A significant role for the circadian clock in the long-term water use efficiency of 1 Arabidopsis 2 3

Abstract In plants, water use efficiency is a complex trait derived from numerous physiological and developmental characteristics. Here, we investigated the involvement of circadian regulation in long-term water use efficiency. Circadian rhythms are generated by the circadian oscillator, which provides a cellular measure of the time of day. In plants, the circadian oscillator contributes to the regulation of many aspects of physiology, including stomatal opening, the rate of photosynthesis, carbohydrate metabolism and developmental processes. We investigated in Arabidopsis the impact upon whole plant, long-term water use efficiency of the misregulation of genes encoding a large number of components of the circadian oscillator, identifying a major role for the circadian oscillator in plant water use. This appears to be due to contributions of the circadian clock to the control of transpiration and biomass accumulation. We also identified that the circadian oscillator specifically within guard cells contributes to both long-term water use efficiency and dehydration tolerance. Our experiments indicate that knowledge of circadian regulation will be important for developing future crops that use less water. One-sentence summary The circadian clock in Arabidopsis makes an important contribution to long-term water use efficiency.


Introduction 40
World population growth is increasing the demand for fresh water for agriculture, with 41 climate change predicted to exacerbate this competition for water resources (Ruggiero et al., 42 2017). One strategy to sustainably increase agricultural production involves the 43 improvement of crop water use (Condon et al., 2004;Xoconostle-Cazares et al., 2010;Hu 44 and Xiong, 2014;Ruggiero et al., 2017). Since up to 97% of water taken up from the soil by 45 plants is lost through stomatal transpiration (Yoo et al., 2009;Na and Metzger, 2014), the 46 manipulation of transpiration represents an excellent candidate for designing crops with 47 increased water use efficiency. 48 Plant water loss can be manipulated through changes in the regulation of stomatal opening 49 and by altering stomatal density and patterning (Pei et al., 1998;Hugouvieux et al., 2001;50 Schroeder et al., 2001;Hetherington and Woodward, 2003;Yoo et al., 2010;Lawson and 51 Blatt, 2014;Franks et al., 2015;Caine et al., 2019). In addition to stomatal responses to 52 environmental cues such as light, temperature and phytohormones, there are circadian 53 rhythms of stomatal opening (Gorton et al., 1989;Hennessey and Field, 1991). Circadian 54 rhythms are self-sustaining biological cycles with a period of about 24 h. These rhythms are 55 thought to adapt plants to daily cycles of light and dark, by anticipating daily changes in the 56 environment and co-ordinating cellular processes. In higher plants, circadian rhythms are 57 generated by several interlocked transcription-translation feedback loops known as the 58 circadian oscillator (Hsu and Harmer, 2014). The phase of the circadian oscillator is adjusted 59 continuously to match the phase of the environment through the process of entrainment, in 60 response to light, temperature and metabolic cues (Somers et al., 1998;Millar, 2004;61 Salomé and McClung, 2005;Haydon et al., 2013). Additionally, the circadian oscillator 62 communicates an estimate of the time of day to circadian-regulated features of the cell, 63 initially through transcriptional regulation (Harmer et al., 2000). The known circadian 64 oscillator controls circadian rhythms of stomatal opening because mutations that alter the 65 circadian period or cause circadian arrhythmia lead to equivalent alterations in the circadian 66 rhythm of stomatal opening (Somers et al., 1998;Dodd et al., 2004;Dodd et al., 2005). The 67 circadian oscillator is also involved in the responses of guard cells to environmental cues 68 such as drought and low temperature (Dodd et al., 2006;Legnaioli et al., 2009). 