Looking at shrubs (more) carefully: epigenetic mosaicism boosts variance of offspring traits

Carlos M. Herrera, Mónica Medrano and Conchita Alonso discuss their recent paper: Ecological significance of intraplant variation: Epigenetic mosaicism in Lavandula latifolia plants predicts extant and transgenerational variability of fecundity-related traits.

Plants are nonunitary entities

Charles Darwin emphasized long ago that, given the constellation of features which set them apart, terrestrial plants and animals are “most remote in the scale of nature”. Possibly one of the least obvious and only seldom acknowledged, but most influential, difference between the two major kingdoms arises from their radical dissimilarity in developmental mode. In animals, reproductive and somatic cell lineages diverge early in embryogenesis. In vascular plants, in contrast, cells at growing tips (technically ‘meristems’) are developmentally undifferentiated and can produce either reproductive or vegetative structures. This lifelong totipotency of meristematic cells is a distinctive plant feature. It also brings about another crucial trait, namely the modular construction of plant bodies by continual organogenesis and reiterated production of homologous structures. The reiteration of homologous, functionally-equivalent structures is a truly quintessential, ancestral feature of the body plan of vascular plants, yet its significance in plant evolution is underappreciated, possibly a consequence of the zoocentric flavour of modern evolutionary synthesis (Mayr and Provine 1980).

The implications of the repetition of elements in the construction of plants was lucidly envisioned by Charles Darwin’s famous grandfather Erasmus Darwin (1731-1802), who, in his work Phytologia or the Philosophy of Agriculture and Gardening published in 1800, already highlighted the notion that individual plants are nonunitary entities: ‘…every bud of a tree is an individual vegetable being; and that a tree therefore is a family or swarm of individual plants’.

In recent times, the evolutionary significance of the nonunitary condition of individual plants has been addressed from diverse perspectives (Herrera 2009, 2017). One of these revolves around the hypothesis that the modular basis of plant development will lead, among other, to (i) the preservation and hierarchical expansion among modules of the genetic or epigenetic mutations possibly arising as the plant grows; and (ii) the resulting genetic or epigenetic mosaicism will induce heterogeneity in the plant’s progeny, or in other words, genomically distinct plant modules will produce phenotypically distinct progenies (Herrera et al. 2019, 2021). Our paper on the ecological significance of epigenetic mosaicism in wild lavender (Lavandula latifolia) sits in this broad conceptual framework.

Epigenetic mosaics have consequences for the plant’s progeny

After having previously found that genomic methylation and fecundity varied co-ordinately across different modules of individual lavender shrubs (Alonso et al. 2018), and that subindividual epigenetic mosaics are the outcome of steady epigenetic diversification taking place over the lifetimes of individual plants (Herrera et al. 2021) (Fig. 1), we thought the time was ripe for addressing a crucial question which, if answered positively, would round up our research on subindividual variation in lavender. The question: Does the epigenetic mosaics that arise within single plants in the course of their ontogeny have any repercussion for the next generation? In practical terms, this was tantamount to asking, ‘Are the characteristics of progenies produced by epigenotypically distinct modules of the same shrub predictably related to the specific epigenotype of the module which produced the seeds?’

Fig. 1. Adult lavender plant showing the characteristic branching architecture of the species in the wild (left), and reconstruction of the genealogical diversification in multilocus epigenotypes which takes place within an individual plant with age, taken from Herrera et al. (2021) (right; colour variation is mapped to multilocus epigenotype variation and tips of the genealogical tree correspond to growing tips of the shrub).

To answer this question we resorted to that venerable, time-tested and proven method consisting of setting a common garden in the greenhouse. A large number of lavender seeds produced naturally in the field whose origin was known in detail (maternal plant and specific part within the maternal plant) were sown and kept under identical environmental conditions. By subsequently monitoring seedling emergence, survival and growth we were able to characterize the progenies from different plant parts (we called them ‘subprogenies’) with regard to seed germination rate, time to germination, seedling dry mass and susceptibility to fungal disease (Fig. 2). Prior to the greenhouse experiment, we had obtained detailed information on the epigenotypes of each of the plant modules which produced every subprogeny by applying molecular methods that allowed to detect intraplant variation in the cytosine methylation status of anonymous epigenetic markers. This enabled us to relate variation among the subprogenies of the same plant to variation in epigenotype among the plant modules which had produced the seeds.

*Fig. 2. Wild lavender seedlings in the greenhouse experiment.*

The two most important results of our study were that phenotypic variation among subprogenies produced by different modules of the same mother plant could be explained by the epigenotypic differences among those modules, and that the more epigenotypically variable was a mother plant, the more phenotypically variable was its progeny. The transgenerational effects of maternal features on progeny traits have been thoroughly documented for many plants over the years, but the possibility that epigenetically-defined sectors of the same maternal individual can produce progenies with distinct phenotypes has not been explicitly envisaged before.

Epigenetic boosting of ecological diversity: further level discovered

As considerable empirical evidence has shown over the last few decades, the variations in DNA sequence are not the only molecular mechanism possessing the capacity to create phenotypic, and thus ecological variance in plants. Potentially heritable epigenetic changes, such as those involving DNA cytosine methylation, can also become an additional layer inducing phenotypic variation through effects on gene expression, transposon activity, and plant growth and development (see, e.g., Richards et al. 2017, Alonso et al., 2018, Balao et al. 2018, and references therein). Understanding the ecological causes and consequences of epigenetic variation is a fundamental ecological question (Sutherland et al. 2013), and epigenetic variations among species, populations and individuals enhance variance in ecologically-relevant plant traits (e.g., fecundity, leaf traits, susceptibility to herbivores; Herrera and Bazaga 2011, Medrano et al 2014).

By showing that epigenetic variation within lavender plants acts enhancing the variance of progeny traits, our study has revealed for the first time that the subindividual level is yet another layer of epigenetic variation which can boost the diversity of ecologically relevant plant traits in populations (Fig. 3). Among other possible effects, the diversifying influence of epigenotypic mosaicism on progeny traits which are important for population recruitment can act enhancing offspring establishment in heterogeneous or environmentally unpredictable habitats through diversifying bet-hedging mechanisms.