Cyano cyclopropyl ABA analogs were provided by Bayer (Division Crop Science, Höchst, Germany). All CCPs, except CCP8, were synthesized as described (Frackenpohl et al., 2018). For the synthesis of CCP8, ethyl-(2Z)-6-(1-cyanocyclopropyl)-3-ethyl-6-hydroxy-7-methyloct-2-en-4-ynoate, copper (I) iodide (7 mg, 0.37 mmol) and bis (triphenylphosphine) palladium (II) chloride (19 mg, 0.03 mmol) were charged under argon, and toluene (3 mL) together with ethyl (2Z)-3-iodopent-2-enoate (233 mg, 0.9 mmol; column chromatography purification of a crude product as described by Frackenpohl et al., 2018) were added at room temperature. After stirring for 10 min, 1-(3-hydroxy-4-methylpent-1-yn-3-yl) cyclopropanecarbonitrile (150 mg, 0.9 mmol; synthesis as describe by Frackenpohl et al., 2018) dissolved in toluene (3 mL) and diisopropylamine (0.4 mL, 2.8 mmol) were added. The resulting reaction mixture was stirred for 3 h prior to addition of water. The aqueous phase was extracted repeatedly with dichloromethane. The combined organic phases were dried over magnesium sulphate, filtered, and concentrated under reduced pressure. Purification of the crude product by silica gel chromatography using ethyl acetate and heptane (Frackenpohl et al., 2018) recovered the ABA agonist CCP8 (167 mg, 63% of theory) as a colourless oil. ¹H NMR (400 MHz, CDCl₃): 6.03 (s, 1H), 4.21-4.15 (q, 2H), 2.68 (br. s, 1H, OH), 2.50-2.43 (sept, 1H), 2.30-2.26 (q, 2H), 1.47-1.41 (m, 2H), 1.33-1.30 (m, 1H), 1.30-1.28 (t, 3H), 1.25-1.21 (m, 1H), 1.16-1.11 (m, 9H). LC-MS (ret.-time, min, [M⁺/log p]) 2.84 [289.30/1.39]. HRMS-ESI: calcd for C₁₇H₂₃NO₃ ⁺ [M+H]⁺ 289.1678, found 289.1690.

(S)-ABA was purchased from CHEMOS (www.chemos.de). Other chemicals were of the highest purity available and obtained from SigmaAldrich (www.sigmaaldrich.com), ROTH (www.carlroth.com), and J. T. Baker (http://www.avantormaterials.com) unless otherwise indicated. The Arabidopsis (Arabidopsis thaliana) accession Col-0 was received from the Nottingham Arabidopsis Stock Center. The elite wheat (Triticum aestivum) cultivar, Elixer, was provided from Bayerische Landesanstalt für Landwirtschaft, Freising. ABA and CCPs were dissolved in DMSO and diluted to designated concentrations before administration. For de-esterification, 0.2 µmol CCP was treated with 4 nkat pig liver esterase (Sigma-Aldrich, Burlington, MA, USA) for 2 h at 25°C according to supplier’s instruction.

Arabidopsis seeds were sown on 1/2 x MS (Murashige and Skoog) solid medium followed by stratification at 4°C for two days to promote germination as reported (Yang et al., 2016). Briefly, the germinated seeds were incubated for 7 days at 60 µmol m^-2 s^-1 PAR and 22°C. Single seedlings were transferred to 0.2 L pots filled with soil (Classic Profi Substrate, Einheitserde Werkverband, Sinntal-Altengronau, Germany) containing 3 mg nitrogen fertilizer per gram of dry soil. All plants were grown under an 8 h light regime with 150 µmol m^-2 s^-1 PAR, 22°C temperature, 50% relative humidity during the day, and 17°C and 60% at night in a plant growth cabinet (Conviron E15, Winnipeg, Canada). The short-day light regime inhibits flowering and thus allows assessment of vegetative leaf growth during the long-lasting drought experiment as well as under well-water conditions. The wheat kernels were sown directly in 0.5 L pots filled with soil as above that covered the seeds by approximately 1 cm. Wheat plants were grown as Arabidopsis with the difference of a 16 h light regime.

