Elevated CO2 and Reactive Oxygen Species in Stomatal Closure

Plant guard cell is essential for photosynthesis and transpiration. The aperture of stomata is sensitive to various environment factors. Carbon dioxide (CO2) is an important regulator of stomatal movement, and its signaling includes the perception, transduction and gene expression. The intersections with many other signal transduction pathways make the regulation of CO2 more complex. High levels of CO2 trigger stomata closure, and reactive oxygen species (ROS) as the key component has been demonstrated function in this regulation. Additional research is required to understand the underlying molecular mechanisms, especially for the detailed signal factors related with ROS in this response. This review focuses on Arabidopsis stomatal closure induced by high-level CO2, and summarizes current knowledge of the role of ROS involved in this process.


Introduction
Stomata, the tiny pores formed by pairs of guard cells on the surfaces of plant leaves, are the main portals for gas exchange, water transpiration and pathogen invasion in plants. Plants adapt to changes in the external environment by regulating the aperture of stomata. The movement of guard cells is regulated by a variety of environmental factors, including water status, light, carbon dioxide (CO 2 ) concentration and pathogen attack, as well as internal signals such as phytohormones, calcium and reactive oxygen species [1,2]. CO 2 , as an important regulator of stomatal movement and development, greatly affects the growth and biomass of plants. With the development of modern industry and the increase in human activities, the CO 2 content in the atmosphere is increasing year by year. In January 2021, data from the Mauna Loa Observatory show that the atmospheric background CO 2 content will exceed 417 parts per million (ppm) (https://www.metoffice.gov. uk/research/climate/seasonal-to-decadal/long-range/forecasts/co2-forecast (accessed on 22 February 2021)). The continuous rise of CO2 in the atmosphere resulted in exponential increasing of intercellular CO 2 levels in leaf cells [3][4][5]. The profound effect of CO 2 on stomatal development has been confirmed using plant fossils with CO 2 [6,7]. In addition to changes in development, the intercellular CO 2 concentration also influences stomatal responses [8,9]. Although CO 2 signaling is well established to regulate stomata movement, its molecular mechanisms are not fully understood.
Reactive oxygen species (ROS) are major regulators of stomatal movement, particularly in response to abiotic and biotic stress [18,19]. ROS, particularly hydrogen peroxide (H 2 O 2 )

Importance of CO 2 Regulation of Stomatal Conductance
Under dark conditions, respiration leads to a rapid increase in intercellular CO 2 concentration ([CO 2 ] cys ) in plant leaves, exceeding 600 ppm [22]. However, under light conditions, [CO 2 ] cys drops below 200 ppm [22]. High environmental CO 2 concentration induces stomatal closure, whereas low environmental CO 2 concentration triggers stomatal opening. Despite its importance, little is known about the molecular mechanisms underlying the initial response of stomata to CO 2 . CO 2 negatively affects stomatal conductance, or the rate of passage of materials through stomata, by reducing both the number of stomata per unit of leaf area and the stomatal apertures. Decreases in stomatal conductance lead to decreases in transpiration, reducing the loss of water from leaves, which is beneficial to plant water conservation [23]. In the case of relatively high CO 2 levels, fewer stomata and smaller apertures reduce the heat evapotranspirative cooling capacity of leaves, which aggravates heat stress under water shortage [24,25]. Notably, heat stress greatly reduces plant resistance and crop yields globally, affecting agricultural production and possibly nutrient content [26,27]. By contrast, greater stomatal conductance improves crop yield [28][29][30]. Therefore, under conditions of sufficient water, higher CO 2 concentrations reduce stomatal conductance, which may contribute to suboptimal yields [25].
In some species, CO 2 concentration also affects leaf development and leaf area. These responses directly affect leaf water conservation capacity and plant biomass [25]. Future research is needed to explore the molecular mechanism by which CO 2 regulates plant gas exchange.

