they imply that cues are easily obtained by the parents at the time of reproduction. First, there might be a strong correlation between maternal and offspring environments, as is often the case in seasonal environments (Figure Ia, dashed arrows). For example, in some insects, deteriorating maternal environmental conditions indicate onset of winter and select for the production of resting eggs or diapausing larvae that over-winter and resume development in spring [20]. Use of parental, rather than direct environmental, cues are particularly likely to evolve when this reduces the risk of a mismatch between offspring phenotype and selection on that phenotype, and when plasticity is expressed at very early life stages when offspring sensory organs are poorly developed. Furthermore, induction of major developmental changes (such as presence/absence of wings in aphids or helmets in Daphnia) requires less reorganization if they occur earlier rather than late in development. Second, the environment in which the offspring develop might be a direct consequence of the parental phenotype (Figure Ib, large solid arrows). This is the case, for example, when mothers oviposit on different hosts [20] or when maternal allocation decisions (e.g. clutch size and sex ratio) affect optimal offspring development [25,47]. Thus, offspring obtain both their environment and the optimal solution to the environment as parental effects [48]. In many animals with parental care, the maternal phenotype is the most important environmental component for a prolonged period of development, favoring co-adaptation of parental phenotype and offspring development [49]. Third, the genetic constitution of the offspring might be affected by maternal or paternal phenotypes, for example, via mate choice (Figure Ic, dashed arrows). If genotypes have different developmental optima, selection should favor parental transmission of nongenetic developmental resources that match the genetic constitution of the offspring [24,50]. For example, adaptive sex ratio adjustment frequently involves simultaneous adjustment of offspring genotype (both via mate choice and via sex manipulation in species with genetic sex determination) and developmental conditions via parental effects [24,50].
Figure I. Yellow beetles have high fitness on the light background and red beetles have high fitness on the dark background. Solid arrows represent parental transmission of developmental factors, and dashed arrows represent other direct and indirect relationships between environments and phenotypes.
Costs and benefits of parental effects The strength of selection on parental and offspring strategies depends on the costs and benefits involved. Many parental effects, such as investment in egg size or parental care, will reduce the future reproductive success of parents and therefore come at a cost. Other effects might entail minor costs for the parents, such as active or passive transmission of trace elements or hormones [13]. Parental effects can also entail costs for the offspring. For example, in birds, offspring have to compete for food by exhibiting costly behaviors, which reduces the amount of ingested food that can be directly converted to growth [23]. The costs of parental effects are important because they influence the level of parent–offspring conflict via the tradeoff between current and future reproductive success of parents, and because they offset the benefits of offspring strategies that manipulate parents. Thus, the evolutionary dynamics of parent–offspring interactions might differ depending on whether the parental effect involves costly investment or cheap manipulation. Transmission of costly resources When offspring cannot exercise any control over parental investment, parents might be able to impose their own optimal allocation to offspring within and between broods, despite negative effects on offspring [14]. For example, insect females do not always oviposit on hosts that are optimal for offspring development, but rather optimize maternal fitness via a tradeoff between search time and feeding time [22]. Similarly, females are expected to adjust 433
Review
Trends in Ecology and Evolution Vol.23 No.8
Box 2. Adaptive transgenerational plasticity in plants Environments are frequently heterogeneous at the population level, with corresponding variation in phenotypic optima. Dispersal or active habitat choice might cause nonrandom associations between the environment experienced by parents and offspring. For example, in many species of plants, seed dispersal is limited, suggesting that the maternal environment provides cues about the selective pressure experienced by the offspring [14,21]. In some cases, this might lead to genetic assimilation of developmental plasticity and local adaptation at a small spatial scale [4,51]. However, when pollen dispersal occurs at significantly longer distances than does dispersal of seeds, the paternally inherited genotype does not provide accurate cues to optimal phenotypic development [12,52,53]. Thus, phenotypic plasticity in response to the local environment should be favored. One possible response is that the offspring phenotype is modulated in response to the maternal environment, that is, a maternal effect on seedling development [12,21]. Recent studies of the herb Campanulastrum americanum (Figure I) have provided substantial evidence for a positive correlation between offspring and maternal, but not paternal, light conditions during growth [12]. Previous studies have shown that seeds from plants grown under good light conditions have an increased probability of germinating in the fall and, hence, of developing as an annual [54]. This might be adaptive if it increases population growth rate under favorable conditions. Indeed, a factorial field experiment [54] showed that seeds grown in the same environment as their mother had increased survival compared to seeds grown in the opposite environment. Furthermore, despite the fact that the maternal environment did not have a significant effect on adult fitness, the projected population growth was higher when environments were matched as a result of differential maternal effects on offspring survival and life history [54]. Parental effects on germination could be favored even in the face of occasional seed dispersal between habitats, as the negative effect will only persist in one generation. The documentation of adaptive transgenerational plasticity in this and other plants implies exciting opportunities for addressing the underlying mechanisms of transgenerational plasticity, and investigating to what extent such plasticity involves epigenetic programming that could span more than one generation [42,43].
egg size in relation to both current and future reproductive returns, even if decreasing egg size has negative effects on offspring fitness [14]. Such differential allocation in relation to current and future reproductive returns are common in animals such as birds and insects [24] and suggests that the adaptive value of parental effects must be addressed from a life-history perspective [14,25]. When offspring are able to influence the resource allocation of their parents, however, the evolutionarily stable parental effect might not reflect the parental optimum. Most theoretical work on parent–offspring conflict over negotiable resources (i.e. when both offspring and parents can respond to each others’ strategies) has been inspired by maternal or paternal food provisioning in birds and mammals [17,23]. However, several assumptions and predictions of the theory have also recently been put to the test using invertebrate systems with parental care [26,27], and the models might have broader applications. A common theoretical outcome of parent–offspring conflict over costly parental effects is that parents tend to provide more resources than is optimal for them [17]. The extent of the deviation from the parental optimum is set by the degree to which increased resource provisioning reduces (costly) offspring solicitation, and the degree to which solicitation affects parental provisioning [17,28], both of which can evolve. Empirical tests of whether the level 434
Figure I. Campanulastrum americanum. Picture by Laura Galloway.
