Stomatal CO2 responses at sub- and above-ambient CO2 levels employ different pathways in Arabidopsis

Abstract Stomatal pores that control plant CO2 uptake and water loss affect global carbon and water cycles. In the era of increasing atmospheric CO2 levels and vapor pressure deficit (VPD), it is essential to understand how these stimuli affect stomatal behavior. Whether stomatal responses to sub-ambient and above-ambient CO2 levels are governed by the same regulators and depend on VPD remains unknown. We studied stomatal conductance responses in Arabidopsis (Arabidopsis thaliana) stomatal signaling mutants under conditions where CO2 levels were either increased from sub-ambient to ambient (400 ppm) or from ambient to above-ambient levels under normal or elevated VPD. We found that guard cell signaling components involved in CO2-induced stomatal closure have different roles in the sub-ambient and above-ambient CO2 levels. The CO2-specific regulators prominently affected sub-ambient CO2 responses, whereas the lack of guard cell slow-type anion channel SLOW ANION CHANNEL-ASSOCIATED 1 (SLAC1) more strongly affected the speed of above-ambient CO2-induced stomatal closure. Elevated VPD caused lower stomatal conductance in all studied genotypes and CO2 transitions, as well as faster CO2-responsiveness in some studied genotypes and CO2 transitions. Our results highlight the importance of experimental setups in interpreting stomatal CO2-responsiveness, as stomatal movements under different CO2 concentration ranges are controlled by distinct mechanisms. Elevated CO2 and VPD responses may also interact. Hence, multi-factor treatments are needed to understand how plants integrate different environmental signals and translate them into stomatal responses.


Introduction
Atmospheric CO 2 concentration has nearly doubled within the past 150 years.As a result, the global average temperature has increased by 1.5 °C and relative air humidity decreased across many vegetated areas (Yuan et al. 2019).Increased air temperature and decreased air humidity lead to rising values of vapor pressure deficit (VPD), the difference between actual and saturated air vapor pressures, and this increases transpiration from plants.In order to survive such conditions, plants need to adjust their water management by regulating stomatal conductance.Each stoma is composed of two guard cells and an opening formed between them; the size of the opening is controlled by increasing or decreasing guard cell turgor pressure.Guard cells adjust their turgor pressure in response to various abiotic stimuli to balance water loss and CO 2 uptake for photosynthesis.Understanding how stomata respond to changing CO 2 concentrations and increasing VPD is needed for breeding climate-ready crops.
Plant stomata close in response to elevated CO 2 concentration and open in response to decreased CO 2 concentration.Ultimately, elevated CO 2 activates the S-type anion channel SLOW ANION CHANNEL-ASSOCIATED 1 (SLAC1), which causes rapid stomatal closure (Negi et al. 2008;Vahisalu et al. 2008).CO 2 can enter guard cells through the PLASMA MEMBRANE INTRINSIC PROTEIN 2 (PIP2) plasma membrane channel or by transmembrane diffusion (Katsuhara and Hanba 2008;Wang et al. 2016).Carbonic anhydrases BETA CARBONIC ANHYDRASE 1 (CA1) and BETA CARBONIC ANHYDRASE 4 (CA4) accelerate the conversion of intracellular CO 2 to bicarbonate (HCO 3 − ), which can act as a second messenger (Hu et al. 2010).In guard cells, CO 2 /HCO 3 − promotes interaction between the protein kinases MITOGEN-ACTIVATED PROTEIN KINASE 4 (MPK4)/MITOGEN-ACTIVATED PROTEIN KINASE 12 (MPK12), and the Raf-like kinase HIGH LEAF TEMPERATURE 1 (HT1), leading to HT1 inhibition, which is an essential step in the regulation of stomatal responses to CO 2 (Hashimoto et al. 2006;Hashimoto-Sugimoto et al. 2016;Hõrak et al. 2016;Jakobson et al. 2016;Takahashi et al. 2022;Yeh et al. 2023).HT1 phosphorylates the CONVERGENCE OF BLUE LIGHT AND CO2 (CBC1)/CONVERGENCE OF BLUE LIGHT AND CO2 2 (CBC2) Raf-like kinases that function downstream of HT1 and these Raf kinases can inhibit the S-type anion channel activation via a currently unknown mechanism (Hõrak et al. 2016;Hiyama et al. 2017;Hayashi et al. 2020).Stomata in HT1-deficient plants do not respond to CO 2 concentration changes while carbonic anhydrase, MPK12 and SLAC1 mutants exhibit impaired stomatal CO 2 responses (Hashimoto et al. 2006;Vahisalu et al. 2008;Hu et al. 2010;Hashimoto-Sugimoto et al. 2016;Hõrak et al. 2016;Jakobson et al. 2016).Elevated VPD increases transpiration and reduces epidermal turgor that due to mechanical interactions between guard and epidermal cells in angiosperms leads to faster light-induced stomatal opening (Mott et al. 1999;Pichaco et al. 2024).To prevent wilting, stomata close under elevated VPD.Abscisic acid (ABA) is a drought-induced plant stress hormone and an important stomatal regulator (Cutler et al. 2010).VPD has a direct effect on ABA concentration as increased ABA levels in angiosperms were observed 20 min after increasing VPD from 0.7 to 1.5 kPa (McAdam and Brodribb 2015), promoting the conclusion that ABA may be involved in elevated VPD-induced stomatal closure in angiosperms (McAdam et al. 2016).The protein kinase OPEN STOMATA 1 (OST1) and the leucine-rich receptor-like pseudokinase GUARD CELL HYDROGEN PEROXIDE-RESISTANT 1 (GHR1) are activated in the presence of ABA and trigger anion efflux through the major guard cell slow-type anion channel SLAC1 (Brandt et al. 2012;Hua et al. 2012;Sierla et al. 2018).OST1 and GHR1 are both involved in elevated VPD-and CO 2 -induced stomatal closure response (Xue et al. 2011;Merilo et al. 2018;Sierla et al. 2018;Hsu et al. 2021;Jalakas et al. 2021b).
