Summary PhD thesis Andres Hagmayer: Benefits and drawbacks of the placenta in live-bearing fish

Understanding how ecology influences the evolution of complex life-history traits is of fundamental interest to ecologists and evolutionary biologists alike. The placenta is an excellent example of a complex life-history trait that has evolved repeatedly – and continues to evolve to date – throughout the animal kingdom. In the live-bearing fish family Poeciliidae alone, the placenta evolved at least nine times independently. This repeated evolution and ongoing elaboration appears to point towards an adaptive advantage to specific environmental conditions; however, this potential benefit is currently insufficiently understood. In this thesis, I attempt to shed light on the causes and consequences of placental evolution by drawing on insights from the live-bearing fish family Poeciliidae.

In Chapter 2, we studied maternal causes and consequences of embryo provisioning during gestation in the placental live-bearing fish species Poeciliopsis retropinna from Costa Rica. Specifically, we examined how maternal traits (i.e. body fat, lean mass, and length) relate to pre- (i.e. allocation to the egg prior to fertilization) and post-fertilization (i.e. allocation to the embryo during pregnancy) maternal provisioning and how this ultimately affects offspring size and body composition at birth. We showed that maternal traits strongly correlate with embryo size and body composition throughout pregnancy, though different maternal traits are differently related to pre- and post-fertilization maternal provisioning. From our results, we conclude that (i) embryo size and offspring size at birth are plastic phenotypic traits that are predicted by the maternal phenotype, and (ii) maternal provisioning is not constant throughout pregnancy, but depends on the developmental stage of the embryos.
In Chapter 3, we studied 28 natural populations of the placental live-bearing fish species P. retropinna in Costa Rica to test a key prediction of the locomotor cost hypothesis, which is that the placenta evolves in high ‘performance-demanding’ environments by gradually shifting maternal investment from pre- to post-fertilization, thereby reducing the reproductive burden experienced by females during pregnancy and improving their locomotor performance. We confirmed two key predictions of this hypothesis, which are that (i) placentas evolve in natural populations in response to high predation conditions and (ii) that this significantly reduces a female’s reproductive burden during pregnancy. This chapter provides the first evidence for an adaptive (ecological) explanation for why the placenta evolves in natural populations. Moreover, it reveals an unexpected fundamental feature of placental animals, which is that an increase in the degree of placentation can lead to a lower reproductive burden without any apparent reproductive cost: i.e. without sacrificing either fecundity or offspring size and quality at birth.

In Chapter 4, we empirically tested the idea that the evolution of placentation and superfetation facilitates a life in ‘performance-demanding’ (e.g. high-flow velocity) environments by reducing the reproductive burden of females during pregnancy, and hence, improving locomotor performance. If true, we predicted placental species and/or species with superfetation to inhabit higher ‘performance-demanding’ (micro)habitats compared to closely related non-placental species or species without superfetation. For this, we used underwater visual census to study diurnal and ontogenetic microhabitat selection in Costa Rican rivers by five sympatric live-bearing fish species (family Poeciliidae) that differ in the absence/presence of placentation and superfetation. Consistent with our predictions, we observed significant interspecific differences in daytime microhabitat use: species with placentation and superfetation were found in deeper and faster-flowing parts of the river, species that lacked both adaptations were confined to shallow slow-flowing areas, and species with one adaptation (i.e. only superfetation) inhabited intermediate areas. This interspecific daytime microhabitat use was strongest in reproductive adults, intermediate in immatures, and absent in juveniles, suggesting that ontogeny influences species-specific microhabitat use. Finally, at night, all fishes, regardless of the species or age-class, congregated in shallow slow-flowing waters to rest (sleep) on the river bottom. Our results suggest that placentation and superfetation may be hitherto unrecognized reproductive features that help to understand differences in ontogenetic and diurnal microhabitat preferences between sympatric live-bearing fish species living in environments characterized by large flow variation.

In Chapter 5, we examined the consequences of a trematode infestation (black spot disease, BSD) in shaping life-history and behavior (boldness trials in the field) in the placental live- bearing fish species P. retropinna from Costa Rica. We proposed that the intimate link the placenta forms between the mother and fetus poses a risk, because parasite infestation may have unfavorable consequences for fetal development. For example, maternal parasite infestation can affect fetal growth in two non-mutually exclusive ways: (i) directly, through infestation of the developing fetus by parasites that can cross the placental barrier; and (ii) indirectly, through the modification of maternal physiology or metabolism to such an extent that it interferes with fetal development. We showed profound variation in parasite infestation among females within populations associated with maternal size and body condition (expressed as maternal fat reserves). Moreover, we found that heavily parasitized females produced smaller and worse-conditioned offspring at birth, possibly because a costly immune response during pregnancy may limit the energy available to (i) nourish developing embryos or (ii) form a well-functioning placenta (indirectly affecting embryo nourishment). However, the infestation rate did not affect an individual’s boldness behavior in the field. Our findings show that in placental live-bearing fish parasite infestation leads to reduced embryo provisioning during pregnancy, resulting in a smaller offspring size and quality at birth potentially with negative implications for offspring fitness.

In Chapter 6, we conducted a 7-week laboratory experiment to examine to what extent maternal food limitation during pregnancy affects offspring size and quality (i.e. body fat) at birth, as well as growth and locomotor performance of offspring after birth in the placental live-bearing fish species Phalloptychus januarius. We showed that maternal food restriction resulted in a decrease in maternal wet mass throughout the experiment, leading to a reduced resource allocation to F1-offspring size and body fat, which can be compensated only in the long-term (i.e. when adult). Although this did not impact the fast-start escape performance of F1-offspring immediately after birth, it negatively affected the postnatal development of locomotor performance during feeding, and thus, body condition and presumably fitness after birth. The negative effects of maternal food restriction on offspring size still occurred in the F2-generation at birth, but disappeared one week after birth. This chapter suggests that placentotrophy in poeciliids is likely a maladaptive strategy in fluctuating resource environments, because sudden reductions in maternal food availability during pregnancy result in smaller offspring with a lower locomotor performance during early life.

Finally, in Chapter 7, I put the findings of this thesis into a wider scientific context. First, I restate the different hypotheses for the evolution of the placenta together with the currently available evidence for each. Second, because superfetation is commonly found in Poeciliidae and thought to co-evolve with the evolution of the placenta, I developed an R-package to simulate and graphically illustrate the potential advantages of having superfetation, as well as both, superfetation and placentation. The simulation shows that the co-evolution of placentation and superfetation might be favored by the additive effects of both traits on the reproductive burden of females during pregnancy. Third, I discuss the potential disadvantages of having a placenta in response to various adverse environmental conditions. I conclude that although having a placenta offers a selective advantage to females in high ‘performance-demanding’ environments by improving swimming performance, the intimate link the placenta forms between mother and fetus also poses a risk: maternal exposure to adverse environmental conditions (e.g. malnutrition, parasite infestation) is likely to have unfavorable consequences for fetal development. Finally, I present an outlook on future research by addressing possible steps and challenges to deepen and expand the knowledge on the causes and consequences of placental evolution.