The yeast cultivation and the life luciferase assay were performed as described (Ruschhaupt et al., 2019). Briefly, the reporter yeast expressing an ABA-dependent luciferase gene (LUC) was transformed with different combinations of ABA signaling components as indicated. Golden gate cloning (Binder et al., 2014) was used to construct pYEAST_LII_5‐6_CEN_URA vector with TDH3 promoter and TDH1 terminator (tTDH1), pYEAST_LII_1‐2_CEN_LEU vector with GK1 promoter and tTDH1, and pYEAST_LII_3‐4_CEN_TRP with the GAL1 promoter (pGAL1) and tTDH1 to express RCARs, PP2Cs, and OST1 (SnRK2.6), respectively (Ruschhaupt et al., 2019). The expression of ABF2 in yeast relies on a modified pGREG503 vector, in which pGAL1 (Jansen et al., 2005) was replaced by the ADH1 promoter.

Colonies of freshly transformed cells were inoculated in 0.5 mL supplemented synthetic dextrose (SD) medium and incubated at 30°C in a gyratory shaker (Thermoshake-THO500, Gerhardt, Königswinter, Germany) at 200 rpm for 16-18 h. After overnight growth, the cells were harvested and washed in SD medium followed by inoculation in SD medium containing 2% galactose for induction of SnRK2 kinases under the control of the GAL1 promoter and supplementation with ligands at indicated concentrations, to an OD600 of 0.2 in flat-bottom 96-well plates (CytoOne, Starlab, Hamburg, Germany). Yeast cultures were incubated at 30°C and 200 rpm for 16-18 h before determination of LUC activity and cell density. Light emission of yeasts was detected in a HIDEX plate luminometer (HIDEX, Turku, Finland) after injection of 100 µL freshly prepared luciferin substrate (1 mM D-luciferin, Promega) provided in 100 mM citric acid, pH 3 (Leskinen et al., 2003) for 5 s and was corrected for background. The LUC activity was normalized to the cell density (OD600) measured in 96-well flat plates (Sarstedt, Nümbrecht, Germany). LUC activity is given as the ratio of light emission to OD600 in counts per second.

Expression, purification, and analysis of RCAR and PP2C proteins were performed essentially as described (Ruschhaupt et al., 2019). Briefly, the ABA receptors RCAR1, RACR8, and RCAR11 were expressed with an aminoterminal histidine tag using the pSEVA431_LII_6xHIS::RCAR1/RCAR8/RCAR11. ABA coreceptors were expressed as histidine-tagged fusions with the maltose-binding protein using the pSEVA431_LII_6xHIS::MBP::ABI2/HAB1. All constructs were transformed and expressed in Escherichia coli strain BL21 and purified to homogeneity according to supplier’s instruction (GE Healthcare, Chicago, USA). The receptor assay was performed with 100 nM PP2C and a two-fold molar excess of RCAR. Values are means ± SD of three replicates. In the assay, the phosphatase activity of ABI2 and HAB1 showed in the presence of RCARs and saturating ABA levels (1 mM) a residual activity of 10% and 4%, respectively. The maximum inhibition was set to 100% in the figures. The phosphatase activity was measured using 4-methylumbelliferyl-phosphate as a substrate (Meinhard and Grill, 2001). The increase in fluorescence was recorded over 25 min with excitation and emission wavelengths at 360 and 460 nm, respectively, using a Synergy 2 plate reader (Bio-Tec, Winooski, USA).

Preparation of Arabidopsis protoplasts, transfection of the effector DNA, and analysis of LUC activity were performed as described (Tischer et al., 2017). In essence, 10⁵ mesophyll protoplasts from leaves of the ABA deficient mutant aba2-1 (Schwartz et al., 1997) were transfected with 4 µg DNA of the pRD29B::LUC construct as an ABA-responsive reporter and 3 µg of p35S::GUS (β-glucuronidase reporter gene) expression cassette as an internal control for normalization. In addition, 3 µg p35S::RCAR11 or the empty vector with the same expression cassette were co-expressed with 0.1 µg p35S::ABI2 effector constructs (Tischer et al., 2017), and the ligand-dependent induction of ABA-responsive reporter expression were assessed after incubation at 22°C. The expression of LUC and GUS was determined 18 h after transfection and ligand administration.

