The increase of crop productivity face to climate change is a major societal challenge. Yield losses due to abiotic constraints, in an example of water deficit, range from 50 to 80% and even exceeded those caused by biotic stresses. In terms of fundamental research, drought responses of model plants and crops have long been studied for the purpose of improvement and/or survival under the conditions of severe water deficit (SD). However, this tolerance/survival predicts neither better growth capacity nor the enhancement of productivity. In the last decade, the global vision of research focused on genes, which results in the elevation of biomass and the yield of seeds under the conditions of moderate water stress (Skirycz et al., 2011; Raza et al., 2019; Zhao et al., 2020). To cope up with water deficit, plants have developed different strategies such as escape, avoidance, and osmotic adjustment strategies (Arve et al., 2011; Fang and Xiong, 2015; Ilyas et al., 2020). The escape strategy deals with the completion of the plant life cycle at the onset of drought, which reflects a high degree of developmental plasticity (Chaves et al., 2003; Lawlor, 2013; Marín-de la Rosa et al., 2019). The avoidance strategy deals with the ability of plants to grow during periods of drought and to maintain their water status by: (1) the limitation of water loss by reducing leaf surface area and stomatal conductance (SC), and, by consequence, transpiration; (2) the optimization of water supply with deeper roots, which requires the reallocation of photoassimilates in favor of root growth, thereby increasing their water absorption capacity; and (3) the storage of water in specialized organs (Griffiths and Males, 2017). The osmotic adjustment strategy is used by plants to cope up with water deficit by accumulating compatible osmolytes (proline, sucrose, polyols, etc.) to maintain the water potential of the cells and also the turgor pressure to support their growth (Hare and Cress, 1997; Hare et al., 1998; Xiong and Zhu, 2002; Moore et al., 2008; Lawlor, 2013; Ilyas et al., 2020). In this context, the accumulation of soluble sugars has been observed in the leaves of Arabidopsis thaliana during water deficit (Taji et al., 2002; Hummel et al., 2010; Mewis et al., 2012; Sperdouli and Moustakas, 2012). Under mild and moderate water deficit (MD), SC decreases, but a partial closure of stomata does not prevent photosynthesis activity. However, the efficiency of photosynthesis is reduced under more severe water stress (Lawlor, 2002; Lawlor and Cornic, 2002; Muller et al., 2011; Zhao et al., 2020).