Mechanism of High CO 2 -Induced Stomatal Closure
Although our understanding of plant responses to environmental CO 2 has advanced, the mechanisms underlying CO 2 signal perception and transduction are still not clear. After entering the cell, CO 2 is sensed by receptors on the plasma membrane. In animals, CO 2 perception is believed to be a vestigial sensory mechanism [31] associated with carbonic anhydrase [32]. Adenylyl cyclase enzymes play important roles in CO 2 sensing in mammalian and fungal systems [31]. Plant homologs of αor γcarbonic anhydrases have not been identified, while β-carbonic anhydrases have been characterized [10,33]. β-carbonic anhydrases have been proposed to carry out the CO 2 receptor function in plants [31], which is important in the stomatal movement response to changes in CO 2 concentration in Arabidopsis. Of the six Arabidopsis thaliana β-carbonic anhydrases, βCA4 is localized at the plasma membrane, while βCA1 is mainly targeted to chloroplasts when transiently expressed in Nicotiana benthamiana cells [10,34]. βCA1 and βCA4 convert CO 2 into HCO 3 − , and both function in CO 2 -induced stomatal movement. The expression of either βCA4 at the plasma membrane or βCA1 in chloroplasts in guard cells of ca1 ca4 double-mutant plants restored CO 2 -induced stomatal closure [35]. The expression of animal carbonic anhydrase in the double mutant (ca1 ca4) similarly restored CO 2 -induced stomatal closure [10,35]. These data suggest that the catalytic activity of carbonic anhydrase is involved in the signaling pathway of CO 2 stimulation of closure, and HCO 3 − and/or H + may act as second messengers in this signal transduction in plants [10]. The apparent conservation of carbonic anhydrase activity between plants and animals implies that they sense CO 2 by similar mechanisms.
Protein HT1 has been characterized as an important factor in CO 2 signaling. As the first CO 2 -response mutant, ht1 has been identified by thermal imaging, which has higher leaf temperature and reduced stomatal CO 2 response [12]. HT1 encodes a putative MAP-KKK kinase that negatively regulates stomatal responses to changes in CO 2 levels, but does not affect abscisic acid (ABA) signaling [12], suggesting that there is an ABA-independent signaling pathway in CO 2 -regulated stomatal response. ca1 ca4 ht1-2 triple mutants exhibit a similar CO 2 response as ht1-2, suggesting that HT1 acts downstream of βCA1 and βCA4 in CO 2 -induced stomatal movement [10]. βCA4 interacts with PIP2;1, which might transport apoplastic CO 2 [36]. The OST1 protein kinase is required for CO 2 -induced stomatal closing [37]. A recent study proposed that HT1 deactivates OST1 [11]. However, another study found HT1 to be epistatic to OST1 in a high-CO 2 -induced stomatal closing assay [38]. Since the function of OST1 in stomatal closure is mainly through regulating ion channels, the research of anion channels in response to high levels of CO 2 is helpful for establishing a comprehensive signal pathway.
The role of HCO 3 − in regulating stomatal movement has been studied using patch clamp experiments. Stomatal closure requires the activation of anion channels in guard cells. Increasing the concentration of HCO 3 − in the cytoplasm activates anion channels [37]. The MATE transporter protein RHC1 is a receptor for HCO 3 − [11]. Increased CO 2 concentration enhances anion channel activity in guard cells [10]. S-type anion channels in the plasma membrane of guard cells might provide a central control mechanism for stomatal closing [39]. The S-type anion channel SLAC1 plays a crucial role in stomatal movement during signaling [40]. SLAC1 is activated by elevated intracellular HCO 3 − levels in guard cells [37]. Nevertheless, the N terminus is important for ABA signaling, and the SLAC1 transmembrane domain responds to CO 2 but not the N terminus or C terminus [41]. Recent studies have shown that residue R256 of SLAC1 is required for elevated CO 2 induced stomatal closure, but not for ABA-induced stomatal closing [42]. Both OST1 and GUARD CELL HYDROGEN PEROXIDE-RESISTANT1 (GHR1) activate SLAC1 in oocytes [43,44]; this activation is inhibited by HT1 [14]. MPK12 counteracts this inhibition [14]. In a recent natural variation study, MPK12 was also identified as a key player in the Arabidopsis Cvi-0 accession for guard cell CO 2 signaling [15]. These findings suggest that MPK12 inhibits HT1, and HT1 inhibits OST/GHR1-activated SLAC1. Unlike HT1, which mainly functions in CO 2 signaling, MPK12 is also activated by ABA or H 2 O 2 [45]. Co-expressing βCA4 and PIP2;1 with OST1-SLAC1 or CPK6/23-SLAC1 in oocytes activates SLAC1 via extracellular CO 2 [36]. These findings suggest that in the presence of various protein kinases, SLAC1 can be regulated by cytosolic CO 2 /HCO 3 − that has been transported by PIP2;1 and converted by βCA4 in oocytes. Till now, a simple regulation pathway for CO 2 -induced stomatal closure has been disclosed.
