Abstracts (first author)

Invited Speaker 

Is variation in chemical defence parasitism on a public good?

Author(s): Mappes J


An individual's chemical defences (toxins) contribute to a ‘common good’ by educating predators about the distastefulness of the population, hence deterring future attacks on the toxic individuals themselves and other individuals of similar appearance. Defensive chemicals are found in both simple and complex organisms including bacteria, fungi, animals and plants. Variation in defensive chemicals both within and among prey populations has been reported repeatedly in chemical ecology literature but it has received far too little attention from evolutionary ecologists. If some individuals lack defences (‘automimics’) they may exploit the common protection profiting from reduced attack rates but paying no individual cost of toxicity themselves. I will discuss whether variation in chemical defence and automimicry are examples of “environmental noise” or whether we need to seek evolutionary explanations for them. I will also discuss examples where understanding the dynamics of deception (e.g. Batesian mimicry) may have important consequences for successful conservation management.

Abstracts (coauthor)


Polymorphic warning signals are puzzling since positive frequency-dependent selection should promote monomorphic warning coloration. We studied predation pressure in the aposematic moth Parasemia plantaginis by using artificial prey resembling white and yellow male colour morphs in five separate populations. We tested if predation was influenced by: 1) natural frequencies of colour morphs; 2) number of interspecific Lepidopterans sharing similar coloration, and; 3) predator community composition. Predation on yellows was lower than whites’ regardless of their local frequency. The number of white interspecifics increased the attack risk of whites and decreased it on yellows, whereas yellow interpecifics lowered predation on both morphs. Interestingly, predation pressure was dependent on predator community composition: Yellows suffered less attacks when Paridae were abundant, whereas whites suffered less attacks when Prunellidae were abundant. Our results suggest spatial heterogeneity in prey and predator community composition can generate geographic mosaic selection facilitating the evolution of polymorphic warning signals.


Conspicuous colour patterns may function as interspecific signals in the context of predation by warning predators about prey unprofitability (aposematism). This strategy relies on the ability of predators to learn the association between colouration and unprofitability, and the subsequent avoidance of the unprofitable prey. Frequency-dependent selection is expected to favour uniformity and act against variability in aposematic signals. However, variation in aposematic species occurs in many taxa suggesting that signal variation may serve other purposes or be under variable selective pressures. Although the fundamental assumptions of aposematism have been well supported by mathematical models and experiments in controlled laboratory setups, their implications in the natural environment of both predators and prey have been until recently greatly overlooked. Predators are supposed to learn simpler patterns easier. Because variation in aposematic signals may imply morph-specific attack rates, aposematism as an anti-predator strategy could be overall less effective for individuals with complex colour patterns, especially when exposed to naïve predators or when invading novel environments, unless there were associated differences in detectability. We tested that hypothesis using wax models of the polymorphic, aposematic poison frog Dendrobates tinctorius placed in the wild, in a site where the actual frogs do not occur. We found that over time aposematic prey get less attacks than cryptic prey, but there were no differences in the attack rate between simple and complex morphs. However, complex morphs seemed to be more difficult to detect than simple ones. We suggest that wild predators are able to generalise aposematic colour patterns. Complex patterns may compensate being difficult to learn by being less detectable, which may contribute to the maintenance of the great intra-populational variation in colour patterns in this species.


Opportunistic pathogens generally face two vastly different environments - within the host and outside host. One mechanism allowing for adaptation to alternating environments is switching between phenotypes (phenotypic plasticity). The opportunistic fish pathogen Flavobacterium columnare can be found from natural waters and from fish farms and it exhibits two reversible colony morphologies; a non-virulent “rough” and a virulent “rhizoid” morphology. As compared to natural waters, fish farms can be considered as extreme environments in terms of available host resources, but also in terms of stress caused by chemical and antibiotic treatments. Fish farms could thus be expected to impose higher selection pressures for coping between the within and outside host environment, and to select for increased phenotypic plasticity. To test these ideas we measured growth parameters of rhizoid and rough colony morphotypes of F. columnare isolates both from natural waters and from disease outbreaks at fish farms in different resource concentrations and temperatures, and tested their virulence with a zebrafish challenge model. We found that the non-virulent “rough” morphotypes had a higher growth rate and lower virulence than the “rhizoid” morphotypes, but only if the isolate was originating from the fish farms. This suggests that phenotypic plasticity between two morphotypes of opportunistic pathogen and their characteristic traits is clearly selected for in fish farms rather than in the natural environment.