In plants, subcellular sugar partitioning and storage require the activity of sugar transporters (SUTs) located on diverse membranes especially vacuoles (Hedrich et al., 2015). SUTs are classified into three main families: sucrose transporters/SUTs (SUCs/SUTs), sugar will eventually be exported transporters (SWEET), and monosaccharide transporters (MSTs). SUC/SUT carriers are mainly H⁺/sucrose symporters localized on the plasma membrane (Meyer et al., 2004; Sauer, 2007; Sivitz et al., 2007, 2008; Reinders et al., 2012), except SUT4 carriers (Type III SUTs), which are targeted to the tonoplast (Endler et al., 2006; Reinders et al., 2008, 2012; Eom et al., 2011). The SWEET family has been more recently identified. The family members characterized as uniporters are involved in the flux of sucrose, glucose, and/or fructose (Chen et al., 2010; Eom et al., 2015; Jeena et al., 2019). AtSWEET2, AtSWEET16, and AtSWEET17 have been described as vacuolar facilitators with a predominant expression in root vacuoles (Guo et al., 2014). AtSWEET2 has been described to transport the glucose analog 2-deoxyglucose. It is expressed in the cortex and epidermis and is probably involved in the sequestration of glucose in root vacuoles, thereby preventing sugar secretion into the rhizosphere (Chen et al., 2015). AtSWEET16 transports glucose, fructose, and sucrose; it is expressed in vascular parenchyma cells and is downregulated by sugars, cold, osmotic stress, and low nitrogen. AtSWEET16 overexpression improves germination, growth efficiency, and tolerance against freezing temperatures (Klemens et al., 2013). AtSWEET17 has first been described as a leaf vacuolar exporter specific for fructose (Chardon et al., 2013). It is also strongly expressed in root cortex cells, by acting as a bidirectional fructose facilitator to mediate fructose uptake into the vacuole for the storage or export of fructose to the cytosol to maintain cell metabolism (Guo et al., 2014). The family of MSTs is divided into seven subfamilies. Three of them have the members that are located on the tonoplast: tonoplastic monosaccharide transporters/tonoplast sugar transporters (TMTs/TSTs), vacuolar glucose transporters (VGTs), and early response to dehydration six-like (ESL) (Büttner, 2007; Slewinski, 2011). AtTMT1 and AtTMT2 have been suggested to function as H⁺/glucose antiporters (Wormit et al., 2006) and later on have also been characterized as H⁺/sucrose antiporters (Schulz et al., 2011). Expressed in most plant tissues, they are regulated by sugars and abiotic stresses such as cold, salt, and drought. The latter is corroborated in TMT1/TMT2 knockout lines, which have shown reduced capacity to accumulate glucose and fructose under cold stress (Wormit et al., 2006). The overexpression of TMT1 in A. thaliana modifies the subcellular sugar compartmentation and increases sugar export capacity from the source to sink tissue leading to improved seed production (Wingenter et al., 2010). AtVGT1, the only characterized member of this subfamily, is expressed at low levels in green tissues and strongly in pollen. It has been localized on the tonoplast and described to be involved in an energy-dependent transport of glucose and fructose. atvgt1 knockout mutants show a decrease in seed germination and a delay in flowering, which imply the involvement of this transporter in developmental processes (Aluri and Büttner, 2007). In A. thaliana, 19 ESL transporters have been identified. Although they constitute the largest MST subfamily in Arabidopsis, their role is poorly understood. To get insights into this issue, Slawinski et al. (2021) applied an evolutionary genomics approach to study ESL transporters in the genome of 63 species representative of the plant kingdom, from algae to angiosperms, and suggested that these transporters might have emerged from a common streptophyte ancestor, and have evolved through diversification events and tandem duplications. Phylogenetic analysis of the 519 identified protein sequences showed that these ESL transporters belong to three subgroups, ESL1, ESL2, and ESL3, and based on their phylogenetic position, a new nomenclature has been suggested to reflect their evolution. The expression profiling of Arabidopsis ESL genes in different organs (leaves, roots, flower buds, flowers, and siliques) has shown that ESL1 and ESL2 genes are, respectively, moderately and weakly expressed, whereas ESL3 display a high expression diversity. This study has demonstrated differential expression patterns between the copies of the ESL tandem pairs, suggesting a functional diversification of these transporters (Slawinski et al., 2021).

Because the identification of the first ESL gene (ESL3.08/ERD6 – At1g08930) via differential screening of a complementary DNA (cDNA) library prepared with A. thaliana plants exposed to dehydration (Kiyosue et al., 1994), only few ESL genes have been characterized. These transporters are involved in plant development and responses to various abiotic stresses. In 4-week-old A. thaliana plants, the level of AtESL3.08/ERD6 messenger RNA (mRNA) seems very low but is transiently increased in response to dehydration and cold (4°C) (Kiyosue et al., 1998). AtESL3.08/ERD6 expression in leaves and roots is similar under normal growth conditions, but decreases in leaves under high salinity (250 mM NaCl) conditions and after 100 μM abscisic acid (ABA) treatment (Yamada et al., 2010). At present, AtESL3.08/ERD6 is considered to take part in the redistribution of sugars, thereby protecting plant cells from the detrimental effects of dehydration and cold stresses (Kiyosue et al., 1998).

