Evolution of Lymnaea stagnalis inbreeding depression under pesticide chronic exposure
Résumé
1. Introduction
Ecological and genetic factors may modulate the toxicity of chemicals to natural populations. In particular,
inbred populations may be more sensitive to pollutants than their outbred conspecific counterparts, due to
high homozygosity, loss of genetic diversity, or increased genetic load. On the other hand, strong inbreeding,
such as expected in highly selfing hermaphrodites, may have a positive effect on population response to
chronic (multigeneration) exposure to toxicants. First, selfing may facilitate the response to the selection
exerted by a toxicant, through a transmission of coadapted gene complexes more efficient than under
random mating. Furthermore, evolutionary theory predicts that spontaneous deleterious mutations are
preferentially purged under selfing, whereas they are maintained for longer time in outcrossing populations,
as masked in heterozygotes [1]. Thus, high and longlasting inbreeding may have positive effects on fitness,
and possibly on the ability to face chronic environmental stress. We tested these hypotheses in the
freshwater snail Lymnaea stagnalis, using a three-generation exposure to a pro-oxidant herbicide.
2. Materials and methods
A total of 16 lineages were derived from a highly diverse genetic pool stemming from five different original
populations [2]. Half of these lines were maintained in control conditions, and the other half exposed to 200
µg/L diquat during 5 weeks (five successive treatments, from the age of one week). After this period,
exposed lines were further reared as their control counterparts in standard conditions (charcoal-filtered
dechlorinated water, 14L/10D photoperiod, 20±1°C temperature). Snail density and food supply (organic
lettuce) were adjusted during the life cycle. Short before maturity, snails from the selfing lineages were
isolated in order to prevent outbreeding, whereas snails from outcrossing lineages were maintained in
groups and temporarily isolated for egg-laying. This protocol was implemented for three successive
generations (Fig. 1).
Fig.1. Experimental design. G0 = wild-caught adults. G1-G5: laboratory-born generations. Mating
system (SELF= selfing, OUT=outcrossing) crossed with exposure to controle conditions
(Unselected, in green), or to a pesticide (Selected, in red). Fitness (W) and relative performances
(RP-Unsel, RP-Sel).Chemical exposure was crossed with the mating system, i.e., in each treatment (control, diquat), four
lineages were forced to self-fertilize and four other lineages were allowed to outcross. After three
generations, inbreeding depression (ID), the fitness decrease of selfed progeny relative to the outcrossed
one, was estimated using fitness-related traits (life history traits) and the following formula:
where, Wselfi and Wouti are the performances at trait i under selfing and outcrossing, respectively. ID
variance was estimated by bootstrapping individual values within each condition [3], and compared between
diquat exposure and control conditions.
3. Results and discussion
Life history traits evolution was globally similar among control and diquat exposed lines. Inbreeding
depression was mostly due to parental fecundity, hatching rate and juvenile size in the next generation
(Fig.2). Inbreeding depression was slightly higher under diquat exposure (61%) than under benign conditions
(53%), suggesting that selfing did not improve the response to selection by diquat. However, the difference
was not significant (bootstrap test). This result, added to the fact that mortality during direct exposure to
diquat was significantly higher in selfing lines (not shown), suggests an aggravation of the toxic effects under
inbreeding. However, selfed lineages also showed an increase in trait variance, especially under diquat
exposure. Some inbred lines performed better than some outcrossed lines, suggesting that the response to
selection might in fact depend on lineage genetic composition.
Fig 2. Relative performances (WSELF/WOUT) at various life history traits measured on G4 adults and
G5 juveniles of L. stagnalis. The blue dashed line indicates equivalent trait value under selfing and
outcrossing.
4. Conclusions
This study showed that the evolutionary response of L. stagnalis to selection by diquat was not improved by
self-fertilization. Highly inbred lines were more sensitive than their outcrossed counterparts to diquat chronic
exposure. In terms of toxicity testing, the use of strains maintained for generations in the laboratory may be
questionable, due to high inbreeding and depleted genetic variability. To circumvent this drawback and
improve ecological risk assessment, we recommend the use of at least two distinct original populations.
5. References
[1] Charlesworth D, Willis JH. 2009. The genetics of inbreeding depression. Nature Reviews Genetics 10: 783-796.
[2] Bouétard A., Côte J., Besnard A.L., Collinet M., Coutellec M.A., 2014. Environmental versus anthropogenic effects on
population adaptive divergence in the freshwater snail Lymnaea stagnalis. PloS one 9(9), e106670.
[3] Coutellec M.A. & Caquet T. 2011. Heterosis and inbreeding depression in bottleneck populations : a test in the
hermaphroditic freshwater snail Lymnaea stagnalis. Journal of evolutionary biology 24:2248-2257.
Acknowledgement – This work was funded by the ANR project ESHAP (coord. Patrice David, CNRS).
Rearing and experimentations were performed at INRA U3E.
Domaines
Sciences du Vivant [q-bio]Origine | Fichiers produits par l'(les) auteur(s) |
---|