Skip to main content
NCBI home page
Journal of Experimental Botany logoLink to Journal of Experimental Botany
. 2024 Apr 17;75(21):6856–6871. doi: 10.1093/jxb/erae168

GABA does not regulate stomatal CO2 signalling in Arabidopsis

Adriane Piechatzek 1,2, Xueying Feng 3,4, Na Sai 5,6, Changyu Yi 7, Bhavna Hurgobin 8, Mathew Lewsey 9,10,11, Johannes Herrmann 12, Marcus Dittrich 13,14, Peter Ache 15, Tobias Müller 16, Johannes Kromdijk 17, Rainer Hedrich 18, Bo Xu 19,20,21,✉, Matthew Gilliham 22,23,24,✉
Editor: Tracy Lawson25
PMCID: PMC11565201  PMID: 38628155

Abstract

Optimal stomatal regulation is important for plant adaptation to changing environmental conditions and for maintaining crop yield. The guard cell signal γ-aminobutyric acid (GABA) is produced from glutamate by glutamate decarboxylase (GAD) during a reaction that generates CO2 as a by-product. Here, we investigated a putative connection between GABA signalling and the more clearly defined CO2 signalling pathway in guard cells. The GABA-deficient mutant Arabidopsis lines gad2-1, gad2-2, and gad1/2/4/5 were examined for stomatal sensitivity to various CO2 concentrations. Our findings show a phenotypical discrepancy between the allelic mutant lines gad2-1 and gad2-2—a weakened CO2 response in gad2-1 (GABI_474_E05) in contrast to a wild-type response in gad2-2 (SALK_028819) and gad1/2/4/5. Through transcriptomic and genomic investigation, we traced the response of gad2-1 to a deletion of full-length Mitogen-activated protein kinase 12 (MPK12) in the GABI-KAT line, thereafter renamed as gad2-1*. Guard cell-specific complementation of MPK12 in gad2-1* restored the wild-type CO2 phenotype, which confirms the proposed importance of MPK12 in CO2 sensitivity. Additionally, we found that stomatal opening under low atmospheric CO2 occurs independently of the GABA-modulated opening channel ALUMINIUM-ACTIVATED MALATE TRANSPORTER 9 (ALMT9). Our results demonstrate that GABA has a role in modulating the rate of stomatal opening and closing, but not in response to CO2per se.

Keywords: ALMTs, CO2 signalling, GABA, genomics, MPK12, stomatal movement

The hypothesis that GABA is involved in CO2 signalling is disproven; decreased stomatal CO2 sensitivity of a commonly used GABA-deficient mutant is attributed to a mutation in the CO2 sensor MPK12.

Introduction

Pairs of guard cells delineate a stomatal pore to regulate the rate of plant gas exchange, such as emission of oxygen (O2) and water (H2O) vapour and absorption of carbon dioxide (CO2). These processes define a plant’s carbon gain and water use efficiency (WUE; the ratio of carbon gain per water loss), which are factors essential for underpinning crop yield and drought tolerance (Leakey et al., 2019).

The non-proteinogenic amino acid γ-aminobutyric acid (GABA) has been shown to encode a plant signal that modulates gas exchange and stress tolerance through negative regulation of stomatal opening (Xu et al., 2021a). GABA is produced predominantly via the GABA shunt pathway, which bypasses reactions that are ordinarily confined to the mitochondrial-based Krebs cycle, converting the C5 amino acid glutamate into the C4 amino acid GABA through to succinate (C4), thereby generating carbon dioxide (CO2) as a by-product within the cytoplasm (Fait et al., 2008; Li et al., 2017). The GABA biosynthesis enzyme glutamate decarboxylase 2 (GAD2) is detectable in nearly all plant tissues, and was found to be the main GABA biosynthesis enzyme in leaves (Mekonnen et al., 2016; Xu et al., 2021a). In total, five GAD isoforms exist in Arabidopsis thaliana. GAD1 is mainly expressed in root tissue and GAD5 in male gametes, whilst GAD3 and GAD4 expression is low under non-stressed conditions (Scholz et al., 2015). Conversely, in response to various stresses like cold, drought, hypoxia, pathogen attack, and salt stress, the expression of GAD4 is significantly increased (Kaplan et al., 2007; Miyashita and Good, 2008; Urano et al., 2009; Renault et al., 2010; Deng et al., 2020), while GAD3 is stimulated by combined heat and high light stress (Balfagón et al., 2022). The absence of GAD2 results in a drastic reduction in GABA in leaf tissue, which makes GAD2 knockout lines (gad2) excellent models for investigating the signalling role of GABA in guard cells, in comparison to the other gad knockout lines.

To date, it is known that GABA modulates stomatal movement by negatively regulating so-called ALUMINIUM-ACTIVATED MALATE TRANSPORTERS (ALMTs; Ramesh et al., 2015; Xu et al., 2021a), with ALMTs such as ALMT9 and ALMT12/QUICK ANION CHANNEL 1 (QUAC1) not activated by aluminium (Meyer et al., 2010; De Angeli et al., 2012). Xu et al. (2021a) demonstrated that plants lacking GABA synthesis in leaves display enhanced stomatal opening due to guard cell ALMT9 deregulation. ALMT9 has been identified as a tonoplast-localised anion channel that mediates stomatal opening (De Angeli et al., 2013). Plasma-membrane located ALMT12/QUAC1 could potentially be involved in GABA regulation of the stomatal closure processes but needs further verification (Xu et al., 2021a, 2021b). However, it is unknown whether additional signalling events also contribute to the inhibition of ALMTs by GABA. Signalling pathways often have elements of crosstalk, due to the complex nature of cellular processes never functioning in isolation, rather they are integrated in a dense signalling web (Taiz et al., 2015). In this context, Xu et al. (2021a) discovered that GABA application significantly affects stomatal responses to the signals abscisic acid (ABA), hydrogen peroxide (H2O2) and coronatine, so GABA may well interact with multiple abiotic and biotic stress signalling pathways. In addition, as GABA impacts were also detected under standard light/dark cycles, GABA is likely to modulate standard responses to the environment (Xu et al., 2021a).

Another stomatal regulator is CO2, which is consumed by plants as a fuel molecule for photosynthetic carbon assimilation. In high concentrations, CO2 is known to stimulate stomatal closure by indirectly activating specific anion channels, including ALMT12/QUAC1 (Meyer et al., 2010) and SLOW ANION CHANNEL-ASSOCIATED 1 (SLAC1) (Negi et al., 2008). Stomatal closure in response to elevated CO2 is a way of plants increasing their water use efficiency. This is because under high CO2 conditions, plants are sufficiently supplied with CO2 for photosynthesis, and can therefore afford to promote stomatal closure in order to minimize water loss through transpiration (Lawson et al., 2014). On the other hand, low CO2 concentrations signal plants to open stomata (Roelfsema and Hedrich, 2005; Hiyama et al., 2017).

Besides ALMT12/QUAC1 and SLAC1, β-CARBONIC ANHYDRASE 1 (β-CA1) and β-CA4, as well as MITOGEN-ACTIVATED PROTEIN KINASE 12 (MPK12) are also crucial signalling components of the CO2 signal transduction pathway. The CO2-binding proteins β-CA1 and β-CA4 were found to initiate the high CO2 signalling pathway by accelerating the conversion of CO2 to bicarbonate (Hu et al., 2010). Plants lacking these carbonic anhydrases had impaired CO2 responsiveness, as revealed by the double mutant ca1/ca4 (Hu et al., 2010). In the absence of MPK12, a negative regulator of high CO2-induced stomatal closure, known as HIGH LEAF TEMPERATURE 1 (HT1), constitutively suppresses stomatal closure (Hashimoto et al., 2006; Hõrak et al., 2016; Jakobson et al., 2016; Yeh et al., 2023).

Here, we sought to reveal the effect of GABA depletion on stomatal CO2 sensitivity. We present the results of gas exchange measurements in control (400 ppm), low (100 ppm), and high (800 ppm) CO2-treated GABA-deficient mutant lines in combination with sequencing-based approaches. Based on evidence from this study, we link a genomic deletion of MPK12 to reduced CO2 responsiveness in one of our GABA-deficient lines, currently known in the literature as gad2-1 but renamed gad2-1* in this article in consequence of the findings presented here, while other GABA mutants did not have a CO2-response phenotype. Therefore, we can eliminate the possibility of GABA deficiency being associated with a decrease in CO2 sensitivity. At the same time, we show that the loss of ALMT9 does not compromise CO2-induced stomatal opening, and present data that imply links between mutations in MPK12/GAD2 and the deregulation of essential guard cell-related genes.

Materials and methods

Plant materials and growth conditions

In all experiments, wild-type Arabidopsis thaliana ecotype Columbia (Col-0) and mutant lines generated in the Col-0 background were used. The two allelic mutant lines gad2-1* (GABI_474E05, originally known as gad2-1 in the literature) and gad2-2 (SALK_028819), as well as additional transgenic lines, such as mpk12-3 (SAIL_543_F07), ca1/ca4 (SALK_106570 and WiscDsLox508D11), almt6-1 (GABI_259D05) and almt6-2 (FLAG_425D02) were obtained from the Arabidopsis Biological Resource Centre (ABRC, Ohio, USA). The transgenic line almt9-1 (SALK_055490) was described in a previous study (De Angeli et al., 2013). The transgenic mutant line gad1/2/4/5 was generated by crossing gad1-1 (SALK_017810), gad4 (SALK_106240), gad5 (SALK_203883), and gad2-1* (GABI_474E05), and was obtained from Shuqun Zhang (Deng et al., 2020). The mutant line gad2-1*/GC1::MPK12 was generated through transformation of gad2-1* with the guard cell-specific promoter GC1 (Yang et al., 2008) fused to full-length MPK12. All primers used for the genotyping PCR and cloning reactions are listed in the Supplementary Table S1. Unless noted, all plants were raised in hydroponics for 5–6 weeks under short-day growth conditions [10 h light (~100 µmol photons m−2 s−1)/14 h dark, an average temperature of 22 °C and 50–60% relative humidity] following methods described in Conn et al. (2013).