Since stomatal movement is always associated with oscillations of cytosolic [Ca 2+ ] cyt [46], research on the stomata response to high levels of CO 2 [47] has also identified the involvement of calcium ions in Commelina communis [48,49] and Arabidopsis thaliana [2,50]. This conclusion is based on experiments in which Ca 2+ accumulated in guard cells subjected to elevated CO 2 concentration, and high CO 2 -induced stomatal closure was impaired in the presence of Ca 2+ chelators, such as 1,2-Bis(2-Aminophenoxy)ethane-N,N,N ,N -tetraacetic acid (BAPTA) or ethylenedinitrilotetraacetic acid (EDTA) [48][49][50]. Exposure to reduced CO 2 concentration triggered more [Ca 2+ ] cyt transients in guard cells than did exposure to elevated CO 2 concentration [2]. As guard cells produce "spontaneous" cytoplasmic Ca 2+ transients, and Ca 2+ is required for high CO 2 -induced stomatal closure, elevated CO 2 concentration might enhance the sensitivity of stomatal closing mechanisms to [Ca 2+ ] cyt [51]. In agreement with this hypothesis, CO 2 -derived bicarbonate enhances the Ca 2+ sensitivity of the S-type anion channel activation in guard cells [37]. The increase in [Ca 2+ ] cyt tran-sients in response to higher CO 2 levels implicates this second messenger in the complex mechanism underlying CO 2 -regulated stomatal movement, an idea that may need more extensive research.
Although an ABA-independent pathway has been demonstrated in CO 2 -regulated stomatal response, ABA signaling components have also been implicated, suggesting that a shared regulatory molecular mechanism may involve ABA and CO 2 in stomatal movement. At both low and high CO 2 concentrations, the [Ca 2+ ] cyt of the ABA-insensitive mutant gca2 changes at a similar rate, and the mutant exhibits strongly attenuated stomatal closure in response to increased CO 2 in leaves [2]. The gca2 mutant also exhibits an altered ABA-induced [Ca 2+ ] cyt pattern in guard cells [46], suggesting that Ca 2+ alteration mediated by GCA2 (GROWTH CONTROLLED BY ABSCISSIC ACID 2) functions downstream or at the convergence point of CO 2 and ABA signaling. Therefore, the same factors as in ABA signaling may have also been implicated in CO 2 -induced stomatal closure, suggesting that the same mechanism exists for stomatal closure, which may be helpful for understanding CO 2 signaling.
In Arabidopsis guard cells, there are two main isoforms of NADPH oxidases, AtRBOHF and AtRBOHD, which are regulated by ABA-dependent processes [60]. ABA-triggered stomatal response is significantly reduced in the atrbohF mutant, a phenotype that is enhanced in the atrbohD atrbohF double mutant but absent in the atrbohD single mutant [60]. Due to its obvious role in pathogen-triggered ROS burst, RBOHD is more commonly recognized to function in plant immune defense [61] and might therefore play a role in stomatal movement. Both NADPH oxidases have been demonstrated to function in guard cell CO 2 responses by staining atrbohD atrbohF with H2DCFDA, and the CO 2 -induced ROS burst requires ABA signaling [21], which implies that the two oxidases are essential for both ABA and CO 2 -induced stomatal response.
Apoplastic ROS are also produced by other oxidases [62]. For example, in Vicia faba and Arabidopsis thaliana, copper amine oxidase and polyamine oxidases are involved in the production of the ROS in the process of ABA-and ethylene-induced stomatal closure [63,64]. The involvement of these oxidases in apoplastic ROS production comes from inhibitor studies, and further research is needed to understand their specific and detailed function and significance in ROS-induced stomatal regulation.