69 Circadian rhythms are often studied under conditions of constant light. However, the 70 circadian oscillator is also important for the regulation of stomatal opening under cycles of 71 light and dark. For example, overexpression of the circadian oscillator component CCA1 72 (CCA1-ox) alters the daily regulation of stomatal opening such that stomatal conductance 73 increases steadily throughout the photoperiod (Dodd et al., 2005). In comparison, in wild 74 type plants stomatal conductance remains relatively uniform during the photoperiod and is 75 substantially lower than CCA1-ox (Dodd et al., 2005). This suggests that misregulation of the 76 circadian oscillator might alter plant water use under cycles of light and dark. 77 Overexpression of CCA1 specifically within guard cells, using a guard cell specific promoter, 78 alters flowering time and drought response phenotypes under cycles of light and dark 79 (Hassidim et al., 2017). Like constitutive CCA1 overexpression (Dodd et al., 2005), CCA1 80 overexpression specifically within guard cells generally causes greater stomatal opening 81 during the light period (Hassidim et al., 2017). Therefore, the circadian oscillator within guard 82 cells is important for the daily regulation of stomatal opening (Hassidim et al., 2017). 83 Modelling suggests that under light/dark cycles, the circadian oscillator contributes at the 84 canopy scale to daily rhythms in stomatal aperture and carbon assimilation in bean and 85 cotton (Resco de Dios et al., 2016). 86 The contribution of the circadian oscillator to both stomatal opening and growth and biomass 87 accumulation (Dodd et al., 2005;Graf et al., 2010) suggests that the circadian oscillator 88 might make an important contribution to water use efficiency (WUE). WUE is the ratio of 89 carbon dioxide incorporated through photosynthesis into biomass to the amount of water lost 90 through transpiration. At the single leaf level, instantaneous, intrinsic WUE is often 91 measured with gas exchange techniques and expressed as net CO 2 assimilation per unit of 92 water transpired (Vialet-Chabrand et al., 2016;Ruggiero et al., 2017;Ferguson et al., 2018). 93 However, such measurements do not provide an accurate representation of WUE over the 94 plant lifetime, which is influenced by features such as leaf position, dark respiration, and time 95 of day changes in instantaneous WUE (Condon et al., 2004;Tomás et al., 2014;Medrano et 96 al., 2015;Ferguson et al., 2018). It is important to note that WUE is not a drought resistance 97 trait (Blum, 2009). 98 Given that the circadian oscillator affects stomatal opening and biomass accumulation 99 (Gorton et al., 1989;Hennessey and Field, 1991;Dodd et al., 2005;Edwards and Weinig, 100 2010;Graf et al., 2010;Edwards et al., 2012), we hypothesized that specific components of 101 the circadian oscillator might make an important contribution to long-term WUE of plants. 102 unclear (Fig. 1). We also included the che-2 mutant in our initial analysis, but inconsistency 130 of its WUE phenotype between experimental repeats led us to exclude the data. WUE was 131 also altered by changing the expression of the energy signalling components TPS1 and 132 KIN10 that participate in inputs to the circadian oscillator (Shin et al., 2017;Frank et al., 133 2018) (Fig. 1). Therefore, correct expression of these circadian clock-associated genes 134 contributes to long-term WUE of Arabidopsis. 135 We were interested to determine whether the WUE alterations caused by misregulation of 136 circadian oscillator gene expression arose from changes in either biomass accumulation or 137 transpiration. There was no clear evidence that a change in one of these parameters alone 138 underlies the altered WUE phenotypes ( Fig. 2A). This suggests that the altered WUE of lines 139 with misregulated circadian clock genes is due to the net effect of altered biomass 140 accumulation and altered transpiration in these genotypes ( Fig. 2A). 