Skip to main content
Dryad logo

Data from: Genotypic variation in the induction and persistence of transgenerational responses to seasonal cues


Alvarez, Mariano; Bleich, Andrew; Donohue, Kathleen (2020), Data from: Genotypic variation in the induction and persistence of transgenerational responses to seasonal cues, Dryad, Dataset,


Phenotypes respond to environments experienced directly by an individual, via phenotypic plasticity, or to the environment experienced by ancestors, via transgenerational environmental effects. The adaptive value of environmental effects depends not only on the strength and direction of the induced response, but also on how long the response persists within and across generations, and how stably it is expressed across environments that are encountered subsequently. Little is known about the genetic basis of those distinct components, or even whether they exhibit genetic variation. We tested for genetic differences in the inducibility, temporal persistence, and environmental stability of transgenerational environmental effects in Arabidopsis thaliana. Genetic variation existed in the inducibility of transgenerational effects on traits expressed across the life cycle. Surprisingly, the persistence of transgenerational effects into the third generation was uncorrelated with their induction in the second generation. While environmental effects for some traits in some genotypes weakened over successive generations, others were stronger or even in the opposite direction in more distant generations. Therefore, transgenerational effects in more distant generations are not merely caused by the retention or dissipation of those expressed in prior generations, but they may be genetically independent traits with the potential to evolve independently.


We used eight accessions of A. thaliana (Ag0, Col0, Cvi0, Cvi1, Ler, Ler2, Ws1, and Wa1; Table S1), which were chosen because they are the parents of recombinant inbred lines that can be used in future genetic studies, and because they are known to differ in phenological traits such as germination and flowering time. As such, these genotypes are not a random representation of global genetic variants in A. thaliana; however, documenting genetic differences among these genotypes in the focal traits studied in this experiment would establish that the focal phenotypes have a genetic basis, and that those phenotypes can be studied in more detail using the genetic resources derived from these lines. Significant differences among accessions in these aspects of transgenerational plasticity would demonstrate that populations have evolved differences in these traits and that genetic variation for these traits is available for future evolution if populations admix in the future, as they have in the past (Beck et al. 2008; The 1001 Genomes Consortium 2016; Palacio‐Lopez et al. 2017). All of these genotypes germinated after 3‐day cold stratification and were able to flower without vernalization.

See Figure S1 for a schematic of the experimental design. To supply the seed stock for this experiment, seeds of these accessions were bulked at 22°C with 12 h light/12 h dark in Environmental Growth Chambers (EGC, Ohio, USA). Seeds from these plants were planted into each of two diel thermal regimes: “warm” and “cold.” Warm conditions consisted of 12 h light at 24°C followed by 12 h dark at 16°C. Cold conditions consisted of 12 h light at 18°C followed by 12 h dark at 10°C. Seeds of a given genotype were deposited on the soil surface of 2‐inch square pots and stratified for 4 days in dark 4°C conditions to stimulate germination. Plants were thinned to a single focal individual within each pot when plants reached the four‐leaf stage. At 4 weeks, half the pots from each treatment were vernalized at 4°C for 7 weeks, whereas the other half did not receive vernalization. This produced a factorial design of vernalization, diel temperature, and genotype, with 12 replicates of each genotype in each combinatorial condition.

After vernalized individuals were returned to their temperature treatments, all individuals were permitted to self‐fertilize and senesce (defined as >90% yellow/brown siliques) before water was withheld. Seeds were collected from each individual separately. Seeds were stored dry at room temperature in a desiccator (Scienceware, USA) until they were used.

Seeds from each experimental treatment (imposed in Generation 1) were used for germination assays (described below) and additional seeds (after‐ripened for 6 months) were planted into a common environmental condition: 22°C with 12 h light/12 h dark and were not vernalized, with 12 replicates per genotype from each treatment. The following postgermination morphological traits were measured: length of longest leaf at bolting and leaf number at bolting. The following phenological traits were measured (with time of planting as the starting point): days to bolting, days to flowering, and days to the appearance of first mature fruit. Seeds were harvested from these second‐generation plants, and these seeds were used in germination assays (see below) and for planting the third generation. One genotype, Cvi1, had extremely low levels of germination in generation 2 when parents were vernalized, and generated only a single replicate, showing a strong effect of parental environment. Multiple seeds from this single replicate were planted into replicate pots for the third generation, creating 12 replicates. However, because these third‐generation replicates are pseudo‐replicated and do not represent independent samples of the first‐generation treatment, we interpret any genotype effects using this genotype with caution.

Usage Notes

All files are provided as R scripts. Data is mirrored at


National Science Foundation, Award: DEB‐1556855