PtrVCS2 Regulates Drought Resistance by Changing Vessel Morphology and Stomatal Closure in Populus trichocarpa

Drought has severe effects on plant growth, forest productivity, and survival throughout the world. Understanding the molecular regulation of drought resistance in forest trees can enable effective strategic engineering of novel drought-resistant genotypes of tree species. In this study, we identified a gene, PtrVCS2, encoding a zinc finger (ZF) protein of the ZF-homeodomain transcription factor in Populus trichocarpa (Black Cottonwood) Torr. & A. Gray. ex Hook. Overexpression of PtrVCS2 (OE-PtrVCS2) in P. trichocarpa resulted in reduced growth, a higher proportion of smaller stem vessels, and strong drought-resistance phenotypes. Stomatal movement experiments revealed that the OE-PtrVCS2 transgenics showed lower stomata apertures than wild-type plants under drought conditions. RNA-seq analysis of the OE-PtrVCS2 transgenics showed that PtrVCS2 regulates the expression of multiple genes involved in regulation of stomatal opening and closing, particularly the PtrSULTR3;1-1 gene, and several genes related to cell wall biosynthesis, such as PtrFLA11-12 and PtrPR3-3. Moreover, we found that the water use efficiency of the OE-PtrVCS2 transgenic plants was consistently higher than that of wild type plants when subjected to chronic drought stress. Taken together, our results suggest that PtrVCS2 plays a positive role in improving drought adaptability and resistance in P. trichocarpa.


Introduction
In addition to being a source of food, fuel, and materials, forests can be used to sequester carbon dioxide through photosynthesis, affecting soils and weather around the world [1]. The vascular cambium of trees differentiates into xylem and phloem, in which xylem mainly transports water and mineral nutrients [2]. However, drought, causing loss in xylem hydraulic conductivity, critically impairs metabolic and hydraulic function [3]. When the water potential of root cells is lower than that of soil, water enters the cells in the root cortex through symplastic movement and moves into the vascular system [4]. However, under drought stress, insufficient water enters the vascular system, and gas enters a single vessel cell, forming an embolism, which travels from one vessel to another. Embolisms can cause serious damage to water transport [5]. It is usually measured by the percentage loss of conductance (PLC) [6]. The broken water column causes the water to not be transported to shoot, which can result in desiccation and mortality [7]. To cope with drought stress, trees have developed many strategies, such as reducing water loss and protecting the water transport capacity of the plant [8].
In response to the biotic stress of bacteria, fungi, viruses, and insects, plants often protect themselves by regulating some ion signals and hormone signals [9,10]. Similarly, signaling of hormones, such as abscisic acid (ABA), jasmonic acid (JA), brassinosteroid (BR), The transgenic line #3 (OE-PtrVCS2-3×FLAG #3) with the highest levels of transgene overexpression was selected and propagated with the wild type (WT) and maintained in a greenhouse. Stem cross-sections analysis revealed that the xylem vessels in OE-PtrVCS2-3×FLAG #3 plants were smaller than those in wild-type plants from the 5th stem internode ( Figure 1B, C) where stem secondary growth begins. The vessel number per unit area in OE-PtrVCS2-3×FLAG #3 plants was greater than that in wild-type plants ( Figure 1D), which resulted in an increase in the total vessel lumen area (void area) in OE-PtrVCS2-3×FLAG #3 plants ( Figure 1E).