141 We hypothesised that variations in WUE might be explained by specific circadian 142 phenotypes in the mutants and overexpressors that we tested. For example, mutations in 143 clock genes expressed with a particular set of phases might have a pronounced effect on 144 WUE. Likewise, the nature of the circadian period change or flowering time change resulting 145 from misexpression of each oscillator component might be associated with certain changes 146 in WUE. To test this, we related the data from our WUE screen to the circadian phase of 147 expression of each mutated or overexpressed gene. We also compared the direction of 148 change of WUE to the period and flowering time phenotypes that arise from each mutant or 149 overexpressor (Fowler et al., 1999;Schultz et al., 2001;Doyle et al., 2002;Nakamichi et al., 150 2002;Yanovsky and Kay, 2002;Imaizumi et al., 2003;Más et al., 2003;Murakami et al., 151 2004;Farré et al., 2005;Hazen et al., 2005;Baena-González et al., 2007;Streitner et al., 152 2008;Wang et al., 2008;Baudry et al., 2010;Nakamichi et al., 2010;Rawat et al., 2011;153 Wahl et al., 2013;Hsu and Harmer, 2014). We note that the phenotypes reported by these 154 studies were often identified under constant conditions, whereas our experiments occurred 155 under light/dark cycles. 156 There was no obvious relationship between the circadian phenotypes that are caused by 157 each mutant or overexpressor investigated and the WUE of each of these lines (Fig. 2B,C,158 D). For example, mutating morning-phased circadian oscillator components can either 159 decrease or increase WUE (Fig. 2B). Mutants that cause long circadian periods and short 160 circadian periods can both increase and decrease WUE (Fig. 2C). Furthermore, mutants and 161 overexpressors that cause both early and delayed flowering can each increase and 162 decrease WUE (Fig. 2D). 163 Circadian regulation of water use efficiency combines multiple traits 164 Mutation or overexpression of components of the circadian oscillator can cause changes in 165 the development of Arabidopsis, such as alterations in rosette size, leaf shape and petiole 166 length ( Fig. 3A) (Zagotta et al., 1992;Schaffer et al., 1998;Wang and Tobin, 1998;Dodd et 167 al., 2005;Ruts et al., 2012;Rubin et al., 2018). These changes are likely to have 168 implications for gas exchange because, for example, spatially separated leaves are 169 predicted to transpire more water (Bridge et al., 2013). We investigated whether the changes 170 in WUE that were identified by our screen might arise from differences in rosette architecture 171 between the circadian clock-associated mutants and overexpressors and the wild types. 172 There was a weak positive correlation between rosette leaf surface area and WUE (r = 173 0.400; r 2 = 0.160; p < 0.001) (Fig. 3B). Therefore, approximately 16% of variability in WUE 174 can be explained by the variations in rosette leaf surface area that arise from misregulation 175 of the circadian oscillator. 176 In comparison, rosette leaf surface area was strongly correlated with each of the individual 177 parameters of water used and dry biomass accumulated. The variation in rosette surface 178 area accounted for 83% of the variability in water transpired across the genotypes (Fig. 3C). 179 Furthermore, the variation in rosette surface area accounted for 73% of the variability in 180 biomass accumulation across the genotypes (Fig. 3D), which is unsurprising given that 181 larger leaves are likely to contain more biomass. This demonstrates that one way that 182 circadian regulation affects WUE is through the influence of the circadian oscillator upon 183 plant development and rosette architecture, but this variation in leaf area does not account 184 for the majority of the influence of circadian regulation upon WUE. It also further supports 185 the notion that the influence of the circadian oscillator upon WUE is complex, and cannot be 186 explained by variation in one of water use or biomass accumulation alone. 