When plants transition from water to land, they became more exposed to carbon dioxide (CO₂) and arid conditions. The evolution of stomata enabled plants to respond to the increased CO₂ availability and limit water loss, ensuring their survival on land. Stomata are micropores on plant surface that are surrounded by two guard cells. These stomata open and close to facilitate gas exchange between plant inner tissues and the environment. Optimal gas exchange and water usage require efficient allocation of leaf surface to stomata. Improper stomatal patterning and density result in a change in mesophyll tissues and the epidermal area, affecting photosynthesis (Chen et al., 2016). Stomatal development, therefore, is tightly regulated to ensure there is no loss in photosynthesis efficiency.

Stomata are found in most land plants (Duckett and Pressel, 2018), and the basic morphology is highly conserved (Chen et al., 2017). Most land plants, including Arabidopsis and grasses, undergo stomatal lineage progression and bear similar molecular components despite their substantial differences in morphology and arrangements (Raissig et al., 2016; Hepworth et al., 2018; Conklin et al., 2019; McKown and Bergmann, 2020). In dicotyledonous Arabidopsis, stomatal development initiates from a subset of protodermal cells, meristemoid mother cells (MMCs), which gain stem-cell fate (Figure 1). These MMCs undergo several rounds of amplifying division, where asymmetric division occurs in an inward spiral manner. Each round of asymmetric division yields a small daughter cell, a meristemoid, and a large daughter cell, a stomatal lineage ground cell (SLGC). Meristemoids become guard mother cells (GMCs) which further undergo a single round of symmetric division and differentiation. This process produces two mirror-symmetric guard cells with a pore in the center. SLGCs can continue with spacing division, in which they divide asymmetrically to generate satellite stomata, or they can differentiate into pavement cells (Han and Torii, 2016; Lee and Bergmann, 2019). The stomatal complexes of monocotyledonous grass species are distinct from those in eudicot. Dumbbell-shaped guard cells are accompanied by subsidiary cells, arranged in parallel to the leaf vein, and develop without stem cell-like meristemoid cells. The initiation of stomatal lineage by an asymmetric division of the precursor cells directly produces GMCs. The neighboring cells of the newly formed GMCs acquire subsidiary mother cell (SMC) fate and establish polarity toward the adjacent GMC. After the SMCs asymmetrically divide and differentiate into subsidiary cells, the GMCs divide symmetrically to generate a pair of guard cells. Two guard cells and subsidiary cells together form a dumbbell-shaped stomatal complex (Hepworth et al., 2018; Nunes et al., 2020).

Stomatal development is sequentially controlled by three basic helix–loop–helix (bHLH) transcription factors, SPEECHLESS (SPCH), MUTE, and FAMA. These transcription factors work together with their heterodimeric partners SCRM (ICE1) and SCRM2 (Ohashi-Ito and Bergmann, 2006; MacAlister et al., 2007; Pillitteri et al., 2007; Kanaoka et al., 2008; Figure 1). The SPCH/SCRM heterodimer determines stomatal fate and maintains the proliferation step by promoting asymmetric division (MacAlister et al., 2007). The MUTE/SCRM dimer terminates the proliferative state initiated by SPCH/SCRM and drives differentiation of meristemoids to GMCs (Pillitteri et al., 2007). The FAMA/SCRM dimer induces guard cell differentiation and restricts the last symmetric division to form a pair of guard cells (Ohashi-Ito and Bergmann, 2006). Two closely related R2R3 MYB transcription factor genes, FOUR LIPS (FLP) and MYB88, function in a semi-redundant manner with FAMA in differentiation and coordination of the terminal symmetric cell division (Geisler et al., 1998).

bHLH genes regulating stomatal development are well conserved across the land plants (Liu et al., 2009). However, their function is alternatively wired among different species to accommodate the fundamental difference in the patterning and shape of stomatal complexes. In particular, two copies of SPCH that function redundantly at the entry of stomatal lineage are found in Brachypodium and rice (Raissig et al., 2016). In contrast to Arabidopsis, MUTE protein in grass species can move from GMCs to the adjacent cells (Raissig et al., 2017; Wang et al., 2019; Wu et al., 2019). This lateral movement of MUTE establishes the identity of SMCs and drives the differentiation of subsidiary cells in a non-cell-autonomous fashion (Nunes et al., 2020).

Expression profiling of stomatal lineage cells at different developmental stages has revealed dynamic changes in the transcriptomes (Pillitteri et al., 2011; Adrian et al., 2015; Zhu et al., 2020; Ho et al., 2021). Moreover, direct target genes of the three master transcription factors have been identified (Hachez et al., 2011; Lau et al., 2014; Han et al., 2018). The recent advance in single-cell RNA-sequencing technology enables profiling spatiotemporal gene expression at the level of the individual stomatal lineage cells (Liu et al., 2020; Lopez-Anido et al., 2021). The systems-level analyses of stomatal lineage cells have deepened our understanding of how the developmental stage-specific transcriptional factors fulfill the developmental program.

