Pores for Thought: Can Genetic Manipulation of Stomatal Density Protect Future Rice Yields?

Abstract

Rice (Oryza sativa L.) contributes to the diets of around 3.5 billion people every day and is consumed more than any other plant. Alarmingly, climate predictions suggest that the frequency of severe drought and high-temperature events will increase, and this is set to threaten the global rice supply. In this review, we consider whether water or heat stresses in crops - especially rice - could be mitigated through alterations to stomata; minute pores on the plant epidermis that permit carbon acquisition and regulate water loss. In the short-term, water loss is controlled via alterations to the degree of stomatal "openness", or, in the longer-term, by altering the number (or density) of stomata that form. A range of molecular components contribute to the regulation of stomatal density, including transcription factors, plasma membrane-associated proteins and intercellular and extracellular signaling molecules. Much of our existing knowledge relating to stomatal development comes from research conducted on the model plant, Arabidopsis thaliana. However, due to the importance of cereal crops to global food supply, studies on grass stomata have expanded in recent years, with molecular-level discoveries underscoring several divergent developmental and morphological features. Cultivation of rice is particularly water-intensive, and there is interest in developing varieties that require less water yet still maintain grain yields. This could be achieved by manipulating stomatal development; a crop with fewer stomata might be more conservative in its water use and therefore more capable of surviving periods of water stress. However, decreasing stomatal density might restrict the rate of CO₂ uptake and reduce the extent of evaporative cooling, potentially leading to detrimental effects on yields. Thus, the extent to which crop yields in the future climate will be affected by increasing or decreasing stomatal density should be determined. Here, our current understanding of the regulation of stomatal development is summarised, focusing particularly on the genetic mechanisms that have recently been described for rice and other grasses. Application of this knowledge toward the creation of "climate-ready" rice is discussed, with attention drawn to the lesser-studied molecular elements whose contributions to the complexity of grass stomatal development must be understood to advance efforts.

Keywords: crop physiology; crop yields; grasses; rice; stomatal density; stomatal development.

Figure 1

Comparison of stomatal morphology and development in the leaves of the eudicot Arabidopsis thaliana and monocot grasses. (A) Comparison of guard cell (GC) morphology: kidney-shaped GCs of A. thaliana (left) and dumbbell-shaped GCs typical of monocot grasses (right). GCs and subsidiary cells (SCs) are depicted in dark green and blue respectively. (B) Stomatal development in A. thaliana. Meristemoid mother cells (MMCs) (1) divide asymmetrically to form meristemoids (green) (2). This process of asymmetric division is repeated in the self-renewing meristemoid phase (3). Note, asymmetric spacing divisions of MMCs occur around this developmental stage, with newly formed satellite meristemoids forming at least one-cell away from pre-existing meristemoids and stomata. Asymmetric divisions are then terminated and meristemoids transition to guard mother cell (GMC) state (4). GMCs undergo a single symmetric division to produce a pair of GCs (5) which mature to define the stomatal aperture amid the tessellated pavement cells of the leaf epidermis (6). (C) Stomatal development in grasses (modelled from barley, Hordeum vulgare). At the beginning of stomatal development, protodermal cells start to proliferate at a faster rate than surrounding cell files (1). Protodermal cells undergo just one round of asymmetric division, generating smaller GMCs (green) and larger epidermal cells (2). GMCs become enlarged and provide cues to induce asymmetric divisions in flanking subsidiary mother cells (SMCs) (3) which result in the production of SCs (blue) (4). Analogous to eudicot stomatal development, GMCs undergo a single symmetric division (5) to produce a pair of daughter cells that differentiate into mature GCs surrounded by columnar pavement cells (6).