Gas exchange measurement

Whole rosette leaves of 4–5-week-old plants were analysed in a LI-6400XT Portable photosynthesis system (LI-COR Biosciences, USA) fitted with a 6400-17 Whole Plant Arabidopsis Chamber. Prior to the gas exchange measurements, hydroponic-grown plants were adapted to light with an intensity of 350 µmol photons m−2 s−1 for 1 h (30W LED Panel, 6500K Cool white, Arlec, China). Transpiration rates of whole intact plants were recorded in response to a variety of CO2 concentrations (100 ppm, 400 ppm, and 800 ppm CO2) inside the LI-6400XT Portable Photosynthesis System under the following chamber conditions: 350 μmol photons m−2 s−1 light intensity with a portion of 10% blue light (6400-18 RBG Light Source), 50–60% relative humidity, an average temperature of ~22°C and an airflow rate of 350 μmol s−1. All transpiration rates were normalized to leaf rosette areas. Kinetic analysis of stomatal movement was performed to obtain stomatal half times in response to changes in CO2 concentration. The calculation for half-time responses was based on the Michaelis-Menten Model, where Km is the Michaelis Menten constant and is equal to the transpiration rates at which half of the maximum velocity is reached. The percentages of high CO2-induced stomatal closure were calculated as follows: 100 × (stomatal width [400 ppm CO2] – stomatal width [800 ppm CO2])/ stomatal width [400 ppm CO2].

Stomatal aperture measurement

Five- to six-week-old hydroponic-grown plants were treated with either 400 or 800 ppm CO2 inside a LI-6400XT Portable photosynthesis system (LI-COR Biosciences) under conditions as described above. The production and analysis of epidermal peels followed a previously described method (Xu et al., 2021a). Images of stomata were captured using an Axiophot Pol Photomicroscope (Carl Zeiss, Germany) with a 20× objective lens, and stomata were measured using either the image processing program ImageJ or a newly developed stomata auto-measuring system, known as StomaAI (Sai et al., 2023).

Reverse transcription PCR

Total RNA was isolated from whole rosette leaves of 5-week-old plants using TRIzol reagent (Invitrogen, USA) and was synthesised to cDNA using SuperScript III Reverse Transcriptase (Invitrogen). The cDNA was then amplified using Phire® Green Hot Start II PCR Master Mix enzyme (Thermo Fisher, USA) and primer pairs specific to MPK12, GAD2, and the housekeeping gene Actin2 (Supplementary Table S1), following the manufacturer’s instructions.

Microarray analysis

Microarray hybridization and data processing were conducted at Julius-von-Sachs-Institute for Biosciences (Department of Molecular Plant Physiology and Biophysics - Botany I, University of Würzburg, Germany), following the method described by Dittrich et al. (2019). RNA integrity and concentration were determined using an RNA 6000 Labchip on an Agilent 2100 BioAnalyzer (Version C.01.069) at ACRF Cancer Genomics Facility (SA Pathology and University of South Australia).

RNA sequencing

RNA extraction from whole rosette leaves as well as the RNA integrity and concentration determination were conducted as described above.

RNA sequencing was performed at Animal Plant and Soil Sciences (La Trobe University, Australia). In brief, RNA-seq libraries were constructed using the TruSeq Stranded mRNA Library Prep Kit in accordance with the manufacturer’s instructions (Illumina) and sequenced on a NextSeq500 system (Illumina) as 75 bp single-end reads with at least 30 million reads per sample. Reads quality was examined using FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc), and low-quality reads and adapters were trimmed using Trim Galore (https://github.com/FelixKrueger/TrimGalore). To obtain transcript abundances as transcripts per million (TPM) and estimated counts, trimmed reads were mapped to the Arabidopsis reference transcriptome (Araport 11) using Salmon (Cheng et al., 2017; Patro et al., 2017). Gene-level TPM and count estimates were obtained using tximport (Soneson et al., 2016). Genes with total counts more than 10 in all samples were retained for differential gene expression analysis using DESeq2, and genes with a false discovery rate <0.05 were considered as differentially expressed genes (Love et al., 2014). Gene ontology enrichment analysis was performed using clusterProfiler (Yu et al., 2012). A Principal component analysis (PCA) was calculated, where the raw count of the genes were normalized using Variance Stabilizing Transformation (VST) from the DESeq2 R package. The variance of each gene was calculated across all sequenced samples (Col-0, gad2-1*, gad2-2 with three replicates for each genotype); genes with a higher variance were more differentially expressed across the samples, while genes with low variance were less differentially expressed (e.g. housekeeping gene). The variance of the genes was ranked with descending order, and the first 500 genes were used for the PCA analysis and plotted using the ggplot2 package in R.

Whole genome resequencing

To extract genomic DNA (gDNA), whole rosette leaves from 5-week-old plants were disrupted in liquid nitrogen and mixed with 750 µl DNA extraction buffer and 750 µl phenol:chloroform:iso-amyl alcohol (25:24:1). After centrifugation at maximal speed for 15 min, the upper fraction was treated with 10 μl autoclaved RNase (50 mg ml–1) at 4 °C for 3 d. After the sample had been assessed as RNA-free by agarose gel electrophoresis, the rest of the gDNA solution was mixed with 720 µl phenol:chloroform:iso-amyl alcohol (25:24:1) and centrifuged for 10 min at 25 °C.

RNase treatment was conducted in between two phenol:chloroform:isoamyl alcohol extractions to maximize DNA recovery (Healey et al., 2014). After the upper aqueous phase had been isolated and mixed with 504 µl of 3 M sodium acetate (pH 4.8):isopropanol (1:10), the DNA was allowed to precipitate overnight at −20 °C, washed with ice-cold 75% ethanol and then dissolved in 50 µl nuclease-free H2O. Validation of the insertion lines was performed as described by Narsai et al. (2017). Whole genome resequencing was performed to confirm whether only single insertions were present in the T-DNA insertion lines and that the insertions occurred at the target loci/genes. Genomic libraries were constructed with the Nextera DNA Library Preparation kit (Illumina) according to the manufacturer’s instructions. Libraries were enriched for large inserts by size selection with a 0.5× SPRI beads clean up. Sequencing in paired-end mode (75 bp) on an Illumina NextSeq 500 yielded 27–45 M reads per library. The quality of the raw data was assessed using FASTQC (Andrews, 2010) v.0.11.8 and multiQC (Ewels et al., 2016) v.1.8. Reads were trimmed with Trim Galore! (https://github.com/FelixKrueger/TrimGalore) v.0.6.3 and mapped with bowtie2 (Langmead and Salzberg, 2012) v. 2.4.1 against a modified version of the TAIR10 (Col-0 ecotype) genome comprising TAIR10, pROK2, and pAC161 T-DNA vector sequences as supplementary chromosomes (Ülker et al., 2008). In order to identify T-DNA insertion sites, read pairs with one read mapping to the TAIR10 genome and its mate mapping to the T-DNA vector were extracted from the mappings with the following awk commands: ‘$3 == “pROK2_T-DNA_only” && $7! = “=”’ (for T-DNA insertion lines in pROK2 vector) and ‘$3 == “pAC161_T-DNA” && $7! = “=”’ (for T-DNA insertion lines in pAC161 vector). Read mappings were visualized using JBrowse (Buels et al., 2016) v.1.16.9.

Plasmid construction and plant transformation

For guard cell-specific complementation, gad2-1* from the GABI_Kat collection was complemented with MPK12 driven by the guard cell promoter GC1. The coding sequences of MPK12 (At5G05440) and pGC1 [At1g22690; Yang et al. (2008)] were separately amplified from Arabidopsis thaliana using Phusion® High-Fidelity PCR Master Mix enzyme (Thermo Fisher) with primer sets that are listed in Supplementary Table S1. Both PCR fragments were combined with each other through overlap PCR. The PCR product was then inserted into a pCR®8/GW/TOPO® vector (Invitrogen) at EcoRI restriction sites, with the resulting pCR8-pGC1-MPK12 construct cloned into the T-DNA destination vector pMDC107 via an LR reaction using an LR Clonase™ II Enzyme mix (Invitrogen). The recombinant plasmid was introduced into competent Escherichia coli cells via heat-shock transformation. After the mutation had been verified by Sanger sequencing, the plasmids were transformed into the Agrobacterium tumefaciens strain AGL1 using the freeze-thaw method (Weigel and Glazebrook, 2006). For plant transformation, flowering gad2-1* mutant lines were floral-dipped into a suspension that contained successfully transformed Agrobacterium cells, in accordance to a method described by Zhang et al. (2006). For gas exchange measurements, T2 plants that had been selected on half-strength Murashige and Skoog medium containing hygromycin (25 µg ml−1) and were later transferred to hydroponics were used. The presence of MPK12 in these plants was confirmed via RT–PCR as described above, using the primers listed in Supplementary Table S1.

Statistical analysis

Unless noted, GraphPad Prism (Version 9.0.0 for Windows) was used for statistical analysis of the data presented in this study. Under the assumption of a Gaussian distribution, one-way Tukey’s multiple comparisons test or unpaired t-tests were applied on the data for mean comparisons between different genotypes. If the data were not normally distributed, a one-way ANOVA Dunn’s multiple comparison test was conducted. For statistical analysis of the ‘change in transpiration rate’ data, two-way ANOVA Tukey’s multiple comparison test was used. All data are presented as means ±SEM or box plot, as indicated.