Thermal imaging was carried out as described (Yang et al., 2016). An assessment of the efficacy of ABA and CCPs was conducted by using 35-day-old Arabidopsis and 30-day-old wheat plants. Stock solutions (10 mM) of CCPs and ABA dissolved in dimethyl sulfoxide (DMSO) were diluted to the respective ligand concentrations (10, 30, 100 µM) in water containing 0.01% (v/v) Tween 20. Aqueous Tween 20 solutions supplemented with 0.1%, 0.3%, and 1% DMSO served for mock treatments. The solutions were sprayed onto the leaf surface until the run-off of droplets. One day after application (unless otherwise indicated), thermal imaging was conducted using the InfraTec instrument (IR-TCM 384, InfraTec, Dresden, Germany). The leaf temperature of Arabidopsis was determined as the average temperature of all leaves. The leaf temperature of wheat was determined as described (Yang et al., 2019).

The progressive drought experiment was performed as described (Yang et al., 2016). Briefly, 18-day-old Arabidopsis plantlets grown under well-watered and short-day conditions (8 h light with 150 µmol m^-2 s^-1 PAR, 22°C, 50% relative humidity, and 16 h dark with 17°C, 60% relative humidity) were transferred to 0.2 L pots loaded with water-saturated soil (33.4 ± 0.4 g dry soil with 148 ± 2 g water). At the onset of the progressive drought (day 0), water was withheld, and a sealed soil surface prevented evaporation. A slowly increasing water deficit was achieved by transpirational water loss from plants. The progressive drought experiment lasted for eight weeks until mock-treated plants had consumed all the plant-usable water. At day 18, 32, and 43 the ligand solutions as indicated were administered onto the growing plants. For measuring growth and water consumption, leaf rosettes were regularly photographed, and the pots were weighed. The projected leaf area of Arabidopsis rosettes was analyzed by using Photoshop Elements software (Adobe, San Jose, USA). The consumed water was calculated as the difference in pot weight between day 0 and the designated days. The retained water was expressed as the difference between the pot weight at the end of the drought (day 57) and the weight of the loaded dry soil. Above-ground material was harvested at day 57, and the biomass was determined after drying the material for three days at 60°C to achieve constant weight. WUE was calculated as the ratio of above-ground dry biomass to consumed water. The data for the mock- and ABA-treated plants in the progressive drought experiment were already published (Yang et al., 2019).

All data were analyzed using Microsoft Excel 365 or Origin 2020. Curve fitting in Figure 1 , and Supplementary Figures S2 , S3 was achieved by using the Hill1 function. Statistical analysis was assessed by one-way ANOVA (Tukey test).

In this study, we used yeast combined with analyses of transient signaling in plant cells and whole plant physiology for assessment of nine structurally similar cyano cyclopropyl compounds ( Figure 1A ) including the lead structure CCP1. The other CCPs differ from CCP1 in their side chains and can be subgrouped accordingly ( Figure 1A ). CCP2 and CCP3 feature an ethyl group at C3 instead of the methyl group in CCP1 of the terpenoid side chain ( Figure 1A , see CCP1 for C numbering), while CCP4 to CCP6 have at this position a trifluorinated methyl group (-CF3). In CCP7 to CCP9, the trans-ene of the side chain is replaced by an enyne group. In addition, CCP3 to CCP9 are ester derivates and in CCP5 and CCP6 the isopropyl group of CCP1 is replaced by a cyclopropyl and cyclopentyl moiety, respectively.

The reconstruction of the ABA signaling pathway in yeast offers an avenue for characterization of specific ligand-receptor complexes. The pathway consists of the ABA receptor such as RCAR11, a protein phosphatase type 2C coreceptor, e.g. ABI2, SnRK2-type protein kinase such as OST1 acting as a response mediator that phosphorylates and activates ABA-responsive transcription factors, e.g. ABF2, to drive expression of a LUC reporter gene ( Figure 1B ) in an ABA-regulated manner in yeast (Ruschhaupt et al., 2019).

Using this yeast line, 30 µM exogenous ABA or CCP1 induced the LUC reporter approximately eightfold compared to mock-treated cells ( Figure 1C ). However, only CCP2 among the other CCPs revealed an ABA-agonistic activity ( Figure 1C ). The inactive CCP3 and CCP9 are all ester derivates. The free carboxyl group of ABA is critical for binding to the ABA receptor (Santiago et al., 2009). Hence, we attempted to generate the free carboxyl group of these CCPs by incubating the compounds in the presence of an unspecific esterase, pig liver esterase (PLE) (Musidlowska-Persson and Bornscheuer, 2003). The degree of ester-bond cleavage by PLE differed among the synthetic agonists from complete de-esterification of CCP3 that convert into CCP2, to approximately 25% for CCP8 ( Supplementary Figure S1 ). Examining those enzyme-treated CCPs in the yeast system showed for CCP3 a reporter activation comparable to esterase-treated CCP2 and ABA, while the other CCPs revealed only minor ABA agonistic activity ( Figure 1C ).