Overexpressing PtrVCS2 Gene Results in More and Smaller Vessels in Xylem Tissue
We previously generated transgenic P. trichocarpa plants overexpressing PtrVCS2 3× FLAG (OE-PtrVCS2-3× FLAG) under the control of a CaMV 35S promoter ( Figure 1A The transgenic line #3 (OE-PtrVCS2-3× FLAG #3) with the highest levels of transgene ove expression was selected and propagated with the wild type (WT) and maintained in greenhouse. Stem cross-sections analysis revealed that the xylem vessels in OE-PtrVCS2 3× FLAG #3 plants were smaller than those in wild-type plants from the 5th stem internod ( Figure 1B, C) where stem secondary growth begins. The vessel number per unit area i OE-PtrVCS2-3× FLAG #3 plants was greater than that in wild-type plants ( Figure 1D which resulted in an increase in the total vessel lumen area (void area) in OE-PtrVCS2 3× FLAG #3 plants ( Figure 1E). , total number of vessel/ mm 2 (D), and total area (µm 2 ) of vessels/ mm 2 (E) using vessel cells from (B). Error bars represent standard errors for three independent replicates with at least 200 OE-PtrVCS2 transgenic vessel cells and 200 WT plant vessel cells for each genotype in each replicate, and asterisks indicate significant differences between the OE-PtrVCS2 transgenics and WT plants. * p < 0.05, ** p < 0.01 (Student's t-test).

Overexpressing PtrVCS2 Gene Improves Drought Resistance
Next, we tested the drought response of OE-PtrVCS2-3×FLAG #3 and wild-type plants. Four-month-old OE-PtrVCS2 and wild-type plants were subjected to drought treatments by withholding water. When plants were adequately watered, the soil was saturated, having 50% volumetric soil water content (VWC) (Figure 2A). In the absence of water replenishment, the VWC decreased to 30%, and the leaves of wild-type plants wilted, while most of the OE-PtrVCS2 transgenics remained turgid (Figure 2A). Under the severe drought condition with 15% VWC, the wild-type plants showed severe wilting symptoms, compared with the OE-PtrVCS2 transgenics ( Figure 2A). After rehydration in 3 days, the survival rate for the stressed plants was estimated ( Figure 2A). Most of the wild-type plants did not recover, with an 18.4% survival rate ( Figure 2B). By contrast, the OE-PtrVCS2 transgenics recovered rapidly, with a survival rate of 62.2% ( Figure 2B).

Overexpressing PtrVCS2 Gene Improves Drought Resistance
Next, we tested the drought response of OE-PtrVCS2-3× FLAG #3 and wild-type plants. Four-month-old OE-PtrVCS2 and wild-type plants were subjected to drought treatments by withholding water. When plants were adequately watered, the soil was saturated, having 50% volumetric soil water content (VWC) (Figure 2A). In the absence of water replenishment, the VWC decreased to 30%, and the leaves of wild-type plants wilted, while most of the OE-PtrVCS2 transgenics remained turgid (Figure 2A). Under the severe drought condition with 15% VWC, the wild-type plants showed severe wilting symptoms, compared with the OE-PtrVCS2 transgenics (Figure 2A). After rehydration in 3 days, the survival rate for the stressed plants was estimated ( Figure 2A). Most of the wild-type plants did not recover, with an 18.4% survival rate ( Figure 2B). By contrast, the OE-PtrVCS2 transgenics recovered rapidly, with a survival rate of 62.2% ( Figure 2B).  In the statistical analysis, the average percentage of PLC was calculated from three independent experiments and standard errors were calculated from three independent experiments. Asterisks indicate significant differences between the OE-PtrVCS2 transgenics and WT plants. ** p < 0.01 (Student's t-test).
Although the leaves of the OE-PtrVCS2 plants were smaller than those of the wild-type plants, the number of functional leaves of the transgenics were much greater than those of the wild-type plants under drought stress conditions (Supplemental Figure S1). We measured PLC in the stems of the OE-PtrVCS2 and wild-type plants. Under the drought condition with 15% VWC, the OE-PtrVCS2 transgenics have lower PLC (19.1%, Figure 2C) compared with wild-type plants (52.9%, Figure 2C).