of parental investment is at the parental or offspring optimum (or at neither) are hampered by the difficulty of measuring the parental fitness returns of variable resource provisioning to specific offspring or clutches, as both the returns from the current brood and returns from future broods need to be taken into account. Relatively novel model systems, such as burying beetles [26,27], might prove to be highly useful in this context. Finally, parental effects might also be sensitive to conflicts over resource investment between parents [29,30]. This complicates the evolutionary dynamics and can even lead to the evolution of parent-of-origin effects on gene expression (i.e. genomic imprinting [31]). For example, when the paternal and offspring optimum of maternal investment is higher than that of the mother, and offspring can influence the amount of resource investment, genes that increase maternal provisioning when inherited paternally will be positively selected [31,32]. Indeed, imprinting seems to be most common in mammals, perhaps because of the high potential for offspring manipulation of maternal resource provisioning. Transmission of manipulative signals When parental effects are cost free in terms of parental acquisition, production and transmission of developmental factors, the evolutionary outcome can be very different
Review from when they are costly to parents. Consider a parental strategy that can induce a plastic response in the offspring without any direct costs to the parents. One example might be hormone allocation to eggs or embryos [13,33], but the argument should apply to any parental effect that does not compromise future reproductive success of parents. Selection can initially favor the use of manipulative signals, such as parental hormone allocation, even if it has an immediate negative effect on the fitness of individual offspring. For example, parents might use manipulative signals to mediate sibling conflict under asymmetric offspring competition by reducing the competitive ability of specific offspring within the brood, thereby increasing parental fitness [23]. Alternatively, selection can favor parental reinforcement of sibling competitive asymmetry to increase the likelihood of survival of the most valuable offspring or to reduce the cost of parental investment [17,23]. However, selection on offspring will favor reduced offspring sensitivity or response to parental hormones, thereby generating an evolutionary arms race between parent and offspring. Evolutionary stability could be reached when offspring have lost the sensitivity to parental hormones altogether. Alternatively, the costs to parents might eventually outweigh the benefits, for example owing to negative pleiotropic effects of increased hormone levels on the physiology or behavior of parents [33]. Both conditions should result in a parental effect
Trends in Ecology and Evolution
Vol.23 No.8
closer to the offspring optimum. On the contrary, there might be constraints on the ability of the offspring to ignore the parental manipulation, which would allow for an outcome closer to the parental optimum. For hormones, this could arise as a result of a necessity for offspring to be sensitive to their own, endogenous hormones required for normal development, which could be indistinguishable from those produced and transmitted by the parents [33,34]. Selection on offspring will also favor the use of existing parental strategies according to the optimum of the offspring, not that of the parents. This might again lead to a parental effect close to the offspring optimum. However, selection on parents is then expected to reduce the transmission of information across generations, either via changes in the parental phenotype or via a temporal or spatial separation of generations (e.g. reduction in the duration of parental care), both of which reduce the availability of cues to offspring and lead to a reduction in the strength of the parental effects. Although explicit models of parental effects are rare, the arguments outlined above question the validity of some proposed adaptive scenarios, whereby parents are suggested to be able to impose their own optima using manipulative signals (as has been suggested for hormonemediated maternal effects [35]; see Ref. [33] for discussion). Furthermore, they might suggest that parental effects
Box 3. Resolving constraints: parental effects and sexual divergence Divergent selection on males and females implies that optimal development will differ for the sexes. Reproducing females thus face a dilemma when simultaneous production of the sexes constrains the opportunity for sex-specific maternal effects [15,47,55]. This dilemma can be resolved in several ways, including sex ratio bias, clutch size reduction or elimination of maternal effects, for example by reducing the period of parental care. However, these responses might compromise female fecundity or offspring survival, suggesting that alternative mechanisms to minimize the developmental overlap between the sexes will be under positive selection. Once they have evolved, such mechanisms might set the stage for considerable maternal effects on the ontogenetic divergence between the sexes [15,56,57]. Resolution of developmental constraints and the evolution of adaptive sex-specific parental effects are best exemplified by studies of the house finch Carpodacus mexicanus (Figure I) in North America [15,50,58–61]. Using a heuristic model of egg development and analyses of the daily uptake of maternally provided nutrients by the growing oocytes, it has been possible to show that oocytes destined to become male or female differ in their timing and duration of growth [58]. This leads to temporal clusters of oocytes that will develop into the same sex, which minimizes developmental overlap between the sexes [59–61] and ensures sex-specific accumulation of yolk components (e.g. androgens [50]). The regulation of these processes could occur through a single hormone, prolactin, via its effects on the onset of incubation behavior, plasma steroid levels, oocyte growth and follicular steroidogenesis [58,62]. Furthermore, the house finch studies suggest that the concentration of a small number of key developmental factors directly affect meiosis and therefore the sex of the resulting ovum (Z or W) via its effect on chromosome segregation during the first meiotic division [61,63]. This mechanism of sex biasing is in accordance with predictions that parental or environmental effects with sex-specific fitness consequences are likely to be co-opted as sex-determining factors [4,64]. The ability for flexible adjustment of sex-specific growth and development via maternal effects has facilitated the rapid colonization
of North America by the house finch in two steps: first, by increasing developmental plasticity in the initial phase of invasion and, second, via subsequent accommodation of plastic but precise parental effects to enable matching of offspring phenotype to local conditions [15,65,66]. Thus, parental effects might provide a key to colonization success of novel habitats in birds, as well as in other animals [4,67– 69].
Figure I. House finch Carpodacus mexicanus. Picture by Alex Badyaev.
435
Review based on the transmission of manipulative signals are inherently unstable unless there are limits to evolutionary responses in parental manipulation or development of offspring counterstrategies. Such limits could arise from several types of constraint, including physical, genetic and developmental constraints. Thus, understanding the evolutionary dynamics of parental effects requires researchers to integrate proximate and ultimate levels of explanation [15] (Box 3). Integrating proximate and ultimate levels of explanation in parental effects research An evolutionary response to current selection requires that novel input or modification of responses can be incorporated into existing developmental systems. Thus, development might constrain or bias evolution [4,36]. Furthermore, theoretical models of parental effects under situations of conflict have shown that the outcome is sensitive to assumptions regarding the ability of parents and offspring to develop counterresponses to novel strategies exhibited by the other party, both at ontogenetic and evolutionary timescales [17,37]. Empirical studies of proximate mechanisms might therefore provide insights into ultimate questions by showing how such constraints can be resolved, in a way that maintains the fecundity and survival of parents (Box 3). At the phenotypic level, parental effects are potentially sensitive to bias, limitations or constraints during several steps, from information acquisition by parents to the incorporation of developmental factors by offspring. For example, adaptive parental effects could be constrained owing to species-specific patterns of reproductive investment and parental care [15,34]. In mammals, the intimate physiological connection between mothers and offspring could make parental effects that are detrimental to offspring more likely by increasing the risk of negative carryover effects of strategies favored by strong parental selection. For example, maternal strategies that ensure her survival during stress, including reduced resource provisioning to offspring or increased levels of circulating steroids [38], could have immediate negative effects on offspring of her current brood. This is particularly likely if there are constraints on evolutionary responses of offspring that allow them to cope with variation in maternal phenotype, which shows the importance of mechanistic studies of development to understand evolutionary outcomes [33]. Close physical or physiological association between parents and offspring also increases the risk that parental exposure to novel environments or substances, such as endocrine disruptors or pollutants, will also have negative effects on offspring (e.g. [39]). Although parental effects generally are stronger early in life [7,40,41], they can have long-lasting consequences. This can be because of priming effects on phenotypic organization [4,38] or via lasting modification of gene expression via DNA methylation (i.e. epigenetic effects [1,42]). Intriguingly, studies on both plants and animals have shown that epigenetic effects not only remain stable within individuals, but also that they can be inherited across generations [1,39,42,43]. Thus, the presence of parental effects does not necessarily rely upon substantial 436
Trends in Ecology and Evolution Vol.23 No.8
contributions over and above genome transmission, suggesting that paternal effects might be more common than previously believed [44]. Furthermore, stable environmentally induced changes in gene expression provide a mechanism whereby short-term exposure to environmental conditions can have phenotypic consequences for more than two generations [1,42,43]. However, more transient parental effects can also have consequences for future generations, in particular if they affect mate preferences or sexual attractiveness of offspring (e.g. [45,46]). Documenting the proximate mechanisms of parental effects and their stability during transmission is therefore important to understand their adaptive value and to predict their effects on population dynamics. Concluding remarks The sheer number of studies of parental effects in virtually all major taxa is testimony to their importance. Furthermore, parental effects are fundamental to the ongoing integration of development, ecology and evolution [1–6], as they represent a particular form of plasticity arising from transmission of non-genetic developmental factors. To what extent parental effects involve permanent changes to gene expression [42,43] or only transient modifications of development remains poorly understood. However, if they are indeed stable, this could have far-reaching consequences for our understanding of the non-genetic transmission of form [1,42]. Progress in the field will require both the development of theoretical models that can be Box 4. Some challenges for the future Theoretical Extending models of within-generation plasticity to encompass transgenerational phenomena in systems with and without conflicts. Investigating the role of costs for both parents and offspring for the evolutionary dynamics of parental effects. Incorporating mechanistic assumptions to establish how constraints influence evolutionary outcomes. Addressing when epigenetic effects transmitted for more than one generation can be favored by selection. Developing models that link parental effects, population dynamics and evolution, for example, via parental effects on offspring attractiveness or reproductive output. Modeling the role of parental effects during species invasions and its consequences for the rate and direction of evolution during colonization of novel habitats. Empirical Putting models of the evolutionary outcome of parental effects with or without conflicts to the test in natural settings. Establishing comparative patterns of the strength and direction of parental effects for parental and offspring fitness in relation to key predictors, such as environmental heterogeneity and conflicting selection pressures. Using artificial selection experiments to tease apart the relative contribution of evolution of parental versus offspring strategies for the evolution of parental effects. Exploring mechanisms of parental effects and their consequences for evolutionary responses of parental and offspring strategies. Testing the stability of epigenetic inheritance across environmental contexts. Addressing how parental effects can affect population dynamics and facilitate range expansion or colonization of novel environments.