Apoplastic ROS production initiates the activation of plasma membrane Ca 2+ channels, leading to an increase in [Ca 2+ ] cyt levels [51]. The molecular identity of these inducible plasma membrane Ca 2+ channels is not clear. In the cytosol, Ca 2+ stimulates the activation of NADPH oxidases either directly by binding to their cytoplasmic EF hands (the EF hand is a helix-loop-helix structural domain or motif found in a large family of calcium-binding proteins) [65] or indirectly by affecting their phosphorylation by CALCIUM DEPENDENT PROTEIN KINASES (CPKs) [66]. Upon Ca 2+ binding, CALCINEURIN-B LIKE PROTEINS (CBLs) interact with CPKs and CBL-interacting PROTEIN KINASES (CIPKs) [66], and a particular complex CBL1/CBL9-CIPK26 is formed, which then phosphorylates and activates RBOHF [67]. The increase in [Ca 2+ ] cyt is also sensed in the chloroplasts, where a thylakoid membrane-associated Ca 2+ -binding protein, CALCIUM SENSING RECEPTOR (CAS), is activated through an unidentified mechanism [51]. The activation of CAS is responsible for a chloroplast ROS burst and the release of Ca 2+ from thylakoids [68][69][70][71], both of which contribute to cytoplasmic Ca 2+ oscillations, apoplastic Ca 2+ -induced stomatal closure during plant immune defense [69]. The drought sensitivity of the Arabidopsis cas mutant is caused by the improper closure of stomata [72], which highlights the importance of chloroplastic Ca 2+ signaling in stomatal regulation [51].
Although there is clear evidence for the involvement of ROS in the regulation of stomatal aperture, it is not known how the ROS signals are sensed in the guard cell apoplast. Identification of the ROS and redox sensors is a major challenge in plant ROS research. In guard cells, only a few ROS sensing mechanisms are involved in stomatal regulation: redox regulation of the GHR1 apoplastic domain [44], of OST1 [73] and of CPK1 [74]. GHR1 is proposed to be involved in the perception of apoplastic ROS, a plasma membrane-associated atypical leucine-rich repeat receptor-like protein kinase. The apoplastic C-terminal domain of GHR1 has two conserved cysteines (C-57 and C-66) that are necessary for the correct function of the protein [44]. GHR1, as an inactive pseudokinase that mediates the activation of SLAC1 via interacting with CALCIUM-DEPENDENT PROTEIN KINASE 3 (CDPK3), could potentially act in stomatal closure as a scaffolding component [75]. Therefore, GHR1 is implicated as a central regulator of guard cell CO 2 and early ABA responses for stomatal movement by mediating ROS signaling. However, more detailed research is demanded to elucidate the mechanism of ROS in stomatal movement induced by different stimuli.

Function of ROS Signaling in CO 2 -Induced Stomatal Closure
Despite their importance, not much effort has been dedicated to understanding the function of ROS in CO 2 -regulated stomatal behavior. Treatment with high concentrations of HCO 3 − induces ROS production and promotes plasma membrane-localized NADPH oxidase-dependent stomatal closure [76], suggesting that ROS play a role in CO 2 -induced stomatal closure.
Chater et al. concluded that ROS are required for stomatal closure induced by high CO 2 concentrations, as no stomatal closure was observed in rbohD rbohF double mutants [21,77], but there was a strong ROS decrease in guard cells at elevated CO 2 concentrations. These observations could be explained by a reduction in the oxygenase activity of RuBisCO at high CO 2 concentration, and hence a reduction in ROS production by glycolate oxidase activity linked to photorespiration [78]. These findings suggest that ROS generated by NADPH oxidase play an important role in high-CO 2 -induced decreases in stomatal aperture.
Cell wall-bound SHAM-sensitive peroxidases also take part in apoplastic ROS production around guard cells [57,58]. Two cell-wall peroxidase-encoding genes, PRX33 and PRX34, which are highly and preferentially expressed in guard cells, are also involved in ROS production in CO 2 -induced stomatal closure [77]. Pharmacological and genetic studies show that ROS generated by both NADPH oxidases and cell-wall peroxidases contribute to high-CO 2 -induced stomatal closure. The high-CO 2 -induced reduction in stomatal aperture is efficiently abolished by either the NADHP oxidase inhibitor DPI or the cell-wall peroxidase inhibitor SHAM, suggesting that cell-wall peroxidases function in high-CO 2 -induced stomatal closure. Similar to rbohD rbohF double mutants, stomatal apertures of both prx33-3 and prx34-2 mutant lines fail to close in response to high CO 2 . ROS accumulation is not triggered by high CO 2 in prx33-3 prx34-2 or rbohD rbohF mutants, in marked contrast to the >50% ROS increase in wild type. These results indicate that in addition to RBOHD and RBOHF, cell-wall peroxidases including PRX33 and PRX34 play an essential role in high-CO 2 -induced ROS generation [77]. However, it is not certain whether other oxidase proteins also participate in CO 2 -induced ROS accumulation, and more high-CO 2 -induced ROS generation related proteins need to be identified.