CCP1 and CCP2 emerged as two synthetic ABA agonists with comparable activity to ABA when interacting with subgroup III receptor RCAR11. To extend these findings and to achieve a better assessment of ligand efficacy and specificity, both CCPs and ABA were analyzed for their dose-dependent activation of the ABA response pathway in yeast. In addition to RCAR11-ABI2, we also expressed in yeast the receptor RCAR1/PYL9-ABI2 and RCAR8/PYL5-ABI2 belonging to receptor subgroups I and II, respectively ( Figures 2A–C ).

Both CCPs activated the ABA response similar or even stronger than ABA via RCAR11-ABI2 ( Figure 2A ). In the presence of RCAR1 and RCAR8, however, the responsiveness towards both CCPs was clearly reduced compared to ABA ( Figures 2B, C ). With RCAR1, CCP1 and CCP2 required approximately 30x and >30x higher exogenous concentrations to achieve a comparable reporter expression than 1 µM ABA. In the case of RCAR8, CCP1 and CCP2 clearly differed in their capacity to regulate the ABA response with approximately half the ABA induction and no significant reporter regulation, respectively, at 1 µM exogenous ligand concentration. To achieve an induction value comparable to 1 µM ABA, about 10x and >30x higher levels of CCP1 and CCP2, respectively, were required. Exchanging ABI2 for the coreceptor HAB1 yielded similar results ( Figures 2D–F ). Both CCPs are selectively discriminated by the different receptor complexes with CCP1 being more active than CCP2 in the presence of RCAR1 and RCAR8.

To corroborate the ABA agonistic activity of CCPs, we next examined the compounds in vitro with purified receptor complexes. First, ABA, CCP1, and CCP2 were titrated to purified ABI2 together with RCAR1, RCAR8, or RCAR11 ( Figures 3A–C ). In agreement with the yeast results, CCP1 and CCP2 inhibited ABI2 in the presence of RCAR11 comparable to ABA but RCAR1 strongly discriminated against the synthetic ABA mimics ( Figures 3A, B ). For 100 nM ABI2 and 200 nM RCAR1, the half-maximal inhibitory concentration (IC50) of ABA was 100 nM whereas for CCP1 and CCP2 the values were approximately 100-fold higher ( Figure 3B ). For RCAR8-ABI2, IC50 values of CCP1 and CCP2 were around 10- and 30-fold higher than for ABA ( Figure 3C ). Assessment of the inhibition of HAB1 phosphatase activity by three RCARs and both CCPs generated similar results except a lower IC50 value for CCP2 in the presence of RCAR1 ( Figures 3D–F ). The esterified CCP3 to CCP9 compounds had no or little activity in the in vitro receptor assay ( Supplementary Figure S2 ). Removal of the ester group by esterase treatment ( Supplementary Figure S1 ) resulted in ABA-agonistic activity of CCP3 to CCP9 ( Supplementary Figure S3 ). CCP3 was completely de-esterified to CCP2 and showed comparable activity as CCP2 ( Supplementary Figures S3A–C ). CCP4, CCP5, and CCP6 with a trifluorinated methyl group displayed a slightly lower activity than ABA with the RCAR11-ABI2 receptor complex, while for RCAR1 and RCAR8, IC50 values were more than 1000-fold and 100-fold higher than ABA, respectively ( Supplementary Figures S3D–F ). Enyne-containing CCP7 to CCP9 had more than 100-fold lower activity for all three receptors ( Supplementary Figures S3G–I ). Taken together, this biochemical analysis is in agreement with the yeast data revealing that CCP1, CCP2, and de-esterified CCP3 are the most active CCPs that regulate the RCAR11-ABI2 complex indistinguishable from ABA. However, CCPs are less active than ABA in other ABA receptor-coreceptor combinations.