Review used to design and interpret empirical studies as well as explicit tests of adaptive scenarios in ecological settings (Box 4). Empirical tests of the adaptive value of parental effects are perhaps most promising in organisms such as plants where environmental heterogeneity often can be accurately described, conflicts are relatively weak or where there is limited scope for offspring manipulation of parental strategies [14,21]. Similarly, studies of egg or seed size are often freed from the potential for evolution of offspring counterstrategies, making those systems tractable for tests of adaptive scenarios. Parental effects arising from epigenetic modification via manipulative signals or cues in systems with conflicts are currently the least well understood. At the same time, they might provide some of the best opportunities for addressing the role of developmental plasticity and epigenetic inheritance for the origin and maintenance of complex adaptations more generally. Thus, the study of parental effects has the potential to become a prime example of how combining methods and concepts from different biological disciplines can lead to fundamental insights into evolutionary processes. Acknowledgements I am grateful to Alex Badyaev, Caroline Isaksson, Dustin Marshall, Geoff While and two anonymous reviewers for stimulating discussions and helpful suggestions on the manuscript. The Wenner-Gren Foundations and the Australian Research Council provided financial support.
References 1 Jablonka, E. and Lamb, M.J. (2005) Evolution in Four Dimensions. Genetic, Epigenetic, Behavioral and Symbolic Variation in the History of Life. MIT Press 2 Oyama, S. (2000) The Ontogeny of Information. Duke University Press 3 Laland, K.N. and Sterelny, K. (2006) Seven reasons (not) to neglect niche construction. Evolution Int. J. Org. Evolution 60, 1751–1762 4 West-Eberhard, M.J. (2003) Developmental Plasticity and Evolution. Oxford University Press 5 Sultan, S.E. (2007) Development in context: the timely emergence of eco-devo. Trends Ecol. Evol. 22, 575–582 6 Pigliucci, M. (2007) Do we need an extended evolutionary synthesis? Evolution Int. J. Org. Evolution 61, 2743–2749 7 Mousseau, T.A. and Fox, C.W. (1998) Maternal Effects as Adaptations. Oxford University Press 8 Kirkpatrick, M. and Lande, R. (1989) The evolution of maternal characters. Evolution Int. J. Org. Evolution 43, 485–503 9 Cheverud, J.M. and Moore, A.J. (1994) Quantitative genetics and the role of the environment provided by relatives. In Quantitative Genetic Studies of Behavioral Evolution (Boake, C.R.B., ed.), pp. 67–100, University of Chicago Press 10 Wolf, J.B. (2000) Gene interactions from maternal effects. Evolution Int. J. Org. Evolution 54, 1882–1898 11 Ra¨sa¨nen, K. and Kruuk, L.E.B. (2007) Maternal effects and evolution at ecological time scales. Funct. Ecol. 21, 408–421 12 Galloway, L.F. (2005) Maternal effects provide phenotypic adaptation to local environmental conditions. New Phytol. 166, 93–99 13 Groothuis, T.G.G. et al. (2005) Maternal hormones as a tool to adjust offspring phenotype in avian species. Neurosci. Biobehav. Rev. 29, 329– 352 14 Marshall, D. and Uller, T. (2007) When is a maternal effect ‘adaptive’? Oikos 116, 1957–1963 15 Badyaev, A.V. (2005) Maternal inheritance and rapid evolution of sexual size dimorphism: passive effects or active strategies? Am. Nat. 166, S17–S30 16 Godfray, H.J. (1995) Evolutionary theory of parent-offspring conflict. Nature 376, 133–138 17 Parker, G.A. et al. (2002) Intrafamilial conflict and parental investment: a synthesis. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 295–307
Trends in Ecology and Evolution
Vol.23 No.8
18 Schlichting, C.D. and Pigliucci, M. (1998) Phenotypic Evolution. A Reaction Norm Perspective, Sinauer Associates 19 Berrigan, D. and Scheiner, S.M. (2004) Modeling the evolution of phenotypic plasticity. In Phenotypic Plasticity. Functional and Conceptual Approaches (Dewitt, T.J. and Scheiner, S.M., eds), pp. 82–97, Oxford University Press 20 Tauber, M.J. et al. (1986) Seasonal Adaptations in Insects, Oxford University Press 21 Donohue, K.J. and Schmitt, J. (1998) Maternal environmental effects: adaptive plasticity? In Maternal Effects as Adaptations (Mousseau, T.A. and Fox, C.W., eds), pp. 137–158, Oxford University Press 22 Mayhew, P.J. (2001) Herbivore host choice and optimal bad motherhood. Trends Ecol. Evol. 16, 165–167 23 Mock, D.W. and Parker, G.A. (1997) The Evolution of Sibling Rivalry, Oxford University Press 24 Sheldon, B.C. (2000) Differential allocation: tests, mechanisms and implications. Trends Ecol. Evol. 15, 397–402 25 Plaistow, S.J. et al. (2007) How to put all your eggs in one basket: empirical patterns of offspring provisioning throughout a mother’s lifetime. Am. Nat. 170, 520–529 26 Smiseth, P.T. and Moore, A.J. (2007) Signalling of hunger by senior and junior larvae in asynchronous broods of a burying beetle. Anim. Behav. 74, 699–705 27 Smiseth, P.T. et al. (2007) Interaction between parental care and sibling competition: parents enhance offspring growth and exacerbate sibling competition. Evolution Int. J. Org. Evolution 61, 2331–2339 28 Royle, N.J. et al. (2004) Parental investment and family dynamics: interactions between theory and empirical tests. Popul. Ecol. 46, 231– 241 29 Wedell, N. et al. (2006) Sexual conflict and life histories. Anim. Behav. 71, 999–1011 30 Lessells, C.M. (2006) The evolutionary outcome of sexual conflict. Philos. Trans. R. Soc. Lond. B Biol. Sci. 361, 301–317 31 Haig, D. (2000) The kinship theory of genomic imprinting. Annu. Rev. Ecol. Syst. 31, 9–32 32 de Jong, T.J. and Scott, R.J. (2007) Parental conflict does not necessarily lead to the evolution of imprinting. Trends Plant Sci. 12, 439–443 33 Mu¨ller, W. et al. (2007) Manipulative signals in family conflict? On the function of maternal yolk hormones in birds. Am. Nat. 169, E84–E96 34 Uller, T. et al. (2007) Consequences of maternal yolk testosterone for offspring development and survival: experimental test in a lizard. Funct. Ecol. 21, 544–551 35 Schwabl, H. et al. (1997) A hormonal mechanism of parental favouritism. Nature 386, 231 36 Arthur, W. (2004) The effect of development on the direction of evolution: toward a twenty-first century consensus. Evol. Dev. 6, 282–288 37 Johnstone, R.A. (1996) Begging signals and parent-offspring conflict: do parents always win? Proc. Biol. Sci. 263, 1677–1681 38 Meaney, M.J. et al. (2007) Epigenetic mechanisms of perinatal programming of hypothalamic-pituitary-adrenal function and health. Trends Mol. Med. 13, 269–277 39 Crews, D. et al. (2007) Transgenerational epigenetic imprints on mate preference. Proc. Natl. Acad. Sci. U. S. A. 104, 5942–5946 40 Wilson, A.J. and Reale, D. (2006) Ontogeny of additive and maternal genetic effects: lessons from domestic mammals. Am. Nat. 167, E23–E38 41 Lindholm, A.K. et al. (2006) Where do all the maternal effects go? Variation in offspring body size through ontogeny in the live-bearing fish Poecilia parae. Biol. Lett 2, 586–589 42 Jablonka, E. and Lamb, M.J. (1995) Epigenetic Inheritance and Evolution. Oxford University Press 43 Molinier, J. et al. (2006) Transgeneration memory of stress in plants. Nature 442, 1046–1049 44 Bonduriansky, R. and Head, M. (2007) Maternal and paternal condition effects on offspring phenotype in Telostylinus angusticollis (Diptera: Neriidae). J. Evol. Biol. 20, 2379–2388 45 Qvarnstro¨m, A. and Price, T.D. (2001) Maternal effects, paternal effects, and sexual selection. Trends Ecol. Evol. 16, 95–100 46 Uller, T. et al. (2005) Female egg investment in relation to male sexual traits and the potential for transgenerational effects in sexual selection. Behav. Ecol. Sociobiol. 57, 584–590
437