Results

CO2 sensitivity is abolished in gad2-1*

To explore the effect of GABA manipulation on CO2-induced stomatal movement, gas exchange measurements were conducted in intact wild-type (Col-0) plants and the loss-of-function mutant line GABI_474E05 gad2-1*. Transpiration rates were monitored in a time-course experiment in response to varying CO2 concentrations (Fig. 1A). In wild-type plants, these measurements revealed a significant increase in leaf transpiration rates (0.272±0.022 mmol m−2 s−1 in total) in response to low CO2 (100 ppm, changed from atmospheric CO2 400 ppm) and a significant drop (P<0.0001) in transpiration rates upon treatment with high (800 ppm) CO2 (0.384±0.055 mmol m−2 s−1 in total; Fig. 1B). A subsequent reduction to 400 ppm CO2 (control CO2 condition) brought transpiration rates back to their initial level. In contrast, transpiration rates in gad2-1* increased by only 0.092±0.017 mmol m−2 s−1 and decreased by 0.098±0.016 mmol m−2 s−1 upon low/high CO2 conditions, respectively, revealing a reduced CO2 sensitivity. The time-resolved CO2 responses of mutant lines that are impaired in the CO2 signalling pathway (known as ca1/ca4 and mpk12-3) were recorded for comparison with the gad2-1* phenotype (Fig. 1A). Similarly, CO2 responses of both mutant lines deviated considerably from the wild-type CO2 response. The stomatal phenotype of mpk12-3 resembled the weakened CO2 response of gad2-1*, while the ca1/ca4 transpiration rates were even higher and slightly more variable (Fig. 1).

Fig. 1.

Fig. 1.

Open in a new tab

CO2-responsiveness in gad2-1* and CO2 signalling mutants is impaired. (A) CO2 response curves illustrate time courses of transpiration rates in response to control (400 ppm), low (100 ppm), and elevated (800 ppm) CO2 concentrations in intact 4–5-week-old plants of A. thaliana wild-type Col-0 (n=10), the GABA-deficient transgenic line gad2-1* (n=7), the CO2 signalling mutant lines mpk12-13 (n=3), and ca1/ca4 (n=5). (B) Steady-state transpiration rates in wild-type, gad2-1*, mpk12-13, and ca1/ca4 plants, based on the CO2 response curves in (A). Pooled data from two to three experimental series are shown and are plotted with box and whiskers (B): the box illustrates the median, the 25th and 75th percentiles, while the whiskers indicate the minimum and maximum values. Error bars represent ±SEM (A). Statistical differences were calculated using two-way ANOVA (B); *P<0.05, ***P<0.001, ****P<0.0001.

Higher order GAD mutants are CO2-sensitive

In order to clarify the effect of knocking out additional GAD genes in gad2-1*, the CO2 response of the quadruple mutant gad1/2/4/5 with T-DNA insertions in four GAD genes (GAD1, GAD2, GAD4, and GAD5; Supplementary Fig. S1) was examined. Note that gad1/2/4/5 harbours the same T-DNA insertion within the GAD2 gene as gad2-1* (GABI_474E05; Deng et al., 2020), since it has been generated through crossing of gad single mutants, including gad2-1* from the GABI-Kat seed collection.

Transpiration rates in gad1/2/4/5 changed in response to alterations in atmospheric CO2 concentrations, similar to that observed in wild-type plants (Fig. 2A). When CO2 concentration was reduced from 400 ppm to a low of 100 ppm, the transpiration rates of both wild-type and gad1/2/4/5 plants increased; whereas raising the CO2 concentration from 100 ppm to 800 ppm their transpiration rates were both decreased (by 0.462±0.092 mmol m–2 s–1 in gad1/2/4/5 and by 0.384±0.055 mmol m–2 s–1 in the wild type; Fig. 2B). This response differed from that of gad2-1*, which exhibited only a subtle reduction in transpiration rate – by 0.092±0.017 mmol m−2 s−1 – when the CO2 concentration was reduced from 400 ppm to 100 ppm (Fig. 2B). Also, stomatal opening kinetics (Kopen) were comparable between wild-type plants and the quadruple mutant in response to different CO2 concentrations; interestingly, gad1/2/4/5 showed significantly faster stomatal closure than wild-type plants upon high (800 ppm) CO2 treatment (Fig. 2A-C). It is unclear whether the accelerated stomatal closure (Kclose) observed in the gad1/2/4/5 mutant is linked to higher reactive oxygen species (ROS) accumulation in the guard cells of this mutant, compared with wild-type plants (Xu et al., 2024), especially given that ROS act as stomatal closure signals in response to high CO2 (Shi et al., 2015; He et al., 2020). Here, the CO2 treatments did not affect the transpiration rates of the gad2-1* mutant (Figs 1B, 2B); furthermore, the stomatal response to different CO2 concentrations was extremely slow in gad2-1* (e.g. Kopen ≈ 160 min upon a decrease in CO2 concentration from 400 ppm to 100 ppm) or its speed could not be calculated (e.g. Kclose not calculable upon an increase in CO2 concentration from 100 ppm to 800 ppm). Consequently, the stomatal kinetics of gad2-1* were not used for comparisons with those of other genotypes within this study. The transpiration rates obtained were in accordance with the stomatal aperture measurement in all three lines (Fig. 2A, D). The stomatal apertures were reduced in gad1/2/4/5 (31.44%) and wild-type (24.93%) plants in response to 800 ppm CO2; in contrast, the stomatal apertures gad2-1* were only slightly reduced (9.57%) (Fig. 2E). It is noted that these three different genotypes have been previously found to have comparable stomatal densities on leaf surfaces (Xu et al., 2024).

Fig. 2.

Fig. 2.

Open in a new tab

CO2 phenotype of gad1/2/4/5 contrasts with CO2 insensitivity of gad2-1*. (A) Time courses of transpiration rates in response to control (400 ppm), reduced (100 ppm), and elevated (800 ppm) CO2 concentrations in intact 4–5-week-old A. thaliana wild-type (Col-0; n=9), GABA-deficient gad2-1* (n=10), and gad1/2/4/5 (n=7) plants. (B) On the basis of the data presented in (A), changes in transpiration rates during specific time periods were determined and are illustrated. For calculating these changes, transpiration rates at earlier time points were subtracted from transpiration rates at later points in time. Time points are numbered and denoted by small black arrows in (A). (C) Half times of stomatal opening and closing in response to each CO2 concentration was determined on the basis of the time-lapse transpiration rates illustrated in (A). The calculation for the stomatal half-response times is based on the Michaelis-Menten Model, where Km (Kopen and Kclose) is the Michaelis-Menten constant and equals the half maximum velocity of the respective transpiration response. (D) Stomatal pore sizes on abaxial leaf surfaces in Col-0 (n=1260), gad2-1* (n=908), and gad1/2/4/5 (n=1176) upon exposure to 800 ppm CO2 are shown. (E) Percentages of stomatal aperture reduction in response to 800 ppm CO2 are based on the stomatal aperture data illustrated in (D). Pooled data from at least two independent experiments are shown. Data were plotted with box and whiskers (D): the box illustrates the median, the 25th and 75th percentiles, while the whiskers indicate the minimum and maximum values. Error bars represent ±SEM (A–C, E). Statistical differences were calculated using two-way ANOVA (B), Student’s t test (C), or one-way ANOVA (D, E); *P<0.05, **P<0.01, ****P<0.0001.

Allelic mutant lines gad2-1* and gad2-2 reveal opposing CO2 responses

To investigate whether the decreased CO2 sensitivity in gad2-1* is linked to the knockout of GAD2, we evaluated the CO2 response of another allelic mutant of GAD2, known as gad2-2. Like gad2-1*, this mutant line has been generated by a T-DNA insertion in GAD2 but from a different seed stock, resulting in defective leaf GABA production (Xu et al., 2021a). While gad2-1* has been generated by a T-DNA insertion in exon 6, gad2-2 contained a T-DNA insertion in the second intron of GAD2 (Supplementary Fig. S1) (Xu et al., 2021a).

As in gad2-1* (P=0.0268), transpiration rates (Fig. 3A) were significantly higher (P=0.0043) in gad2-2 than in the wild type at 400 ppm CO2 (control condition that is close to ambient CO2). However, gad2-2 revealed a wild-type-like response in terms of both CO2-dependent changes in transpiration rates and stomatal kinetics (Fig. 3A–C). Transpiration rates of gad2-2 decreased by 0.429±0.030 mmol m−2 s−1 when CO2 was changed from 100 to 800 ppm compared with 0.055±0.023 mmol m−2 s−1 in gad2-1* and 0.427±0.029 mmol m−2 s−1 in the wild type (Fig. 3B). Transpiration rates decreased by 1.067±0.068 mmol m−2 s−1 in gad2-2, and by 0.967±0.078 mmol m−2 s−1 in the wild type, whereas transpiration rates decreased only by 0.058±0.014 mmol m−2 s−1 in gad2-1* (Fig. 3B). Stomatal aperture measurements were performed in gad2-2 concomitantly with Col-0, gad2-1*, and gad1/2/4/5 plants. To facilitate the comparison, stomatal aperture for Col-0 and gad2-1* has been replotted in Fig. 3D, E, showing the effect of 800 ppm CO2 on stomatal pores of gad2-2. The measurements revealed a larger CO2-stimulated (800 ppm) reduction in stomatal apertures in gad2-2 (36.09%) and the wild type (24.93%) and only a minor reduction in gad2-1* (9.57%; Fig. 3E).

Fig. 3.

Fig. 3.

