Parasites and behaviour in guppies

OK, Jacques just told us we should blog about our work, so I’ll try and explain what I’m actually doing- or trying to do at least. I recently joined the St-Pée lab as an assistant lecturer, after a postdoc with Andrew Hendry and Simon Reader in Montréal. This is where I discovered how fishes and particularly guppies are a wonderful model to test some exciting evolutionary concepts. Why guppies? Well, these fishes are an iconic (and pretty cute) model because they evolve rapidly in response to different ecological conditions. For instance, there are different lineages having evolved with or without predators, and with or without parasites in several Caribbean rivers, giving us access to totally independent instances of potential evolutionary divergence between contrasting environments. So they are a wonderful species model to test the parallel evolution of phenotypic responses to environmental variations in the wild (and you can go to the Caribbean islands to study them, which is not too bad either).

Fig 1. Nice Trinidadian guppies (Picture from Zandona et al. Funct. Ecol 25:5)

The effect of predators was well characterized in this system. For instance, guppies having evolved with predators are more tenacious and have a higher pace-of life (meaning they reproduce sooner and more but die younger) than guppies having evolved without predators. This makes sense. If you have a high probability of an early death, you should reproduce as quickly as possible. But a big question remained. Do parasites have the same kind of effect or not? It is not obvious. After all, even if parasites can affect their host fitness by weakening them, they are not supposed to kill them, or they might die as well. So, do parasites matter and are they able to affect the evolutionary trajectories of their hosts?  Andrew and coworkers set up a nice set of studies to test these hypotheses on morphological and life-history traits. Below is a picture of the kind of parasite they choose to focus on: these nasty critters are called Gyrodactylus, they fix themselves on the skin of fishes using their hooks, and feed from the mucus of their poor hosts. When the fish is not juicy anymore, they can jump on another fish through fish-to-fish contact. Lovely, don’t you think?

Figure 2: Gyrodactylus ectoparasite. Beware. Photo: Eve Zeyl

As for me, I was particularly interested in how these fascinating little parasites could affect the evolution of their host behavior. I knew from my PhD on birds how parasites could enhance host immune system across generations, but it seemed to me that they should also affect the evolution of their host behavior. After all, the best way of avoiding the pathogenic effects of parasites is to avoid encountering them. It may even spare you the costs of mounting a costly immune response. So I was lucky enough to convince the Fyssen foundation to fund my project, and this is how I ended testing how parasites could drive the evolution of personality traits (consistent behavioral traits across contexts) and syndromes (suites of interrelated behaviors) in guppies. We focused on boldness (reaction towards stressful situations), activity and sociability, and the link between those behaviors (syndromes). Indeed, when you evolve with parasites, maybe you should be shyer to avoid being exposed to parasites and/or evolve a lower tendency to join conspecifics since they may infect you. In a nutshell, parasites might shape the different sets of behaviors you should display towards new situations and conspecifics. To test those hypotheses, we compared different guppy populations having evolved with or without predators and with or without parasites in two independent replicate rivers in Trinidad. That’s how I ended up fishing guppies in the wonderful rivers of Trinidad in the Caribs (+30°C) and testing their behavior  in the lab in Montréal (-30°C) (yes, I let you imagine how hard it was to get off the plane).

Fig 3. Sampling guppies in the Marianne river (my favorite river). Picture: Felipe Perez.

To be honest, I expected behavior to vary greatly and not necessarily in a predictable manner. But to my surprise, fish displayed consistent and quite parallel differences in their behaviors between different rivers, even after two generations raised in the lab. Look at those nice graphs (thank you Amandine).

Fig 4: Shyness level (measured as the latency to leave a shelter in seconds) in guppies having evolved in contrasted evolutionary regimes in two independent rivers (Marianne and Aripo).