Review 47 Uller, T. (2006) Sex-specific sibling interactions and offspring fitness in vertebrates: patterns and implications for maternal sex ratios. Biol. Rev. Camb. Philos. Soc. 81, 207–217 48 Lock, J.E. et al. (2007) Coadaptation of prenatal and postnatal maternal effects. Am. Nat. 170, 709–718 49 Wolf, J.B. and Brodie, E.D., III (1998) Coadaptation of parental and offspring characters. Evolution Int. J. Org. Evolution 52, 299–308 50 Badyaev, A.V. et al. (2006) Evolution of sex-biased maternal effects in birds: III. Adjustment of ovulation order can enable sex-specific allocation of hormones, carotenoids, and vitamins. J. Evol. Biol. 19, 1044–1057 51 Linhart, Y.B. and Grandt, M.C. (1996) Evolutionary significance of local genetic differentiation in plants. Annu. Rev. Ecol. Syst. 27, 237– 277 52 Leimar, O. et al. (2006) A new perspective on developmental plasticity and the principles of adaptive morph determination. Am. Nat. 167, 367–376 53 Leimar, O. Environmental and genetic cues in the evolution of phenotypic polymorphism. Evol. Ecol. (in press). 54 Galloway, L.F. and Etterson, J.R. (2007) Transgenerational plasticity is adaptive in the wild. Science 318, 1134–1136 55 Langkilde, T. and Shine, R. (2005) Different optimal offspring sizes for sons versus daughters may favor the evolution of temperaturedependent sex determination in viviparous lizards. Evolution Int. J. Org. Evolution 59, 2275–2280 56 Badyaev, A.V. (2002) Growing apart: an ontogenetic perspective on the evolution of sexual size dimorphism. Trends Ecol. Evol. 17, 369–378 57 Moore, C.L. (1995) Maternal contributions to mammalian reproductive development and the divergence of males and females. Adv. Stud. Behav. 24, 47–118 58 Badyaev, A.V. et al. (2005) Adaptive sex differences in growth of preovulation oocytes in a passerine bird. Proc. Biol. Sci. 272, 2165–2172
438
Trends in Ecology and Evolution Vol.23 No.8 59 Young, R.L. and Badyaev, A.V. (2004) Evolution of sex-biased maternal effects in birds: I. Sex-specific resource allocation among simultaneously growing oocytes. J. Evol. Biol. 17, 1355–1366 60 Badyaev, A.V. et al. (2006) Evolution of sex-biased maternal effects in birds: II. Contrasting sex-specific oocyte clustering in native and recently established populations. J. Evol. Biol. 19, 909–921 61 Badyaev, A.V. et al. (2008) Evolution of sex-biased maternal effects in birds: IV. Intra-ovarian growth dynamics can link sex-determination and sex-specific acquisition of resources. J. Evol. Biol. 21, 449–460 62 Sockman, K.W. et al. (2006) Orchestration of avian reproductive effort: an integration of the ultimate and proximate bases for flexibility in clutch size, incubation behaviour, and yolk androgen deposition. Biol. Rev. Camb. Philos. Soc. 81, 629–666 63 Rutkowska, J. and Badyaev, A.V. (2008) Meiotic drive and sex determination: molecular and cytological mechanisms of sex ratio adjustment in birds. Philos. Trans. R. Soc. Lond. B Biol. Sci. 363, 1675–1686 64 Uller, T. et al. (2007) Evolution of sex ratios and sex-determining systems. Trends Ecol. Evol. 22, 292–297 65 Badyaev, A.V. et al. (2002) Sex-biased hatching order and adaptive population divergence in a passerine bird. Science 295, 316–318 66 Badyaev, A.V. and Oh, K.P. (2008) Environmental induction and phenotypic retention of adaptive maternal effects. BMC Evol. Biol. 8, 3 67 Badyaev, A.V. (2005) Role of stress in evolution: from individual adaptability to evolutionary adaptation. In Variation: A Central Concept in Biology (Hallgrimsson, B. and Hall, B.K., eds), pp. 277– 302, Elsevier Academic Press 68 Carroll, S.P. and Fox, C.W. (2007) Dissecting the evolutionary impacts of plant invasions: bugs and beetles as native guides. Glob. Change Biol. 13, 1644–1657 69 Cote, J. et al. (2007) Mother-offspring competition promotes colonization success. Proc. Natl. Acad. Sci. U. S. A. 104, 9703–9708
Developmental plasticity and the evolution of parental effects Tobias Uller Edward Grey Institute, Department of Zoology, University of Oxford, Oxford OX1 3PS, UK
One of the outstanding challenges for evolutionary biologists is to understand how developmental plasticity can influence the evolutionary process. Developmental plasticity frequently involves parental effects, which might enable adaptive and context-dependent transgenerational transmission of phenotypic strategies. However, parent–offspring conflict will frequently result in parental effects that are suboptimal for parents, offspring or both. The fitness consequences of parental effects at evolutionary equilibrium will depend on how conflicts can be resolved by modifications of developmental processes, suggesting that proximate studies of development can inform ultimate questions. Furthermore, recent studies of plants and animals show how studies of parental effects in an ecological context provide important insights into the origin and evolution of adaptation under variable environmental conditions. Developmental plasticity and parental effects There is a growing awareness that evolutionary biologists need to redirect their focus away from a narrow genecentered view and toward developmental aspects of phenotypic evolution, to fully understand the evolution of organismal form and function [1–6]. This change in focus requires us first to extend our view of heredity to include epigenetic inheritance [1–3] and second to address the role of developmental plasticity in initiating and directing adaptive evolution [4,5]. Together, the transmission of non-genetic developmental factors and developmental plasticity result in an effect of the parental phenotype on offspring phenotype that cannot solely be ascribed to inherited genes. Thus, parental effects [7] represent a form of plasticity that spans generations. Although parental effects initially received scant treatment by ecologists and evolutionary biologists, there has been a rapid accumulation of studies since a landmark publication 10 years ago [7]. This book has been instrumental for the general acceptance of parental (in particular maternal) effects as an important factor in evolution. However, many limitations to our understanding of evolution of parental effects still remain. For example, whereas theoretical work has focused on the consequences of parental effects for the short-term response of traits to selection [8–11], empirical studies tend mainly to be concerned with whether or not parental effects are adaptive [7,12–14]. Thus, it remains unclear under what circumstances parental effects are selected for, Corresponding author: Uller, T. ([email protected])
432
and the relative contribution of evolution of parental and offspring strategies [14,15]. This review attempts to take a step toward filling this gap by outlining a framework for the evolution of parental effects. It does so by integrating selection, parent–offspring conflict [16,17], developmental plasticity and proximate mechanisms of interactions between parents, offspring and their environments. Such an approach helps to clarify the circumstances that will select for the parental and offspring strategies that comprise parental effects, and to estimate their fitness consequences. Importantly, the approach also shows that understanding the evolution of parental effects requires knowledge both of the selection pressures involved and of the ways in which conflicts and constraints can be resolved. Evolution when everyone agrees: adaptive transgenerational phenotypic plasticity Parental effects represent a form of phenotypic plasticity spanning generations. Models of within-generation phenotypic plasticity generally predict that plasticity is favored when (i) there is environmental heterogeneity (spatial or temporal), (ii) there are cues that reliably predict future environmental conditions and, hence, selective regimes and (iii) the cost of plasticity is low [18,19]. Applied to the context of parental effects, this suggests that parentally induced plasticity is favored when (i) there is temporal or spatial fluctuation in environments across generations, (ii) offspring environmental conditions are predictable from the parental environmental conditions or the parental phenotype and (iii) the cost of obtaining information and responding appropriately is low for both parents and offspring. Furthermore, the level of parent– offspring conflict should be low, because otherwise the optimal level of plasticity will differ for parents and Glossary Development: phenotypic change during the lifetime of an individual. Developmental plasticity: the ability of a genotype to give rise to multiple phenotypes in different environments. Epigenetic inheritance: according to the broad definition, epigenetic inheritance means transgenerational transmission of form by other means than DNA. According to the narrow definition it implies heritable variation in DNA expression or activity, for example via persistent DNA methylation. Parent–offspring conflict: evolutionary conflict arising from different fitness optima in parents and offspring. Parental effect: an effect of the parental phenotype on offspring phenotype that cannot solely be ascribed to the parental or offspring genome, non-parental components of the environment or the interaction between these. Phenotypic plasticity: synonymous with developmental plasticity.
