Computational protocol: Is the Success of Plant Invasions the Result of Rapid Adaptive Evolution in Seed Traits? Evidence from a Latitudinal Rainfall Gradient

Similar protocols

Protocol publication

[…] We conducted a reciprocal transplant experiment with plants originated of five populations from the two studied localities (La Serena and Coyhaique) to investigate possible evolutionary change in ESCP and its adaptive consequences. At each of these five populations in each locality, we collected seeds from 40 maternal plants, and germinated in a room with a photon flux density (PFD) of 470 (±135) μmol m-2s-1 at 20°C ± 2 in Petri dishes. Five days after the appearance of the first true leaf, seedlings were transferred to growth chamber at 10°C with a photon flux density (PFD) of 250 (±12) μmol m-2s-1 and 16/8 h light/dark photoperiod. This temperature was chosen because this is an intermediate of annual mean temperature of both localities (). Ten 1-month-old seedlings from each population, each one from a different mother plant, were transplanted directly in a garden plot and randomly assigned following a grid design in both, La Serena and Coyhaique localities (n = 100 individuals from each locality). Each group of ten individuals (population) was separated by 1 m between each point of transplant. The plants were watered once every day only for the first week in order to reduce the death of plants due to the transplants. Once per week, over 10 weeks, survival was recorded for each plant maintaining the recording separated by locality of origin. All surviving plants were harvested after 10 weeks and total plant biomass was obtained after the whole plants were oven-dried at 70°C for 72 h. Considering that neither survival percentage [one-way ANOVA = F1,8 = 24,56; p = 0.65 and F1,8 = 16,33; p = 0.75 for La Serena and Coyhaique, respectively] nor biomass [one-way ANOVA = F1,8 = 41,56; p = 0.71 and F1,8 = 79,13; p = 0.21 for La Serena and Coyhaique, respectively] differed between populations from the same locality of origin, we merged the results by locality. Survival percentage and biomass at 10 weeks of T. officinale individuals from both localities were compared by one-way ANOVA (STATISTICA 7.0). Comparisons were conducted independently for each locality where individuals were transplanted. [...] Total DNA was extracted from dry foliar tissue of plants previously assessed in each locality according to cetyl-trimethyl-ammonium bromide (CTAB) method (), following the protocol described in . Final DNA concentrations were quantified in a NanoDrop® spectrophotometer (ThermoFisher, United States), and their integrities verified by electrophoresis in a 1% agarose gel. The genetic diversity and population structure were estimated using Amplified Fragment Length Polymorphism (AFLP). The AFLP protocol was performed following the description of . Genomic DNA (∼250 ng) was digested in a total volume of 25 μL using EcoRI (NEB) and MseI (NEB) restriction enzymes (1U each) for 2 h at 37°C, followed by 15 min at 70°C. The resulting DNA fragments were then ligated with the corresponding EcoRI (5 pmol) and MseI (50 pmol) adapters using T4 DNA ligase (1U) and 1 × ligation buffer (Roche) for 3 h at 37°C. Preselective amplification reactions were performed using 1 μL of digested-ligated DNA in a total volume of 20 μl. The mixture contained 1 × PCR buffer, 1.5 mM MgCl2, 0.25 mM of each dNTP, 0.5 μM Eco + A primer, 0.5 μM MseI + C primer and 1U of AmpliTaq Gold® DNA polymerase. PCR amplification was carried out with a profile of 20 cycles of denaturation 30 s at 94°C, annealing 1 min at 56°C, and extension 1 min at 72°C. After PCR amplification, amplification products were diluted 1:10 with distilled water and stored at -10°C. Selective amplifications were performed using a 1 μL of diluted preselective amplification as a template in reactions containing 1x PCR buffer, 1.5 mM MgCl2, 0.25 mM of each dNTP, 0.5 μM Eco + ANN (fluorescent) primer, 0.5 μM Mse + CNNN (or Mse + CNNN) primer, and 1U of Platinum Taq DNA polymerase (Invitrogen Brazil). Selective amplifications started with a touchdown step of 2 min at 94°C, then 10 cycles of 30 s at 94°C, 30 s at 65°C (1°C decrease each cycle) and 1 min at 72°C. This was followed by 30 cycles of 30 s at 94°C, 30 s at 56°C and 1 min at 72°C, ending with a final elongation of 30 min at 72°C.Following , we used three selective primers combinations: E + AGC/Mse + CAAT, Eco + AAT/Mse + CAAC and EcoAGC/Mse + GATG that produced 85, 61 and 50 fragments, respectively. All profiles were run in an automatic sequencer (Applied Biosystems 3120, 16 capillaries) at the Molecular Biology Laboratory of the Pontificia Universidad Católica de Chile. For each pair of primers combination, fragments with sizes between 80 and 450 bp, and intensities > 200, were first selected using GeneMarker 2.4.0 (Soft Genetics). These fragments were coded in a presence/absence (1/0) matrix. The chosen fragments were, besides polymorphic, present in no less than 5% (nor more than 95%) of the individuals, and error rates of replication were lower than 8%. AMaRe (AFLP Matrix Reduction) method, outlined in , was used to calculate reliability and error rate of the matrix. The average unbiased expected heterozigosity (HE), percentage of polymorphic loci (% PL; a loci was considered polymorphic if at least in 5% of the sample the less frequent state was observed) and Shannon’s index of diversity (I) were estimated using GENALEX ().Population genetic structure was investigated though Discriminant Analysis of Principal Components (DAPC) using the adegenet R-package (; ), and hierarchical analysis of molecular variation (AMOVA). First, DAPC analysis was applied to the AFLP data to visualize the potential clustering of individuals, which together with the pair-wise FST values between localities, was used as the criteria to define a population structure to be tested in the analysis of molecular variance (AMOVA). Both, FST and AMOVA analysis were performed in Arlequin 3.5 (), all other analysis were carried out on the R language and environment for statistical computing v3.1.3 ().We explored the relationships between genetic differentiation among localities (pairwise FST) and geographical distance (Euclidean), differences in mean annual rainfall and differences in seed coat to endosperm proportion. The level of correlation between pairs of distance matrices was estimated using Mantel test’s (). The significance for each correlation coefficient between matrix pairs was obtained from the distribution of 9999 randomizations as supported by the “ade4” R-package (). [...] Asexual reproduction (agamospermy) is common in dandelions (). Thus, in its native range of distribution natural populations can be formed either by co-occurring sexual diploids and apomictic triploids, or by exclusively by apomictic triploids (; ; ). To assess whether the studied populations are formed by few successful clonal lineages or by several multi-locus lineages, we estimated the extent of clonal reproduction in our AFLP dataset using the software GenoDive 2.0b17 (). We choose GenoDive because it has been extensively employed to estimate clonal diversity on marine organisms (e.g., ) and plants (e.g., ; ) using AFLP markers. Briefly, GenoDive uses a “threshold” of genetic distance between pairs of individuals to distinguish clones from non-clones. Below that threshold of genetic dissimilarity, samples are considered to represent a single clone. In other words, this threshold indicates the maximum distance that is allowed between a pair of individuals to still be considered clonemates (individuals from the same clonal lineage). GenoDive assumes that random mating within populations, i.e., its tests whether the allelic frequencies deviate from these that are expected under random mating. To do this, GenoDive tests the null hypothesis that the observed clonal diversity is due to sexual reproduction by randomizing alleles over individuals and comparing the observed clonal diversity with that of the randomized dataset uses (). It is important to note that other processes than asexual reproduction such as self-fertilization can also lead to identical genotypes, and therefore, under random mating clones can also be produced. Since scoring errors and mutations may cause individuals from the same clonemate to have a pairwise distance larger than zero, choosing an appropriate threshold is crucial to perform accurate clonal assignments (; ). While too low thresholds overestimates the estimates of clonal diversity, choosing this value too high results in its underestimation.Our dataset consisted of a total of 89 individuals from 5 populations genotyped with AFLP’s (Supplementary Table ). To reduce the scoring error rates in clone assignment we previously eliminated from our dataset all loci containing missing genotypes (45 loci). In addition, scoring errors were checked using the software AMARE (), This analysis revealed that scoring error rate was reduced from 5% in the total data set (196 loci) to 2% in the subset used to estimate the levels of clonality (151 loci). After this procedure we selected this subset of 151 AFLP bands to evaluate clonal diversity. Specifically we made a clone assignment analysis to determine the number of multilocus lineages (MLLs) within each population and the clonal fraction (i.e., the ratio of MLLs to samples) which was calculated as (N – MLLs) / N. The analyses were run considering the following parameters: (i) a threshold of 6 loci (i.e., pairs of individuals that differed in a maximum of 6 loci were considered as clonemates), (ii) an infinite allele model, (iii) 1,000 permutations and (iv) randomizing of alleles over individuals within populations, and over individuals over all populations. Since most studies to date have reported threshold values between 2 and 4% (e.g., ; ; ), we used a threshold of 6 for our dataset as it represents a 4% (6/151 loci = 0.039) of AFLP genetic dissimilarity between pairs of individuals. […]

Pipeline specifications

Software tools Statistica, GeneMarker, GenAlEx, adegenet, Arlequin, Genodive
Applications Miscellaneous, Population genetic analysis
Organisms Taraxacum officinale