Figure 2

Simplified models of bHLH-mediated stomatal cell lineage transitions in Arabidopsis thaliana and grasses. Five basic helix-loop-helix (bHLH) transcription factors, SPEECHLESS (SPCH), MUTE, FAMA, and their heterodimeric partners SCREAM (SCRM) and SCRM2 exhibit control over stomatal cell type transitions and are conserved across land plants. (A) Stomatal development in A. thaliana. SPCH directs asymmetric divisions of meristemoid mother cells (MMCs) and promotes the meristemoid self-renewing divisions. Epidermal patterning factor 2 (EPF2) and EPF-like protein 9 (EPFL9) are antagonistic regulators of SPCH; competing to suppress and promote its activity respectively. MUTE terminates the meristemoid self-renewing state and promotes guard mother cell (GMC) differentiation and GMC symmetric division. EPF1 primarily targets MUTE; by restraining its activity extra symmetric divisions of GMCs are prevented. EPF1, like EPF2, is assumed to compete with EPFL9 for the regulation of MUTE. FAMA also prevents ectopic symmetric divisions of GMCs and ultimately promotes guard cell (GC) maturation together with the closely related MYB proteins FOUR LIPS (FLP) and MYB88. At each stage of transition, SPCH, MUTE and FAMA form obligate heterodimer complexes with SCRM or SCRM2. (B) The molecular control of stomatal development is ‘alternatively wired’ in monocotyledonous grasses such as rice (Oryza sativa) and Brachypodium (Brachypodium distachyon). It has been suggested that the transcription factors SHORTROOOT1 and SHORTROOT2 (SHR1/2) are involved in the establishment of stomatal lineage cell files (dashed arrow). SPCH has been duplicated and both homologues act to induce asymmetric divisions in these stomatal lineage cells, with SPCH2 the more predominant actor. MUTE promotes GMC differentiation and is involved in the formation of subsidiary cells (SCs) in neighboring subsidiary mother cells (SMCs). As in Arabidopsis, FAMA inhibits supernumerary symmetric divisions of GMCs, probably in combination with a single FLP protein. SCRM and SCRM2 also appear to form heterodimer complexes with the aforementioned bHLHs throughout the lineage, as is observed in Arabidopsis. SHR1/2 and their partner transcription factors SCARECROW1/2 (SCR1/2) exert regulatory influence at each stage of transition. EPFs/EPFLs are also present during grass stomatal development and, while they can demonstrably constrain or promote development, exactly where and how they regulate the grass bHLHs has not been determined (dashed arrows).

Figure 3

Phylogenetic analysis of SPEECHLESS (SPCH), MUTE, and FAMA peptide sequences in Arabidopsis (Arabidopsis thaliana), rice (Oryza sativa), and Brachypodium (Brachypodium distachyon). Peptide sequences were obtained via BLAST searches of the rice and Brachypodium genomes using Phytozome v12. All peptide sequences with expect values (E-values) < 1 x 10^-30 against AtSPCH, AtMUTE, and AtFAMA were used in the analysis. Stomatal bHLH peptide sequences PpSMF1 and PpSMF2 from the non-vascular moss Physcomitrella patens were included for evolutionary context; obtained from Chater et al. (2017). A total of 18 peptide sequences were involved in the phylogenetic analysis. Peptide sequences were aligned using the MUSCLE algorithm (Edgar, 2004). The phylogenetic tree was constructed using the neighbor-joining method (Saitou and Nei, 1987). The optimal tree with the sum of branch length = 4.05347046 is shown. A bootstrap test was performed (1,000 replicates) and the percentage of replicate trees that clustered taxa into the conformation displayed is shown adjacent to the nodes. Evolutionary distances between peptide sequences (in units of amino acid substitutions per site) were calculated using the Poisson correction method (Zuckerkandl and Pauling, 1965). The tree is drawn to scale, with branch lengths proportionate to the evolutionary distances used to infer the tree. All evolutionary analyses were performed in MEGA6. From left to right, species name, gene accession number and gene name (if characterised) are labeled at the branch tips. The clades comprising SPCH, MUTE, and FAMA orthologues are coloured red, blue and green respectively.