Many plants and animals advertise their unpalatability through warning signals in the form of colour and shape. A trade-off is inevitably faced to either allocate resources to signal efficiency or to other processes such as thermoregulation (i.e. melanin production). In the case where organisms undergo different phenotypes throughout their lifespan, such as Lepidopterans, it is unclear if allocating resources to warning signals in one phenotype can transfer fitness advantages to the next phenotype. Here we address this question by rearing full-sib tiger moth larvae (Parasemia plantaginis) in high and low temperature conditions and follow their warning signal development until adulthood. Subsequently, we tested for fitness differences as measured by survival and adult heating and metabolic rates. Our analyses showed that larvae reared at low temperatures had higher survival rate and decreased their signal size along its development, producing more melanised body segments. However, adults reared in higher temperatures had higher amount of melanin in their thorax. No clear differences were observed in the amount of melanin in the fore and hind wings between treatments. Adults reared at high temperature had a faster heating rate. On the other hand, adults reared at low temperature needed longer heating time, but had higher body temperature when flying was engaged. No differences were found between metabolic rates between adults of both treatments. Overall, our results suggest that allocating resources to maximise fitness during one life phase or phenotypic stage, does not necessarily translates into higher fitness in the following stages.


Recent studies on invertebrate immunity show that insects possess mechanisms to protect themselves from a secondary infection of a pathogen they have previously encountered. This suggests that invertebrates are able to differentiate between pathogens and mount more specific defence responses than previously thought. However, mechanisms for this specificity remain largely unknown. Here, we report a strain-specific immune priming response in the lepidopteran Parasemia plantaginis with enthomopathogenic Serratia marcescens bacteria. A major proportion of the moth larvae survived a septic injury with S. marcescens strain to which they had been previously orally exposed. Priming with another strain of S. marcescens and a similarly gram-negative, but harmless, Escherichia coli both failed to provide the protection later in life. We also show that production of caseinolytic proteases is likely a very important virulence factor in orally induced infection in this system. The findings have implications on the evolution of immune system in relatively short-lived insects, as well as on the selective processes entomopathogens may encounter in the arms race with their hosts.

Post-mating sexual selection and maintenance of colour polymorphism in an aposematic species

Author(s): Charge, R, Lindstedt C, Hämäläinen L, Övermark E, Wedell N, Mappes J


Predation is assumed to select for signal uniformity and conspicuousness in aposematic species, but colour polymorphism is surprisingly common. Sexual selection can maintain colour polymorphism, but only a few studies have examined its role in the maintenance of polymorphism in aposematic species. Previous studies in the aposematic Wood Tiger moth, Parasemia plantaginis, suggest that the conspicuous yellow male morph is better defended against predators, whereas females seem to prefer to mate with the white morph offering one potential explanation for the maintenance of colour polymorphism. Here we test the possible direct benefits that female could gain by mating with white males by examining whether white males provide females with bigger spermatophores and more sperm and whether white males recover faster after mating than yellow males. Spermatophores of either white or yellow males were collected from recently mated females. To measure male recovery rate, the amount of fertile sperm stored in males’ reproductive tract was assessed in mated males over several days post-copulation. We found that white males transferred bigger spermatophores than yellow males, but this difference was found only in older males. No difference in the pattern of sperm recovery was observed between male morphs. Contrary to our prediction, we found slightly higher number of fertile sperm stored in the reproductive tract of yellow males compared to white males, regardless of the recovery period. One explanation is that white males invest more in sperm competition by producing more non-fertile sperm that may suppress female receptivity. This would result in bigger spermatophores but with reduced number of fertile sperm. White males might also transfer other compounds such as defensive chemicals into the spermatophore that may benefit females. Altogether our results suggest that fertilization benefits do not solely explain how colour polymorphism is maintained in P. plantaginis.


The maintenance of trait polymorphisms in aposematic organisms is an evolutionary puzzle. The advantage of predator learning should drive warning signals of prey towards monomorphism, yet there are several examples of polymorphic aposematic species. Solving the riddle of polymorphism in such challenging cases would therefore represent an important contribution to our understanding of polymorphism maintenance in nature. Here, we examine the role of frequency dependent selection (FDS) on the maintenance of color polymorphism in an aposematic species. While models of frequency dependent selection are common, empirical evaluations are rare. The aposematic wood tiger moth (Parasemia plantaginis) presents two distinct male color morphs (yellow and white). Some populations have an admixture of both morphs whereas a few populations present only one of them. Research has previously shown that the more conspicuous yellow morph has a survival advantage under predation, which increases with morph frequency (i.e. positive FDS). This advantage and positive FDS should lead the yellow morph to fixation, yet this is not the case. We hypothesized that counteracting reproductive FDS (negative or positive) could maintain the polymorphism. First, we tested for reproductive FDS in the species using semi-natural enclosures set with three different frequencies of male color morphs (control, white-biased, and yellow-biased) to examine if the rare or common morph has a mating advantage. The experiment showed positive FDS in mating success for both morphs. Second, we formulated a theoretical model to test whether these results, and previous results on survival, can explain the maintenance of the observed polymorphism. The model showed that our patterns of positive frequency dependence can lead to polymorphism if considered in a spatial context, where variation in interacting ecological factors (e.g. predation), combined with a very small amount of gene flow can maintain the polymorphism.