To what extent the CCP capacity for the ABA response activation in yeast is transferable to plants was assessed by employing plant cells and whole plant physiology. Firstly, we analyzed CCPs for stimulating ABA signaling in leaf protoplasts of the Arabidopsis mutant, aba2-1. ABA2 catalyzes a key step in ABA biosynthesis (Gonzalez-Guzman et al., 2002). Hence, aba2-1 has low levels of endogenous ABA and ABA signaling is sensitively stimulated by administration of exogenous ABA (Tischer et al., 2017). Most diene-containing CCPs, CCP1 to CCP6, with the exception of CCP3 showed similar or higher activity than ABA to induce ABA signaling ( Figures 4A–C ). CCP5 was the most active compound with up to 60% higher induction rates than ABA at different ligand concentrations. Unexpectedly, the ABA-like activity of CCP3 was not observed in the protoplast assay. Since CCP3 is the ester form of CCP2 and CCP2 did show ABA-like activity in protoplasts, we anticipated that the de-esterified CCP3 should induce ABA-responsive reporter expression. Indeed, de-esterified CCP3 induced ABA-responsive reporter expression in protoplasts, similar to CCP2 and higher than ABA ( Supplementary Figure S4 ). The enyne-type CCPs, CCP7 to CCP9, were not or moderately active ( Figure 4D ). The transient signaling analysis confirmed the high activity of CCP1 and CCP2, and it identified the up to now inconspicuous CCP5 as a very potent compound.

To complement our first plant analysis of CCPs, the compounds were assayed for their efficiency in reducing leaf transpiration. Transpiration reduces the leaf temperature, which can be monitored by infrared thermal imaging (Morison et al., 2008). Our previous studies identified aqueous solutions containing 10, 30, and 100 µM ABA as suitable for changing transpiration in Arabidopsis leaves within one day (Yang et al., 2019). Hence, these concentrations of ABA and CCPs including the mock treatment were applied onto leaves of Arabidopsis grown under well-watered conditions. As reported, a dose-dependent increase in leaf temperature was observed for ABA-treated plants compared to the mock-exposed controls ( Figure 5A ). CCP administration also increased the leaf temperatures and the CCP values were compared to the temperature increase achieved by 10 µM ABA ( Figure 5B ). Most CCPs were more active than ABA. Specifically, CCP1 to CCP3 and CCP5 were the most potent. It took about 100 µM ABA to achieve a temperature increase, i.e. reduction in stomatal transpiration, similar to 10 µM CCP1 and CCP2. Interestingly, CCP3 was almost as active as the non-esterified CCP2. The CF3-containing CCPs displayed variable activity, with CCP5 bearing a cyclopropyl group at C6 being more efficacious than ABA and even exceeding the efficacy of CCP1 to CCP3, whereas CCP4 with an isopropyl group at the same position showed a moderately higher activity compared to ABA. CCP6 with a bulky cyclopentyl group at C6 displayed a somewhat lower efficacy. The enyne-type CCPs (CCP7 to CCP9) increased leaf temperatures less efficient or comparably to ABA with a clear correlation of longer alkyl ester groups impairing ABA-like activity. These results indicate several diene CCPs are by an order of magnitude more active than ABA in controlling stomatal aperture of Arabidopsis.

The higher activity of CCPs in the transpiration control possibly reflects improved uptake and higher metabolic stability of synthetic ABA mimics compared to the natural ligand. To corroborate this idea, the persistence of the minimized transpiration was examined by recording for 19 days the effect of a single application of diene CCPs (100 µM; Figures 5C–E ). Among the analyzed chemistry, CCP1, CCP2, CCP3, and CCP5 were more active and persistent than ABA. We next chose the most active CCPs, CCP1 and CCP5, to assess their equivalency with ABA by lowering concentrations of both CCPs and comparing their effectiveness in controlling transpiration ( Supplementary Figure S5 ). The ABA-inhibited transpiration reached an optimum between two and four days after administration. Subsequently, the leaf temperature dropped and reached starting values after eight days with 10 µM ABA. The 10 and 30 µM ligand treatment with CCP1 and CCP5 resulted, however, in a more persistent response for 10 days and a stronger response than with 100 µM ABA, revealing an approximately three times longer lasting effect and more than tenfold higher potency in this assay. Nineteen days after administration, plants treated with 30 µM CCP1 still had a 1.5°C higher leaf temperature clearly exceeding the maximum temperature increase triggered by 100 µM ABA ( Figure 5D and Supplementary Figure S5 ). The results emphasize the improved potency of several diene-type CCPs in mediating long-lasting transpiration changes, more effective than ABA by at least one order of magnitude.