Understanding CO 2 -induced plant stomatal closure responses is essential for future plant breeding.Due to changing climate conditions, it is also important to understand if and how stomatal CO 2 regulation is affected by elevated VPD levels.Previous work in grasses suggests that elevated VPD levels reduce both stomatal conductance and stomatal sensitivity to CO 2 concentration changes (Morison and Gifford 1983) but the interactions of CO 2 and humidity responses in dicots remain poorly understood.
Under light, CO 2 concentration inside the leaf is usually below ambient levels and this causes stomatal opening.Stomata close when CO 2 concentration inside the leaf increases to ambient levels, and an additional rise in CO 2 concentration to above-ambient levels causes further stomatal closure (Brodribb et al. 2009;Hõrak et al. 2017).Thus, stomatal closure response exists within subambient as well as in above-ambient CO 2 levels; however, it is not clear whether these responses are controlled by the same regulators.In some studies, CO 2 -induced stomatal closure is defined as the response to an increase in CO 2 concentration from ambient to above-ambient levels (Franks and Britton-Harper 2016;Hõrak et al. 2017), while some studies use a [CO 2 ] change from subambient to above-ambient levels (Azoulay-Shemer et al. 2015;Chater et al. 2015).Data from previous studies comparing CO 2 responses in ferns and angiosperms suggest that stomatal responses to CO 2 are different, when changing CO 2 levels in the sub-ambient or above-ambient ranges (Brodribb et al. 2009;Hõrak et al. 2017).To address the underlying mechanisms of CO 2 -induced stomatal closure at different CO 2 transitions under normal and elevated VPD conditions, we studied plants deficient either in guard cell anion channel SLAC1 and its activation (slac1-3, ghr1-3, and ost1-3) or in the CO 2 -specific stomatal signaling branch regulated by MPK12 and HT1 kinases (mpk12-4, ht1-2, and ht1-8D) and carbonic anhydrases CA1 and CA4 (ca1ca4).Our results show that different stomatal regulators have a different degree of importance in CO 2 -induced stomatal closure in subambient and above-ambient CO 2 levels and are also differently affected by VPD.

Stomatal closure kinetics are different between sub-ambient to ambient and ambient to above-ambient [CO 2 ] transitions
We analyzed stomatal responses to CO 2 in the sub-ambient and above-ambient concentration ranges in the model plant Arabidopsis (Arabidopsis thaliana) to clarify whether these responses are controlled by the same or by different regulators.Four different CO 2 transition sequences were used (Fig. 1, A, C, E, and G); in two setups, we applied high VPD (2.3 kPa) as an additional factor before and throughout CO 2 treatments (Fig. 1, A and E).Experiments were started with ambient (400 ppm) CO 2 concentration and each consecutive CO 2 treatment lasted for 2 h.This approach allowed us to investigate plant stomatal response to different CO 2 transitions-stomatal opening in response to CO 2 transition from 400 to 100 ppm (from here on referred to as 400-100; Fig. 1, A, C, and G), from 800 to 400 ppm (800-400; Fig. 1C), or from 800 to 100 ppm (800-100; Fig. 1E), and stomatal closure in response to CO 2 transition from 100 to 400 ppm (100-400; Fig. 1A), from 400 to 800 ppm (400-800; Fig. 1, A, C, and E), or from 100 to 800 ppm (100-800; Fig. 1G).We analyzed the amplitude and speed of stomatal responses, whereas the speed was defined here as stomatal 75% response time (see Methods and Supplementary Fig. S1).