AtESL3.07/ESL1 (ERD six-like-1—At1g08920) is a tandem duplicated gene of AtESL3.08/ERD6 and are located on chromosome 1. In 3-week-old A. thaliana plants, AtESL3.07/ESL1 expression is induced after 1-h treatment by dehydration, high salinity (250 mM NaCl), and ABA (100 μM). Without stress, this expression is higher in roots than in leaves and it is highly induced in roots under high salinity and ABA supplementation. AtESL3.07/ESL1 protein has been described as a low-affinity facilitator capable of transporting different hexoses such as glucose, fructose, galactose, mannose, and xylose (Yamada et al., 2010).

AtESL3.13/AtSFP1 and AtESL3.14/AtSFP2 (sugar-porter family protein 1 and 2—At5g27350 and At5g27360) are the two tandem duplicated genes located on chromosome 5. These two genes display different expression patterns during leaf development. AtESL3.13/AtSFP1 is detected in seedlings 9 days after germination but not in mature plant organs (leaves, flowers, flower buds, stem, and roots), while AtESL3.14/AtSFP2 is expressed in all tested organs and in seedlings. Further analysis shows that only AtESL3.13/AtSFP1 is induced during leaf senescence while the expression of AtESL3.14/AtSFP2 remains stable during this process (Quirino et al., 2001).

AtESL1.02/ERDL6 (early responsive to dehydration-like six—At1g75220) is a vacuolar H⁺/glucose symporter involved in the export of glucose from the vacuole to cytosol (Klemens et al., 2014). This expression is regulated by the mobilization of vacuolar carbohydrate reserves: an upregulation by darkness, heat stress, and wounding and a downregulation under cold stress and glucose supply. atesl1.02 mutant presents a higher vacuolar glucose content (>90%) than the wild-type mutant (86%), and cell lines overexpressing ESL1.02/ERDL6 have glucose content lower than that of wild plants. These data indicate that ESL1.02/ERDL6 would act as an exporter of vacuolar glucose during developmental phases with a high metabolic turnover, such as seed germination, abscission zones, and wound responses (Poschet et al., 2011). In apple, MdERDL6-1, a homolog to ESL1.02/ERDL6 in Arabidopsis, has recently been identified. It is highly expressed in fruits, and its encoded protein seems to be localized on the tonoplast, where it acts as a low-affinity H⁺/glucose symporter. In an inverse relation to AtERDL6, the overexpression of MdERDL6-1 in apple and tomato leaves and fruits leads to an increase rather than a decrease in sugar levels (Zhu et al., 2021).

AtESL2.01/AtZIF2 (zinc-induced facilitator—At2g48020) is a tonoplastic transporter involved in the vacuolar sequestration of zinc at the root level. Transporter function loss causes hypersensitivity to Zn, whereas its overexpression increases tolerance (Remy et al., 2014). Until now, AtESL2.01/AtZIF2 capacity of transporter to transport sugars is not demonstrated.

More recently, Desrut et al. (2020) have shown that six AtESL genes are differentially regulated in the presence of plant growth-promoting rhizobacteria (PGPR): ERD6-like7, ERD6-like12, and ERD6-like16, as well as ERD6-like18, are downregulated in shoots and roots, respectively, whereas ERD6-like13 and ERD6-like15 are upregulated in roots in the response of PGPR.

Furthermore, Breia et al. (2020) have shown that the grapevine VvERD6l13 is induced in the grape berries upon infection by necrotrophic or biotrophic pathogens such as Botrytis cinerea and Erysiphe necator. VvERD6l13 is the first ESL described to be localized in the plasma membrane as demonstrated in tobacco epidermal cells. More surprisingly, using yeast as a heterologous system, VvERD6l13 has been characterized as a low-affinity H⁺/sucrose symporter.