Open in a new tab

Allelic GAD2 loss-of-function mutants reveal contrary stomatal CO2 responses. (A) Time courses of transpiration rates in response to control (400 ppm), elevated (800 ppm), and reduced (100 ppm) CO2 concentrations in intact 4–5-week-old A. thaliana wild-type (Col-0; n=7), GABA-deficient gad2-1* (n=6), and gad2-2 (n=7) plants. (B) On the basis of the data presented in (A), changes in transpiration rates during specific time periods were determined and are illustrated. For calculating these changes, transpiration rates at earlier time points were subtracted from transpiration rates at later points in time. Time points are numbered and denoted by small black arrows in (A). (C) Half times of stomatal opening and closing in response to each CO2 concentration were determined on the basis of the time-lapse transpiration rates illustrated in (A). The calculation for the stomatal half-response times is based on the Michaelis-Menten model, where Km (Kopen and Kclose) is the Michaelis-Menten constant and equals the half maximum velocity of the respective transpiration response. (D) Stomatal pore sizes on abaxial leaf surfaces in Col-0 (n=1260), gad2-1* (n=908), and gad2-2 (n=773) upon exposure to 800 ppm CO2 are shown. (E) Percentages of stomatal aperture reduction in response to 800 ppm CO2 are based on the stomatal aperture data illustrated in (D). Pooled data from at least two independent experiments are shown. Data were plotted with box and whiskers (D): the box illustrates the median, the 25th and 75th percentiles, while the whiskers indicate the minimum and maximum values. Error bars represent ±SEM (A–C, E). Statistical differences were calculated using two-way ANOVA (B), Student’s t tests (C), or one-way ANOVA (D, E); **P<0.01, ****P<0.0001.

Whole genomic sequencing analyses reveal a genomic deletion in gad2-1*

The mutant lines gad2-1* and gad2-2 had been expected to carry a single T-DNA at two different positions of the same gene (GAD2; AT1G65960). Nonetheless, they were found to display contrasting CO2 responses. Moreover, the presence of additional T-DNA insertions within the GAD genes correlated with the renewed CO2 sensitivity in gad1/2/4/5, despite the gad2-1* background.

To decipher the genetic cause, whole genome sequencing (WGS) was performed in the wild type (Col-0) and the GABA-deficient mutant lines gad2-1*, gad2-2 and gad1/2/4/5. For the identification of T-DNA insertion sites in the genome of these lines, discordant reads pairs were extracted, where one mate aligned to the TAIR10 reference genome, and the other mate aligned to the respective T-DNA transformation vector sequence (pROK2 or pAC16). With this approach, all T-DNA insertion sites were confirmed.

Specifically, the resulting data identified an expected T-DNA insertion in the sixth exon of GAD2 in gad2-1* and one in the second intron of GAD2 in gad2-2 (Supplementary Fig. S1). At the same time, genomic profiling of the gad2-1* genome also unveiled a non-specific deletion of 4750 bp in the MPK12 (AT2G46070) region (18 945 445 bp to 18 950 195 bp) on chromosome 2 of this mutant line (Fig. 4). This deletion did not only comprise full-length MPK12 but also full-length BYPASS2 (BPS2; AT2G46080), which is adjacent to MPK12. In all the lines tested, gad2-1* was the only one in which these two genes had been removed. The transgenic mutant line gad2-2 contained only the target-site mutation.

Fig. 4.

Fig. 4.

Open in a new tab

Whole genome sequencing (WGS) uncovers an additional mutation in gad2-1*. Genome browser view of a non-T-DNA insertion deletion in the MPK12 (AT2G46070) region (from 18 945 445 bp to 18 950 195 bp) on chromosome 2 of the transgenic line gad2-1* as revealed by WGS (Illumina). This deletion was absent in the wild type (Col-0), gad2-2, and gad1/2/4/5, and included another gene (BPS2; AT2G46080) adjacent to MPK12.

As expected, the quadruple mutant gad1/2/4/5 was found to harbour T-DNA insertions in the GAD1 (AT5G17330), GAD2 (AT1G65960), GAD4 (AT2G02010) and GAD5 (AT3G17760) region (Supplementary Fig. S2A−D). Note that in the GAD5 region only one flanking end of the T-DNA insertion was recovered and sequenced. The sequencing data also confirmed that the T-DNA insertion in GAD2 was the same in the quadruple mutant and gad2-1* (Supplementary Fig. S2A). All these specified mutations were absent in the background genome of wild-type ecotype Columbia 0 (Col-0).

As the Arabidopsis ecotype Cape Verde Islands (Cvi‐0) also displays reduced CO2 responsiveness and more open stomata than Col-0 due to a mutation in MPK12 (Jakobson et al., 2016), the gad2-1* genome was examined for a potential Cvi contamination by mapping the unmapped reads to the Cvi genome (including T-DNA vector sequences). No evidence for accidental Cvi contamination was found in the gad2-1* genome, although some reads could still map (with many SNPs and Indels) to the Cvi genome. However, this was also the case for all the other lines (including Col-0), which did not reveal the Cvi phenotype (Supplementary Fig. S3).

Global transcriptomic analysis indicates the absence of MPK12 and BPS2 transcription in gad2-1*

To further explore the impact of the 4750 bp fragment deletion in gad2-1* mutant, the global transcriptional profiles in the leaves of wild-type (Col-0), gad2-1* and gad1/2/4/5 plants using RNA–seq and microarray analysis were investigated. Differential expression analysis was conducted using Deseq2 (Love et al., 2014), where lowly expressed genes (total count number less than 10) were filtered out and genes with P adj. <0.05 were considered as differentially expressed genes using the Wald test. A PCA separated the three different lines in accordance with their transcription profiles (Supplementary Fig. S4).

Both gad2-1* and gad2-2 possess many genes that were differentially expressed relative to the wild type (Supplementary Fig. S5). Pairwise comparison of differentially expressed genes (DEGs) identified 1536 DEGs between the wild type and gad2-1* (643 up- and 893 down-regulated), and 892 DEGs between the wild type and gad2-2 (263 up- and 629 down-regulated). These genes were mostly enriched for functions related to hypoxia and oxygen responses (Supplementary Fig. S6). Numerous genes were also differentially expressed between gad2-1* and gad2-2. The mutant line gad2-1* had more DEGs than gad2-2 in relation to the wild type (false discovery rate <0.05; Supplementary Fig. S5A). However, only three genes were conversely expressed between the two mutants (Supplementary Fig. S5B, C). The allele-specific DEGs were mainly functionally enriched for glucosinolate/glycoside metabolism (Supplementary Fig. S6).

Next, some CO2-sensing related genes were examined more closely, and it was found that the GAD2 expression level was significantly lower (P<0.0001) in both gad2-1* and gad2-2 in comparison with the wild type. However, it was also discovered that GAD2 was less expressed in gad2-2 than in gad2-1*. Furthermore, the data revealed that the adjacent gene to GAD2, known as THIOREDOXIN-DEPENDENT PEROXIDASE 2 (TPX2; AT1G65970), was strongly up-regulated in gad2-1* in relation to the wild type and gad2-2, while no considerable difference was detected for TPX1 (AT1G65980; neighbour gene of TPX2) expression. On the other hand, MPK12 and BPS2 expression was not detectable in gad2-1*, but present within gad2-2 (Fig. 5).

Fig. 5.

Fig. 5.

Open in a new tab

List of genes related to stomatal regulation in GABA-deficient mutant lines by RNA–seq analysis. Average log2(transcripts per million [TPM]+1) values of differentially expressed genes — GLUTAMATE DECARBOXYLASE 2 (GAD2), MITOGEN-ACTIVATED PROTEIN KINASE 12 (MPK12), BYPASS 2 (BPS2), THIOREDOXIN-DEPENDENT PEROXIDASE 2 (TPX2), SLAC1 HOMOLOGUE 3 (SLAH3), ALUMINIUM-ACTIVATED MALATE TRANSPORTER 12/ QUICKLY ACTIVATING ANION CHANNEL 1 (ALMT12/QUAC1), PYRABACTIN RESISTANCE-LIKE 4 (PYL4) and PYL5, and non-differentially expressed genes which are, inter alia, involved in stomatal closure responses, including genes that are related to high CO2 signalling in guard cells [CARBONIC ANHYDRASE 1 (CA1), CA4, MITOGEN-ACTIVATED PROTEIN KINASE 4 (MPK4), GUARD CELL HYDROGEN PEROXIDE-RESISTANT 1 (GHR1) and HIGH TEMPERATURE 1 (HT1)] in leaves of the wild type (Col-0), gad2-1*, and gad2-2, based on RNA–seq analysis. Genes with P adj. <0.05 were considered as differentially expressed using Wald test; P adj. was calculated using the Benjamini-Hochberg method. Individual data points were plotted with the median and minimum and maximum values; ****P<0.0001, ***P<0.001, **P<0.01 and *P<0.05.

These findings are consistent with the gene expression data from our microarray examination of leaf material of Col-0, gad2-1*, and gad1/2/4/5. The resulting data, presented as fold changes in Table 1, revealed the down-regulation of GAD2 expression and concurrent up-regulation of TPX2 expression in gad2-1* and gad1/2/4/5. Moreover, a significant down-regulation of MPK12 and BPS2 was detected in gad2-1* in comparison to the wild type and gad1/2/4/5.

Table 1.

Microarray analysis in GABA-deficient mutant lines.