Most research on the adaptive significance of warning signals has focused on the colouration and patterns of prey animals. However, behaviour, odour and body shape can also have signal functions and thereby reduce predators’ willingness to attack defended prey. For example, European vipers all have a distinctive triangular head shape and they are all venomous. Several non-venomous snakes are known to flatten their heads (head triangulation) when disturbed. Also many Lepidopteran larvae enhance their resemblance to tree vipers by concealing their heads and inflating their thorax or abdomen to express a false, sometimes triangular-shaped head. Even though anecdotal evidence of significance of snake head mimicry is dated back to the Henry Bates (1862), the role of body shape recognition is rarely experimentally investigated. Here we present data from field experiments and show that the triangular head shape can be recognized and avoided by predators. We also discuss the significance of this finding on population dynamics of snakes and its application their conservation. The smooth snake (Coronella austriaca) is non-venomous endangered species. By head triangulation it mimics vipers (Vipera sp.) which are not always protected by law. Because vipers are heavily killed by humans, it is possible that this asymmetric conservation program will be flawed because deceptive mimicry only works if the relative density and frequency of model species is higher than mimics. Based on the experimental evidence, we suggest that vipers should be protected at least in the locations where they co-exist with endangered mimic species.

Fluctuating temperature leads to evolution of thermal generalism and pre-adaptation to novel environments

Author(s): Ketola, T, Mikonranta L, Zhang J, Saarinen K, Friman V, Örmälä A, Mappes J, Laakso J


Climate change scenarios do not only expect elevated temperatures but also increased temperature fluctuations. Environmental fluctuations are suggested to select for low levels of plasticity in fitness that is also hypothesized to increase organisms’ ability to invade novel environments and affect virulence of pathogens. We tested these hypotheses and show that across a range of temperatures, opportunistic bacterial pathogen Serratia marcescens that evolved in fluctuating temperature (daily variation between 24 and 38 °C, mean 31 °C), outperforms strains that evolved in constant temperature (31° C) across all measured temperatures. Their better growth was also evident in novel environments with parasitic viruses and predatory protozoans. However, the strains from fluctuating environment were less virulent to Drosophila melanogaster host. Therefore, whilst supporting the hypothesis that evolution in fluctuating environments is paired with tolerance to several novel environments, our results show that adapting to fluctuating environments can also be costly in terms of reduced virulence. Together these results suggest that thermal fluctuations driven by the climate change could affect not only species thermal tolerance but also species’ invasiveness and virulence.


Flavobacterium columnare is a gram-negative bacterial pathogen that causes columnaris disease in freshwater aquaculture. Columnaris outbreaks occur at fish farms during summer months and may cause mortality up to 100 %. Virulence of environmental isolates of F. columnare has been found to be lower than those isolated during disease outbreaks at fish farms. In order to understand factors selecting for the higher virulence at fish farms, we studied if the bacterial dose, exposure time (transient or continuous), or nutrients have an effect on the virulence of F. columnare. Three F. columnare strains were used in two separate experiments: a non-virulent strain B398 isolated from the lake and two virulent strains from disease outbreaks (B185 and B67). In the first experiment zebrafish (Danio rerio) and rainbow trout (Oncorhynchus mykiss) fingerlings were individually infected with bath immersion (transient challenge) with 9 different doses of bacterial strains B185, B398, and a mixture of these strains. In the second experiment the bacteria (strains B185, B398 and B67) were added in three doses directly into aquaria (continuous challenge) where zebrafish and rainbow trout were maintained. Longevity of fish was monitored for five days in both experiments, and the infection verified by bacterial culture from gills.
We found bacterial dose to have a positive effect on mortality of both fish species. Increase in nutrients had a significantly positive effect on columnaris infection and fish mortality. The non-virulent strain was able to infect the fish when introduced in continuous exposure, but not in transient challenge. Our results suggest that the continuous exposure to bacteria at fish farms combined with a high nutrient level can promote virulence also in environmental non-virulent bacteria. In addition, the zebrafish can be used as a functional model host to study F. columnare virulence and infection dynamics in the laboratory.


Chairman: Octávio S. Paulo
Tel: 00 351 217500614 direct
Tel: 00 351 217500000 ext22359
Fax: 00 351 217500028
email: mail@eseb2013.com


XIV Congress of the European Society for Evolutionary Biology

Organization Team
Department of Animal Biology (DBA)
Faculty of Sciences of the University of Lisbon
P-1749-016 Lisbon


Computational Biology & Population Genomics Group