Mitogen-Activated Protein Kinase (MAPK) cascades consist of YODA (YDA), MKK4/5/7/9, and MPK3/6. These cascades act upstream of the developmental stage-specific transcription factors (Bergmann et al., 2004; Wang et al., 2007). The activation of the MAPK cascade destabilizes SPCH and MUTE (Lampard et al., 2008; Qi et al., 2017). YDA-MPK3/6 MAPK signaling also plays a substantial role in establishing the cellular polarity required for asymmetric division and stomatal fate specification (Zhang et al., 2015, 2016). The molecular switch that either activates or suppresses the MAPK cascade, at least in part, consists of secreted epidermal patterning factor (EPF) peptides. EPF peptides are perceived by their corresponding cell-surface receptor complexes composed of the proteins belonging to the ERECTA family and their co-receptors TMM and SERK families (Torii, 2012). EPFs and their cognate receptors are conserved throughout the land plants (Caine et al., 2016; Hepworth et al., 2018). Functional studies of monocot species suggest that EPF1 and EPF2 inhibit, but EPFL9 enhances stomatal development, indicating that they are the functional ortholog of Arabidopsis EPFs (Franks et al., 2015; Hepworth et al., 2015; Wang et al., 2016; Hughes et al., 2017; Caine et al., 2019; Dunn et al., 2019; Mohammed et al., 2019). In addition, PANGLOSS1 and PANGLOSS 2 are Leucine-Rich-Repeat receptors in grasses and promote polarization of asymmetric SMC division (Cartwright et al., 2009; Facette and Smith, 2012; Facette et al., 2015). This result implies divergent receptor complexes perceive the extrinsic signal to modulate stomatal development.

Stomatal development is tightly controlled to ensure growth and survival and allows the plant to adapt to environmental changes. The current global climate changes could directly affect stomatal development and function and, thus, plant growth. CO₂ is a gas that exacerbates the greenhouse effect and, therefore, global climate change (Lindsey, 2020). Temperature increases are often accompanied by drought stress due to enhanced water evaporation and changing rain patterns. Both excessive heat and drought stress negatively impact plant production. A significant loss (80–90%) in grain yield can result from warm temperatures during the plant reproductive stage (Hatfield and Prueger, 2015). Given the role of stomata in plant growth, further investigation of stomatal development and their response to environmental changes would provide a strategy that could enhance plant performance and productivity despite global climate change (Dow and Bergmann, 2014; Buckley et al., 2019).

Plant survival dictates that the level of expression and the activity of the master transcription factors in each state of stomatal development must be tightly regulated. SPCH expression is modulated directly by upstream transcription factors. These factors are induced by light, drought, and heat (Klermund et al., 2016; Lau et al., 2018; Qi et al., 2019). SPCH activity is crucial to determine the fate of bi-potent stomatal lineage cells after asymmetric division and is substantially modulated by various kinases. MAPK signaling cascades inhibit the SPCH activity (Lampard et al., 2008; Gudesblat et al., 2012; Yang et al., 2015; Han et al., 2020). The mechanism by which the expression and activity MUTE and FAMA are regulated, in contrast, is not fully understood yet.

The signaling pathway upstream of the bHLH transcription factors is initiated by the EPF-family ligands, ERECTA-family receptor-like kinases, TMM as a co-receptor, and the downstream MAPK signaling cascade. The EPF ligands belong to the peptide family secreted to the apoplast (Herrmann and Torii, 2020). Structural and biochemical analysis revealed that heterodimerization between the ER-family receptors and TMM is required for EPF1 and EPF2 ligand perception, but was not required for EPFL4/5/6 (Lin et al., 2017), which indicates that ligand-based selectivity of receptor heterodimerization specifies the downstream biological process. ER receptor-TMM modules can also associate with the co-receptors BRI1-ASSOCIATED RECEPTOR KINASE/SOMATIC EMBRYOGENESIS RECEPTOR KINASE (BAK/SERK) to form a ternary receptor complex. This complex contributes to the stabilization of a specific ligand–receptor pair and hence, confers higher specificity in signaling (Meng and Yao, 2015).