Stomatal closure responses had clearly different 75% response times in Col-0 wild-type plants at different CO 2 transitions: the 400-800 stomatal closure was faster than 100-400 closure under normal VPD (Fig. 1, A and B).The rapid 400-800 response was consistent throughout all experimental setups (Fig. 1, A to F).The observed different kinetics of these CO 2 -induced stomatal closure responses suggest that they could be regulated by different components.Stomatal opening responses to sub-ambient CO 2 levels, 400-100 (Fig. 1, A to D, G, and H) and 800-100 (Fig. 1, E and F) had relatively slow 75% response times, whereas the 800-400 response, opening from above-ambient to ambient CO 2 levels, had lower response range and faster 75% response time (Fig. 1, C and  D).This suggests that stomatal opening to sub-ambient CO 2 concentrations is a much stronger, albeit slower, response than stomatal opening during the above-ambient to ambient [CO 2 ] change.

High VPD but not the order of CO 2 transitions affects stomatal CO 2 -response kinetics
To study whether high VPD affects CO 2 responses, we conducted the 400-100-400-800 and the 400-800-100 CO 2 transitions under conditions where plants were first acclimatized to increased VPD (2.3 kPa) for ∼3 h and then subjected to changes in CO 2 levels under the elevated VPD conditions.Under such conditions, both the 400-100 and 800-100 stomatal opening responses were significantly faster than under normal VPD (Fig. 1, A, B, E, and F).Stomatal closure in sub-ambient to ambient CO 2 concentration range was also enhanced under high VPD conditions, resulting in a shorter stomatal 75% response time during the 100-400 ppm [CO 2 ] transition compared with normal VPD (Fig. 1, A  and B), whereas we were not able to detect an effect of high VPD on the 400-800 ppm [CO 2 ] transition 75% response time (Fig. 1,  A, B, E, and F).These results suggest that during high VPD stress, plant stomata could be primed for faster movements in the subambient to ambient CO 2 concentration range.
In our first experiments, the 400-100 opening stimulus was applied before the 400-800 closure.To test whether the exposure to subambient CO 2 levels had an effect on stomatal CO 2 -responsiveness, we also applied the CO 2 transitions in reverse order (400-800-400-100, Fig. 1C).Stomatal 75% response times during the 400-100 transition appeared slightly shorter when it was the last transition (Fig. 1, C  and D) than when it was the first (Fig. 1, A, B, G, and H), whereas the 400-800 transition response speed was unaffected by previous treatments (Fig. 1, A, B, and C to F).Thus, there were no major effects of the order of CO 2 concentration transitions on stomatal CO 2 -response kinetics.
CO 2 concentration-dependent stomatal CO 2 responses | 609 The CO 2 -specific pathway components MPK12 and carbonic anhydrases are more involved in sub-ambient to ambient than in above-ambient CO 2 -induced stomatal closure To better understand the role of the CO 2 -signalling module comprising MPK12, HT1 and carbonic anhydrases CA1 and CA4 in stomatal CO 2 responses at different CO 2 levels, we analyzed the mpk12-4, ht1-2, ht1-8D, and ca1a4 in a similar experimental set-up as described for wild-type plants in Fig. 1A.All the CO 2 -signalling mutants showed very little stomatal closure in magnitude compared with wild-type plants, whereas 75% response time was significantly reduced only in the ca1ca4 mutant in response to the 100-400 transition (Fig. 2, A, C, and E).The 75% response times for the ht1-2 and ht1-8D mutants in this CO 2 transition are not informative due to hardly any stomatal response in these mutants (Fig. 2, A, C, and E).The response of mpk12-4 to the 100-400 transition was as fast as in wild-type, but significantly smaller in magnitude (Fig. 2, A, C, and E).In response to the 400-800 transition, the HT1 mutants had a very weak stomatal response with small magnitude and slow speed, whereas stomatal closure in the mpk12-4 and ca1ca4 plants was slower than in WT, but larger in magnitude (Fig. 2, A, C, and E).Very weak responses of the HT1 mutants suggest that HT1 is necessary to initiate stomatal responses to CO 2 concentration changes in both ambient and subambient levels.MPK12 and carbonic anhydrase mutants display stronger stomatal responses in the above-ambient than subambient CO 2 range (Fig. 2, A, C, and E), suggesting that respective signaling components have a more important role in sub-ambient compared with the above-ambient CO 2 concentrations.
Elevated VPD led to lower stomatal conductance in all studied mutants (Fig. 2, A and B), and thus smaller magnitudes of CO 2 -induced changes in stomatal conductance (Fig. 2, C  and D).Elevated VPD had some effects on the patterns of CO 2 -responsiveness in the CO 2 -signalling mutants (Fig. 2, A to F).The ca1ca4 and mpk12-4 stomatal response times were similar to WT in the 400-800 transition under elevated VPD (Fig. 2F), and the mpk12-4 mutant had a significantly slower than wild-type response to the 100-400 transition only under elevated VPD.Thus, the relatively larger degree of importance of MPK12 and carbonic anhydrases in regulating stomatal responses in the sub-ambient CO 2 ranges was more pronounced under elevated VPD conditions.