187 Circadian regulation within guard cells alone contributes to water use efficiency 188 Next, we identified that the circadian oscillator within guard cells contributes to WUE. There 189 is evidence that guard cells contain a circadian oscillator that regulates stomatal opening 190 (Gorton et al., 1989;Hassidim et al., 2017). To investigate the contribution of the guard cell 191 circadian oscillator to WUE, we overexpressed two circadian oscillator components (CCA1,192 TOC1) in guard cells, using two guard cell-specific promoters (GC1, MYB60) for each of 193 CCA1 and TOC1 ( Fig. 4A) (Cominelli et al., 2005;Galbiati et al., 2008;Yang et al., 2008;194 Nagy et al., 2009;Meyer et al., 2010;Cominelli et al., 2011;Bauer et al., 2013;Rusconi et 195 al., 2013). GC1 is a strong guard cell-specific promoter that is relatively unresponsive to a 196 To further verify the guard cell-specific overexpression of CCA1 and TOC1 in the GCS 210 plants, we examined CCA1 and TOC1 transcript accumulation within guard cells. Under 211 constant light conditions, we measured CCA1 transcript accumulation in epidermal peels at 212 dusk (when CCA1 transcript abundance is normally low in the wild type) and TOC1 213 transcript accumulation at dawn (when TOC1 transcript abundance is normally low in the 214 wild type). Guard cell CCA1 overexpressors had greater CCA1 transcript abundance in 215 epidermal peels at dusk than the wild type (GC: t₄ = -2.233, p>0.05; MC: t₄ = -7.409, p = 216 0.002) (Fig. S2D), and guard cell TOC1 overexpressors had greater TOC1 transcript 217 abundance at dawn than the wild type (GT: t₄ = -6.636, p = 0.003; MT: t₄ = -2.736, p = 218 0.050) (Fig. S2D). These data indicate that CCA1 and TOC1 were overexpressed within the 219 guard cells of the guard cell-specific CCA1 or TOC1 overexpressor plants that we 220 generated, respectively. 221 We investigated the effect on WUE of overexpression of CCA1 and TOC1 within guard cells. 222 Two independent GC1::CCA1 lines (GC-1 and GC-2) were significantly more water use 223 efficient than the wild type (GC-1: p < 0.001; GC-2: p = 0.002) (Fig. 4B). GC-1 and GC-2 224 were 8% and 4% more water use efficient than the wild type, respectively (Fig. 4B). Two 225 independent MYB60::CCA1 lines also had numerically higher WUE than the wild type, but 226 this was not statistically significant (p > 0.05) (Fig. 4B). In contrast, overexpression of TOC1 227 in guard cells with both the GC1 and MYB60 promoters did not alter WUE (p > 0.05) (Fig. 228 4B). Together, these data suggest that overexpressing CCA1 in guard cells can increase 229 whole plant long-term WUE. 230 A previous study identified that constitutive overexpression of TOC1 (TOC1-ox) reduces the 231 dehydration tolerance of seedlings (Legnaioli et al., 2009). We wished to determine whether 232 this altered dehydration tolerance is due specifically to the circadian oscillator within guard 233 cells. Using a similar experimental system to Legnaioli et al. 2009, we found that 234 MYB60::CCA1 and GC1::CCA1 increase dehydration survival (Fig. 4C). In contrast, 235 GC1::TOC1 and MYB60::TOC1 had decreased dehydration survival relative to the wild type 236 ( Fig. 4C). This suggests that overexpressing CCA1 or TOC1 in guard cells can increase or 237 decrease survival to dehydration under constant light conditions, respectively. 238 Like MYB60::CCA1 and GC1::CCA1, more seedlings constitutively overexpressing CCA1 239 (CCA1-ox) survived dehydration under our experimental conditions (Fig. 4C). Similarly, like 240 GC1::TOC1 and MYB60::TOC1, more seedlings overexpressing TOC1 constitutively (TOC1-241 ox) were killed by dehydration (Fig. 4C). Therefore, manipulation of the expression of these 242 clock genes in guard cell and whole plants causes similar phenotypes, with some 243 differences in magnitude (Fig. 4C). One interpretation is that altered dehydration survival in 244 CCA1-ox and TOC1-ox seedlings might be partly or wholly due to the circadian clock that is 245 specifically within guard cells. Because the stomatal density was unaltered relative to the 246 wild type in the guard cell overexpressors of CCA1 and TOC1 (Fig. 4D, E), the WUE and 247 dehydration survival phenotypes that we identified might be due to alterations in processes 248 within guard cells rather than due to altered stomatal density. 249

Discussion 250
