Pollution breaks down the expression of genetic structure

Pollution breaks down the expression of genetic structure

Isofemale line technique

In this study we applied the isofemale line technique (ILT), commonly used on wild populations of Drosophila sp. (review by David et al., 2005). The relationship between narrow-sense heritability (h 2 ) and the coefficient of intraclass correlation (t, provided by ILT), is not so obvious but t does give a better heritability estimate than the half-sib design or the parent-offspring regression method. For example, the technique provides measures of about 95 ± 15% of the heritability for morphometric traits in natural population of D. melanogaster (David et al., 2005), thus yielding a close estimation of additive genetic variance. It is a useful tool to rapidly estimate genetic variability and genetic correlation of traits and to compare these genetic parameters between different traits and different environments (or even different populations). Furthermore, ILT can be easily applied to C. elegans because of its androdioecious breeding system (i.e. self-fertilization of hermaphrodites and facultative outcross with males). Indeed after mating, each isogenic line can be conserved as an inbred line. To facilitate reading, the term heritability was used throughout the text instead of the coefficient of intraclass correlation. 

Nematodes culturing 

Here, we used a stock population of C. eleganscomposed of a mixture of 16 wild isolates (Teotónio et al., 2012) in order to obtain a large genetic diversity. The population, composed of more than 30% of males, was maintained in laboratory conditions during 140 generations, where recombination-selection equilibrium was mostly achieved without significant loss of genetic diversity (Teotónio et al., 2012). However, we used different laboratory conditions for our needs. We used 500 individuals in a 9 cm diameter Petri dish with NGM-modified agar (use of HEPES 72 Article II – Effects on genetic structure buffer, article I). We produced six replicated experimental populations. Plates were seeded with 1 ml of a 20:1 mixture of Escherichia coli strain OP50 (OD600nm of 3) as the food source. Every 3 days we washed the nematodes (about 20,000 individuals) off the plates with an M9-modified solution (use of HEPES buffer) for each replicate and kept each replicate in 15 ml falcon tubes. The number of individuals in a tube was estimated with five sample drops of 5 µl, following Teotónio et al. (2012), and then the volume corresponding to 500 individuals, from all developmental stages, was placed in a fresh Petri dish. Nematodes were cultured throughout the experiment at 20°C and 80% RH.

Creation of lines 

After repeating this protocol about forty times (~ 40 generations despite the discrete overlapping of developmental stages), we selected 100 gravid hermaphrodites, and placed them on a new Petri dish, for each of the six replicates. Potential evolutionary changes caused by such changing conditions, should be similar in both the treatment and control environments. We allowed the hermaphrodites to lay eggs for a 2-hour period, and then transferred approximately 200 embryos on a new Petri dish (200 embryos per replicate). After 40 hours, when sexual differentiation is visually possible but individuals are not yet mature (i.e. L3 larval stage; Sulston et al., 1980), we randomly picked a single hermaphrodite and a single male, from different Petri dishes, and placed them separately (by couples) in 6 well tissue-plates. We created 14 couples. NGM in wells were seeded with 150 µl of a 5:1 mixture of OP50. After leaving plates in a laminar flow hood one hour to allow them to dry, plates were then top-exposed to UV doses for 90 s to stop bacterial growth (Bio-Link Crosslinker; λ=254 nm; intensity = 200 µwatt.cm-2 ). The main aim of this UV treatment was to avoid different bacterial growth in control and polluted plates. Reproduction by mating produces a high frequency of male in the progeny, whereas hermaphrodite selffertilization produces only 0.1% of males (Ward & Carrel, 1979 ; Prahlad et al., 2003 ; Teotónio et al., 2006). Therefore, 72 hours after the separation of the couple, we checked if the ratio of males and hermaphrodites in the progeny was approximately 1:1, to guaranty that mating between the hermaphrodite and the male was successful. We then picked one L3 stage hermaphrodite in each of the progeny of 14 couples to initiate 14 isogenic lines. This was done 72 hours after couple creation to avoid the risk of self-fertilization that could occur within the few hours after mating Article II – Effects on genetic structure 73 (Hart, 2006). Each line reproduced by self-fertilization for two generations in a 6 cm Petri dish, composed of the same environment as presented above. 