Stomatal opening response to 400-100 transition in all of the studied CO 2 -signaling mutants was lower in magnitude and slower in response time, irrespective of VPD, whereas the difference in response time between wild-type and mutants was larger under elevated VPD (Fig. 2, A to F).Thus, stomatal opening response to . CO 2 pathway mutants retain response patterns under regular and elevated vapor pressure deficit (VPD) conditions.A and B) Stomatal response to CO 2 concentration changes from 400 to 100 parts per million (ppm), 100 to 400 ppm, and 400 to 800 ppm under regular (A) and elevated (B) VPD conditions, mean stomatal conductance ± SEM is shown.C, D) Boxplot of stomatal conductance (g s ) change (mmol m −2 s −1 ) in response to CO 2 concentration changes from 400 to 100 ppm, 100 to 400 ppm, and 400 to 800 ppm, respectively.E, F) Boxplot of 75% response time (min) of stomatal response to CO 2 concentration changes from 400 to 100 ppm, 100 to 400 ppm, and 400 to 800 ppm, respectively.C to F) Boxes represent 25% to 75% quartiles and median as the horizontal lines, whiskers indicate the smallest and largest values, and points show individual plant values.Statistically significantly different groups are marked with different letters (One-way ANOVA with Tukey post hoc test, P < 0.05).A to F) Sample size was 5 for all plant lines.VPD during experiments was 0.9 kPa in (A, C, E) and 2.3 kPa in (B, D, F).Start of the first treatment was between 11:30 and 12:30.Col-0 data are the same as used in Fig. 1, and experiments with Col-0 and mutant lines were done together.
CO 2 concentration-dependent stomatal CO 2 responses | 611 sub-ambient [CO 2 ] involves all of the CO 2 -signaling pathway components represented in this study, including the carbonic anhydrases CA1 and CA4, and the HT1 and MPK12 kinases.

SLAC1 and GHR1 are more important for above-ambient CO 2 -induced stomatal closure
The SLAC1 anion channel is a major component in the activation of stomatal closure response.Thus, we examined CO 2 responses across different CO 2 concentration ranges in plants deficient in the SLAC1 anion channel (slac1-3) and in OST1 or GHR1: proteins involved in SLAC1 activation (ost1-3, ghr1-3; Fig. 3).Stomatal 75% response time of slac1-3 plants was longer in both, 100-400 and 400-800 CO 2 transitions (Fig. 3, A and E) but the latter was more affected as the 75% response time difference compared with wildtype plants was notably larger in the 400-800 transition.Yet, the magnitude of stomatal closure between wild-type Col-0 and slac1-3 was only different in the 100-400 transition (Fig. 3C), while their 400-800 response magnitude was similar.Under elevated VPD, slac1-3 stomatal response amplitude was similar to wild-type both during the 100-400 and 400-800 transitions, although stomatal response times of slac1-3 were still significantly longer than wild-type on both transitions (Fig. 3, D and F).Together, these results show that SLAC1 is important in stomatal closure in both sub-ambient and above-ambient [CO 2 ] ranges, but response speed tends to be more severely impacted in the 400-800 transition, suggesting a more prominent role for SLAC1 in ensuring fast aboveambient CO 2 -induced stomatal closure.
Plants deficient in SLAC1 activation via OST1 or GHR1 also showed impaired CO 2 -induced stomatal closure both in the 100-400 and in the 400-800 transitions (Fig. 3, A, C, and E).The ost1-3 mutant had long stomatal response times in both the 100-400 and 400-800 transitions, whereas the ghr1-3 response was similar to wild type in the 100-400 transition, but slower and very small in magnitude during the 400-800 transition (Fig. 3, A, C and E).Thus, both SLAC1-activating proteins are involved in CO 2 responses in all tested concentration ranges, but GHR1, like SLAC1, appears to contribute more toward above-ambient CO 2 -induced stomatal closure.
Sub-ambient CO 2 -induced stomatal opening response magnitude was similar to wild-type in all of the studied anion channel activation mutants, irrespective of VPD (Fig. 3, C and D).Stomatal 75% response time under regular VPD was similar to wild-type in all but ost1-3 (Fig. 3E), whereas under elevated VPD conditions, both ghr1-3 and ost1-3 had longer stomatal opening 75% response times (Fig. 3F).Therefore, regulation of SLAC1 is less important for sub-ambient CO 2 -induced stomatal opening than the guard cell CO 2 -specific signaling pathway.Only under elevated VPD, the 400-100 transition response was slower compared to the wild-type in ghr1-3 (Fig. 3F), indicating an interaction between CO 2 and VPD signaling in stomatal opening responses.