In this study, we characterized the integrative physiological response of A. thaliana Col-0 plants to gradual water deprivation and established the following phases of the water status of rosette leaves, i.e., “early mild water deficit” (EM), “MD,” “SD,” and “wilting” (Wi). To explore the involvement of AtESL genes in plant response to water deficit, we evaluated growth, physiological, and biochemical parameters in parallel with the levels of AtESL expression. Furthermore, transfer DNA (T-DNA) insertional mutants for the four ESL genes upregulated in response to water deficit were identified, and characterized under well-watering, water deprivation, and re-watering. To get global insights into sugar exchanges during water deficit between two subcellular compartments, vacuole, and cytosol, the expression profiles of known SUTs located on the tonoplast (AtTMT, AtVGT, AtSUC, and AtSWEET) in leaf cells were established, as well as those of two vacuolar invertases (AtβFRUCT).

The present study aims to correlate the expression of AtESL SUTs in the response of A. thaliana to water deficit intensity, in terms of growth, physiological, and leaf sugar metabolic behavior, to identify their role in plant response to water deficit. Using 35-day-old Arabidopsis plants and inducing water stress by withholding water, we were able to define four different phases in the stress period and analyzed, for each phase, the regulation of AtESL transporter genes (Figure 11).

In the EM phase (WC ranged between 98 and 92.2%), the physiological parameters and sugar contents for WD plants did not display any significant differences compared to those of WW plants. Thus, the progressive leaf water loss (5.8%) has not yet affected the plant phenotype (Figures 1–3). However, AtESL3.08/ERD6 gene repression corresponded to the onset of EM phase and was maintained all along the water-deficit period. Therefore, this ESL may be considered as the earliest responsive gene, which might be regulated by initial changes in water potential before the appearance of any changes in growth and physiological parameters. However, even though AtESL3.08/ERD6 has been described not only as the first identified ESL but also as an early drought responsive gene, our results disagree with these previous reports in the fact that the expression of AtESL3.08/ERD6 was described to be induced after few hours of dehydration (Kiyosue et al., 1998; Seki et al., 2002; Yamada et al., 2010). This discrepancy might be explained by the fact that in those previous studies water deficit was performed via dehydration instead of water withholding as in our experiments. Furthermore, we were able to show that, in WW soil cultured plants, AtESL3.08/ERD6 was the highest expressing ESL in Arabidopsis organs, such as leaves, roots, buds, flowers, and siliques and that its strongest expression level is observed in roots (Slawinski et al., 2021). Using ERD6pro:GUS and ERD6pro:TM-GFP transgenic plants, Yamada et al. (2010) have demonstrated a preferential expression of ESL3.08/ERD6 in epidermal and cortex cells. ESL3.08/ERD6 root localization suggests that its rapid repression may be a part or a consequence of the early drought perception in roots.

In the MD phase, WC decreased from 92.2 to 90% (i.e., a net 2.2% water loss), thereby reaching a cumulative water loss of 8%, which caused stomatal closure as an early plant response to maintain the plant water state under reduced soil water potential (Costa França et al., 2000; Flexas et al., 2004; Blouin et al., 2007; Harb et al., 2010; Arve et al., 2011). ABA is known as the major plant hormone with a double function involved not only in stomatal closure, but also in the prevention of stomatal opening (Liu et al., 2005; Seki et al., 2007; Arve et al., 2011; Živanović et al., 2020). The steps that occur between the sensing of drought and the production of ABA are complex vary among plant species and are still not completely understood. Various signals, such as hydraulic pressure, ROS/Ca²⁺ waves, and peptides, contribute to this response to drought (Takahashi et al., 2020). In Arabidopsis, the peptide CLAVATA3/EMBRYO-SURROUNDING REGION-RELATED25 (CLE25) showed to function as a long-distance signal from roots to shoots, and causes ABA production in leaves (Christmann and Grill, 2018). In our experiments, the MD phase was associated with the induction of three SUT genes ESL3.03, ESL3.05/ESL3, and ESL3.07/ESL1, with an enhancement of the repression of ESL1.01 and the maintained repression of ESL3.08/ERD6. These events marked also the onset of glucose and sucrose accumulation in leaves. Our results imply that the three induced ESLs (ESL3.03, ESL3.05/ESL3, and ESL 3.07/ESL1) might be involved in the accumulation of sugars. Furthermore, we were able to show that among the three induced genes only ESL3.07/ESL1 was clearly induced when 750 μM ABA was sprayed on leaves (Figure 6). A similar induction was also observed in leaves and roots, 5 h after 100 μM ABA treatment (Yamada et al., 2010). This suggests that, in response to water deficit, the induction of ESL3.07/ESL1 in leaves might involve an ABA-dependent pathway, whereas the induction of ESL3.03 and of ESL3.05/ESL3 implies an ABA-independent pathway. It is interesting to note that in the MD phase, the expression of ESL3.07/ESL1 and ESL3.08/ERD6 were evolving in opposite ways. In fact, the gene expression data analysis from Genevestigator databases showed that ESL3.07/ESL1 is highly induced by ABA treatment, whereas ESL3.08/ERD6 is strongly repressed. According to these data, Yamada et al. (2010) have shown that the expression of ESL3.08/ERD6 is reduced in leaves and roots after ABA treatment. These results corroborate the idea of a differential regulation by ABA and water deficit of these two tandem duplicated genes and highlight the possible mechanisms of sub-functionalization.