Contamination design and measurement of traits 

At the second generation of self-fertilization, we transferred 20 gravid hermaphrodites of each line in a new Petri dish. We considered this moment as t = 0. After 2 hours of egg-laying, eggs were individually placed in 12-well of a tissue culture plate. Twelve eggs per line and per environment were placed in three different NGM media: control, uranium (addition of 1.1 mM U) and high salt concentration (addition of 308 mM NaCl). Wells were seeded with 75 µl of a 5:1 mixture of OP50 (same UV treatment as for 6-well plates). In our experiment, the term “isogenic” does not correspond exactly to our lines, as there is heterozygosity in this population (Teotónio et al., 2012). However, individuals have reproduced twice by self-fertilization before the experiment; we thus expected a high reduction of heterozygosity (Hedrick, 1980 ; Reed & Frankham, 2003 ; Honnay et al., 2007). Individuals growing in salt showed a slower development than in the control and in uranium. Therefore, individuals were transferred twice into a new well at 96h in the control and uranium environment or 168h in the salt environment and then again 36h later. Hatched progeny were counted the day following each transfer to measure brood size, an index of fertility. We photographed individuals using a stereomicroscope (Olympus SZX12, 1.6 x 90 magnification) with a computer-connected camera (Nikon D5000). Using a micrometer scale measure and ImageJ software (Rasband, 2012), body length was measured at 72 and 144h. These two hours correspond approximately to the population’s onset and to the end of the egg-laying period in a control environment, respectively (pers. obs.). These two measures will be called early growth and late growth thereafter. Survival was estimated by counting the number of parental individuals still alive at 144h. 

Genetic parameters estimation 

We estimated the genetic parameters by using a Bayesian model approach in the MCMCglmm package for generalized linear mixed-effects model (Hadfield, 2010) in the R software (R Development Core Team, 2012). We fitted multivariate generalized linear mixed-effects models for different traits within the same environment (a quadrivariate model per environment) and for 74 Article II – Effects on genetic structure the same trait across different environments (a trivariate model per trait). So, models did not contain fixed effects but we included lines as random effects in order to estimate between- and within-line variance (respectively Vband Vw) for each trait. We modelled the survival with a binary error structure and a Gaussian one for the other traits. Expanding the binomial data into binary data allows us to fit a model in which we can estimate the correlation between survival and the others traits. Estimations were generated by a Bayesian approach to obtain the entire posterior distributions of (co)variance matrices of traits rather than only point estimates. To avoid any biased results caused by the fact that traits differ in their means, we rescaled the traits prior to analysis by dividing the mean by twice its standard deviation (Gelman, 2008). After several priors tested, we retained proper prior (nu = k – 1 + 0.002) with a very low variance parameter (V = diag(k)*Vp*0.05), where Vp is the phenotypic variance, kthe dimension of V (i.e. number of traits). There is one exception for survival for which we fixed the environmental (also called residual) variance to 1. For multivariate analyses with four different traits we allowed models to estimate different random and environmental variances, and covariances between the pairs of traits. In the models analysing the same trait across the three different environments, traits were measured on different individuals, and we thus fixed the environmental covariance to zero (see appendix C Table C1 for comparison of models with or without genetic covariance between traits allowed in prior). Despite the use of slightly informative priors used in the models, starting with low variance and any kind of covariance allowed us to say that the generated (co)variances were tangible. After verifying the convergence of parameters values (i.e. number of iterations, burn-in phase and thinning) and autocorrelation issues, we retained 2 500 000 iterations with a burn-in phase of 500 000, for a total of 1 000 samples for each analysis (Hadfield, 2010). We estimated heritability in the multivariate models for traits within environments using t = nVb / (nVb + (n – 1)Vw), where n was the number of lines (David et al., 2005). Following David et al. (2005) we also estimated genetic correlations without correcting for variance and covariance within-line, although n was less than 20. Using this correction would generate, posterior distribution for correlations containing values greater than 1. We used the posterior mode of the distribution of heritability, genetic and phenotypic correlations as quantitative genetic parameter estimates. For each trait we also estimated phenotypic (VP = Vw + Vb), environmental (VE = VP – VG) and genetic variances (VG = tVP) and genetic covariance (Cov) between traits was calculated as Cov1,2 = rG VG1VG2 , where rG corresponds to genetic correlation between traits 1 Article II – Effects on genetic structure 75 and 2 and VG1 and VG2 to the genetic variance for trait 1 and 2, respectively (see appendix C Figure C1). We considered that each estimate was significantly different from zero when its 95% Bayesian posterior intervals of highest density (HPDIs) did not include zero. We also tested the differences of an individual’s traits and genetic parameters in different environments. We considered differences to be significant between two environments when 95% HPDIs of subtraction between the posterior distributions of the trait in these two environments did not overlap 0.