Shifting CO 2 levels from 100 to 800 ppm masks the differences between stomatal behavior in 100-400 and 400-800 ppm [CO 2 ] transitions
To further study how stomatal movements differ depending on CO 2 concentrations, we did additional gas-exchange measurements, during which CO 2 concentration was changed directly from sub-ambient (100 ppm) to above-ambient (800 ppm).Previously, in Figs. 2 and 3 we saw different stomatal response characteristics for 100-400 and 400-800 CO 2 transitions for different mutants that we failed to detect during the 100-800 CO 2 transition (Fig. 4).For example, mpk12-4 stomatal response was small in amplitude in the 100-400 transition, but much larger in the 400-800 transition (Fig. 2, A and C), whereas in the 100-800 transition the mpk12-4 stomatal response amplitude was similar to wild type (Fig. 4A).Similarly, in the ca1ca4 plants 100-400 stomatal response was smaller and 400-800 response larger than in wild-type in amplitude (Fig. 2C), yet the 100-800 response magnitude was similar to wild-type (Fig. 4C).The ghr1-3 mutant had slower response to the 400-800 transition (Fig. 3E), whereas its 75% response time was similar to wild-type in the 100-800 transition (Fig. 4F).These results demonstrate that differences in stomatal responses between plant lines can remain undiscovered depending on CO 2 concentrations that are used for experiments.
Similar to CO 2 -induced stomatal closure experiments, [CO 2 ] ranges were important also in stomatal opening assays (Figs.2-3 vs Fig. 5).For example, response to the 400-100 transition in mpk12-4 was slower and smaller in magnitude compared with wildtype plants (Fig. 2, C and E), but in the 800-100 experiments, mpk12-4 had normal response amplitude and wild-type-like 75% response time (Fig. 5, C and E).The slac1-3 plants had similar to wild-type 400-100 stomatal opening speed and magnitude (Fig. 3, C and E), but were slower in the 800-100 response (Fig. 5F).We also observed a significantly slower opening response to the 800-400 transition in slac1-3 (Supplementary Fig. S2F), which indicates that the reduced stomatal opening speed of the 800-100 response in this mutant is caused by slower opening in the 800-400 range.Thus, although wild-type plants have no discernible differences between the 400-100 and the 800-100 stomatal opening responses (Fig. 1, A, B, E, and F), the molecular mechanisms are at least partly different for these CO 2 transitions.
We combined information from previously analyzed CO 2 transitions in the mutants used in our study with our results (Table 1).Our findings mostly confirm previous results, where available, with the exception of some differences in response amplitude in the slac1-3, ost1-3, and ca1ca4 mutants that can be explained by longer treatment duration in this work that allowed the slower stomatal responses of these mutants to reach amplitudes similar to wild-type plants.We also found a faster stomatal response to sub-ambient to above-ambient CO 2 transition in the ht1-2, potentially explained by different parameters used to describe response speed in different studies.Our experiments add information on the ambient to sub-ambient, above-ambient to ambient, subambient to above-ambient, above-ambient to sub-ambient, and sub-ambient to ambient CO 2 transitions that have not been systematically addressed before in all the mutants analyzed here.

Discussion
Here, we show that stomatal closure in response to an increase in CO 2 concentration, which occurs both at sub-ambient and at above-ambient CO 2 concentration ranges, is regulated by both the CO 2 -specific and SLAC1-related pathways under all CO 2 ranges, but these pathways have a different degree of importance under different CO 2 concentration ranges.Stomatal closure in response to the 400-800 CO 2 concentration range is faster (Fig. 1B), while the 100-400 response has greater amplitude (Figs.1A  and 2C).These differences may be explained by higher guard cell volume and larger stomatal apertures under lower CO 2 levels, leading to slower responses with larger amplitudes, similar to slower responses of larger stomata (Drake et al. 2013;Kübarsepp et al. 2020).However, the relative contribution of the components involved in stomatal CO 2 -signalling is also different in the 400-800 and 100-400 stomatal closure responses (Figs. 2 to 3).Increasing CO 2 abruptly from 100 to 800 masked the presence of two VPD 0.9 kPa VPD 0.9 kPa 400 100 400 800 [CO ] ppm response patterns in anion channel activation mutants are affected by vapor pressure deficit (VPD) conditions.A and B) Stomatal response to CO 2 concentration changes from 400 to 100 parts per million (ppm), 100 to 400 ppm, and 400 to 800 ppm in regular (A) and elevated (B) VPD conditions, mean stomatal conductance ± SEM is shown.C, D) Boxplot of stomatal conductance (g s ) change (mmol m −2 s −1 ) in response to CO 2 concentration changes from 400 to 100 ppm, 100 to 400 ppm, and 400 to 800 ppm, respectively.E, F) Boxplot of 75% response time (min) of stomatal response to CO 2 concentration changes from 400 to 100 ppm, 100 to 400 ppm, and 400 to 800 ppm, respectively.C to F) Boxes represent 25% to 75% quartiles and median as the horizontal lines, whiskers indicate the smallest and largest values, and points show individual plant values.Statistically significantly different groups are marked with different letters (One-way ANOVA with Tukey post hoc test, P < 0.05).A to F) Sample size was 5 for all plant lines.VPD during experiments was 0.9 kPa in (A, C, E) and 2.3 kPa in (B, D, F).Start of the first treatment was between 11:30 and 12:30.Col-0 data are the same as used in Fig. 1, and experiments with Col-0 and mutant lines were done together.The HT1 kinase is required for plant stomatal CO 2 signaling; plants with impaired HT1 function are nearly insensitive to all CO 2 concentration changes (Hashimoto et al. 2006;Hashimoto-Sugimoto et al. 2016;Hõrak et al. 2016).HT1 together with MPK12 or MPK4 forms a primary CO 2 sensing complex, where CO 2 /bicarbonate triggers the interaction of MPKs with HT1, and this leads to inhibition of the HT1 kinase activity (Takahashi et al. 2022).In our experiments, mpk12-4 plants had disrupted stomatal response amplitude to the  A to F) Sample size was 14 for Col-0; 6 for ht1-2; 7 for ht1-8D; 8 for ghr1-3, mpk12-4 and ost1-3; 9 for slac1-3 and ca1ca4.VPD during experiments was 0.9 kPa in (A to F). Start of the first treatment was between 11:30 and 12:30.Col-0 data are the same as used in Fig. 1, and experiments with Col-0 and mutant lines were done together.