PtrVCS2 Regulates Stomatal Closure-Related Genes
Stomatal movement experiments revealed that the OE-PtrVCS2 transgenics showed lower stomata aperture than wild-type plants under the drought condition with 30% VWC ( Figure 3A-E). These results suggested that overexpression of PtrVCS2 may improve drought resistance by regulating stomatal aperture.
To explore the molecular regulation of PtrVCS2 in drought resistance of P. trichocarpa plants, we performed RNA-seq analysis of WT and OE-PtrVCS2 plants with stem-developing xylem (SDX) tissues under the drought condition with 30% VWC (Figure 2). We identified 172 differentially expressed genes (DEGs) in the OE-PtrVCS2 transgenics with 103 up-regulated and 69 down-regulated genes ( Figure 3F) (Supplemental Dataset S1). Of these DEGs, the expression levels of multiple genes involved in regulation of stomatal opening and closing were obviously changed (Supplemental Table S1). The expression levels of these stomatal closure-related genes were analyzed in leaves of WT and OE-PtrVCS2 plants with drought treatments. We found that the expression levels of five genes, whose homologous genes such as PtrSULTR3;1-1, PtrNAPP-4, PtrEDS1-1, and PtrSERK2-2 which have functions in promoting stomatal closure, were highly induced by drought stress ( Figure 3G). Considering that induction by drought stress is particularly apparent in PtrSULTR3;1-1, we speculated that PtrVCS2 may control stomatal closure by regulating the expression of the PtrSULTR3;1-1 gene in P. trichocarpa.

PtrVCS2 Regulates the Genes Related to Cell Wall Biosynthesis
Scanning electron micrographs show that elevated PtrVCS2 expression resulted in thicker fiber and vessel cell walls in the xylem of the OE-PtrVCS2 transgenics compared with the WT plants ( Figure 4A). Next, we conducted gene ontology (GO) analysis ( Figure 4B), with 172 DEGs identified from the RNA-seq of WT and OE-PtrVCS2 plants under drought conditions, and found three genes: PtrPR3-3, PtrFLA11-12, and a novel one, related to cell wall organization or biogenesis (Supplementary Table S2). Moreover, we found that PtrVCS2 regulates three genes, PtrGH9B1-1, PtrLAC44, and PtrLAC45 under drought conditions, and their homologous genes function in regulating cellulose and lignin biosynthesis.  The error bars represent one SE of the three biological replicates. "*"," **" denote significant differences: tion. The error bars represent one SE of the three biological replicates. "*"," **" denote significant differences: * p < 0.05, ** p < 0.01 (Student's t-test).

PtrVCS2 Regulates the Genes Related to Cell Wall Biosynthesis
Scanning electron micrographs show that elevated PtrVCS2 expression resulted in thicker fiber and vessel cell walls in the xylem of the OE-PtrVCS2 transgenics compared with the WT plants ( Figure 4A). Next, we conducted gene ontology (GO) analysis ( Figure  4B), with 172 DEGs identified from the RNA-seq of WT and OE-PtrVCS2 plants under drought conditions, and found three genes: PtrPR3-3, PtrFLA11-12, and a novel one, related to cell wall organization or biogenesis (Supplementary Table S2). Moreover, we found that PtrVCS2 regulates three genes, PtrGH9B1-1, PtrLAC44, and PtrLAC45 under drought conditions, and their homologous genes function in regulating cellulose and lignin biosynthesis.

Overexpressing PtrVCS2 Gene Improves the Adaptability of P. trichocarpa to Chronic Drought Stress
Chronic drought, a continuous stress, inhibits the biomass accumulation of plants. To further elucidate the role of PtrVCS2 in response to drought stress, we employed a chronic water deficit assay ( Figure 5A). PtrVCS2 transgenic plants were imaged by SEM with the 12th internode imaged at ×1000 (u panels) and ×3000 (lower panels) magnification. Bars = 20 μm. (B) Gene ontology categories re to cell wall biosynthesis.