Proteins that regulate ROS homeostasis in CO 2 -induced stomatal closure have also been screened. High-CO 2 -induced ROS production and stomatal closure in big mutants are compromised [77]. The BIG/CIS1 protein is involved in diverse processes, including auxin transport, light and hormonal signaling, vesicle trafficking, endocytosis, phosphate deficiency tolerance, and the dynamic adjustment of circadian period [79][80][81][82][83][84][85][86]. He et al. found that cis1 mutants are compromised in both elevated CO 2 -induced closure and reduction in stomatal density [87]. cis1 mutants have a significantly lower leaf surface temperature compared with the wild type when exposed for 40 min to 1500 ppm CO 2 . S-type anion channel activity is disrupted by HCO 3 − treatment compared with the wild type [87]. This work suggests that the signaling pathway for ROS in CO 2 -regulated stomatal movement may include other signals, and unraveling the control mechanism will depend on identifying more of these molecules.

ROS Are the Nodes of CO 2 and ABA Signaling during Stomatal Movement
Since both ABA-and high CO 2 -induced stomatal closure involve the activation of SLAC1 in guard cells, there may be a convergence between ABA and CO 2 signaling. Several mutations causing stomatal ABA insensitivity, such as abi1-1, abi2-1 [88,89], ost1 and ghr1 [14,88], impair stomatal responses to high CO 2 concentrations, indicating that there is a relationship between ABA and CO 2 . Stomata of the gca2 mutant are insensitive to both ABA and high CO 2 signaling [2,46]. ABA receptors, PYR/PYL/RCARs, are also involved in CO 2 signaling, as the inactivation of several of these proteins impairs stomatal closure in elevated CO 2 concentrations [21,88], warranting further identification of the role of these receptors in CO 2 signaling.
Although several key components of ABA signaling are connected with stomatal responses to high CO 2 concentration, suggesting that an ABA-dependent pathway participates in CO 2 signaling, ABA-independent components also exist. ABA-induced stomatal closure is completely functional in the mutants of HT1 and MPK12, whereas these plants are deficient in CO 2 -induced stomatal movements [14,15]. Moreover, experiments which aimed to dissect which parts of the SLAC1 anion channel are important for ABA-and for high-CO 2 -induced stomatal closure showed that transgenic plants expressing the SLAC1 anion channel without both C-and N-terminal regions still respond to changes in CO 2 concentration, but remain ABA-insensitive. Thus, the ABA-induced activation of SLAC1 appears to involve its C-and N-terminal regions, whereas high-CO 2 -induced stomatal closure appears to rely only on the transmembrane region [41].
Current knowledge of both ABA-and high-CO 2 -induced stomatal closure suggests at least three partially overlapping pathways: (1) the direct perception of HCO 3 − by SLAC1 in the presence of protein kinases that activate anion channel SLAC1 [36]; (2) the ABAindependent pathway; and (3) the ABA-dependent component that partially mediates high-CO 2 -induced stomatal closure [21,37,88]. ABA signaling that activates OST1 and CPKs by the suppression of PP2Cs could enhance SLAC1 sensitivity to HCO 3 − , and directly trigger SLAC1 anion currents, although this hypothesis needs to be verified in future research. When the CO 2 supply for plants is sufficient, it is possible that plants under water stress react to increased [CO 2 ] cyt faster and stronger than plants with satisfactory water supply, thus saving water in leaves. This could explain the importance of ABA signaling for CO 2 -induced stomatal movements, which would enable plants to adapt to changing environmental conditions. These reports suggest that ABA and CO 2 signaling pathways in stomatal closure are not totally the same, but special factors or regulation systems are recruited.