100-400 CO 2 transition, while the 400-800 amplitude was unaffected (Fig. 2, A and C).In Tõldsepp et al (2018) mpk12 mpk4GC doublemutants, where MPK4 expression is suppressed only in guard cells, had no 400-800 CO 2 response, yet mpk4GC single-mutants responded to the 400-800 CO 2 transition similar to wild-type plants.Takahashi et al. (2022) showed MPK12 mutants to have the 400-800 response similar to the current study, although neither of these studies tested the 100-400 response.Thus, it seems that the MPK12/MPK4-HT1 complex largely loses functionality if HT1 is impaired (Fig. 2), same happens if both MPK12 and MPK4 are missing from guard cells (Tõldsepp et al. 2018), but losing only MPK12 preferentially affects the 100-400 CO 2 transition (Fig. 2).Therefore, while in the 400-800 transition, the lack of MPK12 is likely compensated by MPK4, and MPK4 cannot effectively replace the function of MPK12 at ht1-2 ca1ca4 mpk12-4 ht1-8d Col-0 ghr1-3 slac1-3 ost1-3 F 400 800 100 [CO ] ppm Stomatal opening in 800 to 100 parts per million (ppm) CO 2 transition masks different responses present in sub-ambient and above-ambient CO 2 concentrations.A and B) Stomatal response to CO 2 concentration changes from 400 to 800 ppm and 800 to 100 ppm, mean stomatal conductance ± SEM is shown.C, D) Boxplot of stomatal conductance (g s ) change (mmol m −2 s −1 ) in response to CO 2 concentration changes from 400 to 800 ppm and 800 to 100 ppm, respectively.E, F) Boxplot of 75% response time (min) of stomatal response to CO 2 concentration changes from 400 to 800 ppm and 800 to 100 ppm, respectively.C to F) Boxes represent 25% to 75% quartiles and median as the horizontal lines, whiskers indicate the smallest and largest values, and points show individual plant values.Statistically significantly different groups are marked with different letters (One-way ANOVA with Tukey post hoc test, P < 0.05).A to F) Sample size was 5 for all plant lines.VPD during experiments was 0.9 kPa in (A to F). Start of the first treatment was between 11:30 and 12:30.Col-0 data are the same as used in Fig. 1, experiments with Col-0 and mutant lines were done together.sub-ambient CO 2 levels.This could mean that while MPK12 and MPK4 both can form a CO 2 /bicarbonate sensing complex with HT1, their affinity for CO 2 /bicarbonate may be different.
CA1 and CA4 also affected stomatal responsiveness more in the 100-400 CO 2 range, with slow and shallow stomatal response in the ca1ca4 mutant (Fig. 2).The strong stomatal closure in above-ambient 400-800 CO 2 transition in this mutant (Fig. 2, Supplementary Fig. S2, A, C and E) might be related to increased autonomous CO 2 conversion to HCO 3 − in the elevated CO 2 environment due to shifting of the reaction balance toward bicarbonate production under above-ambient CO 2 levels or by the involvement of other carbonic anhydrases, as recently demonstrated (Sun et al. 2022).SLAC1 is important for stomatal closure responses to both elevated CO 2 transitions, but compared with wild type, the response speed of slac1-3 in the 400-800 transition was more affected than in the 100-400 transition (Fig. 3, A and E).In slac1-3 plants, there was small in magnitude and slow stomatal closure and opening in the 400-800-400 CO 2 transitions, yet stomatal opening in response to the 400-100 transition was strong and similar in speed to wild-type plants (Supplementary Fig. S2, B, D and F), further supporting the major role of SLAC1 in stomatal responses under ambient to above-ambient CO 2 concentration changes.In the 100-400 CO 2 transition, stomatal responsiveness may be partly compensated by other ion channels that are functional in slac1-3.In addition to the S-type anion channels like SLAC1, stomatal closure is also affected by R-type anion channels, such as QUICK-ACTIVATING ANION CHANNEL 1 (QUAC1, Meyer et al. 2010;Imes et al. 2013).Jalakas et al. (2021a) demonstrated that the quac1-1 slac1-3 double mutant and quac1-1 slac1-3 slah1-3 triple-mutant had no response to the 400-800 CO 2 transition, while individually slac1-3 had a weak response and quac1-1 stomatal response was similar to wild-type plants.SLAC1 HOMOLOGUE 3 (SLAH3) is another S-type anion channel contributing to stomatal closure (Zhang et al. 2016) and could potentially compensate for the lack of SLAC1 in the 100-400 CO 2 response, although it does not affect the 400-800 stomatal CO 2 response (Jalakas et al. 2021a).Future studies should address the 100-400 CO 2 response in mutants deficient in major S-and R-type anion channels to better understand their potential role in the 100-400 CO 2 response.