Pervasive influence of the circadian oscillator upon water use efficiency 251 Our data indicate that the circadian oscillator is important for regulating the long-term WUE 252 of Arabidopsis. Misregulation of several functional subsections of the circadian oscillator 253 altered the WUE of Arabidopsis. Misexpression of morning (PRR7, PRR9, CCA1), late day 254 (GI, PRR5) and evening (TOC1, ZTL, ELF3) components of the circadian oscillator all 255 perturb WUE under our experimental conditions ( Fig. 1, Fig. 2B). Additionally, altered 256 expression of TEJ and GRP7 also alters WUE (Fig. 1). Therefore, oscillator components that 257 impact WUE are not confined to a specific region or expression phase of the multi-loop 258 circadian oscillator. Misexpression of genes encoding some proteins that provide 259 environmental inputs to the circadian oscillator (ELF3, TPS1, ZTL, KIN10; (Covington et al., 260 2001;Kim et al., 2007;Shin et al., 2017;Frank et al., 2018)) also alters WUE. Together, this 261 suggests that the entire circadian oscillator influences WUE, and that alterations in water 262 use that are caused by mutations to the circadian oscillator are not confined to a specific 263 sub-loop of the circadian oscillator or restricted to its input or output pathways. One 264 explanation for these circadian-system wide alterations in WUE relates to the nature of 265 feedback within the circadian oscillator. The complex feedback and interconnectivity of the 266 circadian oscillator means that individual components of the circadian oscillator that directly 267 influence stomatal function or water use are likely to be altered by mutations that are distal 268 to that component. Therefore, if correct circadian timing is required for optimum water use 269 efficiency, multiple components of the circadian oscillator are likely to influence water use found that tps1-11 and tps1-12 had lower long-term WUE than the wild type ( Fig. 1). 277 Reduced biomass accumulation in tps1-11 and tps1-12 ( Fig broad range of phenotypes that are altered in tps1-11, tps1-12 and KIN10-ox 6.5 indicates 282 that these genotypes might alter WUE through mechanisms other than circadian regulation. 283

Multiple physiological causes of altered WUE in circadian oscillator mutants 312
Our data suggest that changes in WUE caused by misexpression of circadian clock 313 components might be due to a combination of physiological factors. Some mutants or 314 overexpressors tested alter biomass accumulation, whilst others predominantly alter water 315 loss (Fig. 2), so mutations to the circadian oscillator did not alter water use by specifically 316 altering one of carbon assimilation or transpiration. This is consistent with previous work 317 demonstrating that both stomatal opening and CO 2 fixation is perturbed in circadian 318 arrhythmic plants under light/dark cycles (Dodd et al., 2005), and with the findings that daily 319 carbohydrate management is dependent upon correct circadian regulation (Graf et al., 320 2010). We speculate that delayed or advanced stomatal and photosynthetic responses to 321 the day-night cycle might occur in circadian period mutants, because period mutants 322 inaccurately anticipate the onset of dawn (Dodd et al., 2014). Circadian clock mutants might 323 also affect WUE by changing the sensitivity of stomatal movements and photosynthesis to 324 environmental transitions, because there is circadian gating of the responses of both 325 stomata and photosynthesis to environmental cues (Dodd et al., 2006;Kinoshita et al., 2011;326 Litthauer et al., 2015;Joo et al., 2017;Cano-Ramirez et al., 2018). Some effects of the 327 circadian oscillator upon WUE arise from alterations in leaf size that occur in some circadian 328 oscillator mutants (Fig. 3A  between the SpeI and NotI restriction sites in the pGREENII0229 binary vector (Hellens et 373 al., 2000). The GC1 upstream sequence (-1894 to -190) or MYB60 upstream sequence (-374 1724 to -429) was then ligated between the KpnI and ApaI restriction sites of 375 pGREENII0229. Finally, the CCA1 coding sequence or TOC1 coding sequence, obtained 376 using RT-PCR, was ligated between the restriction XhoI and XmaI sites. Primers used are 377 identified in Table S2. Constructs were transformed into Col-0 wild type Arabidopsis using 378 transformation with Agrobacterium tumefaciens strain GV3101. Transformants were 379 identified by screening for phosphinothricin resistance, then further validated using genomic 380 DNA PCR. Homozygous lines were identified via phosphinothricin (BASTA) resistance, and 381 two independently transformed homozygous lines were investigated in detail per genotype. 