Fc Col-0/gad2-1* Fc Col-0/gad1/2/4/5 Fc gad2-1*/gad1/2/4/5
AT1G65960 GAD2 15.08**** 11.91**** –1.27
 P-value 8.6805E-11 1.54454E-09 0.757157463
AT2G46070 MPK12 21.05**** 1.05 20.01****
 P-value 2.20746E-17 0.955018824 3.11481E-17
AT2G46080 BPS2 189.98**** 1.03 –183.7****
 P-value 6.32731E-12 0.997708397 7.1887E-12
AT1G65970 TPX2 –15.19**** –22.71**** –1.49
 P-value 5.3474E-08 5.23735E-09 0.683301215
Open in a new tab

****represents a significant difference (P<0.0001).

Fold changes (fc) of differentially expressed genes in the wild type (Col-0), gad2-1* and gad1/2/4/5, based on a microarray experiment. Positive and negative fold changes indicate whether genes are up- or down-regulated. GAD2 (AT1G65960) is down-regulated in gad2-1* and gad1/2/4/5 compared with the wild type, while MPK12 (AT2G46070) and its neighbouring gene BPS2 (AT2G46080) were significantly lower in gad2-1* in comparison to the wild type and gad1/2/4/5. Gene expression of a neighbouring gene of GAD2, known as TPX2 (AT1G65970), was highly up-regulated in gad2-1* and gad1/2/4/5. No significant differences in PFK1 (AT4G29220) expression levels were identified. Only fold-changes with adjusted P-values ≤0.05 are shown.

We investigated whether the mutations in GAD2, MPK12, and BPS2 affected the expression levels of other genes that encode key elements of stomatal regulation (Fig. 5). On the one hand, similar expression patterns were detected for GAD2 and SLAC1 Homologue 3 (SLAH3) in gad2-1* and gad2-2. Moreover, the microarray experiment did not find differences in SLAH3 expression between gad2-1* and gad1/2/4/5. On the other hand, ALMT12/QUAC1 and SLAC1 were found to be only misexpressed in gad2-1* but not in gad2-2. While ALMT12/QUAC1 and SLAC1 regulate both CO2- and ABA-induced stomatal closure, the anion channel SLAH3 has only been identified as an ABA signalling component (Geiger et al., 2011). In contrast to stomatal closure-related ALMT12/QUAC1, the expression of ALMT9, which is required for stomatal opening, was not found to be differentially expressed between two gad2 mutant alleles. As the loss of ALMT9 did not impair the stomatal response to CO2 (Supplementary Fig. S7), our data suggest that GABA does not contribute to stomatal CO2 sensing through the regulation of ALMT9.

We additionally examined the gene expression levels of (further) genes that are proposed to be involved in CO2 signalling processes, such as HIGH TEMPERATURE 1 (HT1, Hashimoto et al., 2006), CARBONIC ANHYDRASE 1 (CA1), CA4 (Zhou et al., 2020), MPK4 (Tõldsepp et al., 2018), GUARD CELL HYDROGEN PEROXIDE-RESISTANT 1 (GHR1, Sierla et al., 2018), PYRABACTIN RESISTANCE-LIKE 4 (PYL4) and PYL5 (Dittrich et al., 2019). Among all (potential) CO2 signalling-related genes examined in this study, PYL4 and PYL5 were the only genes that were differentially expressed according to P adj.<0.05 (Wald test). The ABA receptors PYL4 and PYL5 had been demonstrated to be crucial for stomatal CO2 sensing (Dittrich et al., 2019), however, a subsequent report have questioned this finding (Zhang et al., 2020), so examining their role with respect to GABA and potential interaction with ABA signalling will be interesting (Xu et al., 2021a, 2021b). HT1 might appear on first sight to be down-regulated in gad2-1*, but it should be noted that the corresponding P-value did not meet the pre-specified level of statistical significance (P adj. <0.05; Fig. 5).

Taken together, the expression of genes related to CO2 signalling was not found to be differentially expressed between the different genotypes. Only genes that are also relevant for other signalling cascades, like the ABA signalling pathway, appear to be differentially expressed between the different lines. Thereby, the question arises if the differences in transcript abundance are based on differences in the detected expression levels of TPX2, MPK12 and/or GAD2. Hence, we compared the expression patterns of corresponding genes across the three genotypes (wild type, gad2-1* and gad2-2; Fig. 5). SLAH3 expression could possibly be dependent on GAD2, as both SLAH3 and GAD2 were found to be down-regulated in gad2-1* and gad2-2. On the other hand, the up-regulation of SLAC1 and ALMT12/QUAC1 might correlate with MPK12 down-regulation in gad2-1*. These gene expression levels are merely clues that require follow up experiments for verification (Fig. 5). This is especially relevant, as these levels were not confirmed by the microarray analysis in gad2-1* and gad1/2/4/5.

CO2 sensitivity in gad2-1* is restored by guard cell-specific MPK12 complementation

To ascertain whether CO2 responsiveness can be recovered in gad2-1* by driving MPK12 expression in its guard cells, we transformed gad2-1* with full-length MPK12 fused to the guard cell-promoter GC1. The transformation success was confirmed via RT–PCR amplification of MPK12 in corresponding plants (gad2-1*/GC1::MPK12;Supplementary Fig. S8).

Gas exchange measurements were performed in gad2-1*/GC1::MPK12, which were measured concomitantly with Col-0, gad2-1* and gad2-2 plants. For ease of comparison, the Col-0 and gad2-1* data have been replotted in Fig. 6, which illustrates the CO2 response of gad2-1*/GC1::MPK12. Clearly, gad2-1*/GC1::MPK12 revealed an increased CO2 sensitivity compared with gad2-1*, as it adjusted transpiration rates in accordance to altered CO2 concentrations (Fig. 6A). It responded to both low (100 ppm) and high (800 ppm) CO2. Not only was the speed of stomatal responses to different CO2 concentrations nearly back to normal levels, but also the magnitude of change in transpiration rates was significantly higher (t2-t1: P=0.0419; t4-t3: P<0.0001; t6-t5: P<0.001) in gad2-1*/GC1::MPK12 than in gad2-1* (Fig. 6B, C).

Fig. 6.

Fig. 6.

Open in a new tab

CO2 responsiveness was recovered by MPK12 complementation in guard cells of gad2-1*. (A) Time courses of transpiration rates in response to control (400 ppm), reduced (100 ppm) and elevated (800 ppm) CO2 concentrations in intact 4–5-week-old A. thaliana wild-type plants (Col-0; n=7), the transgenic lines gad2-1* (n=6) and gad2-1*/GC1::MPK12 (n=6). Of note, the same Col-0 and gad2-1* data are presented here as in Fig. 3 since it was acquired at the same time. (B) On the basis of the data presented in (A), changes in transpiration rates during specific time periods were determined and are illustrated. For calculating these changes, transpiration rates at earlier time points were subtracted from transpiration rates at later points in time. Time points are numbered and denoted by small black arrows in (A). (C) Half times of stomatal opening and closing in response to each CO2 concentration was determined on the basis of the time-lapse transpiration rates illustrated in (A). The calculation for the stomatal half-response times is based on the Michaelis-Menten model, where Km (Kopen and Kclose) is the Michaelis-Menten constant and equals the half maximum velocity of the respective transpiration response. Error bars in all diagrams represent ±SEM. Statistical differences were calculated using two-way ANOVA (B) or Student’s t tests (C); **P<0.01, ***P<0.001, ****P<0.0001.

However, the CO2 response of gad2-1*/GC1::MPK12 was not as extensive as that of gad2-2 (Supplementary Fig. S9). Particularly, high CO2-dependent closure was greater in gad2-2 compared with the MPK12 complementation line. This was true for the speed as well as the extent of change in transpiration rates. Overall, transpiration rates in both lines were much greater than in the wild type (Supplementary Fig. S10).

Discussion

Only gad2-1* shows an impaired stomatal CO2 response

GABA has been previously demonstrated to act as a signal in guard cells for the regulation of transpiration. However, the GABA signalling pathway is largely unexplored. Besides, its production from glutamate releases CO2 molecules in the cytosol. Therefore, we explored a possible connection between GABA and CO2 signalling using multiple GABA-deficient gad mutants in several physiological and genetic studies. We analysed the mutant lines gad2-1* (from the GABI-Kat mutant collection) and gad2-2 (from the SALK mutant collection), which harbour T-DNA insertions at two different sites in GAD2, resulting in GABA depletion in leaves. In particular, we note that gad2-1* has been used for stomatal assays in other stomata-related studies (Scholz et al., 2015; Mekonnen et al., 2016; Deng et al., 2020; Xu et al., 2021a).

As the aim was to clarify if GABA plays a role in CO2-dependent stomatal responses, we examined gad2-1* and gad2-2 for CO2 sensitivity by monitoring their transpiration rates in response to alternating CO2 concentrations (high and low CO2). Under standard CO2 conditions (400 ppm CO2), the data indicated a significant increase in transpiration rates and stomatal pores in both mutant lines, which matches the observations of a previous report (Xu et al., 2021a). It was proposed that this is due to direct or indirect deregulation of the vacuolar anion channel ALMT9, and as a consequence, stomata are more open, and these plants have a drastically reduced WUE (Xu et al., 2021b).

However, exposure to drastically increased or decreased (800 ppm or 100 ppm, respectively) CO2 concentrations unveiled significant differences in the stomatal phenotype between both allelic mutant lines (Fig. 3). In wild-type plants, transpiration rates decreased in response to elevated CO2 (800 ppm), whereas they increased under low CO2 (100 ppm); this is a common behaviour of plants that serves to increase their WUE (Lawson and Blatt, 2014). In contrast, transpiration rates in gad2-1* changed only subtly upon different CO2 concentrations, which suggests a reduced CO2 sensitivity in gad2-1*. Surprisingly, both gad2-2 and gad1/2/4/5 exhibited CO2 responses similar to those of the wild-type plants in terms of the changes in transpiration rates and stomatal opening speed, except for the faster CO2-induced stomatal closure observed in the gad1/2/4/5 mutant (Fig. 2). The faster stomatal closure in gad1/2/4/5 by CO2 raises the question about its potential association with higher ROS accumulation in the guard cells, as ROS are known to mediate stomatal closure in response to elevated CO2 (Figs 2, 3) (Shi et al., 2015; He et al., 2020; Xu et al., 2024). The adverse stomatal behaviour of gad2-1* and other gad mutants (gad2-2 and gad1/2/4/5) was backed up by stomatal aperture measurements, which revealed impaired stomatal closure to 800 ppm CO2 in gad2-1* (Figs 2D, 3D).