EPF1 and EPF2 peptide ligands negatively regulate stomatal formation (Hara et al., 2007, 2009; Hunt and Gray, 2009). EPF2 restricts the initiation of stomatal formation by reducing the stability of SPCH (Bergmann et al., 2004; Pillitteri et al., 2011; Robinson et al., 2011). Exaggerated EPF2 signaling leads to a plant leaf epidermis composed of only pavement cells, similar to the spch mutants (Hara et al., 2007; Rowe and Bergmann, 2010; Rychel et al., 2010). EPF2 expression is directly regulated by SPCH and SCRM (Lau et al., 2014; Horst et al., 2015). It is repressed by MUTE activity when the stomatal fate is committed in GMCs. This regulation allows EPF1 to access the common receptor system (Han et al., 2018). Another EPF-family member, EPF1, inhibits stomatal formation by negatively regulating the stability of the MUTE protein (Hara et al., 2007; Lee et al., 2012, 2015; Lin et al., 2017; Qi et al., 2017, 2020). Enhanced EPF1 signaling causes the stomatal lineage to arrest at the meristemoid state, phenocopying the mute mutants. Unlike EPF2, EPF1 expression is not directly regulated by MUTE (Qi et al., 2017). The transcription factors regulating EPF1 expression are unknown.

EPF-like 9 (EPFL9), also known as STOMAGEN, promotes stomatal formation and is highly responsive to environmental cues (Kondo et al., 2010; Sugano et al., 2010). With a highly similar topology structure, STOMAGEN competes with EPF1 and EPF2 for the same receptor complex but does not activate the MAPK cascade (Ohki et al., 2011; Lee et al., 2015; Qi et al., 2017). An auxin-related transcription factor MONOPTEROS (MP) suppresses the expression of STOMAGEN. The light-inducible transcription factor ELONGATED HYPOCOTYL 5 (HY5) induces STOMATAGEN expression (Wang et al., 2021).

SCRM functions as a scaffold protein and associates with MPK3/6 and SPCH (Putarjunan et al., 2019). MPK3/6 first binds to the bipartite SCRM-KRAAM motifs of SCRM, which bridges with SPCH. Then, it triggers the subsequent phosphorylation of SPCH, resulting in the degradation of SPCH-SCRM (Putarjunan et al., 2019). The KRAAM motif is distinct from the putative MAPK docking sites and is uniquely found in SCRMs and their orthologs. This distinction suggests a mechanism for MAPKs conferring substrate specificity in a developmental process.

The localization of the MAPK cascade is unevenly distributed in MMC and in the two daughter cells derived from an asymmetric division. This distribution is facilitated by the intrinsic polarity protein, BASL. BASL arranges the tethering of YDA-MPK signaling components at its polarized crescent of the membrane, which results in elevated SPCH phosphorylation/destabilization to reduce the concentration of SPCH protein in SLGCs (Zhang et al., 2015, 2016). YDA-MAPK signaling must be restricted to allow high division potential in MMCs. This process is mediated by the polarized POLAR-BRASSINOSTEROID-INSENSITIVE2 (BIN2) module at the polarity crescent. Polarized BIN2 can inhibit YDA, leading to the attenuation of the YDA-MAPK signaling (Houbaert et al., 2018; Guo and Dong, 2019). The protein phosphatase BRI1 SUPPRESSPR 1 (BSU1)-LIKE 1 (BSL1) associates with the polarity complex and subsequently triggers the translocation of BIN2 from the polar crescent to the nucleus. It also activates YDA-MAPK signaling in SLGCs. BIN2 in the nucleus directly phosphorylates and destabilizes SPCH. This destabilization further restricts asymmetric division and leads to differentiation (Gudesblat et al., 2012; Guo et al., 2021a,b).

MAP KINASE PHOSPHATASE1 (MKP1) fine-tunes the MAPK signaling by inactivating the MAPK cascade at the early stage of stomatal development (Tamnanloo et al., 2018). The mkp1 mutants undergo the asymmetric entry division but fail to differentiate, resembling the mute mutant phenotype. Genetic analysis puts MKP1 upstream of MPK3/6 and downstream of YDA. Additionally, MPK3/6 signaling is hyperactivated when MKP1 is absent, suggesting MKP1 deactivates MPK3/6. Interestingly, MKP1 expression fully rescues the mkp1mutant phenotype only when the MPK1 expression is driven by the SPCH promoter. This result suggests that MKP1 has a significant role in controlling guard cell fate commitment in early stomatal precursor cells.

Phosphorylation on the SPCH protein usually leads to its degradation. Yet, phosphorylation on Thr49, Thr50, Ser51, and Ser52 of the SPCH protein has been shown to stabilize SPCH (Han et al., 2020). The accumulation of KIN10, a catalytic subunit of the conserved energy sensor kinase SnRK1, can be induced by sucrose and promotes stomatal development. KIN10 is highly expressed in the nuclei of stomatal lineage cells where it phosphorylates and stabilizes SPCH to promote sucrose-induced stomatal formation (Han et al., 2020). Mutations in KIN10 result in decreased stomatal index, while overexpression of KIN10 results in a higher stomatal index. This finding partly explains the previously observed stomatal phenotype when plants are treated with high doses of sucrose (Akita et al., 2013).