In the SD phase (WC from 90 to 86%), although FW and PLA decreased, TW stayed stable and DW still increased as in WW plants (Figure 2). The osmotic pressure (Table 1) rose in parallel with a strong increase of glucose and sucrose content (Figure 4). This suggests that plants, which do not present Wi phenotype, still have the capacity to maintain basic metabolism despite the closure of stomata. Our microscopy observations confirmed that the cells were not only viable but also remained turgid until 47 dps (Figure 1C). Thus, in the SD phase, an important increase in glucose and sucrose amount contributes to an osmotic adjustment of cells to maintain turgid and metabolically active cells (Morgan, 1984; Hare et al., 1998; Taji et al., 2002; Moore et al., 2008; Hummel et al., 2010; Blum, 2017). Leaf expansion is certainly affected by the arrest of cell growth under enhanced dehydration, leading to the drop of cell water potential and the loss of turgidity (Hsiao, 2003; Simonneau et al., 2017; Turner, 2018). Taken together our results agree with the already established concept that the arrest of cell growth responds to a milder water deficit than that required for a photosynthetic rate reduction (dos Santos Gouvêa and Marenco, 2018; Zhao et al., 2020). Consequently, the limitation of carbon assimilation is not the real cause of cell growth arrest. The latter is mainly related to the cell choice to invest metabolites and energy for its osmotic adjustment, the signaling of reduced water potential, and cell wall stiffening to promote plant acclimation. In this context, Mewis et al. (2012) reported that drought stress enriched the nutritional quality of phloem sap in A. thaliana plants with 93–94% WC by increasing the amounts of free amino acids and sugars, especially sucrose. In the SD phase, the expression levels of ESL3.03, ESL3.05/ESL3, and ESL 3.07/ESL1 were higher than those in the MD phase, and the expression of TMT2 was significantly increased. On the contrary, the expression of ESL1.01 and SWEET16 were reduced, and the expression levels of ESL3.08/ERD6, ESL2.01/ZIF2, and ESL3.04 were minimal. This confirms that ESL3.03, ESL3.05/ESL3, and ESL 3.07/ESL1 are involved together with TMT2 in sugar accumulation for an osmotic adjustment.