The loss of stomatal CO2 sensitivity is due to the deletion of MPK12 in gad2-1*

Due to the opposing phenotypes of gad2-1* and other gad mutants, we suspected more than one T-DNA insertion within its genome. To examine if additional mutations existed, we conducted a comprehensive transcriptomic and genomic analysis. Notably, the genetic studies revealed clear genotype-phenotype correlations in these lines. As expected, the microarray and RNA-seq analyses detected the down-regulation of GAD2 in both gad2-1* and gad2-2. Thereby, GAD2 expression was found to be even lower in gad2-2 than gad2-1*, probably due to partial (non-functional) GAD2 transcripts present in the gad2-1* mutant line (Mekonnen, 2013). Intriguingly, both RNA-seq and microarray analysis revealed the down-regulation of MPK12 in gad2-1* compared with the wild-type, gad2-2, and gad1/2/4/5 lines used in this study (Table 1; Fig. 5).

The outcome of the WGS analysis provided an explanation for the specific MPK12 down-regulation in gad2-1*, compared with the other genotypes. Apparently, MPK12 and its neighbouring gene BPS2 are completely removed from the gad2-1* genome (4750 bp in total). According to the TAIR database, BPS2 encodes a protein linked to BPS1, which was found to be involved in the formation of a root-synthesized mobile signal that induces a growth arrest in young Arabidopsis leaves (Van Norman et al., 2011). Silencing of MPK12 is associated with impaired stomatal closure and altered CO2 responsiveness as shown by the loss-of-function mutant mpk12-3 in the present and another, independent study (Jakobson et al., 2016). This occurs because MPK12 suppresses the activity of a negative regulator of high CO2 signalling, known as HT1 (Hõrak et al., 2016; Tõldsepp et al., 2018; Takahashi et al., 2022). When MPK12 is inactive or absent as in gad2-1* or mpk12-3, HT1 constitutively inhibits high CO2-induced stomatal closure, leading to enlarged stomatal apertures (Hõrak et al., 2016; Jakobson et al., 2016). In accordance with this, guard cell-specific MPK12 complementation rescued gad2-1* insensitivity to CO2 (Fig. 6). The stomatal CO2 responses of the MPK12 complementation line were slightly weaker than in the case of gad2-2, likely linked to MPK12 expression not being mediated by its native promoter and only being present in the guard cells of gad2-1*/GC1::MPK12 (Supplementary Fig. S9). The link between CO2 insensitivity and the MPK12 mutation was originally identified in an Arabidopsis accession that originates from the Cape Verde Islands, known as Cvi-0 (Jakobson et al., 2016). This accession appears to be endemic to these islands and it is not known whether its MPK12 mutation resulted from natural selection or genetic drift (Des Marais et al., 2014). The geographic conditions, isolation, and fairly constant vapour-pressure deficit (VPD), as well as moderate temperatures on the Cape Verde Islands have permitted the existence of the accession (Des Marais et al., 2014). We examined the genome of gad2-1* for a potential Cvi-0 contamination, but this was not detected. Interestingly, Jakobson et al. (2016) detected a sequence deletion of a comparable base pair size (4772 bp) in two of their GABI-Kat lines. Both MPK12 and BPS2 had also been eliminated, leading to enlarged stomata and reduced CO2 responsiveness. In contrast, corresponding allelic mutant lines from different mutant collections (SALK and SAIL lines) did not reveal this specific deletion, neither did they display any stomatal abnormalities (Jakobson et al., 2016). It is conceivable that these GABI-Kat lines, including gad2-1*, have been generated in Col-0 background plants that already contained this specific gene deletion. In our study, a GABI-Kat line (almt6-1) and a transgenic line from the INRA-Versailles mutant collection (almt6-2) were examined for CO2-sensitive stomatal responses and MPK12 expression. The data unveiled that both lines express MPK12 and are capable of adjusting stomatal aperture sizes to changing CO2 (Supplementary Fig. S11), suggesting that the specified deletion is not universal to all mutant lines from the GABI-Kat catalogue.

Random DNA removal does not seem rare, as the genome of angiosperms has been found to be very plastic (Devos et al., 2002; Ma and Bennetzen, 2004; Vitte and Bennetzen, 2006; Bennetzen and Wang, 2018). Illegitimate recombination is estimated to be the most common cause for gene loss in Arabidopsis (probably to a greater extent than unequal homologous recombination), and could be a possible explanation for the gene loss in gad2-1* and the other GABI-Kat mutant lines (Ehrlich et al., 1993; Devos et al., 2002).

Surprisingly, the quadruple mutant gad1/2/4/5 was found to reveal a robust CO2 response in contrast to gad2-1*, although it had been produced by crossing gad2-1* with other gad lines. However, our genetic study has uncovered that only gad2-1* lacks MPK12 and BPS2, whereas gad1/2/4/5 has gained both genes. A plausible explanation for this genetic difference is that the gene deletion was lost in the quadruple mutant during the crossing processes.

Both complementation of gad2-1* by MPK12, and the CO2-sensitive phenotypes of gad2-2 and gad1/2/4/5, demonstrate that GABA is not required for stomatal responses to high or low CO2. The enhanced stomatal opening response of gad2-2 to low CO2 conditions seems to be independent of ALMT9, since measurements in almt9-1 revealed a wild type-like CO2 response, indicating that the lack of ALMT9 does not affect CO2 sensitivity. Interestingly, Dellero et al. (2021) discovered large increased concentrations of GABA in Arabidopsis plants that had been subjected to low CO2 conditions for 4 h. Thus, GABA might play certain roles under these conditions. However, it appears that these do not include GABA regulation of low CO2-mediated stomatal movements.

GABA does not contribute to stomatal response to CO2

In summary, we could clearly rule out that GABA deficiency results in decreased CO2 sensitivity by attributing the reduced CO2 responsiveness in gad2-1* to a second, non-target mutation in MPK12. Furthermore, we provide proof that guard cell responses to alterations in atmospheric CO2 are not compromised by loss of ALMT9 channels. Our research has shown that it is crucial to examine multiple independent allelic mutant lines and to combine physiological data with genetic analyses in order to avoid misinterpretation of physiological data.

Supplementary data

The following supplementary data are available at JXB online.

Table S1. Primers used in this study.

Fig. S1. Schematic map of T-DNA insertional sites in the genomes of gad2-1*, gad2-2 and gad1/2/4/5.

Fig. S2. Confirmed T-DNA insertional sites in the genomes of Col-0 and GABA-deficient mutants.

Fig. S3. Examining mutant genomes for Cvi contamination.

Fig. S4. Principal Component Analysis (PCA) of RNA-seq data.

Fig. S5. Differentially expressed gene (DEG) analysis of RNA-seq data.

Fig. S6. Comparison of functional enrichment results from RNA-seq data.

Fig. S7. Time-resolved patterns of transpiration rates in response to different CO2 concentrations in almt9-1.

Fig. S8. Verification of MPK12 expression in gad2-1*/GC1::MPK12.

Fig. S9. CO2 responsiveness is higher in gad2-2 than in gad2-1*/GC1::MPK12.

Fig. S10. Transpiration rates are increased in both gad2-2 and gad2-1*/GC1::MPK12.

Fig. S11. Time-resolved patterns of transpiration rates in response to different CO2 concentrations in almt6 mutant lines.

erae168_suppl_Supplementary_Table_S1_Figures_S1-S11
erae168_suppl_supplementary_table_s1_figures_s1-s11.pdf (1.8MB, pdf)

Acknowledgements

We thank Everard Edwards and Annette Betts from CSIRO Agriculture & Food (Waite Campus, Adelaide) for the kind provision of a LI-6400XT Portable Photosynthesis System, Honghong Hu from Huazhong Agricultural University for providing ca1/ca4 seeds. We appreciate the help of Rosalie Kenyon at ACRF Cancer Genomics Facility (SA Pathology and University of South Australia) for conducting the RNA Labchip assay. We also thank Kylie Neumann and Sandy Khor, Department of Plant Science (University of Adelaide), for the kind provision of Invitrogen Qubit fluorometer.

Contributor Information

Adriane Piechatzek, Plant Transport and Signalling Lab, ARC Centre of Excellence in Plant Energy Biology, Waite Research Institute, Glen Osmond, SA 5064, Australia; School of Agriculture, Food and Wine, Waite Research Precinct, University of Adelaide, Glen Osmond, SA 5064, Australia.

Xueying Feng, Plant Transport and Signalling Lab, ARC Centre of Excellence in Plant Energy Biology, Waite Research Institute, Glen Osmond, SA 5064, Australia; School of Agriculture, Food and Wine, Waite Research Precinct, University of Adelaide, Glen Osmond, SA 5064, Australia.

Na Sai, Plant Transport and Signalling Lab, ARC Centre of Excellence in Plant Energy Biology, Waite Research Institute, Glen Osmond, SA 5064, Australia; School of Agriculture, Food and Wine, Waite Research Precinct, University of Adelaide, Glen Osmond, SA 5064, Australia.

Changyu Yi, La Trobe Institute for Agriculture and Food, AgriBio Building, La Trobe University, Bundoora, VIC 3086, Australia.

Bhavna Hurgobin, La Trobe Institute for Agriculture and Food, AgriBio Building, La Trobe University, Bundoora, VIC 3086, Australia.