Stomatal opening triggered by decreased CO 2 levels, from 400 to 100 and from 800 to 100, had larger magnitude than the 800-400 response (Fig. 1, A, C, E, and G).Responses to 100 ppm final CO 2 concentration had a consistently large amplitude and slow response rate in different experimental setups and were not affected by the order of CO 2 treatments.Thus, stomatal opening in response to sub-ambient CO 2 concentrations is a very prominent response.This is in line with Merilo et al. (2014), where two stimuli with an opposing effect on stomata were simultaneously applied, such as darkness and low CO 2 , or low CO 2 and elevated VPD.In such combinations, Arabidopsis stomata always opened in response to sub-ambient CO 2 levels, further indicating that reduction of CO 2 is a strong and prevailing signal.The sub-ambient CO 2 -induced stomatal opening, similar to the light-induced opening, likely involves H + ATPase activation (Inoue and Kinoshita 2017), changes in sugar and starch metabolism (Flütsch and Santelia 2021), and suppression of stomatal closure via e.g.inhibiting anion channel activation (Marten et al. 2007).The combined activation of all these processes may explain the slow stomatal opening rate in response to sub-ambient CO 2 levels.
Elevated VPD triggers ABA biosynthesis (McAdam et al. 2016), which can potentially increase stomatal responsiveness to CO 2 due to interactions of CO 2 and ABA signaling pathways (Raschke 1975;Merilo et al. 2013;Chater et al. 2015).In previous experiments, elevated VPD has been shown to either increase stomatal responsiveness to elevated CO 2 , potentially via enhanced ABA levels (Bunce 1998), or decrease it, potentially due to reduced stomatal apertures under elevated VPD (Morison and Gifford 1983;Talbott et al. 2003).Our results showed an enhanced stomatal response speed in the 400-100, 100-400, and 800-100 CO 2 concentration transitions but not in the already faster 400-800 transition (Fig. 1, A, B, E, and F).Elevated VPD has been shown to accelerate stomatal opening in light in angiosperms due to reduced back-pressure of epidermal cells on guard cells (Mott et al. 1999;Pichaco et al. 2024), our results of faster stomatal opening in response to sub-ambient [CO 2 ] under elevated VPD are in line with this (Fig. 1, B and E).Additionally, elevated VPD also accelerated stomatal closure responses to elevated CO 2 levels in the sub-ambient to ambient concentration ranges in wild-type Arabidopsis (Fig. 1B).Faster stomatal closure may be caused by increased ABA levels under elevated VPD conditions (McAdam and Brodribb 2015) or may be explained by the smaller steady-state stomatal conductance caused by smaller stomatal apertures under elevated VPD that can adjust faster in response to environmental changes.
Here we have focused on guard cell-specific stomatal CO 2 -signaling components and their different contribution to stomatal closure responses in the sub-ambient to ambient and ambient to aboveambient CO 2 concentration transitions.In addition to these components, it is likely that signal mediators outside guard cells also contribute to different stomatal CO 2 -response characteristics under different CO 2 levels.While guard cells in isolated epidermis can respond to elevated CO 2 levels by a decrease in stomatal aperture (Webb et al. 1996;Chater et al. 2015), signals from mesophyll are needed for strong CO 2 -induced stomatal closure (Mott et al. 2008;Fujita et al. 2013).Mesophyll processes, e.g.photosynthesis and sugar metabolism, are known to impact stomatal behavior (Lawson and Matthews 2020), and their contribution to different stomatal CO 2 -response patterns under different CO 2 concentration ranges merits further study.