Multiple kinases phosphorylate SPCH, resulting in the destabilization of SPCH. Protein phosphatases that are responsible for dephosphorylating SPCH, however, have recently been discovered (Bian et al., 2020). The scaffolding A subunit of the PP2A heterotrimeric complex directly associates with SPCH. Loss-of-function mutations in PP2A or inactivation by PP2A inhibitors results in reduced stomata formation and round pavement cells. This result suggests that PP2A promotes stomatal development by positively regulating SPCH stability. Furthermore, undifferentiated precursors after asymmetric divisions and round pavement cells in the pp2a mutants increase the likelihood that PP2A may regulate other proteins such as MUTE or those involved in pavement cell differentiation.

Given the essential role of stomata in photosynthesis and nutrient uptake, stomatal development is tightly controlled in response to a diverse range of environmental and phytohormonal signals (Qi and Torii, 2018). Plants have developed a highly coordinated regulatory mechanism that balances the need for photosynthesis and transpiration according to their specific environment (Hepworth et al., 2016). Atmospheric CO₂ concentration, water availability, environmental temperature, light intensity, and nutrients all influence the endogenous program (Figure 2).

Significant progress has been made in stomatal development over the past two decades. The discovery of key regulators and a core developmental pathway in stomatal lineage has contributed to the current understanding of stomatal development. Technical advances have identified genome-wide transcriptomes at the bulk-tissue, single-cell types, or single-cell level. ChIP-sequencing data of the master transcription factors are also accumulating. Despite the recent discoveries, there are open questions to be addressed: How are the stomatal lineage cells first established? How is the expression of master transcription factors initiated and coordinated at the chromatin level? How is intrinsic polarity complex targeted to the cortical membrane and integrated into the extrinsic signaling pathway? What are the molecular links between the environmental and hormonal control of stomatal development? Given the wealth of genetic resources and various types of omics data, one should explore novel factors and investigate comprehensive gene networks in stomatal lineage.

Single-cell RNA-seq can be applied to profile responses to environmental or hormonal perturbations, which could reveal whether the cell identity and cells at different developmental stages differentially respond to stress or hormone. Profiling chromatin accessibility such as Assay for Transposase Accessible Chromatin Sequencing and deoxyribonuclease I hypersensitive sequencing at the single-cell level would further the current understanding of the transcriptional program when combined with the single-cell transcriptome. Additionally, advanced live-cell imaging and quantitative image analysis make it easier to keep track of lineage and monitor the cellular events at single-cell resolution, which will help to pave the way to applying the single-cell approach to crop species.

We are currently facing an environmental crisis as global climate change caused by the greenhouse gases accelerates the temperature rise and water shortage. Each of these factors will negatively impact plant ecosystems, sustainability, and food supply. Stomata can directly influence the atmospheric CO₂ level, water evaporation, and even the surface temperature of the earth. Future challenges in stomatal development research, therefore, include addressing the open questions as well as translating discoveries made in Arabidopsis to agriculturally important crops. Many studies have demonstrated that stomatal morphology, density, size, distribution, and specialized cell type in grasses determine stomatal conductance. Stomatal development, then, significantly influences plant behavior in response to environmental changes. Genetic manipulation of stomatal development and physiology may contribute to better survival of plants and maintain agricultural productivity despite unfavorable conditions. Studies of stomatal development in monocots have started mostly by the comparative analysis of the genes in Arabidopsis. Because monocots bear linear arrangement of guard cells in specific cell files and four-celled stomatal complex, which enables “speedy stomata,” the investigation of stomatal development and their regulation is a key to the breeding guidance. As transcriptional regulation is the major mechanism in environmental regulation on stomatal development, the rapidly responding transcription factors could form a useful toolbox for genetic manipulation. The use of CRISPR/Cas9 genome-editing technology in a cell type-specific manner could overcome deleterious effects caused by the ubiquitous disruption of essential genes. It is necessary to collect useful stomatal traits for plants to better adapt to global climate changes, which will allow us to see the outcome of current efforts.

S-KH, JK, and XQ wrote the manuscript. All authors read and approved this version of the manuscript.

The work in the authors’ laboratories was supported by grants from National Research Foundation (2019R1A2C3007376 and 2020R1A4A1019408). S-KH was supported by the Brin Pool Program funded by the Ministry of Science and ICT through the National Research Foundation (2020H1D3A2A02110999). XQ was supported by Rutgers University.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.