382 Guard cell specificity of promoter activity was investigated using GC1::GFP:nos and 383 MYB60::GFP:nos promoter-reporter lines (Sup. Fig. 3A-C), which were created as above 384 with the GFP coding sequence ligated between the XhoI and XmaI restriction sites. Leaf 385 discs (5 mm diameter) from seedlings or mature plants were mounted on microscope slides 386 with dH 2 O, and examined for GFP fluorescence using confocal microscopy (Leica DMI6000). 387 The following settings were used: argon laser at 20% capacity, 488 nm laser at 48% 388 capacity with a bandwidth of 505 nm-515 nm, gain of 1250, offset at 0.2%, 20x or 40x 389 objective, zoom x1 to x4. 390

Measurement of water use efficiency 391
The WUE assay was adapted from Wituszynska et al. Q water (Fig. S3). Each Falcon tube lid had a 2 mm diameter hole drilled in its centre to 396 allow plant growth. The lid was spray-painted black (Hycote) because we found that the 397 orange colour of the Falcon tube lid caused leaf curling (Fig. S3). The system was wrapped 398 in aluminium foil to exclude light (Fig. S3). 10-15 seeds were sown through the Falcon tube 399 lid using a pipette. Following stratification, Falcon tube systems were placed under growth 400 conditions using a randomised experimental design. 7 days after germination, seedlings 401 were trimmed to one per Falcon tube system, and initial Falcon tube weight was recorded. 402 After 6 weeks of growth, rosette leaf surface area was measured by photography (D50; 403 Nikon) and Fiji software, rosette dry weight was measured (4 d at 60°C), and final Falcon 404 tube weight was recorded. Negative controls (Falcon tube systems without plants) were 405 used to assess soil water evaporation over 18 experimental repeats, with an overall mean 406 weight loss of 0.513 g ± 0.004 g over 6 weeks for plant-free Falcon tubes. 407 Plant WUE was calculated as follows: 408 Where d is the rosette dry weight at the end of the experiment (mg), t i and t f are the falcon 409 tube weight at the start and end of the experiment, respectively (g), and e is the amount of 410 water evaporation directly from the compost (g). WUE is derived as mg biomass per ml -1 411 water lost. These calculations assumed that 1 g of weight change was equivalent to a 412 change of 1 ml of water. For each of 3 independent experimental repeats, 15 plants were 413 screened per genotype. Due to variation between the WUE of each background (Fig. S1), 414 the WUE of each circadian oscillator genotype was normalized to its respective background 415 and expressed as a percentage of that background. Statistical comparisons with the wild 416 types were conducted before this normalization. 417

Dehydration tolerance assay 418
This assay was adapted from Legnaioli et al. (2009). For experiments investigating survival 419 to dehydration, surface-sterilized seeds were sown on Petri dishes containing half strength 420 agar and 3% (w/v) sucrose, then stratified for 3 days at 4 °C before transfer to the growth 422 chamber. For these experiments, seedlings were cultivated in MLR-352 growth chambers 423 (Panasonic) at 19°C with photon flux density of 120 µmol m -2 s -1 . 14-day old seedlings were 424 dehydrated on a double layer of filter paper (Fisher Scientific) for 9 h under constant light 425 conditions, watered with sterile dH 2 O, and kept under constant light conditions for a further 426 48 h before being scored for survival. Seedlings with a green apical meristematic region 427 were counted as survivors. 32 seedlings were treated per genotype and within each 428 experimental repeat. 429

Measurement of stomatal density 430
