Assembling the picture of stomatal evolution

Stomata are microscopic pores that allow the exchange of gases between the plant and its environment. They are typically found on aerial surfaces, opening and closing in response to environmental cues to ensure photosynthesis, respiration and transpiration. In addition, stomata also find themselves at the core of a growing body of work that views plant evolution through the lens of a new phylogenetic hypothesis: bryophyte monophyly. Monophyly refers to an inclusive group of organisms descended from a common ancestor and, in this case specifically, the recent resolution of two monophyletic groups of land plants: nonvascular bryophytes and vascular tracheophytes (Fig. 1). This is in contrast to a previously held paradigm whereby the earliest plants would have preceded various bryophyte-like lineages. Fortin and Friedman's work highlights the fundamental role of phylogeny in reconstructing the evolution of plant traits, where stomata are emerging as a fascinating model to understand how bryophyte monophyly may affect our understanding of the nature of the earliest plants (Harris et al., 2020).

The evolution of stomata has been controversial: the homology of stomata in vascular and nonvascular plants has been questioned, along with the number of times they may have evolved or been lost (Ligrone et al., 2012; Harris et al., 2020). Likewise, their ability to respond to a diversity of environmental cues is well characterised in angiosperms and seed plants, but has been intensely debated in other groups, including ferns, lycophytes and mosses (Sussmilch et al., 2017; Clark et al., 2022). Until now, however, it has not been controversial to propose that among extant plants stomata are only found on the diploid, sporophyte generation. Fortin and Friedman challenge this dogma, identifying mucilage-producing pores (‘Hornwort Gametophytic Pores’ (HGPs)) in the gametophyte generation of hornworts as putatively homologous to stomata and argue that they should be considered the first example of stomata in the gametophyte generation of living plants.

To determine whether these structures are, or are not, stomata it is necessary to define what stomata are. Fortin and Friedman provide a broad definition of a stomate which is independent of function: two cells surrounding a pore, which may be able to open or close and may be filled with liquid or gas. In vascular plants (tracheophytes), stomata are broadly distributed and function as previously described, although in certain cases stomata have taken on new, specialised roles or even been lost. In nonvascular plants (bryophytes), stomata are more restricted in their distribution and are believed to primarily function in facilitating desiccation for spore release. Both lineages exhibit variations and elaborations of the simple design of stomata, but all would be encompassed by this definition. Notably, this definition excludes the liverwort air pore, a structure predicted to not be homologous and so confirms the absence of stomata in liverworts, but also the pseudostomata of Sphagnales, which lack a pore (Merced, 2015).

One of the more striking points made by Fortin and Friedman is how frequently HGPs are overlooked in discussions of stomatal evolution (for which I am also at fault). However, the timeliness of this study is highlighted by another recent work which has reemphasised the counter-arguments (Duckett et al., 2024). Duckett et al. (2024) take a hard line that these structures are definitively not stomata and present multiple arguments elaborating why these structures are unlikely to be homologous to guard cells. The majority of the arguments, laid out previously and by Duckett et al. against HGPs being homologous to stomata is that: they occur in the gametophyte generation; they possess different functions; and they possess anatomical differences from the stomata of other lineages. The first argument does not hold, since stomata in the gametophyte are a feature of multiple species of early land plants preserved in the Rhynie Chert (Edwards et al., 1998). Against the second, Fortin and Friedman argue that two structures do not need to have a common function to be homologous. This is clear even within stomata, where floral nectaries and stomata may represent functional divergence of homologous structures within the same plant. Finally, the anatomical differences observed in HGPs are also observed in other instances among plants or could be the result of functional divergence in hornworts. For example, HGPs occur in clusters, while stomata typically are evenly spaced along the epidermis. However, other bryophyte and tracheophyte species demonstrate exceptions to this rule and clusters of stomata are known to occur naturally in multiple species across the angiosperm phylogeny including Begonia (Rudall et al., 2017), where they are hypothesised to form via perturbations of the normal development of regularly spaced stomata, the genetic basis of which is well understood (Dow et al., 2014).

If we accept the possibility that HGPs are homologous to stomata, then, hornworts represent the only extant lineage to possess stomata in the gametophyte generation. That the exception to the rule may occur in hornworts is not surprising. Hornworts are the most species-poor of all the major land plant groups, possessing a scant fossil record and numerous unique morphological traits. In a phylogenetic context, they have been considered in almost every conceivable position, from the earliest branch of land plants to the immediate sister group to vascular plants. While they are now resolved within the monophyletic bryophyte lineage, molecular dating predicts both that hornworts diverged from the remaining bryophytes during the Ordovician (> 444 million years ago) and that the surviving members of the lineage are relatively young (Harris et al., 2022). The large evolutionary and geological distance between hornworts and all other plants may reconcile some of the anatomical and functional differences between HGPs and other types of stomata. Hornwort genomes are also characterised by extensive gene loss which has been found to be important during the evolution of other unique features of hornworts and may contribute to the rewiring of genetic networks (MacLeod et al., 2022).

Another consequence of the possibility of HGPs being homologous to stomata is how we view the earliest land plants more broadly. The alternation of generations refers to the two long-lived and multicellular generations that plants go through, the haploid gametophyte and the diploid sporophyte, and has recently seen renewed interest as one of the outstanding questions of plant evolution (Donoghue et al., 2021; Duckett et al., 2024; Renner & Sokoloff, 2024). The presence of stomata in both the gametophyte and sporophyte generation, as occurs in early plant fossils and may occur in hornworts, may indicate an isomorphic ancestor of plants, although this is currently speculative (Fig. 1). Alternatively, as Fortin and Friedman also point out, it is possible that stomata arose initially in one generation and then were co-opted into the other through ‘intergenerational heterotropy’, where a trait of one generation is expressed in the other.

The homology of stomata among vascular and nonvascular sporophytes has so far been corroborated by genetic analyses in the moss Physcomitrium patens alongside various vascular systems, identifying a common genetic toolkit underlying stomatal development across land plants (Chater et al., 2016). The nature of these structures remains a case not closed and many of the anatomical anomalies outlined by Duckett et al. (2024) will need to be further scrutinised, yet this work has established a fascinating hypothesis and one that is readily testable. The recent establishment of a tractable and transformable model system in Anthoceros (Frangedakis et al., 2021) means that future experiments will be able to determine whether the hornwort HGPs invoke the same developmental toolkit as bona fide stomata. This work may represent another intriguing complication during the evolution of stomata, with greater complexity in their evolutionary history and greater diversity in both form and function than was previously appreciated. Either way, it is clear that alongside the deluge of genomic data, comparative anatomy remains invaluable to the field of plant evolution and should continue to be explored alongside palaeobotanical, molecular and developmental evidence.