Natural autopolyploids: Understanding their formation and establishment

The term “polyploidy”, a state when the cell nucleus possesses more than two haploid sets of chromosomes, was introduced over a century ago. Since then, many monographs and book chapters, as well as tens of thousands of papers, have been published dealing with this phenomenon. This continuous interest testifies to at least two things: (1) a prominent role of polyploidy in plant evolution, ecology, and breeding; and (2) the remaining gaps in our knowledge of this process. Indeed, polyploidy is widespread in plants because all seed plant lineages experienced at least one whole-genome multiplication event (WGM hereafter; Jiao et al., 2011). Furthermore, WGM frequently leads to speciation, i.e., the formation of a new evolutionary lineage that differs from its diploid or lower-ploid ancestor.

Polyploidization is often connected with the emergence of novel traits. This is more easily achievable through WGM associated with interspecific hybridization, since the resulting allopolyploid often expresses intermediate or transgressive traits that can be biologically relevant. However, in such a situation, it is not clear what is the contribution of polyploidization per se, vs. the contribution of hybridization or their mutual interaction. What is well known, however, is that allopolyploidization stabilizes the reproduction of otherwise highly sterile interspecific hybrids by allowing bivalent chromosome pairing in meiosis and formation of functional gametes (Jenczewski and Alix, 2004). In contrast, in autopolyploids, i.e., pure polyploids arising within a single species, multivalent pairing of homoeologous chromosomes causes serious problems with proper chromosome segregation. This in turn frequently results in unbalanced and aneuploid gametes, i.e., those showing a different number of chromosomes than an exact multiple (euploid) of the haploid chromosome set, and hence, reduced fertility (Lv et al., 2024). Reduced fertility may explain why autopolyploids are considered rare in nature and why they have been less studied compared to allopolyploids (Spoelhof et al., 2017). Consequently, most research on autopolyploids has focused on “old” polyploid lineages, where post-polyploidization evolution might mask the direct effect of WGM (e.g., Hollister et al., 2012). Alternatively, synthetic neoautopolyploids have provided important insights into the immediate effect of WGM (Parisod et al., 2010), but have several disadvantages. These include selection of only the fittest genotypes and karyological and physiological instabilities due to the collateral effect of tubulin inhibitors used to create synthetic neoautopolyploids (Münzbergová, 2017). Moreover, the synthetic approach skips the formation of neotriploids, which are considered as a bridge to reproductively more-stable autotetraploids (Ramsey and Schemske, 1998). Therefore, studying natural neoautopolyploids—those that have undergone recent WGM from a diploid progenitor in sympatric contact zones and are still in the process of stabilizing their new chromosomal structure—offers the best opportunity to investigate the pure effects of WGM. Studying natural neoautopolyploids could also shed light on the origin and establishment of neoautopolyploids in nature, which is “perhaps the most poorly understood aspect of polyploid evolution” (Spoelhof et al., 2017).

In this essay, we introduce Pilosella rhodopea (Griseb.) Szeląg (Asteraceae), a natural diploid-autopolyploid complex that we started to study about 15 years ago, originally with a purely taxonomic and biosystematic goal. Over the years, we have accumulated data that might at least partially answer some of the big questions of autopolyploid research such as “Does WGM trigger evolutionary novelty?” or “How do neoautopolyploids establish and persist in the primary contact zone?” These findings provide a basis to address further exciting questions and suggest directions for future in-depth research in neoautopolyploids.

Pilosella rhodopea is an alpine, herbaceous perennial species that is endemic to the Balkans and consists of three main cytotypes, 2x, 3x and 4x, and rare 5x, 6x and aneuploids. Autopolyploids exhibit genetic clustering based on their population origins, suggesting their recurrent in situ formation (Šingliarová et al., 2019). Importantly, all three major cytotypes coexist in almost all populations studied across the species’ range. Given the high incidence of polyploidy in nature, and intensity of polyploid research, such patterns should be observed frequently, but in fact are not (Kolář et al., 2017). It is therefore of interest to ask why the primary contact zones are so rare in nature.