0169-5347/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2008.04.005 Available online 27 June 2008
Review
Trends in Ecology and Evolution
Vol.23 No.8
offspring and fitness might not be maximized for any of the individuals (see below). The above criteria seem to apply to many taxa, and parental effects that increase the fitness of both offspring and parents have indeed been documented in both plants and animals (Boxes 1,2). However, across-generation heterogeneity and selection for developmental plasticity are not sufficient for parental effects to be favored. Cues to what constitutes optimal phenotypic development can be obtained by the offspring either through the parents or through other aspects of offspring environment. Thus, whether the parental phenotype will affect offspring
development should also depend on the reliability of information provided by parents versus that obtained through the non-parental environment. Furthermore, parental effects will only be favored when there is a high likelihood that parental induction will allow an appropriate timing of the adaptive response in offspring (Box 1). For example, in plants and insects, lack of offspring sensory organs that can detect reliable cues early in development might explain why germination, diapause and dispersal strategies are commonly under maternal control, rather than being affected by environmental conditions experienced by the offspring [20,21].
Box 1. Predictability and the evolution of adaptive parental effects
Evolution of parental effects under parent–offspring conflict Parental effects represent the outcome of interactions between parents and offspring. Thus, their evolution will be affected by selection pressures acting on both generations, and parent–offspring conflicts will occur whenever there are different optimal strategies for parents and offspring [16,17]. This means that selection on parents might favor parental effects that reduce offspring fitness [14,22]. By contrast, offspring are under selection to maximize their own fitness, not that of the parents [16]. Thus, the evolutionary dynamics of parent–offspring conflict will depend on the relative costs and benefits to the interacting individuals, the strength of selection and the potential for control. The fitness consequences of parental effects for parents and offspring reflect the evolutionary outcome of this conflict (i.e. who wins, and at what cost?) [16,17,23].
Adaptive parental effects rely upon cues that predict offspring environment. Three scenarios are particularly likely to fulfill this criterion, because they imply that cues are easily obtained by the parents at the time of reproduction. First, there might be a strong correlation between maternal and offspring environments, as is often the case in seasonal environments (Figure Ia, dashed arrows). For example, in some insects, deteriorating maternal environmental conditions indicate onset of winter and select for the production of resting eggs or diapausing larvae that over-winter and resume development in spring [20]. Use of parental, rather than direct environmental, cues are particularly likely to evolve when this reduces the risk of a mismatch between offspring phenotype and selection on that phenotype, and when plasticity is expressed at very early life stages when offspring sensory organs are poorly developed. Furthermore, induction of major developmental changes (such as presence/absence of wings in aphids or helmets in Daphnia) requires less reorganization if they occur earlier rather than late in development. Second, the environment in which the offspring develop might be a direct consequence of the parental phenotype (Figure Ib, large solid arrows). This is the case, for example, when mothers oviposit on different hosts [20] or when maternal allocation decisions (e.g. clutch size and sex ratio) affect optimal offspring development [25,47]. Thus, offspring obtain both their environment and the optimal solution to the environment as parental effects [48]. In many animals with parental care, the maternal phenotype is the most important environmental component for a prolonged period of development, favoring co-adaptation of parental phenotype and offspring development [49]. Third, the genetic constitution of the offspring might be affected by maternal or paternal phenotypes, for example, via mate choice (Figure Ic, dashed arrows). If genotypes have different developmental optima, selection should favor parental transmission of nongenetic developmental resources that match the genetic constitution of the offspring [24,50]. For example, adaptive sex ratio adjustment frequently involves simultaneous adjustment of offspring genotype (both via mate choice and via sex manipulation in species with genetic sex determination) and developmental conditions via parental effects [24,50].
Figure I. Yellow beetles have high fitness on the light background and red beetles have high fitness on the dark background. Solid arrows represent parental transmission of developmental factors, and dashed arrows represent other direct and indirect relationships between environments and phenotypes.
Costs and benefits of parental effects The strength of selection on parental and offspring strategies depends on the costs and benefits involved. Many parental effects, such as investment in egg size or parental care, will reduce the future reproductive success of parents and therefore come at a cost. Other effects might entail minor costs for the parents, such as active or passive transmission of trace elements or hormones [13]. Parental effects can also entail costs for the offspring. For example, in birds, offspring have to compete for food by exhibiting costly behaviors, which reduces the amount of ingested food that can be directly converted to growth [23]. The costs of parental effects are important because they influence the level of parent–offspring conflict via the tradeoff between current and future reproductive success of parents, and because they offset the benefits of offspring strategies that manipulate parents. Thus, the evolutionary dynamics of parent–offspring interactions might differ depending on whether the parental effect involves costly investment or cheap manipulation. Transmission of costly resources When offspring cannot exercise any control over parental investment, parents might be able to impose their own optimal allocation to offspring within and between broods, despite negative effects on offspring [14]. For example, insect females do not always oviposit on hosts that are optimal for offspring development, but rather optimize maternal fitness via a tradeoff between search time and feeding time [22]. Similarly, females are expected to adjust 433
Review
Trends in Ecology and Evolution Vol.23 No.8
Box 2. Adaptive transgenerational plasticity in plants Environments are frequently heterogeneous at the population level, with corresponding variation in phenotypic optima. Dispersal or active habitat choice might cause nonrandom associations between the environment experienced by parents and offspring. For example, in many species of plants, seed dispersal is limited, suggesting that the maternal environment provides cues about the selective pressure experienced by the offspring [14,21]. In some cases, this might lead to genetic assimilation of developmental plasticity and local adaptation at a small spatial scale [4,51]. However, when pollen dispersal occurs at significantly longer distances than does dispersal of seeds, the paternally inherited genotype does not provide accurate cues to optimal phenotypic development [12,52,53]. Thus, phenotypic plasticity in response to the local environment should be favored. One possible response is that the offspring phenotype is modulated in response to the maternal environment, that is, a maternal effect on seedling development [12,21]. Recent studies of the herb Campanulastrum americanum (Figure I) have provided substantial evidence for a positive correlation between offspring and maternal, but not paternal, light conditions during growth [12]. Previous studies have shown that seeds from plants grown under good light conditions have an increased probability of germinating in the fall and, hence, of developing as an annual [54]. This might be adaptive if it increases population growth rate under favorable conditions. Indeed, a factorial field experiment [54] showed that seeds grown in the same environment as their mother had increased survival compared to seeds grown in the opposite environment. Furthermore, despite the fact that the maternal environment did not have a significant effect on adult fitness, the projected population growth was higher when environments were matched as a result of differential maternal effects on offspring survival and life history [54]. Parental effects on germination could be favored even in the face of occasional seed dispersal between habitats, as the negative effect will only persist in one generation. The documentation of adaptive transgenerational plasticity in this and other plants implies exciting opportunities for addressing the underlying mechanisms of transgenerational plasticity, and investigating to what extent such plasticity involves epigenetic programming that could span more than one generation [42,43].
