Where stem cells are situated and how they function as part of a proliferative cell population, or ‘meristem’, to promote the growth and development of plants has been under intensive investigation for decades. The root meristem has become an important model for understanding the nature of stem cell populations in plants, primarily because it is more accessible than other meristems (Dolan et al., 1993; Scheres, 2007). Here, we highlight two recent preprints that update our current understanding of the function and diversity of root meristems.
The root meristem consists of the stem cell niche (the ‘quiescent centre’, QC) and surrounding stem cells and their proliferating daughter cells (transit amplifying cells), which together give rise to the various concentric cell layers of the root. How are cell division and differentiation linked in the Arabidopsis root meristem? Following observations by Rahni et al. (2016) and Rahni and Birnbaum (2019) on the graded duration of the cell cycle in root meristems (i.e. from stem cells to transit amplifying cells), Echevarria et al. (2022 preprint) use live-imaging to assess cell cycle duration in different parts of the root. Specifically, they map the duration of G1 in the PlaCCI (plant cell cycle indicator) line, which expresses CFP-labelled CDT1a – a marker that rapidly accumulates after mitosis and degrades at the G1/S transition. They find that G1 duration decreases further away from the stem cell niche, going from more than 20 h around the QC to 2-4 h at the root meristem boundary. Instead, G2 is a little shorter close to the QC, but not enough to equalise the duration of the cell cycle across the tissue.
The authors also analyse published transcriptomic datasets to identify genes with an expression pattern that correlates with the G1 gradient, looking for enriched transcription factor binding sites in these genes to find potential upstream regulators of such gradients. This search highlights sites for APETALA2-AINTEGUMENTA (AP2-ANT), bHLH and SQUAMOSA PROTEIN BINDING family members within the identified genes. The AP2-ANT family includes PLETHORA (PLT) proteins, which are well-known root stem cell regulators that confer cell growth potential and are expressed in a graded fashion matching the G1 gradient (Aida et al., 2004; Mähönen et al., 2014), suggesting that PLTs might be implicated in the gradient. Indeed, plt1-4, plt2-2 double mutants exhibit a small root meristem and the G1 gradient is abolished in these plants, with all cells having an equal, short G1 duration.
Thus, these PLTs inhibit G1 progression, but how does this square with the previously established role of these genes in regulating cell proliferation potential? The authors propose that PLTs act upstream of an incoherent feed-forward loop, whereby they promote cell proliferation on the one hand and, on the other hand, upregulate a negative inhibitor of cell proliferation. They further use a computational model to show that such regulation is sufficient to explain the G1 duration gradient based on the PLT expression gradient. Finally, they identify a regulatory pathway responsible for the indirect cell division inhibition by PLTs: PLTs upregulate KRP5, which inhibits CDK, which in turn inhibits RBR1, an inhibitor of cell cycle progression.
The manuscript by Echevarria et al. (2022 preprint) makes an important contribution in sharpening our view on how PLT genes function as master regulators of root meristem zonation. As PLTs have been shown to play a role in various other meristems during shoot and root development, it remains to be studied whether their role in modulating the cell cycle is conserved in other developmental contexts.
Another recent trend has been to understand the diversity of root meristems. This approach has been facilitated by the appearance of several new model organisms representing plant lineages that diverged before the evolution of seed plants. One such model is the lycopod Selaginella, the meristem of which differs from that of seed plants in that there is a pyramid-shaped initial cell (IC) in a position that corresponds to the QC and surrounding stem cells of a typical root meristem in seed plants. Furthermore, Selaginella roots form branches by bifurcating the root meristem rather than by forming lateral roots via the generation of new root meristems from the pericycle layer higher up the root, as seed plants do.
In a second preprint, Fang et al. (2022 preprint) investigate how the roots of the lycophyte Selaginella moellendorffii bifurcate. Although it was already known that in the shoot apical meristem of Selaginella, the apical cell divides symmetrically to initiate dichotomy, there were still multiple hypotheses for the underlying mechanism in roots, for example the idea that the original IC disappears before two new meristematic cells become recruited. The authors therefore use a time-course assay to track the dynamics of the IC in the root as it bifurcates. They find that branching happens regularly enough to allow the enrichment of samples that contain a newly bifurcating meristem. The IC is irregularly shaped and hard to identify but, in all samples analysed, the authors find either a single IC, or two very close to each other, suggesting that bifurcation happens after a symmetric division of the IC.
The authors also perform RNA sequencing on root samples from different time points after the first branching event to identify genes that initiate root branching. They first identify genes that are preferentially expressed in the meristem and then identify the subset that is differentially expressed around the branching event. The majority of these genes are differentially expressed around the midpoint, returning to baseline values in the end. The downregulated genes are enriched in cell cycle- and cell division-associated genes, suggesting that the cell cycle slows or arrests around the time of branch initiation. The upregulated genes are enriched in ribosome- and cell growth-associated genes, similar to the genes seen in young meristematic cells in Arabidopsis. The total set of differentially expressed genes contains a group of 40 genes that are known to play a role in Arabidopsis branching, including ARFs, ARR, PINs, CRF1 and various hormone signalling genes. In addition, TMO5 and LHW, which are involved in early vasculature development in Arabidopsis (De Rybel et al., 2013), are upregulated during branching, as are WOX and SCL, which are known stem cell regulators.
The evolutionary context of these findings is interesting. There is a long-standing hypothesis that roots originate from shoots, also in lycophytes (Fujinami et al., 2020; Fujinami et al., 2021; Harrison, 2017; Gensel et al., 2001). The similarity of the branching mechanism in the shoot and root of Selaginella, involving a symmetric division of the stem cell, corroborates this idea. This is a different mechanism from that employed by Lycopodium, in which branching is initiated when multiple cells in the quiescent region become mitotically active (Fujinami et al., 2021); this supports the notion that roots evolved multiple times.
The study on meristem organization in Arabidopsis by Echevarria et al. (2022 preprint) will be interesting to follow up from an evolutionary perspective. Fujinami et al. (2020) have indicated that, in lycophytes, diverse cellular patterns of root meristem are evident. As opposed to the IC characteristic for Selaginella, Lycopodium appears to have a region of low mitotic activity analogous to the QC characteristic of seed plants. It will be interesting to analyse the organization of the zone of proliferating cells in a comparative fashion. Finally, the presence in the Selaginella root meristem of multiple differentially expressed homologues of Arabidopsis genes provides additional support for the idea that a similar genetic toolkit was recruited for root generation in different plant clades. In the future, it will be interesting to analyse how variations in the interactions of these conserved genes might explain the diversity of root architectures.
We are grateful to Karolina Blajecka for assistance.
R.V. is supported by the Gatsby Charitable Foundation (G112566); Y.H. is supported by the Gatsby Charitable Foundation (GAT3395/PR3), the Bill and Melinda Gates Foundation (CASS 2.0), a European Research Council PoC APPLICAL award, a University of Helsinki (Helsingin Yliopisto) award (799992091), Jane ja Aatos Erkon Säätiö, the Academy of Finland Finnish CoE in Tree Biology program (2022-2029) (decision 346139) and an Academy of Finland Academy Professor award (2021-2026) (decision 345137).
The authors declare no competing or financial interests.