Mathew Lewsey, La Trobe Institute for Agriculture and Food, AgriBio Building, La Trobe University, Bundoora, VIC 3086, Australia; Australian Research Council Research Hub for Medicinal Agriculture, AgriBio Building, La Trobe University, Bundoora, VIC 3086, Australia; ARC Centre of Excellence in Plants for Space, AgriBio Building, La Trobe University, Bundoora, VIC 3086, Australia.

Johannes Herrmann, Institute for Molecular Plant Physiology and Biophysics, University of Würzburg, Würzburg 97078, Germany.

Marcus Dittrich, Department of Bioinformatics, University of Würzburg, Würzburg 97078, Germany; Institute of Human Genetics, University of Würzburg, Würzburg 97074, Germany.

Peter Ache, Institute for Molecular Plant Physiology and Biophysics, University of Würzburg, Würzburg 97078, Germany.

Tobias Müller, Department of Bioinformatics, University of Würzburg, Würzburg 97078, Germany.

Johannes Kromdijk, Department of Plant Sciences, University of Cambridge, Downing St., Cambridge, CB2 3EA, UK.

Rainer Hedrich, Institute for Molecular Plant Physiology and Biophysics, University of Würzburg, Würzburg 97078, Germany.

Bo Xu, Plant Transport and Signalling Lab, ARC Centre of Excellence in Plant Energy Biology, Waite Research Institute, Glen Osmond, SA 5064, Australia; School of Agriculture, Food and Wine, Waite Research Precinct, University of Adelaide, Glen Osmond, SA 5064, Australia; ARC Centre of Excellence in Plants for Space, School of Agriculture, Food and Wine & Waite Research Institute, Glen Osmond, SA 5064, Australia.

Matthew Gilliham, Plant Transport and Signalling Lab, ARC Centre of Excellence in Plant Energy Biology, Waite Research Institute, Glen Osmond, SA 5064, Australia; School of Agriculture, Food and Wine, Waite Research Precinct, University of Adelaide, Glen Osmond, SA 5064, Australia; ARC Centre of Excellence in Plants for Space, School of Agriculture, Food and Wine & Waite Research Institute, Glen Osmond, SA 5064, Australia.

Tracy Lawson, University of Essex, UK.

Author contributions

MG, BX, JK, RH, and AP designed the research; AP conducted the bulk of the experiments, with major contributions by BX who was involved in building the GC1::MPK12 vector construct for gad2-1* complementation, performing the Agrobacterium transformation, floral dipping and screening of the complementation lines; XF contributed to the sample preparation for the microarray analysis, which was supervised by AP; TM and MD analysed the microarray data; NS assisted in stomatal aperture measurements; CY conducted the analysis of the RNA–seq data, and BH analysed the WGS data, supervised by ML; AP wrote the article. MG, JK, and RH acquired the funding for this project.

Conflict of interest

The authors declare that they have no conflicts of interest.

Funding

This work was funded by ARC Discovery grant DP210102828 to MG, JK, and RH and ARC Centre of Excellence grant CE230100015 to MG and ML. AP was supported by the GOstralia!/University of Adelaide PhD Scholarship and a School of Agriculture, Food and Wine Short Term Scholarship.

Data availability

Microarray data are available at GEO (https://www.ncbi.nlm.nih.gov/geo/) under accession no. GSE261985. The RNA-seq and Whole Genome Sequencing data are available in the NCBI SRA (https://www.ncbi.nlm.nih.gov/bioproject) under BioProject PRJNA1087439 and PRJNA1087448. All other data supporting the findings of this study are available within the paper and within its supplementary data published online.