In the beginning of the Wi phase (WC ranged between 86 and 76%), all measured physiological parameters decreased except DW, which remained stable. The amount of sucrose in the Wi phase was doubled compared to that in the SD phase and reached a maximum. During this EWi phase, re-watering was allowed to restore all physiological parameters at least to the levels observed in the MD phase. The repression of ESL1.01, ESL2.01/ZIF2, ESL3.04, ESL3.08/ERD6, and SWEET16 genes in the SD phase was maintained at a lower level. In parallel, the expression of six other SUTs (ESL2.02, ESL3.01, ESL3.13/SFP1, TMT1, SWEET17, and SUC4) was decreased. This suggests that sugar transport is highly affected. For a WC lower than 76%, all measured growth and physiological parameters, including DW, dropped down. The WD plants at the LWi step persisted its ability to absorb water even at a WC below 76%. Below this threshold, massive cell plasmolysis was observed in leaf sections and Wi became irreversible for some leaves (Figures 1B,C). Glucose content was still maintained (although a little bit decreased), whereas sucrose content increased 2-fold in comparison to that in SD phase and fructose significantly increased. As the accumulation of soluble sugars has been shown to increase during senescence in several plant species, including Arabidopsis (Quirino et al., 2001; Stessman et al., 2002; Diaz et al., 2005) and that sugars are considered as a metabolic signal to trigger senescence (Pourtau et al., 2006; Rolland et al., 2006), it is tempting to suggest that the wilted plants having a WC lower than 76% were going to enter the senescence stage. However, this might not be really the case as we observed the repression of ESL3.13/SFP1, which has been described to be specifically induced during senescence (Quirino et al., 2001). In the LWi phase, the expression of ESL3.02 and of the vacuolar invertase Atβfruct4 was strongly repressed. Remarkably, among the studied genes, AtESL1.02/ERDL6 was the only one to be specifically induced under SD. This ESL has been described to be a vacuolar proton-driven glucose exporter (Klemens et al., 2013). This gene expression is upregulated when the changes in cell status require rapid energy mobilization through the export of vacuolar sugar reserves (a transition from light to obscurity, a temperature shift from 23 to 37°C and after injury). Inversely, AtESL1.02/ERDL6 expression is downregulated under the conditions favorable for sugar accumulation into the vacuole, such as cold stress and a high concentration of extracellular sugars (Poschet et al., 2011). These data and the characterization of AtESL1.02/ERDL6 as a H⁺/glucose symporter provide evidence for the correlation between AtESL1.0/ERDL6 expression and the sugar status of the cell (Poschet et al., 2011; Klemens et al., 2013). Under strong water deficit, AtESL1.02/ERDL6 may be involved, in glucose efflux from the vacuole to cytosol, which is required for the remobilization of sugars from mature wilted leaves and reallocation to young sink organs.

Taken together, our results demonstrate that among the 17 studied AtESL genes, 12 respond to water deficit, 4 with an increased expression and 8 with a decreased expression. Another important result concerns the fact that tandem duplicated ESL copies did not share the same expression pattern dependent on the plant water status. AtESL2.02 showed a reduced expression at WC < 86% while ESL3.11 was insensitive. AtESL3.02 also showed a reduced expression at WC < 76%, while AtESL3.03 increased as WC was lowered from 92.2 to 90%. AtESL3.05/ESL3 expression increased in the MD phase (92.2% > WC > 90%), while AtESL3.06/ESL2 seemed to be insensitive to water depletion. AtESL3.07/ESL1 was upregulated for WC comprised between 92.2 and 90%, whereas AtESL3.08 was downregulated from the EM phase. AtESL3.13/SFP1 was repressed at WC < 86%, but AtESL3.14/SFP2 was not regulated by water depletion. These differential responses of tandem duplicated ESL genes provide arguments in favor of a high diversification of AtESL roles in the response of plants to water deficit.

To elucidate the roles of the four upregulated AtESL genes using water deficit, T-DNA insertional mutants were identified and characterized under normal growth conditions and water deficit. However, none of the single atesl mutants displayed a specific phenotype under water deficit in comparison to wild-type mutants (Supplementary Figure 2). These results were predictable as the ESL transporters belong to a multigenic subfamily. Moreover, the tonoplast location of other SUTs (SWEETs, SUCs/SUTs, TMTs, and VGT) implies possible compensatory effect(s) between them.