In natural populations of P. rhodopea, all cytotypes form seeds sexually, but autotriploids and autotetraploids produce a significantly lower number of viable seeds than diploids, and many autopolyploid plants are completely sterile. Moreover, from the small proportion of viable seeds they produce, the majority contain aneuploid embryos, while the remaining seeds with euploid embryos arise from the fusion of ploidy-variable euploid gametes, including unreduced ones. These gametes are involved in virtually unrestricted intra- and intercytotype gene flow instantaneously resulting in cytotypically variable progeny (Šingliarová et al., 2023). This pattern therefore challenges the common view that intercytotype interactions are predominantly detrimental because of the production of nonviable progeny resulting from heteroploid crosses (a triploid block mechanism; Kohler et al., 2010). Pilosella rhodopea plants apparently overcome this ploidy-driven reproductive barrier, and intercytotype gene flow actually enhances the production of neoautopolyploids (Figure 1).

We studied incipient neopolyploid formation and intercytotype interactions using flow cytometric seed screening (FCSS), a method originally introduced in apomixis research (Matzk et al., 2000). FCSS allows accurate tracking of gamete ploidy in natural populations if the ploidy of the mother seed plant is known. Therefore, large-scale application of FCSS could answer whether unreduced female gametes are more important for the formation of neoautotriploids than unreduced pollen, which could be lost during the pollination process, and whether autotriploids play a key role in the neopolyploidization process, as has been shown in P. rhodopea.

To become evolutionarily relevant, a newly born neoautopolyploid must establish a reproductive subpopulation within a primary contact zone. This is a critical step because neoautopolyploids immediately face reduced fitness due to chromosomally unbalanced or even aborted gametophytes and/or gametes (see above), and density-dependent hybridization with the more common parental cytotype often resulting in nonviable progeny (minority-cytotype exclusion; Levin, 1975). Although P. rhodopea can, at least partially, mitigate the negative effect of the triploid block, neoautopolyploids produce so few seeds that their establishment and maintenance in a sympatric population would be highly problematic if they relied solely on sexual reproduction—an assumption supported by a simulation study of a mixed-ploidy Pilosella echioides (Chrtek et al., 2017).

Surprisingly, experimental cultivation and field observations revealed that autotriploids and autotetraploids of P. rhodopea showed increased vegetative growth and formation of new root-borne shoots (root sprouting)—a novel trait never observed in diploids. Importantly, such a shift can occur within a single generation of neoautopolyploids (Šingliarová et al., 2023). As a result, WGM increases vegetative reproduction to such an extent that neoautopolyploids can successfully establish and persist, forming the largest known primary contact zone in angiosperms. From this perspective, the emergence of root sprouting after WGM seems to be a real game changer (Figure 1).

The question arises of how can something completely new be created from the same raw genetic material simply by the multiplication of the genome? Increasing the amount of DNA in the nucleus leads to cytotypic changes that are reflected in a larger nucleus and cell size, an altered cell surface-to-volume ratio, but also longer cell cycle duration (Doyle and Coate, 2019). As a direct consequence of changed stoichiometry at the cellular level, WGM could alter intra- and interchromosome interactions, which in turn can modulate both genome-wide and locus-specific expression of genes involved in specific metabolic pathways (Zang et al., 2019). Following this line of reasoning, we hypothesize that enhanced clonality and emergence of root sprouting in P. rhodopea autopolyploids is likely caused by deregulation of genes involved in phytohormone production, because specific cytokinin-to-auxin ratios determine organ growth in shoots and roots (Schaller et al., 2015; Martínková et al., 2023) (Figure 1). Pilosella rhodopea thus provides an excellent diploid-neoautopolyploid system to test putative links between the WGM-driven phenotype and specific components likely involved in its expression, from genes to chromatin and methylome configurations, and to transcriptomic and phytohormone profiles.

To gain real knowledge about the formation and establishment of neoautopolyploids and the immediate effect of WGM, a more intensive search for natural models such as Pilosella rhodopea should be carried out. Quantitative FCSS in taxa with previously reported intrapopulation ploidy variation, coupled with molecular evidence of autopolyploid origin, could reveal other promising neoautopolyploid systems. These, in turn, will allow for integrated research focused on the immediate effect of WGM in order to answer the following fundamental questions: Can the relatively rare occurrence of primary diploid-autopolyploid contact zones in nature be explained by intrinsic genetic constraints? What factors could facilitate the successful establishment of neoautopolyploids in these zones? What is the extent of triploid block in neoautopolyploids? What are the genomic mechanisms, including epigenetic ones, putatively involved in emergence of novel and evolutionarily relevant traits?

P.M. conceived the idea of this essay and wrote the first draft, B.Š. prepared the figure, and both authors contributed to the final draft.