Overexpressing PtrVCS2 Gene Improves the Adaptability of P. trichocarpa to Chronic Drought Stress
Chronic drought, a continuous stress, inhibits the biomass accumulation of pl To further elucidate the role of PtrVCS2 in response to drought stress, we employ chronic water deficit assay ( Figure 5A). Under normal conditions (ND (normal condition, 50% VWC)), the transgenics had higher WUE (10.1 µmol·mmol −1 ) than the WT. After 15-day drought treatments, there was no significant difference in WUE between the OE-PtrVCS2 transgenics and the WT plants. After 30-day drought treatments, the WUE of the OE-PtrVCS2 transgenics (6.9 µmol·mmol −1 ) was higher than that of the WT plants (5.6 µmol·mmol −1 ). After 60-day drought treatments, the WUE of the OE-PtrVCS2 transgenics (6.4 µmol·mmol −1 ) was significantly higher than that of the WT plants (4.6 µmol·mmol −1 ) ( Figure 5B). Moreover, the OE-PtrVCS2 transgenic plants had more leaves ( Figure 5C) and larger base diameters ( Figure 5D) than the WT plants. These results indicated that overexpressing PtrVCS2 could improve the adaptability of P. trichocarpa to chronic drought stress.

Discussion
Changes in the duration and frequency of drought led to a gradual increase in tree mortality [36]. Land drought has become a key global problem [37]. Transpiration forces transport water and solutes absorbed from the roots through the vessel cells to the leaves, and the water deficit can cause the water column in vessels to break, thus forming an embolism that blocks water transport in the stem [7,38]. In this study, we identified a gene in P. trichocarpa, PtrVCS2, which plays a vital role in plant response to drought. Overexpressing PtrVCS2 in P. trichocarpa resulted in reduced growth and enhanced development of more and smaller vessel cells (Figure 1), which promoted the ability of plants to resist embolism ( Figure 2C) [19]. The OE-PtrVCS2 transgenic plants had higher survival rate ( Figure 2B) and lower water conductivity loss ( Figure 2C) compared with wild-type plants under drought conditions, indicating that overexpression of PtrVCS2 is conducive to drought adaptation [39,40]. The overexpression plants had a large number of functional leaves after drought stress, suggesting that these transgenics do not require less water than the wildtype plants under drought stress conditions. Moreover, drought-induced stomatal closure of leaves reduced gas exchange and water loss [12], which protected the OE-PtrVCS2 transgenics from stress.
There is a trade-off between growth and stress responses such that loss of growth and development tends to enhance the stress resistance of plants [41]. Overexpression of PtrVCS2 gene led to reduced growth phenotype of the transgenic plants ( Figure 1A). RNA-seq analysis of the transgenics revealed that two auxin related genes, PtrIAA17-1 (Potri.001G177400) and PtrIAA4-4 (Potri.008G161100), were inhibited by PtrVCS2 (Supplemental Dataset S1). There is one possibility that the inhibition of PtrVCS2 on auxin response genes may be the reason for reduced growth of the transgenic plants. PtrVCS2 encodes a ZF-HD transcription factor, and the homologs of PtrVCS2 in other species were also found to be involved in regulation of plant development. In soybean, GhMIF regulates petal elongation by activating the expression of GEG gene of GASA family members [42]. MIF1 is related to root growth reduction and dwarfism in Amaranthus hypochondriacus L. [43]. Recently, we demonstrated that PtrVCS2 regulates cambium development by directly binding to PtrWOX4a gene promoter through interaction with PtrWOX13 in P. trichocarpa [34]. In this study, we found that PtrVCS2 not only reduces the number of cambium layers but also inhibits the enlargement of xylem vessel cells. The individual vessel cell area was smaller in the OE-PtrVCS2 transgenics than in wild-type plants ( Figure 1C). Small vessels are less prone to embolism [17]; therefore, the loss of xylem conductance of OE-PtrVCS2 transgenics was lower than that of the WT ( Figure 2C). Reduction in the size of the vessel is one approach to improve WUE. The vessel diameter correlated negatively with intrinsic WUE, and the correlation was stronger under water deficit conditions [44]. We speculate that WUE of OE-PtrVCS2 transgenics was also higher than that of WT under drought stress. Although smaller vessels have little embolism, they also transport less water, which is not sufficient for the growth of the plants [4]. We found that the vessel cell number ( Figure 1D) and the area of vessels in the transverse section of the woody stem ( Figure 1E) were significantly increased in the OE-PtrVCS2 transgenics compared with wild-type plants. The increase in the number of vessels solves the problem of water loss caused by the smaller area of the individual cells in OE-PtrVCS2 transgenic plants. Moreover, we demonstrated that PtrVCS2 plays a positive role in improving drought adaptability and resistance in P. trichocarpa (Figures 2 and 5). Previous studies and our findings suggest that PtrVCS2 and its homologs in different species have diverse functions not only in the regulation of plant development but also in plant response to abiotic stresses.