In WW atesl1.02/erdl6 mutant plants, the repression of AtESL1.02/ERDL6 gene resulted in higher leaf glucose content than that in the wild genotype. This difference was observed in WD plants, until the MD phase, and in RW plants. This increase of glucose in atesl1.02/erdl6 mutant happens due to glucose accumulation into the vacuole as AtESL1.02/ERDL6 mutation avoids glucose efflux toward the cytosol (Poschet et al., 2011). In SD and Wi phases, no significant differences in the glucose content could be detected between atesl1.02/erdl6 mutant and Col-0. This indicates that even though the mutant atesl1.02/erdl6 presented higher glucose content in leaves under SD and EWi, it was still able to accumulate glucose involving an AtESL1.02/ERDL6 transport activity. Glucose accumulation in these two phases was certainly due to the activity of ESL3.03, ESL3.05/ESL3, and ESL3.07/ESL1. However, it is worth to note that the expression of AtESL3.07/ESL1 was upregulated from the beginning of the MD phase in the wild genotype, while in atesl1.02/erdl6 mutant its induction was delayed at the Wi phase (WC < 86%). These results suggest that in the mutant atesl1.02/erdl6 the upregulation of AtESL3.07/ESL1 expression may be inhibited by high glucose content in MD and SD phases. In Wi esl1.02/erdl6 plants, the evolution of glucose content is similar to that measured in Col-0, which implies the involvement of other SUTs for glucose export from the vacuole. The slightly induced expression of AtESL3.05/ESL3 in comparison to Col-0 plants suggests that this gene might compensate the repression of AtESL1.02/ERDL6. One can also imagine that in response to SD ESL3.07/ESL1, which is induced at this stage similar to the level in the wild type, could probably be involved in hexose remobilization into the cytosol. As AtESL3.07/ESL1 has been characterized as a low-affinity facilitator for hexose transport (Yamada et al., 2010), our hypothesis implies that the aforementioned transporters could exert a bidirectional control of glucose fluxes between the vacuole and cytosol.

In conclusion, Arabidopsis and other plants respond in a similar way to water deficit with a decrease of WC, SC, FW, PLA, and DW and a concomitant increase of sugars and other compatible osmolytes (Mewis et al., 2012; Sperdouli and Moustakas, 2012; dos Santos Gouvêa and Marenco, 2018; Zhao et al., 2020). Although more than 1,000 genes have been identified to be involved in drought response (Seki et al., 2002; Shinozaki and Yamaguchi-Shinozaki, 2007; Lawlor, 2013; Fang and Xiong, 2015), only a few of them have been characterized in terms of induced tolerance to water depletion, combined with enhanced plant productivity in model plants, as well as in important agricultural crops (Skirycz et al., 2011). In A. thaliana, the most studied genes, whose expression was regulated in response to moderate water stress, are dehydrins, LEA, aquaporins, K⁺ ionic channels, and ESL SUTs. Water depletion responsiveness of ESL genes and their putative localization on the tonoplast, already demonstrated for a few of them, assumed the functional importance of these transporters through their involvement in subcellular sugar partitioning. In our study, the profiling of 17 Arabidopsis ESL genes revealed that 12 are responsive to water deficit, 4 upregulated, and 8 downregulated. These 12 AtESLs genes were differentially expressed depending on the physiologically defined phases of the plant water status. The comparison of the four atesl single mutants and the wild-type Col-0 under well-watering and water-deprivation growth conditions did not reveal any phenotypic difference in response to water deficit. The latter is probably due to the functional redundancy of ESLs and their synergistic actions; therefore, it will be helpful to characterize multiple mutants for ESL and/or other tonoplastic SUTs. The differential expression of each of the tandem duplicated AtESL genes in response to water stress is in favor of their plausible functional diversity. Our results corroborate the hypothesis of the acquisition of new physiological functions by the ESLs, which favors plant plasticity to cope up with environmental constraints.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AI, FD, LS, ML, and RA designed the experiments, discussed the data, and wrote the article. AI, CA, FD, FT, LS, ML, and RA performed the experiments. All authors approved the published version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.