On the left is the shyness level (latency to leave a shelter) of wild-caught guppies (generation F0) from different evolutionary regimes (having evolved with parasites and predators, parasites only, or no parasite and no predator- you can’t have sites with predators only because of the topography of the rivers) within the two independent rivers (Marianne in white and Aripo in yellow). On the right we tested again the behavior of the F2 lab-raised generation (the babies of their babies) to see if behavioral differences stemmed from environmental or genetic-based factors. Well, you can see some variations between the F0 and the F2 generation, which might be due to environmental effects or habituation to captivity across generations. But the differences in behavior between evolutionary regimes remain pretty parallel in the Marianne and the Aripo rivers, even at the F2 generation.

This suggests that whatever the initial genetic background of the populations and the environmental variations between the two independent rivers (those rivers were quite different in a lot of aspects), predators seem to consistently drive the evolution of a genetic-based bolder behavior (this had been already shown before). And more importantly, parasites seem to consistently drive the evolution of a shyer behavior in a parallel way in the two rivers. In other words, parasites seem to repeatably select for a more cautious behavior, maybe because it can decrease the fish chances to encounter a parasite, although this adaptive interpretation remain to be tested. We also tested other behaviors and the links between them, and saw that activity is also affected, as well as shoaling, but in a more complex way. In addition, predators and parasites seemed to decouple the correlations between traits, that is to say, the structure of syndromes. This suggests that predators and parasites not only affect the mean value of behavioral traits, but also impact the whole structure of behavioral packages that individuals can display.

This is pretty exciting, and supports the hypothesis that selection act on suites of behavioral traits as a whole rather than on isolated traits, and that parasites might be a more potent evolutionary force than previously thought. And more importantly, this helps us realizing how parasites can have a central role in shaping the diversity of life. So, next time you see a parasite, think about it. It might hurt you a little bit, but it might also affect the evolution of your gran-gran children. Pretty scary…

Author: Lisa Jacquin

More info: https://sites.google.com/site/jacquinlisa/

References

Jacquin L, Reader S, Matelunna J, Patalas I, Perez-Jvostov F, Hendry A.  Variations in behavioral syndromes across parasitism and predation regimes in wild Trinidadian guppies. In prep for Behavioral Ecology

Gotanda, K.M., Delaire, L.C., Raeymaekers, J.A.M., Pérez-Jvostov, F., Dargent, F., Bentzen, P., Scott, M.E., Fussmann, G.F., Hendry, A.P., 2012. Adding parasites to the guppy-predation story: insights from field surveys. Oecologia 172, 155–166.

Barber, I., Dingemanse, N.J., 2010. Parasitism and the evolutionary ecology of animal personality. Philosophical Transactions of the Royal Society B: Biological Sciences 365, 4077–4088.

When EU cares about fish.

It might not seem obvious from the start, but fish behaviour and population dynamics are a concern for policy makers. Why is it so? Well, fish are both a resource and a marker of ecosystem health. For instance, Atlantic salmon has been exploited for hundreds of years, at different scales and in contrasted ecosystems: freshwater, marine water. They migrate between ecosystems and between nations, even continents. European eel is no different: this species' habitat is approximately a quarter of our planet. Brown trout is another example, being able to exploit sea resources such as plankton, or being capable of surviving in high altitude lakes. 

Yet the potential of these species to adapt to a changing world, as well as to resist our exploitation is in question. What happens if their migratory behaviour changes? What happens if their reproductive behaviour is modified? How much will they be impacted, and in turn, how far will it affect our society?

This is no mystery that much of these resources are now in jeopardy: Atlantic salmon and European eel have strongly declined the past 50 years. Over-exploitation, habitat degradation, and possibly climate change are at wheel here. 

So, there are good reasons for which policy makers should care about fish behaviour and populations. And they do in fact. Between 2009 and 2014, our lab has been part of a large project, the  AARC project, bringing together public information, ecosystemic restoration, academic teaching, and efficient science. Because we are reaching the end of this program, we would like to hint at some of the actions developed by our lab thanks to this EU funding.

Sexy Eels

First, did you know European eels have only a weak genetic control over sexual determination? In this species, it appears that sex is largely environmentally determined, over the course of the life cycle. This question was explored by Benjamin Geffroy during his PhD, and he investigated the dynamics of gonads development in rearing conditions, as well as the aromatase gene expression, an hormone known to control for sexual differentiation in fish [1]. Benjamin obtained some surprising results for some of the eels: while many could be tagged as female or male at a reasonable age and size, some appeared as intersexual with a gonad containing cysts with spermatozoa and pre-vitellogenic oogonia. In addition, the percentage of males was very high, whatever the density tested. While the cause of such dominance of the male phenotype still escapes us, it directly impacts our management practices regarding European eel conservation and stocking decided by the EU.