Plant growth adapts to the changing environments. Under drought conditions, plants alter their physiology to reduce growth and enhance drought resistance for adaptation [39,[45][46][47][48]. Drought adaptation also includes development of cell wall structures that are conducive to water transport, i.e., structures with smaller stem xylem vessels to minimize xylem cavitation (failure of upward water transport) [49,50]. Cell walls limit cells to different shapes and protect the cells in response to abiotic stresses [51,52]. The secondary cell wall acts as a conduit for water and resistance to the tension generated by the transpiration pull [53]. We observed the cross section of the stem by scanning electron microscope and found that the xylem cell wall of the OE-PtrVCS2 plants was thicker than that of the wild-type plants ( Figure 4A). Therefore, it is possible that PtrVCS2 may restrict xylem cell morphological changes by regulating cell wall thickening. Moreover, PtrVCS2 regulated the expression of the cell wall synthesis genes, such as PtrFLA11-12 and PtrPR3-3 ( Figure 4B). The homologous gene of PtrFLA11-12 is associated with cell wall expansion [54], and the homologous gene of PtrPR3-3, as a marker gene of Botrytis cinerea, has been shown to promote disease resistance in plants [55]. In Arabidopsis thaliana, AtGH9B1-1 is involved in cellulose biosynthesis [56]. In Cleome hassleriana L., LACs promote the biosynthesis of lignin components [57]. Together, these results suggest that PtrVCS2 may control xylem cell morphological changes by regulating the expression of these cell wall biosynthesis-related genes in P. trichocarpa. The increase of cell wall thickness increases the density of wood, thus changing the wood properties and increasing the biomass of plants [53,58]. The key traits in the OE-PtrVCS2 transgenics, i.e., reduced plant growth and a larger number of small xylem vessels that are consistent with overexpression of PtrVCS2, are all conducive to drought adaptation. The recovery of plant stems with small and abundant cells and thick cell walls may depend on the recovery of xylem function and the regulation of carbohydrate metabolism in xylem after rehydration after drought stress [59]. These apparent adaptative traits make the OE-PtrVCS2 lines unique as potential sources for the attainment of new knowledge of the complex regulations involved in cell-type biosynthesis in wood formation and growth under stresses.
Hydraulic conductivity is positively correlated with stomatal conductance, and a decrease in water transport leads to stomatal closure [60,61]. In response to drought stress, plants use hormones to trigger stomatal closure, reducing water loss [62]. Using transgenesis ( Figure 2) and RNA-seq ( Figure 3) we found that PtrVCS2 controls stomatal closure by regulating the expression of stomatal closure-related genes in plant response to drought (Supplemental Table S1) under drought conditions [63][64][65][66][67][68][69][70]. Of these genes, the expression levels of PtrSULTR3; 1-1 were significantly increased in xylem and leaves of the OE-PtrVCS2 transgenic plants compared with wild-type plants under drought conditions ( Figure 3G). The homolog of PtrSULTR3;1-1 in Arabidopsis, AtSULTR3, moves into chloroplasts via sulfate transport to promote cysteine synthesis, trigger ABA biosynthesis, and regulate stomatal closure under stress conditions [63]. The most stomata of the OE-PtrVCS2 transgenic plants were closed under drought stress ( Figure 3A,E), suggesting that PtrVCS2 may regulate the expression of these stomatal closure-related genes and thus control water losses through transpiration of the leaves by an ABA-dependent pathway. Additionally, the expression levels of PtrPNP-A-2 were also induced by PtrVCS2, but the relative expression abundance was lower than that of PtrSULTR3; 1-1 ( Figure 3G). The homologue of PtrPNP-A-2 in Arabidopsis, AtPNP-A, as an antagonist of ABA, limits gas exchange in stomata [66]. There is one possibility that PtrSULTR3; 1-1 and PtrPNP-A-2 may play opposite roles in ABA-induced stomatal closure, but we do not know when PtrPNP-A-2 is dominant. Usually, ABA signaling receptors inhibit PP2C phosphatase activity and promote the activation of SnRK2s protease phosphorylation to trigger stomatal closure [71]. Whether PtrVCS2 is involved in this process, what role it plays, and the target of phosphorylation still needs to be further investigated.
Moreover, gene ontology (GO) analysis showed that PtrVCS2 regulates the expression of genes related to the ABA signaling pathway in xylem (Supplemental Figure S2). The ZF-HD transcription factor also responded to abiotic stress and the ABA hormone signal in tomato [72]. Thus, PtrVCS2 may control stomatal closure to improve drought resistance by integrating ABA signals in P. trichocarpa. Further exploration of the connections between the PtrVCS2-mediated control of stomatal closure and ABA signaling should yield new insights into the regulation of drought resistance in P. trichocarpa. In addition to the ABA signaling pathway, PtrVCS2 also regulates the expression of the BR response gene PtrCYP67A2-13 (Potri.009G064800) and the ethylene response gene PtrPR4-3 (Potri.013g041600) (Supplemental Dataset S1). Considering the complexity of the regulatory network of plant stress response [73,74], it is possible that multiple hormone signaling pathways are involved in improving drought resistance of OE-PtrVCS2 plants.
Water deficit often results in the decrease of aboveground biomass and retardation of plant growth [75]. Under chronic drought stress, the leaves of wild plants fall off ( Figure 5A), but there was no obvious leaf loss in OE-PtrVCS2 plants. The number of leaves of OE-PtrVCS2 transgenics was significantly more than that of wild type plants ( Figure 5C). WUE is a key ecosystem attribute of plant water cycles [76]. The OE-PtrVCS2 transgenic plants had strong photosynthesis and high water use efficiency under drought conditions (Supplemental Figure S3), indicating that OE-PtrVCS2 plants produced more biomass for the same amount of water consumption [77]. This is consistent with the increase in stem size and wood yield of OE-PtrVCS2 transgenics ( Figure 5D). We demonstrated that the OE-PtrVCS2 transgenics could maintain high WUE ( Figure 5B) and thus accumulate a large amount of biomass under chronic drought stress condition.
In conclusion, we present a novel gene, PtrVCS2, in regulation of secondary xylem development and drought resistance. Knowledge of PtrVCS2 regulation may help design genetic controls that could maximize beneficial wood traits while minimizing negative effects on growth under drought conditions.