Reduced leaf transpiration allows soil water to be saved and stored for alleviating potential incidences of water deficit. To examine the potential of CCP agonists to contribute to water saving, CCP-treated Arabidopsis was compared to mock- and ABA-challenged plants in their growth, water consumption, and WUE during a progressive drought experiment (Yang et al., 2016). CCP2 and CCP3 were chosen because they are highly efficient in increasing leaf temperature and both compounds just differ structurally in the free and ester-modified carboxyl group, respectively. The experiment started with cease of watering of eighteen-days old plants. At this stage, the total leaf area was smaller than 0.5 cm², and the experiment was finished after more than eight weeks, when all available water for mock-treated plants was consumed. The mock-treated plants reached a maximum in the projected leaf area of approximately 74 cm² after six weeks, followed by wilting and shrinking of the leaf rosette ( Figure 6A ). Throughout the experiment ABA and ABA agonists were three times administered by spraying 10 µM or 30 µM ligand solutions ( Figure 6A ). Growth of leaf rosettes of mock-treated and 10 µM ABA-treated plants was indistinguishable for six weeks, while 30 µM ABA, CCP2, or CCP3 treatment resulted in growth reduction ( Figure 6A ). Growth reduction of the 30 µM CCP3-challenged plants was about 60% compared to the control, suggesting a strong stomatal limitation for CO₂ uptake. Consistently, plants sprayed with 30 µM CCP3 displayed the highest temperature increase, i.e. lowest transpiration, 24 h after the first application of ligands ( Figure 6B ). Reduced leaf growth and transpiration for ABA- or CCP-treated plants, in turn, slowed down the rate of soil water consumption compared to mock-treated plants ( Figure 6C ). Therefore, water deficit and resultant symptoms were expected to occur earlier for mock-treated plants than the ligand-treated plants. Indeed, at day 48, control plants showed first signs of wilting and leaves had a pinkish hue from stress-induced anthocyanin while the 30 µM ABA- and CCP-treated plants were turgescent and the anthocyanin induction in these plants was less pronounced ( Figure 6D ). At the end of the drought, there was still 15.2 ± 1.3 g water retained in the soil for control plants, accounting for about 10% of the initial level ( Figures 6C, E ). On the other hand, 30 µM ABA-treated plants had still about 25% of initial soil water, and approximately 75% in the case of 30 µM CCP3 treatment ( Figure 6E ). The slow growth and low water consumption of the 30 µM CCP3 cohort resulted in a high soil water content at the end of the experiment when the above-ground plant material was harvested for determination of the dry matter. Already the first week after 30 µM CCP3 administration a clear reduction in water consumption was observable while growth was little affected compared to the control ( Figures 6A, C ) supporting a higher potency of the carboxyl-alkylated compound compared to the parent CCP2. The dry-weight biomass of the ABA- and CCP2-exposed plants was similar to the control ( Figure 6F ). CCP3 treatment led to a significant reduction in biomass accumulation of 60% by 30 µM CCP3 exposure compared to the control. In terms of WUE, all ABA and ABA agonists improved the ratio of biomass per water with CCP2 doubling the benefit of ABA to almost 50% WUE increase at 30 µM solute concentration ( Figure 6G ). Taken together, these results demonstrate that foliar application of CCP2 is more effective than ABA in enhancing WUE with minor trade-offs in growth performance.

Arabidopsis is a dicot, whereas most of the staple crops are monocots. The question remains whether the results on CCPs gained with Arabidopsis can be transferred to a major C3 crop. Previous analysis of wheat has revealed increases of leaf temperature and WUE in response to foliar ABA administration albeit requiring higher phytohormone levels than for Arabidopsis (Yang et al., 2019). To test the CCP chemistry, wheat plants were compared in their transpiration response by recording the thermal response of wheat leaves after exposure towards 0.1 mM ABA and ABA agonists ( Figure 7 ). Half the plants of a single pot were mock-treated, the other half challenged to transpiration regulators and compared to pots treated in parallel with ABA ( Figure 7A ). ABA application resulted in an approximately 0.8°C higher leaf temperature compared to the control one day after administration ( Figure 7B ). CCPs also increased the leaf temperature ( Figure 7C ). CCP1 and CCP3 had a higher activity than ABA, CCP4 and CCP5 were somewhat less active, while the other CCPs were comparable to ABA. The efficacy of CCP for rising leaf temperature differs between Arabidopsis and wheat. In both plant species, CCP1 and CCP3 were more potent than ABA, however, wheat appears to have a reduced sensitivity towards CCP5.