Plants were grown for 7-8 weeks on compost mix. Dental paste (Coltene) was applied to the 431 abaxial surface of fully expanded leaves. Transparent nail varnish (Rimmel) was applied to 432 these leaf moulds once they had set, and then peeled away from the mould using clear 433 adhesive tape (Scotch Crystal). Stomatal and pavement cells were counted within an 434 800 µm x 800 µm square at the centre of each leaf half, using an epifluorescence 435 microscope (HAL100; Zeiss) and Volocity (Perkin Elmer) and Fiji software. EvaGreen qPCR mastermix (Solis Biodyne). qRT-PCR primers are provided in Table S3.   can increase WUE. WUE expressed as a percentage of the wild type (normalised to 100%, 521 red reference line). Two to four independent experimental repeats were performed, with data 522 from one representative dataset shown (n = 5 -15). Data for CCA1-ox and TOC1-ox are 523 derived from Fig. 1, for purposes of comparison. Data were analysed with independent 524 samples t-tests, and statistical significance compared to Col-0 is indicated using starring (** 525 = p < 0.01; *** = p < 0.001). (C) Guard cell CCA1 or TOC1 overexpression alters 526 dehydration survival of seedlings compared with the wild type. Data were obtained from 527 three independent experimental repeats (mean; n = 32 per experimental replicate; at least 528 two independent experimental repeats were performed for each genotype). A single 529   Figure 2. Altered WUE of plants with mutations or overexpression of circadian clock associated genes is not caused consistently by variation in one of dry weight, water use, phase of expression of each gene, or resultant altered period or flowering time. Data are derived from Fig. 1 and expressed as a percentage of the respective background (WT, normalised to 100%, red reference line) (n = 5 -15). (A) Altered WUE is not specifically due to altered water use or altered dry weight of screened genotypes, but results from the combination of both.  those for which period and/or flowering time are unknown are included on the right. Studies describing the phase of expression, period and flowering time of the genotypes tested are identified in the main text. We note that the phase of expression and period data used for this analysis were often obtained under constant conditions, in contrast to our experiments occurring under light/dark cycles. Altering circadian-associated gene expression can affect rosette architecture and size, as illustrated for elf3-1, lux-1, and gi-2 in (Col-0 background). Image backgrounds removed for clarity. Variation in rosette leaf surface area across the genotypes investigated explained (B) 16% of variation in WUE (p < 0.001, r = 0.400, r 2 = 0.160), (C) 83% of variation in transpiration (p < 0.001, r = 0.912, r 2 = 0.832) and (D) 73% of variation in rosette dry biomass (p < 0.001, r = 0.857, r 2 = 0.734). Data were analysed using Pearson correlation tests. Water used (ml) Lower WUE (***) than wild type Lower WUE (**) than wild type Lower WUE (*) than wild type No WUE difference from wild type Greater WUE (*) than wild type Greater WUE (***) than wild type B C D

Rosette weight (mg)
Leaf area (cm2) Leaf area (cm2) Leaf area (cm2) Figure 4. Overexpressing CCA1 or TOC1 in guard cells affects WUE and survival of dehydration by seedlings. (A) Constructs used to overexpress CCA1 or TOC1 coding sequence under control of GC1 or MYB60 promoters. (B) Guard cell CCA1 overexpression can increase WUE. WUE expressed as a percentage of the wild type (normalised to 100%, red reference line). Two to four independent experimental repeats were performed, with data from one representative dataset shown (n = 5 -15). Data for CCA1-ox and TOC1-ox are derived from Fig. 1, for purposes of comparison. Data were analysed with independent samples t-tests, and statistical significance compared to Col-0 is indicated using starring (** = p < 0.01; *** = p < 0.001). (C) Guard cell CCA1 or TOC1 overexpression alters dehydration survival of seedlings compared with the wild type. Data were obtained from three independent experimental repeats (mean; n = 32 per experimental replicate; at least two independent experimental repeats were performed for each genotype). A single GC1::TOC1 line is shown here because other lines produced extremely variable data. (D, E) Guard cell CCA1 or TOC1 overexpression does not affect (D) stomatal index nor (E) stomatal density. Two independent experimental repeats were performed, with data from one representative dataset shown (n = 19 -32; mean ± S.E.M.). Data were analysed with ANOVA and Tukey's post hoc tests (NS = p > 0.05). Bar colours identify the whole plant overexpressor control (black), wild type control (dark grey), and guard cell-specific overexpressor genotypes (light grey).