RNA-Seq Assay and Data Analyses
In RNA-seq experiments, 4-month-old wild-type P. trichocarpa plants and OE-PtrVCS2 transgenic plants were placed in a separate 15-cm pot, which were used for drought treatments (30% VWC). The plants were divided into two groups: (1) control (WT), and (2) OE-PtrVCS2 transgenic plants (OE). Before drought treatment, all plants were thoroughly watered until the soil was saturated. Short-term drought treatments were conducted according to established procedures [39] with a few modifications. Plants in saturated soil were placed in a greenhouse to stand, without replenishing water, and soil water content was reduced to 30%. All the plants were harvested at 30% VWC. Stem-developing xylem tissues for RNA-seq were packed in aluminum foil and collected in liquid nitrogen [79,80]. Total RNA was extracted using the Qiagen RNeasy Mini Kit (Qiagen, Hilden, NRW, Germany) as previously described [81]. RNase-Free DNase Set (Qiagen) was used to remove DNA impurities. Purified RNA was used for RNA-seq. RNA-seq library construction of each sample was performed using the Illumina TruSeq RNA sample preparation kit. The quality of libraries was examined using a Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA). Six libraries were sequenced using the Illumina Genome Analyzer, with an average reading length of 100-bp. After the library index sequences from each read were removed by HISAT [82] with parameters (-dta -phred64 unstranded -new-summary -x index -1 read_r1 -2 read_r2 (PE)), the remaining RNA-seq reads were mapped to the P. trichocarpa genome v.4.0. The frequency of raw counts was determined by DIAMOND tools [83][84][85] for all annotated genes. DEGs between the WT and OE samples were identified using DEseq2 [86,87] based on raw counts of mapped RNA-seq reads to annotated genes. Gene ontology (GO) was analyzed using R (3.1.2) statistical software with the Phyper function by Fisher's exact test with FDR multiple test correction (FDR ≤ 0.05).