The overlap of CO 2 and ABA signaling also indicates that ROS production in guard cells can increase in response to high CO 2 concentration, similar to ABA-induced stomatal closure [51]. We previously showed that ROS accumulate in guard cells treated with HCO 3 − or high CO 2 [21,76,90]. A connection between CO 2 and ABA signaling is supported by the absence of ROS accumulation in stomata under elevated CO 2 concentration in the ABA-deficient double mutant nced3 nced5, the triple pyr1 pyl1 pyl4 and quadruple pyr1 pyl1 pyl2 pyl4 mutants [21]. Similar to ABA, elevated CO 2 induces ROS formation by NADPH oxidases [21,90]. In rbohD rbohF double mutants, guard cells are insensitive to CO 2 /HCO 3 − . These mutants also fail to produce ROS in guard cells in response to elevated CO 2 [21,76]. The impaired accumulation of ROS in guard cells and decreased stomatal closure in response to elevated CO 2 concentration are also observed in the tomato mutant rboh1 [90]. Since ROS-related signal components in ABA-induced stomatal closure have been revealed, the involvement of these ABA factors in ROS signaling in response to CO 2 suggests that high-CO 2 and ABA signaling may converge at ROS, and similar downstream components may function in both pathways.

Concluding Remarks and Future Directions
This review summarizes the molecular mechanisms of the ROS signaling network in plant stomatal movement, primarily under high CO 2 concentrations (Figure 1). Elevated [CO 2 ]-mediated stomatal closure requires an increase in ROS. However, great difficulty still exists in understanding the complex interactions within the guard cell signaling networks in response to changes in CO 2 , which include ROS signaling. Recent research has highlighted the complex interplay between CO 2 , redox/ROS signaling and phytohormonal regulation in the control of stomatal movements, and as discussed in this review, there is cross-talk between ABA and CO 2 signaling by ROS during stomatal closure. The importance of guard cells in photosynthesis and transpiration has attracted considerable effort toward understanding how guard cells perceive and transmit signals from the environment. The identification of proteins that can sense and transmit changes in CO 2 /HCO 3 − and molecules that function in regulating ROS signals in guard cells is urgent to improve understanding of the regulation of stomatal movement in response to CO 2 . CO2 has been transported by PIP2;1 and converted into HCO3 -by βCA4, which is located at the plasma membrane and interacts with PIP2;1. The accumulation of cytosolic HCO3 -leads to the suppression of HT1 by MPK4/12 and RHC1. Repressed HT1 inhibits OST1, resulting in the activation of SLAC1. OST1 phosphorylates respiratory burst ROBHs to produce ROS and induces stomatal closure. GHR1 is an inactive pseudokinase that mediates the activation of SLAC1. BIG1 activates SLAC1 when cytosolic HCO3 -accumulates and produces ROS that induce stomatal closure.
Author Contributions: X.M. and L.B. co-wrote this review. All authors have read and agreed to the published version of the manuscript.  Future breakthroughs will most likely come from the development of new methods and technologies that enable real-time imaging of physiological indexes such as cellular localization of ROS, Ca 2+ , and pH in response to various stimuli including CO 2 ; obtaining novel genetic mutants to identify proteins involved in CO 2 response will also be important [47]. Furthermore, plant altered DNA methylation has been implicated in various environmental stress response [91], increasing evidences demonstrated that multiple epigenetic factors and their interactions with hormones, especially auxin will be another direction for understanding high CO 2 -induced stomatal closure [92]. The isolation of methylation-related mutants will be a useful tool for exploring molecular mechanisms. Future research should also seek to understand the complex interactions between various guard cell signaling pathways and how guard cells modulate the interactions among hormones, ROS, CO 2 and Ca 2+ homeostasis. In addition, the translation of such knowledge from model plants to important crop species will be increasingly important for improving stress resistance and increasing yield. CO 2 has been transported by PIP2;1 and converted into HCO 3 − by βCA4, which is located at the plasma membrane and interacts with PIP2;1. The accumulation of cytosolic HCO 3 − leads to the suppression of HT1 by MPK4/12 and RHC1. Repressed HT1 inhibits OST1, resulting in the activation of SLAC1. OST1 phosphorylates respiratory burst ROBHs to produce ROS and induces stomatal closure. GHR1 is an inactive pseudokinase that mediates the activation of SLAC1. BIG1 activates SLAC1 when cytosolic HCO 3 − accumulates and produces ROS that induce stomatal closure.