egg size in relation to both current and future reproductive returns, even if decreasing egg size has negative effects on offspring fitness [14]. Such differential allocation in relation to current and future reproductive returns are common in animals such as birds and insects [24] and suggests that the adaptive value of parental effects must be addressed from a life-history perspective [14,25]. When offspring are able to influence the resource allocation of their parents, however, the evolutionarily stable parental effect might not reflect the parental optimum. Most theoretical work on parent–offspring conflict over negotiable resources (i.e. when both offspring and parents can respond to each others’ strategies) has been inspired by maternal or paternal food provisioning in birds and mammals [17,23]. However, several assumptions and predictions of the theory have also recently been put to the test using invertebrate systems with parental care [26,27], and the models might have broader applications. A common theoretical outcome of parent–offspring conflict over costly parental effects is that parents tend to provide more resources than is optimal for them [17]. The extent of the deviation from the parental optimum is set by the degree to which increased resource provisioning reduces (costly) offspring solicitation, and the degree to which solicitation affects parental provisioning [17,28], both of which can evolve. Empirical tests of whether the level 434
Figure I. Campanulastrum americanum. Picture by Laura Galloway.
of parental investment is at the parental or offspring optimum (or at neither) are hampered by the difficulty of measuring the parental fitness returns of variable resource provisioning to specific offspring or clutches, as both the returns from the current brood and returns from future broods need to be taken into account. Relatively novel model systems, such as burying beetles [26,27], might prove to be highly useful in this context. Finally, parental effects might also be sensitive to conflicts over resource investment between parents [29,30]. This complicates the evolutionary dynamics and can even lead to the evolution of parent-of-origin effects on gene expression (i.e. genomic imprinting [31]). For example, when the paternal and offspring optimum of maternal investment is higher than that of the mother, and offspring can influence the amount of resource investment, genes that increase maternal provisioning when inherited paternally will be positively selected [31,32]. Indeed, imprinting seems to be most common in mammals, perhaps because of the high potential for offspring manipulation of maternal resource provisioning. Transmission of manipulative signals When parental effects are cost free in terms of parental acquisition, production and transmission of developmental factors, the evolutionary outcome can be very different
Review from when they are costly to parents. Consider a parental strategy that can induce a plastic response in the offspring without any direct costs to the parents. One example might be hormone allocation to eggs or embryos [13,33], but the argument should apply to any parental effect that does not compromise future reproductive success of parents. Selection can initially favor the use of manipulative signals, such as parental hormone allocation, even if it has an immediate negative effect on the fitness of individual offspring. For example, parents might use manipulative signals to mediate sibling conflict under asymmetric offspring competition by reducing the competitive ability of specific offspring within the brood, thereby increasing parental fitness [23]. Alternatively, selection can favor parental reinforcement of sibling competitive asymmetry to increase the likelihood of survival of the most valuable offspring or to reduce the cost of parental investment [17,23]. However, selection on offspring will favor reduced offspring sensitivity or response to parental hormones, thereby generating an evolutionary arms race between parent and offspring. Evolutionary stability could be reached when offspring have lost the sensitivity to parental hormones altogether. Alternatively, the costs to parents might eventually outweigh the benefits, for example owing to negative pleiotropic effects of increased hormone levels on the physiology or behavior of parents [33]. Both conditions should result in a parental effect
Trends in Ecology and Evolution
Vol.23 No.8
closer to the offspring optimum. On the contrary, there might be constraints on the ability of the offspring to ignore the parental manipulation, which would allow for an outcome closer to the parental optimum. For hormones, this could arise as a result of a necessity for offspring to be sensitive to their own, endogenous hormones required for normal development, which could be indistinguishable from those produced and transmitted by the parents [33,34]. Selection on offspring will also favor the use of existing parental strategies according to the optimum of the offspring, not that of the parents. This might again lead to a parental effect close to the offspring optimum. However, selection on parents is then expected to reduce the transmission of information across generations, either via changes in the parental phenotype or via a temporal or spatial separation of generations (e.g. reduction in the duration of parental care), both of which reduce the availability of cues to offspring and lead to a reduction in the strength of the parental effects. Although explicit models of parental effects are rare, the arguments outlined above question the validity of some proposed adaptive scenarios, whereby parents are suggested to be able to impose their own optima using manipulative signals (as has been suggested for hormonemediated maternal effects [35]; see Ref. [33] for discussion). Furthermore, they might suggest that parental effects
Box 3. Resolving constraints: parental effects and sexual divergence Divergent selection on males and females implies that optimal development will differ for the sexes. Reproducing females thus face a dilemma when simultaneous production of the sexes constrains the opportunity for sex-specific maternal effects [15,47,55]. This dilemma can be resolved in several ways, including sex ratio bias, clutch size reduction or elimination of maternal effects, for example by reducing the period of parental care. However, these responses might compromise female fecundity or offspring survival, suggesting that alternative mechanisms to minimize the developmental overlap between the sexes will be under positive selection. Once they have evolved, such mechanisms might set the stage for considerable maternal effects on the ontogenetic divergence between the sexes [15,56,57]. Resolution of developmental constraints and the evolution of adaptive sex-specific parental effects are best exemplified by studies of the house finch Carpodacus mexicanus (Figure I) in North America [15,50,58–61]. Using a heuristic model of egg development and analyses of the daily uptake of maternally provided nutrients by the growing oocytes, it has been possible to show that oocytes destined to become male or female differ in their timing and duration of growth [58]. This leads to temporal clusters of oocytes that will develop into the same sex, which minimizes developmental overlap between the sexes [59–61] and ensures sex-specific accumulation of yolk components (e.g. androgens [50]). The regulation of these processes could occur through a single hormone, prolactin, via its effects on the onset of incubation behavior, plasma steroid levels, oocyte growth and follicular steroidogenesis [58,62]. Furthermore, the house finch studies suggest that the concentration of a small number of key developmental factors directly affect meiosis and therefore the sex of the resulting ovum (Z or W) via its effect on chromosome segregation during the first meiotic division [61,63]. This mechanism of sex biasing is in accordance with predictions that parental or environmental effects with sex-specific fitness consequences are likely to be co-opted as sex-determining factors [4,64]. The ability for flexible adjustment of sex-specific growth and development via maternal effects has facilitated the rapid colonization
of North America by the house finch in two steps: first, by increasing developmental plasticity in the initial phase of invasion and, second, via subsequent accommodation of plastic but precise parental effects to enable matching of offspring phenotype to local conditions [15,65,66]. Thus, parental effects might provide a key to colonization success of novel habitats in birds, as well as in other animals [4,67– 69].
Figure I. House finch Carpodacus mexicanus. Picture by Alex Badyaev.