References

  1. Andrews S. 2010. FastQC: a quality control tool for high throughput sequence data. Babraham Institute, Babraham Bioinformatics. Available online at: http://www.bioinformatics.babraham.ac.uk/projects/fastqc. [Google Scholar]
  2. Balfagón D, Gómez-Cadenas A, Rambla JL, Granell A, Ollas C, Mittler R, Zandalinas SI.. 2022. γ-Aminobutyric acid plays a key role in plant acclimation to a combination of high light and heat stress. Plant Physiology 188, 2026–2038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bennetzen JL, Wang X.. 2018. Relationships between gene structure and genome instability in flowering plants. Molecular Plant 11, 407–413. [DOI] [PubMed] [Google Scholar]
  4. Buels R, Yao E, Diesh CM, et al. 2016. JBrowse: a dynamic web platform for genome visualization and analysis. Genome Biology 17, 66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cheng C-Y, Krishnakumar V, Chan AP, Thibaud-Nissen F, Schobel S, Town CD.. 2017. Araport11: a complete reannotation of the Arabidopsis thaliana reference genome. The Plant Journal 89, 789–804. [DOI] [PubMed] [Google Scholar]
  6. Conn SJ, Hocking B, Dayod M, et al. 2013. Protocol: optimising hydroponic growth systems for nutritional and physiological analysis of Arabidopsis thaliana and other plants. Plant Methods 9, 11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. De Angeli A, Zhang JB, Meyer S, Martinoia E.. 2013. AtALMT9 is a malate-activated vacuolar chloride channel required for stomatal opening in Arabidopsis. Nature Communications 4, 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dellero Y, Mauve C, Jossier M, Hodges M.. 2021. The impact of photorespiratory glycolate oxidase activity on Arabidopsis thaliana leaf soluble amino acid pool sizes during acclimation to low atmospheric CO2 concentrations. Metabolites 11, 501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Deng X, Xu X, Liu Y, Zhang Y, Yang L, Zhang S, Xu J.. 2020. Induction of γ-aminobutyric acid plays a positive role to Arabidopsis resistance against Pseudomonas syringae. Journal of Integrative Plant Biology 62, 1797–1812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Des Marais DL, Auchincloss LC, Sukamtoh E, McKay JK, Logan T, Richards JH, Juenger TE.. 2014. Variation in MPK12 affects water use efficiency in Arabidopsis and reveals a pleiotropic link between guard cell size and ABA response. Proceedings of the National Academy of Sciences, USA 111, 2836–2841. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Devos KM, Brown JK, Bennetzen JL.. 2002. Genome size reduction through illegitimate recombination counteracts genome expansion in Arabidopsis. Genome Research 12, 1075–1079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dittrich M, Mueller HM, Bauer H, et al. 2019. The role of Arabidopsis ABA receptors from the PYR/PYL/RCAR family in stomatal acclimation and closure signal integration. Nature Plants 5, 1002–1011. [DOI] [PubMed] [Google Scholar]
  13. Ehrlich SD, Bierne H, d’Alençon E, Vilette D, Petranovic M, Noirot P, Michel B.. 1993. Mechanisms of illegitimate recombination. Gene 135, 161–166. [DOI] [PubMed] [Google Scholar]
  14. Ewels P, Magnusson M, Lundin S, Käller M.. 2016. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics 32, 3047–3048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fait A, Fromm H, Walter D, Galili G, Fernie AR.. 2008. Highway or byway: the metabolic role of the GABA shunt in plants. Trends in Plant Science 13, 14–19. [DOI] [PubMed] [Google Scholar]
  16. Geiger D, Maierhofer T, Al-Rasheid KAS, et al. 2011. Stomatal closure by fast abscisic acid signaling is mediated by the guard cell anion channel SLAH3 and the receptor RCAR1. Science Signaling 4, 12. [DOI] [PubMed] [Google Scholar]
  17. Hashimoto M, Negi J, Young J, Israelsson M, Schroeder JI, Iba K.. 2006. Arabidopsis HT1 kinase controls stomatal movements in response to CO2. Nature Cell Biology 8, 391–397. [DOI] [PubMed] [Google Scholar]
  18. He J, Zhang R-X, Kim DS, Sun P, Liu H, Liu Z, Hetherington AM, Liang Y-K.. 2020. ROS of distinct sources and salicylic acid separate elevated CO2-mediated stomatal movements in Arabidopsis. Frontiers in Plant Science 11, 542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Healey A, Furtado A, Cooper T, Henry RJ.. 2014. Protocol: a simple method for extracting next-generation sequencing quality genomic DNA from recalcitrant plant species. Plant Methods 10, 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hiyama A, Takemiya A, Munemasa S, Okuma E, Sugiyama N, Tada Y, Murata Y, Shimazaki K-i.. 2017. Blue light and CO2 signals converge to regulate light-induced stomatal opening. Nature Communications 8, 1284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Hõrak H, Sierla M, Tõldsepp K, et al. 2016. A dominant mutation in the HT1 kinase uncovers roles of MAP kinases and GHR1 in CO2-induced stomatal closure. The Plant Cell 28, 2493–2509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hu HH, Boisson-Dernier A, Israelsson-Nordstrom M, Bohmer M, Xue SW, Ries A, Godoski J, Kuhn JM, Schroeder JI.. 2010. Carbonic anhydrases are upstream regulators of CO2-controlled stomatal movements in guard cells. Nature Cell Biology 12, 87–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jakobson L, Vaahtera L, Tõldsepp K, et al. 2016. Natural variation in Arabidopsis Cvi-0 accession reveals an important role of MPK12 in guard cell CO2 signaling. PLoS Biology 14, e2000322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kaplan F, Kopka J, Sung DY, Zhao W, Popp M, Porat R, Guy CL.. 2007. Transcript and metabolite profiling during cold acclimation of Arabidopsis reveals an intricate relationship of cold-regulated gene expression with modifications in metabolite content. The Plant Journal 50, 967–981. [DOI] [PubMed] [Google Scholar]
  25. Langmead B, Salzberg SL.. 2012. Fast gapped-read alignment with Bowtie 2. Nature Methods 9, 357–359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lawson T, Blatt MR.. 2014. Stomatal size, speed, and responsiveness impact on photosynthesis and water use efficiency. Plant Physiology 164, 1556–1570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lawson T, Simkin AJ, Kelly G, Granot D.. 2014. Mesophyll photosynthesis and guard cell metabolism impacts on stomatal behaviour. New Phytologist 203, 1064–1081. [DOI] [PubMed] [Google Scholar]
  28. Leakey ADB, Ferguson JN, Pignon CP, Wu A, Jin Z, Hammer GL, Lobell DB.. 2019. Water use efficiency as a constraint and target for improving the resilience and productivity of C3 and C4 crops. Annual Review of Plant Biology 70, 781–808. [DOI] [PubMed] [Google Scholar]
  29. Li Z, Yu JJ, Peng Y, Huang BR.. 2017. Metabolic pathways regulated by abscisic acid, salicylic acid and γ-aminobutyric acid in association with improved drought tolerance in creeping bentgrass (Agrostis stolonifera). Physiologia Plantarum 159, 42–58. [DOI] [PubMed] [Google Scholar]
  30. Love MI, Huber W, Anders S.. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15, 550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ma J, Bennetzen JL.. 2004. Rapid recent growth and divergence of rice nuclear genomes. Proceedings of the National Academy of Sciences, USA 101, 12404–12410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Mekonnen DW. 2013. Phenotypic and chemotypic characterization of GABA-shunt mutants in Arabidopsis thaliana. PhD thesis, University of Cologne (Germany). https://kups.ub.uni-koeln.de/5305/. Accessed April 2022. [Google Scholar]
  33. Mekonnen DW, Flügge UI, Ludewig F.. 2016. Gamma-aminobutyric acid depletion affects stomata closure and drought tolerance of Arabidopsis thaliana. Plant Science 245, 25–34. [DOI] [PubMed] [Google Scholar]
  34. Meyer S, Mumm P, Imes D, Endler A, Weder B, Al-Rasheid KAS, Geiger D, Marten I, Martinoia E, Hedrich R.. 2010. AtALMT12 represents an R-type anion channel required for stomatal movement in Arabidopsis guard cells. The Plant Journal 63, 1054–1062. [DOI] [PubMed] [Google Scholar]
  35. Miyashita Y, Good AG.. 2008. Contribution of the GABA shunt to hypoxia-induced alanine accumulation in roots of Arabidopsis thaliana. Plant and Cell Physiology 49, 92–102. [DOI] [PubMed] [Google Scholar]
  36. Narsai R, Gouil Q, Secco D, Srivastava A, Karpievitch YV, Liew LC, Lister R, Lewsey MG, Whelan J.. 2017. Extensive transcriptomic and epigenomic remodelling occurs during Arabidopsis thaliana germination. Genome Biology 18, 172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Negi J, Matsuda O, Nagasawa T, Oba Y, Takahashi H, Kawai-Yamada M, Uchimiya H, Hashimoto M, Iba K.. 2008. CO2 regulator SLAC1 and its homologues are essential for anion homeostasis in plant cells. Nature 452, 483–486. [DOI] [PubMed] [Google Scholar]
  38. Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C.. 2017. Salmon provides fast and bias-aware quantification of transcript expression. Nature Methods 14, 417–419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ramesh SA, Tyerman SD, Xu B, et al. 2015. GABA signalling modulates plant growth by directly regulating the activity of plant-specific anion transporters. Nature Communications 6, 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Renault H, Roussel V, El Amrani A, Arzel M, Renault D, Bouchereau A, Deleu C.. 2010. The Arabidopsis pop2-1 mutant reveals the involvement of GABA transaminase in salt stress tolerance. BMC Plant Biology 10, 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Roelfsema MRG, Hedrich R.. 2005. In the light of stomatal opening: new insights into ‘the Watergate’. New Phytologist 167, 665–691. [DOI] [PubMed] [Google Scholar]
  42. Sai N, Bockman JP, Chen H, Watson-Haigh N, Xu B, Feng X, Piechatzek A, Shen C, Gilliham M.. 2023. StomaAI: An efficient and user-friendly tool for measurement of stomatal pores and density using deep computer vision. New Phytologist 238, 904–915. [DOI] [PubMed] [Google Scholar]
  43. Scholz SS, Reichelt M, Mekonnen DW, Ludewig F, Mithofer A.. 2015. Insect herbivory-elicited GABA accumulation in plants is a wound-induced, direct, systemic, and jasmonate-independent defense response. Frontiers in Plant Science 6, 1128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Shi K, Li X, Zhang H, Zhang G, Liu Y, Zhou Y, Xia X, Chen Z, Yu J.. 2015. Guard cell hydrogen peroxide and nitric oxide mediate elevated CO2-induced stomatal movement in tomato. New Phytologist 208, 342–353. [DOI] [PubMed] [Google Scholar]
  45. Sierla M, Hõrak H, Overmyer K, et al. 2018. The receptor-like pseudokinase GHR1 is required for stomatal closure. The Plant Cell 30, 2813–2837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Soneson C, Love MI, Robinson MD.. 2016. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research 4, 1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Taiz L, Murphy A, Monshausen GB, Peer W.. 2015. Signals and signal transduction. In: Taiz L, Zeiger E, Møller IM, Murphy AS, eds. Plant Physiology and Development. Sunderland, Massachusetts: Sinauer Associates Inc, 408–445. [Google Scholar]
  48. Takahashi Y, Bosmans KC, Hsu PK, et al. 2022. Stomatal CO2/bicarbonate sensor consists of two interacting protein kinases, Raf-like HT1 and non-kinase-activity requiring MPK12/MPK4. Science Advances 8, eabq6161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tõldsepp K, Zhang J, Takahashi Y, et al. 2018. Mitogen-activated protein kinases MPK4 and MPK12 are key components mediating CO2-induced stomatal movements. The Plant Journal 96, 1018–1035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Ülker B, Peiter E, Dixon DP, Moffat C, Capper R, Bouché N, Edwards R, Sanders D, Knight H, Knight MR.. 2008. Getting the most out of publicly available T-DNA insertion lines. The Plant Journal 56, 665–677. [DOI] [PubMed] [Google Scholar]
  51. Urano K, Maruyama K, Ogata Y, et al. 2009. Characterization of the ABA-regulated global responses to dehydration in Arabidopsis by metabolomics. The Plant Journal 57, 1065–1078. [DOI] [PubMed] [Google Scholar]
  52. Van Norman J, Murphy C, Sieburth L.. 2011. BYPASS1: synthesis of the mobile root-derived signal requires active root growth and arrests early leaf development. BMC Plant Biology 11, 28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Vitte C, Bennetzen JL.. 2006. Analysis of retrotransposon structural diversity uncovers properties and propensities in angiosperm genome evolution. Proceedings of the National Academy of Sciences, USA 103, 17638–17643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Weigel D, Glazebrook J.. 2006. Transformation of Agrobacterium using the freeze-thaw method. Cold Spring Harbor Protocols 2006, doi: 10.1101/pdb.prot4666. [DOI] [PubMed] [Google Scholar]
  55. Xu B, Feng X, Piechatzek A, Zhang S, Konrad KR, Kromdijk J, Hedrich R, Gilliham M.. 2024. The GABA shunt contributes to ROS homeostasis in guard cells of Arabidopsis. New Phytologist 241, 73–81. [DOI] [PubMed] [Google Scholar]
  56. Xu B, Long Y, Feng X, et al. 2021a. GABA signalling modulates stomatal opening to enhance plant water use efficiency and drought resilience. Nature Communications 12, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Xu B, Sai N, Gilliham M.. 2021b. The emerging role of GABA as a transport regulator and physiological signal. Plant Physiology 187, 2005–2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yang Y, Costa A, Leonhardt N, Siegel RS, Schroeder JI.. 2008. Isolation of a strong Arabidopsis guard cell promoter and its potential as a research tool. Plant Methods 4, 6–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Yeh C-Y, Wang Y-S, Takahashi Y, et al. 2023. MPK12 in stomatal CO2 signaling: function beyond its kinase activity. New Phytologist 239, 146–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Yu G, Wang L-G, Han Y He Q-Y.. 2012. ClusterProfiler: An R package for comparing biological themes among gene clusters. OMICS A Journal of Integrative Biology 16, 284–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhang L, Takahashi Y, Hsu P-K, Kollist H, Merilo E, Krysan PJ, Schroeder JI.. 2020. FRET kinase sensor development reveals SnRK2/OST1 activation by ABA but not by MeJA and high CO2 during stomatal closure. eLife 9, e56351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zhang XR, Henriques R, Lin SS, Niu QW, Chua NH.. 2006. Agrobacterium-mediated transformation of Arabidopsis thaliana using the floral dip method. Nature Protocols 1, 641–646. [DOI] [PubMed] [Google Scholar]
  63. Zhou Y, Vroegop-Vos IA, Van Dijken AJ, Van der Does D, Zipfel C, Pieterse CM, Van Wees SC.. 2020. Carbonic anhydrases CA1 and CA4 function in atmospheric CO2-modulated disease resistance. Planta 251, 1–14. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

erae168_suppl_Supplementary_Table_S1_Figures_S1-S11
erae168_suppl_supplementary_table_s1_figures_s1-s11.pdf (1.8MB, pdf)

Data Availability Statement

Microarray data are available at GEO (https://www.ncbi.nlm.nih.gov/geo/) under accession no. GSE261985. The RNA-seq and Whole Genome Sequencing data are available in the NCBI SRA (https://www.ncbi.nlm.nih.gov/bioproject) under BioProject PRJNA1087439 and PRJNA1087448. All other data supporting the findings of this study are available within the paper and within its supplementary data published online.

Articles from Journal of Experimental Botany are provided here courtesy of Oxford University Press

RESOURCES