RT-qPCR Analyses
Mature leaves of wild-type and OE-PtrVCS2 transgenics after drought stress treatments were harvested for RNA extraction. RNase-Free DNase Set (Qiagen) was used to remove DNA. Purified RNA was stored in the refrigerator at −80 • C. cDNAs were synthesized by reverse transcription using SrimeScript™ RT reagent Kit (TAKARA) according to the manufacturer's protocol. FastStart Universal SYBR Green Master (Roche) was used to formulate the reaction system and an Agilent Mx3000P Real-time PCR System was used for the RT-qPCR experiment. Primers for RT-qPCR are listed in Supplemental Table S3.

Histochemical and Histological Analyses
The 5th-10th internodes of OE-PtrVCS2 transgenics and wild-type plants were harvested and cut into 2-3 mm fragments, and then immersed in 4% formaldehyde following established procedures [39]. The cut wax tape was fixed on the slide and immersed in 2.5% safranin O and 1.25% fast green. The size and area of vessels were measured by using PreciPoint Streaming Software of Precipoint Digital Scanning microscope imaging System (M8). The observation and statistical methods of stomata were conducted as previously described [88]. The images were scanned with ViewPoint Light. IBM SPSS Statistics 19 was used for statistical analyses.

Scanning Electron Micrograph (SEM) Analyses
Fresh stem segments from the 12th internode of OE-PtrVCS2 transgenic and WT plants were plated with gold (Au) for 60 s at 10 mA. Fresh leaves of OE-PtrVCS2 transgenic and WT plants were harvested, and the lower epidermis was coated by nail polish. The leaves were then placed on ice to keep fresh. The leaves were cut into small squares, about 1 cm 2 in area, for observation of the stomatal aperture. The samples were observed under high vacuum at 15 kV using a Nanotech JCM-5000 [89].

Drought Treatments
For short-term drought treatments, 4-month-old plants in saturated soil (50% VWC) were placed in a greenhouse without replenishing water, and soil water content was reduced from 50% to 30% to 15% and then finally rehydrated for 3 days. At least 12 transgenic plants and 12 wild-type plants were used for each drought treatment. Three independent experiments were performed. The screening experiments revealed that 15% VWC caused shoot dieback to 4-month-old wild-type P. trichocarpa plants; this was used for the estimation of the survival rate. After the soil water content was reduced to 15% VWC, all plants were rehydrated for 3 days to estimate their survival rates. For chronic drought treatments, 3-month-old OE-PtrVCS2 and WT plants were subjected to drought treatments by withholding water. After 12-day drought treatments, the VWC was reduced to 40% and maintained for 60 days [90].

Data Availability Statement:
The data supporting the findings of this study are available within the article and its Supplemental Information files.