435
Review based on the transmission of manipulative signals are inherently unstable unless there are limits to evolutionary responses in parental manipulation or development of offspring counterstrategies. Such limits could arise from several types of constraint, including physical, genetic and developmental constraints. Thus, understanding the evolutionary dynamics of parental effects requires researchers to integrate proximate and ultimate levels of explanation [15] (Box 3). Integrating proximate and ultimate levels of explanation in parental effects research An evolutionary response to current selection requires that novel input or modification of responses can be incorporated into existing developmental systems. Thus, development might constrain or bias evolution [4,36]. Furthermore, theoretical models of parental effects under situations of conflict have shown that the outcome is sensitive to assumptions regarding the ability of parents and offspring to develop counterresponses to novel strategies exhibited by the other party, both at ontogenetic and evolutionary timescales [17,37]. Empirical studies of proximate mechanisms might therefore provide insights into ultimate questions by showing how such constraints can be resolved, in a way that maintains the fecundity and survival of parents (Box 3). At the phenotypic level, parental effects are potentially sensitive to bias, limitations or constraints during several steps, from information acquisition by parents to the incorporation of developmental factors by offspring. For example, adaptive parental effects could be constrained owing to species-specific patterns of reproductive investment and parental care [15,34]. In mammals, the intimate physiological connection between mothers and offspring could make parental effects that are detrimental to offspring more likely by increasing the risk of negative carryover effects of strategies favored by strong parental selection. For example, maternal strategies that ensure her survival during stress, including reduced resource provisioning to offspring or increased levels of circulating steroids [38], could have immediate negative effects on offspring of her current brood. This is particularly likely if there are constraints on evolutionary responses of offspring that allow them to cope with variation in maternal phenotype, which shows the importance of mechanistic studies of development to understand evolutionary outcomes [33]. Close physical or physiological association between parents and offspring also increases the risk that parental exposure to novel environments or substances, such as endocrine disruptors or pollutants, will also have negative effects on offspring (e.g. [39]). Although parental effects generally are stronger early in life [7,40,41], they can have long-lasting consequences. This can be because of priming effects on phenotypic organization [4,38] or via lasting modification of gene expression via DNA methylation (i.e. epigenetic effects [1,42]). Intriguingly, studies on both plants and animals have shown that epigenetic effects not only remain stable within individuals, but also that they can be inherited across generations [1,39,42,43]. Thus, the presence of parental effects does not necessarily rely upon substantial 436
Trends in Ecology and Evolution Vol.23 No.8
contributions over and above genome transmission, suggesting that paternal effects might be more common than previously believed [44]. Furthermore, stable environmentally induced changes in gene expression provide a mechanism whereby short-term exposure to environmental conditions can have phenotypic consequences for more than two generations [1,42,43]. However, more transient parental effects can also have consequences for future generations, in particular if they affect mate preferences or sexual attractiveness of offspring (e.g. [45,46]). Documenting the proximate mechanisms of parental effects and their stability during transmission is therefore important to understand their adaptive value and to predict their effects on population dynamics. Concluding remarks The sheer number of studies of parental effects in virtually all major taxa is testimony to their importance. Furthermore, parental effects are fundamental to the ongoing integration of development, ecology and evolution [1–6], as they represent a particular form of plasticity arising from transmission of non-genetic developmental factors. To what extent parental effects involve permanent changes to gene expression [42,43] or only transient modifications of development remains poorly understood. However, if they are indeed stable, this could have far-reaching consequences for our understanding of the non-genetic transmission of form [1,42]. Progress in the field will require both the development of theoretical models that can be Box 4. Some challenges for the future Theoretical Extending models of within-generation plasticity to encompass transgenerational phenomena in systems with and without conflicts. Investigating the role of costs for both parents and offspring for the evolutionary dynamics of parental effects. Incorporating mechanistic assumptions to establish how constraints influence evolutionary outcomes. Addressing when epigenetic effects transmitted for more than one generation can be favored by selection. Developing models that link parental effects, population dynamics and evolution, for example, via parental effects on offspring attractiveness or reproductive output. Modeling the role of parental effects during species invasions and its consequences for the rate and direction of evolution during colonization of novel habitats. Empirical Putting models of the evolutionary outcome of parental effects with or without conflicts to the test in natural settings. Establishing comparative patterns of the strength and direction of parental effects for parental and offspring fitness in relation to key predictors, such as environmental heterogeneity and conflicting selection pressures. Using artificial selection experiments to tease apart the relative contribution of evolution of parental versus offspring strategies for the evolution of parental effects. Exploring mechanisms of parental effects and their consequences for evolutionary responses of parental and offspring strategies. Testing the stability of epigenetic inheritance across environmental contexts. Addressing how parental effects can affect population dynamics and facilitate range expansion or colonization of novel environments.
Review used to design and interpret empirical studies as well as explicit tests of adaptive scenarios in ecological settings (Box 4). Empirical tests of the adaptive value of parental effects are perhaps most promising in organisms such as plants where environmental heterogeneity often can be accurately described, conflicts are relatively weak or where there is limited scope for offspring manipulation of parental strategies [14,21]. Similarly, studies of egg or seed size are often freed from the potential for evolution of offspring counterstrategies, making those systems tractable for tests of adaptive scenarios. Parental effects arising from epigenetic modification via manipulative signals or cues in systems with conflicts are currently the least well understood. At the same time, they might provide some of the best opportunities for addressing the role of developmental plasticity and epigenetic inheritance for the origin and maintenance of complex adaptations more generally. Thus, the study of parental effects has the potential to become a prime example of how combining methods and concepts from different biological disciplines can lead to fundamental insights into evolutionary processes. Acknowledgements I am grateful to Alex Badyaev, Caroline Isaksson, Dustin Marshall, Geoff While and two anonymous reviewers for stimulating discussions and helpful suggestions on the manuscript. The Wenner-Gren Foundations and the Australian Research Council provided financial support.