Example of gonad for an intermediate gonadal phenotype, with both spermatocytes (Sc) and normal oocytes (No) in the same gonad [1].

Salmon on common grounds 

For some species, we are much more advanced at evaluating stocks, and even selective impacts of over-exploitation.  Yet, the diversity of interest for a given resource makes it hard - if not impossible - to find a compromise that may satisfy everyone. Salmon is an iconic incarnation of this problem. Amateur anglers like to catch their yearly salmon during the reproductive migration back in rivers. But professional anglers also want to benefit from this resource when salmon enter the estuaries after their marine migration. On top of that, fishing in high seas is targeting salmon feeding areas around Greenland or Feroes Is. Did we mention naturalists that just love watching salmons jumping upstream waterfalls? And biodiversity managers, that have to make sure that the species remains healthy and viable.

Mélanie Brun tried to tackle this problem, and summoned mathematics as fresh troops to do so [2]. Simply put, she proposed that each stakeholder defined a utility function for the resource: what level or characteristic of the resource is the most important or the least important for them. Of course, the biodiversity managers will not draw the same function as the professional angler, for instance. But Mélanie found a way to objectively locate some common grounds where all stakeholders would see their claim at least partly acknowledged. This decision making tool simply helps at targeting what is of shared importance among the resource users, while ensuring that the resource does not dry out.

An example of utility function for a stakeholder, combining both population size and fish origin data. The considered stakeholder in this case is equally interest in maintaining the population size and in supporting natural regeneration of the population [2].

Can brown trout see into the future?

We often talk about brown trout on this blog, but let us not weaken. Because of climate change, hydraulic dynamics of lotic freshwater are expected to increase in stochasticity: extreme events are predicted as more frequent in the near future. If you are a female trout, you might want your offspring to avoid a cruel fate, such as: being scoured with the redd during a high flow event, or getting dried up because the water level went so low. Additionally, you want to provide them with the best habitat for their development. Elie Chantriaux and Zoé Gauthey investigated these questions, by monitoring female habitat choice, female investment in terms of egg size, and egg survival, conditional on water level during three years on two different rivers.

  

Female brown trout have access to a large range of habitat conditions to bury their eggs. But do they make an optimal choice regarding extreme flow conditions?

 

Their results show that between 60 and 70% of eggs get scoured before hatching, mainly due to high flow events, sometimes due to redd scouring by competing females themselves. But interestingly, the redd depth or its composition in terms of substratum particle size had no effect on the probability to get impacted by scouring. It appears that brown trout early life is mainly dominated by random events, and that females do not have any way - or possibility - to predict these events. So no, female trout cannot see into the future, and could be increasingly impacted by frequent extreme flow events. They still have a last resort: they can fraction their clutch in several batches, and dig several redds. Costly, but probably increasingly efficient if climate experts are right with their predictions. 

So yes, EU cares about fish life, behaviour, populations. In fact, without EU funding, much of our research would not be possible, and this research is often connected to local and regional interest. More actions, educational programs or information can be found directly on the AARC project website:

Références:

[1] Geffroy B., Guiguen Y., Fostier A., Bardonnet A. 2013. New insights regarding gonad development in European eel: evidence for a direct ovarian differentiation. Fish Physiology and Biochemistry, 39:1129-1140.

[2] Brun M. 2011. Aide à la décision pour la conservation des populations de saumon Atlantique (Salmo salar L.). Doctorat de Biologie, Université de Pau et des Pays de l'Adour, 205p.

[3] Chantriaux E. 2014. Phenotype-habitat matching: théorie explicative de la variabilité de taille d'oeufs interponte chez Salmo trutta? Rapport de stage de L3, INRA-UPPA, Saint-Pée sur Nivelle, 30p.