Elevated temperatures caused by climate change increase the evaporative demand of the atmosphere manifested as higher VPD levels, which increases transpiration and triggers stomatal closure to avoid wilting.Together, changes in atmospheric VPD and CO 2 levels are among the greatest agricultural challenges of the future, yet how these stimuli together affect plant stomatal behavior is poorly understood.Here, we show that while elevated VPD negatively affects steady-state stomatal conductances, it has little effect on stomatal CO 2 -responsiveness.Nevertheless, in some genetic backgrounds, we found an interaction between CO 2 and VPD treatments, indicating that the simultaneous effects of these factors on stomatal behavior merit further study.We also show that stomatal response to elevated CO 2 has different kinetics under sub-ambient and above-ambient CO 2 concentration ranges and its known regulators contribute to a different degree under different CO 2 concentration transitions.Thus, to better understand stomatal responses to CO 2 it is necessary to carefully consider CO 2 levels and experimental setups.

Gas-exchange measurements
Measurements of stomatal conductance were carried out with a temperature-controlled custom-built gas-exchange device (Kollist et al. 2007;Hõrak et al. 2017).Plants were inserted into measurement cuvettes and allowed to acclimate for 1 to 2 h at normal air humidity (VPD, 0.9 kPa) or at lower air humidity (VPD, 2.3 kPa), 24 °C air temperature and 400 ppm CO 2 .Experiments with various CO 2 transitions were carried out and in some cases both under normal and at high VPD conditions, as shown in Fig. 1.The first CO 2 treatment was applied approximately at noon (11:30 to 12:30), stomatal conductance was always followed for 2 h under each treatment (Fig. 1).

Data analysis
Magnitude of stomatal closure response was calculated as the absolute difference in stomatal conductance between last time point before treatment and at the end of 2 h of treatment at a given CO 2 level (displayed as "Change in g s ").To describe the kinetic characteristics of stomatal responses, we identified the difference between stomatal conductance at the last time point before treatment and at the time point when maximal change in stomatal conductance had occurred for any given response as the total 100% response and calculated the time when 75% of the total stomatal response was achieved (see also illustration in Supplementary Fig. S1).The 75% response time describes the overall response speed across the 2-h treatment time but not the initial kinetics of the stomatal responses.One-way ANOVA with Tukey post hoc test was used for statistical analyses as indicated in the figure legends, and P < 0.05 was considered statistically significant.Statistical analyses were carried out using Past 4.0 (Hammer et al. 2001) and Statistica 7.1 (Stat.Soft.Inc).

Figure 1 .
Figure 1.Kinetics of CO 2 -induced stomatal responses in wild-type Arabidopsis with regular and elevated vapor pressure deficit (VPD).Col-0 wild-type Arabidopsis stomatal response to sequential changes in CO 2 concentration under regular (A, C, E, G) and elevated (A, E) VPD conditions.Mean stomatal conductance ± standard error of the mean (SEM) is shown.B, D, F, H) Boxplot of 75% response time (min) of stomatal response to CO 2 concentration changes.Boxes represent 25% to 75% quartiles and median as the horizontal lines, whiskers indicate the smallest and largest values, and points show individual plant values.B, F) Statistically significantly different groups are indicated with different letters (Two-way ANOVA with Tukey post hoc test, P < 0.05).D, H) Statistically significantly different groups are indicated with different letters (One-way ANOVA with Tukey post hoc test, P < 0.05).Sample size was 5 in (A, B, C, D, E, F) and 14 in (G, H).Start of the first treatment was between 11:30 and 12:30.
CO 2 concentration-dependent stomatal CO 2 responses | 613 processes with different kinetics (Figs. 2 to 4), indicating the necessity to analyze stomatal closure responses to elevated CO 2 levels separately in the ambient to above-ambient and sub-ambient to ambient CO 2 concentration ranges.

Figure 4 .
Figure 4. CO 2 transition from 100 to 800 parts per million (ppm) masks different responses present in 100 to 400 and 400 to 800 ppm CO 2 transitions.A and B) Stomatal response to CO 2 concentration changes from 400 to 100 ppm and 100 to 800 ppm, mean stomatal conductance ± SEM is shown.C, D) Boxplot of stomatal conductance (g s ) change (mmol m −2 s −1 ) in response to CO 2 concentration changes from 400 to 100 ppm and 100 to 800 ppm, respectively.E, F) Boxplot of 75% response time (min) of stomatal response to CO 2 concentration changes from 400 to 100 ppm and 100 to 800 ppm, respectively.C to F) Boxes represent 25% to 75% quartiles and median as the horizontal lines, whiskers indicate the smallest and largest values, and points show individual plant values.Statistically significantly different groups are marked with different letters (One-way ANOVA with Tukey post hoc test, P < 0.05).A to F) Sample size was 14 for Col-0; 6 for ht1-2; 7 for ht1-8D; 8 for ghr1-3, mpk12-4 and ost1-3; 9 for slac1-3 and ca1ca4.VPD during experiments was 0.9 kPa in (A to F). Start of the first treatment was between 11:30 and 12:30.Col-0 data are the same as used in Fig.1, and experiments with Col-0 and mutant lines were done together.