References 1 Jablonka, E. and Lamb, M.J. (2005) Evolution in Four Dimensions. Genetic, Epigenetic, Behavioral and Symbolic Variation in the History of Life. MIT Press 2 Oyama, S. (2000) The Ontogeny of Information. Duke University Press 3 Laland, K.N. and Sterelny, K. (2006) Seven reasons (not) to neglect niche construction. Evolution Int. J. Org. Evolution 60, 1751–1762 4 West-Eberhard, M.J. (2003) Developmental Plasticity and Evolution. Oxford University Press 5 Sultan, S.E. (2007) Development in context: the timely emergence of eco-devo. Trends Ecol. Evol. 22, 575–582 6 Pigliucci, M. (2007) Do we need an extended evolutionary synthesis? Evolution Int. J. Org. Evolution 61, 2743–2749 7 Mousseau, T.A. and Fox, C.W. (1998) Maternal Effects as Adaptations. Oxford University Press 8 Kirkpatrick, M. and Lande, R. (1989) The evolution of maternal characters. Evolution Int. J. Org. Evolution 43, 485–503 9 Cheverud, J.M. and Moore, A.J. (1994) Quantitative genetics and the role of the environment provided by relatives. In Quantitative Genetic Studies of Behavioral Evolution (Boake, C.R.B., ed.), pp. 67–100, University of Chicago Press 10 Wolf, J.B. (2000) Gene interactions from maternal effects. Evolution Int. J. Org. Evolution 54, 1882–1898 11 Ra¨sa¨nen, K. and Kruuk, L.E.B. (2007) Maternal effects and evolution at ecological time scales. Funct. Ecol. 21, 408–421 12 Galloway, L.F. (2005) Maternal effects provide phenotypic adaptation to local environmental conditions. New Phytol. 166, 93–99 13 Groothuis, T.G.G. et al. (2005) Maternal hormones as a tool to adjust offspring phenotype in avian species. Neurosci. Biobehav. Rev. 29, 329– 352 14 Marshall, D. and Uller, T. (2007) When is a maternal effect ‘adaptive’? Oikos 116, 1957–1963 15 Badyaev, A.V. (2005) Maternal inheritance and rapid evolution of sexual size dimorphism: passive effects or active strategies? Am. Nat. 166, S17–S30 16 Godfray, H.J. (1995) Evolutionary theory of parent-offspring conflict. Nature 376, 133–138 17 Parker, G.A. et al. (2002) Intrafamilial conflict and parental investment: a synthesis. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 295–307
Trends in Ecology and Evolution
Vol.23 No.8
18 Schlichting, C.D. and Pigliucci, M. (1998) Phenotypic Evolution. A Reaction Norm Perspective, Sinauer Associates 19 Berrigan, D. and Scheiner, S.M. (2004) Modeling the evolution of phenotypic plasticity. In Phenotypic Plasticity. Functional and Conceptual Approaches (Dewitt, T.J. and Scheiner, S.M., eds), pp. 82–97, Oxford University Press 20 Tauber, M.J. et al. (1986) Seasonal Adaptations in Insects, Oxford University Press 21 Donohue, K.J. and Schmitt, J. (1998) Maternal environmental effects: adaptive plasticity? In Maternal Effects as Adaptations (Mousseau, T.A. and Fox, C.W., eds), pp. 137–158, Oxford University Press 22 Mayhew, P.J. (2001) Herbivore host choice and optimal bad motherhood. Trends Ecol. Evol. 16, 165–167 23 Mock, D.W. and Parker, G.A. (1997) The Evolution of Sibling Rivalry, Oxford University Press 24 Sheldon, B.C. (2000) Differential allocation: tests, mechanisms and implications. Trends Ecol. Evol. 15, 397–402 25 Plaistow, S.J. et al. (2007) How to put all your eggs in one basket: empirical patterns of offspring provisioning throughout a mother’s lifetime. Am. Nat. 170, 520–529 26 Smiseth, P.T. and Moore, A.J. (2007) Signalling of hunger by senior and junior larvae in asynchronous broods of a burying beetle. Anim. Behav. 74, 699–705 27 Smiseth, P.T. et al. (2007) Interaction between parental care and sibling competition: parents enhance offspring growth and exacerbate sibling competition. Evolution Int. J. Org. Evolution 61, 2331–2339 28 Royle, N.J. et al. (2004) Parental investment and family dynamics: interactions between theory and empirical tests. Popul. Ecol. 46, 231– 241 29 Wedell, N. et al. (2006) Sexual conflict and life histories. Anim. Behav. 71, 999–1011 30 Lessells, C.M. (2006) The evolutionary outcome of sexual conflict. Philos. Trans. R. Soc. Lond. B Biol. Sci. 361, 301–317 31 Haig, D. (2000) The kinship theory of genomic imprinting. Annu. Rev. Ecol. Syst. 31, 9–32 32 de Jong, T.J. and Scott, R.J. (2007) Parental conflict does not necessarily lead to the evolution of imprinting. Trends Plant Sci. 12, 439–443 33 Mu¨ller, W. et al. (2007) Manipulative signals in family conflict? On the function of maternal yolk hormones in birds. Am. Nat. 169, E84–E96 34 Uller, T. et al. (2007) Consequences of maternal yolk testosterone for offspring development and survival: experimental test in a lizard. Funct. Ecol. 21, 544–551 35 Schwabl, H. et al. (1997) A hormonal mechanism of parental favouritism. Nature 386, 231 36 Arthur, W. (2004) The effect of development on the direction of evolution: toward a twenty-first century consensus. Evol. Dev. 6, 282–288 37 Johnstone, R.A. (1996) Begging signals and parent-offspring conflict: do parents always win? Proc. Biol. Sci. 263, 1677–1681 38 Meaney, M.J. et al. (2007) Epigenetic mechanisms of perinatal programming of hypothalamic-pituitary-adrenal function and health. Trends Mol. Med. 13, 269–277 39 Crews, D. et al. (2007) Transgenerational epigenetic imprints on mate preference. Proc. Natl. Acad. Sci. U. S. A. 104, 5942–5946 40 Wilson, A.J. and Reale, D. (2006) Ontogeny of additive and maternal genetic effects: lessons from domestic mammals. Am. Nat. 167, E23–E38 41 Lindholm, A.K. et al. (2006) Where do all the maternal effects go? Variation in offspring body size through ontogeny in the live-bearing fish Poecilia parae. Biol. Lett 2, 586–589 42 Jablonka, E. and Lamb, M.J. (1995) Epigenetic Inheritance and Evolution. Oxford University Press 43 Molinier, J. et al. (2006) Transgeneration memory of stress in plants. Nature 442, 1046–1049 44 Bonduriansky, R. and Head, M. (2007) Maternal and paternal condition effects on offspring phenotype in Telostylinus angusticollis (Diptera: Neriidae). J. Evol. Biol. 20, 2379–2388 45 Qvarnstro¨m, A. and Price, T.D. (2001) Maternal effects, paternal effects, and sexual selection. Trends Ecol. Evol. 16, 95–100 46 Uller, T. et al. (2005) Female egg investment in relation to male sexual traits and the potential for transgenerational effects in sexual selection. Behav. Ecol. Sociobiol. 57, 584–590
437
Review 47 Uller, T. (2006) Sex-specific sibling interactions and offspring fitness in vertebrates: patterns and implications for maternal sex ratios. Biol. Rev. Camb. Philos. Soc. 81, 207–217 48 Lock, J.E. et al. (2007) Coadaptation of prenatal and postnatal maternal effects. Am. Nat. 170, 709–718 49 Wolf, J.B. and Brodie, E.D., III (1998) Coadaptation of parental and offspring characters. Evolution Int. J. Org. Evolution 52, 299–308 50 Badyaev, A.V. et al. (2006) Evolution of sex-biased maternal effects in birds: III. Adjustment of ovulation order can enable sex-specific allocation of hormones, carotenoids, and vitamins. J. Evol. Biol. 19, 1044–1057 51 Linhart, Y.B. and Grandt, M.C. (1996) Evolutionary significance of local genetic differentiation in plants. Annu. Rev. Ecol. Syst. 27, 237– 277 52 Leimar, O. et al. (2006) A new perspective on developmental plasticity and the principles of adaptive morph determination. Am. Nat. 167, 367–376 53 Leimar, O. Environmental and genetic cues in the evolution of phenotypic polymorphism. Evol. Ecol. (in press). 54 Galloway, L.F. and Etterson, J.R. (2007) Transgenerational plasticity is adaptive in the wild. Science 318, 1134–1136 55 Langkilde, T. and Shine, R. (2005) Different optimal offspring sizes for sons versus daughters may favor the evolution of temperaturedependent sex determination in viviparous lizards. Evolution Int. J. Org. Evolution 59, 2275–2280 56 Badyaev, A.V. (2002) Growing apart: an ontogenetic perspective on the evolution of sexual size dimorphism. Trends Ecol. Evol. 17, 369–378 57 Moore, C.L. (1995) Maternal contributions to mammalian reproductive development and the divergence of males and females. Adv. Stud. Behav. 24, 47–118 58 Badyaev, A.V. et al. (2005) Adaptive sex differences in growth of preovulation oocytes in a passerine bird. Proc. Biol. Sci. 272, 2165–2172
438
Trends in Ecology and Evolution Vol.23 No.8 59 Young, R.L. and Badyaev, A.V. (2004) Evolution of sex-biased maternal effects in birds: I. Sex-specific resource allocation among simultaneously growing oocytes. J. Evol. Biol. 17, 1355–1366 60 Badyaev, A.V. et al. (2006) Evolution of sex-biased maternal effects in birds: II. Contrasting sex-specific oocyte clustering in native and recently established populations. J. Evol. Biol. 19, 909–921 61 Badyaev, A.V. et al. (2008) Evolution of sex-biased maternal effects in birds: IV. Intra-ovarian growth dynamics can link sex-determination and sex-specific acquisition of resources. J. Evol. Biol. 21, 449–460 62 Sockman, K.W. et al. (2006) Orchestration of avian reproductive effort: an integration of the ultimate and proximate bases for flexibility in clutch size, incubation behaviour, and yolk androgen deposition. Biol. Rev. Camb. Philos. Soc. 81, 629–666 63 Rutkowska, J. and Badyaev, A.V. (2008) Meiotic drive and sex determination: molecular and cytological mechanisms of sex ratio adjustment in birds. Philos. Trans. R. Soc. Lond. B Biol. Sci. 363, 1675–1686 64 Uller, T. et al. (2007) Evolution of sex ratios and sex-determining systems. Trends Ecol. Evol. 22, 292–297 65 Badyaev, A.V. et al. (2002) Sex-biased hatching order and adaptive population divergence in a passerine bird. Science 295, 316–318 66 Badyaev, A.V. and Oh, K.P. (2008) Environmental induction and phenotypic retention of adaptive maternal effects. BMC Evol. Biol. 8, 3 67 Badyaev, A.V. (2005) Role of stress in evolution: from individual adaptability to evolutionary adaptation. In Variation: A Central Concept in Biology (Hallgrimsson, B. and Hall, B.K., eds), pp. 277– 302, Elsevier Academic Press 68 Carroll, S.P. and Fox, C.W. (2007) Dissecting the evolutionary impacts of plant invasions: bugs and beetles as native guides. Glob. Change Biol. 13, 1644–1657 69 Cote, J. et al. (2007) Mother-offspring competition promotes colonization success. Proc. Natl. Acad. Sci. U. S. A. 104, 9703–9708