The Plant JournalVolume 102, Issue 4 p. 779-796 Original Article Free Access Ectopic expression of a rice plasma membrane intrinsic protein (OsPIP1;3) promotes plant growth and water uptake Siyu Liu, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorTatsuya Fukumoto, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorPatrizia Gena, Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari ‘Aldo Moro’, Bari, ItalySearch for more papers by this authorPeng Feng, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorQi Sun, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorQiang Li, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorTadashi Matsumoto, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorToshiyuki Kaneko, Research Institute for Bioresources, Okayama University, Kurashiki, 710-0046 JapanSearch for more papers by this authorHang Zhang, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorYao Zhang, College of Horticulture, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorShihua Zhong, Department of Biochemistry, the University of Texas Southwestern Medical Center, Dallas, TX, 75390 USASearch for more papers by this authorWeizhong Zeng, Department of Biophysics, the University of Texas Southwestern Medical Center, Dallas, TX, 75390 USASearch for more papers by this authorMaki Katsuhara, Research Institute for Bioresources, Okayama University, Kurashiki, 710-0046 JapanSearch for more papers by this authorYoshichika Kitagawa, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorAoxue Wang, College of Horticulture, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorGiuseppe Calamita, Corresponding Author giuseppe.calamita@uniba.it Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari ‘Aldo Moro’, Bari, Italy For correspondence (e-mail xiaodong.ding@neau.edu.cn; giuseppe.calamita@uniba.it). Search for more papers by this authorXiaodong Ding, Corresponding Author xiaodong.ding@neau.edu.cn Siyu Liu, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorTatsuya Fukumoto, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorPatrizia Gena, Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari ‘Aldo Moro’, Bari, ItalySearch for more papers by this authorPeng Feng, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorQi Sun, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorQiang Li, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorTadashi Matsumoto, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorToshiyuki Kaneko, Research Institute for Bioresources, Okayama University, Kurashiki, 710-0046 JapanSearch for more papers by this authorHang Zhang, Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorYao Zhang, College of Horticulture, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorShihua Zhong, Department of Biochemistry, the University of Texas Southwestern Medical Center, Dallas, TX, 75390 USASearch for more papers by this authorWeizhong Zeng, Department of Biophysics, the University of Texas Southwestern Medical Center, Dallas, TX, 75390 USASearch for more papers by this authorMaki Katsuhara, Research Institute for Bioresources, Okayama University, Kurashiki, 710-0046 JapanSearch for more papers by this authorYoshichika Kitagawa, Graduate School of Bioresource Sciences, Akita Prefectural University, Akita, 010-0195 JapanSearch for more papers by this authorAoxue Wang, College of Horticulture, Northeast Agricultural University, Harbin, 150030 ChinaSearch for more papers by this authorGiuseppe Calamita, Corresponding Author giuseppe.calamita@uniba.it Department of Biosciences, Biotechnologies and Biopharmaceutics, University of Bari ‘Aldo Moro’, Bari, Italy For correspondence (e-mail xiaodong.ding@neau.edu.cn; giuseppe.calamita@uniba.it). Search for more papers by this authorXiaodong Ding, Corresponding Author xiaodong.ding@neau.edu.cn Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat Summary Plasma membrane intrinsic proteins (PIPs) are known to be major facilitators of the movement of a number of substrates across cell membranes. From a drought-resistant cultivar of Oryza sativa (rice), we isolated an OsPIP1;3 gene single-nucleotide polymorphism (SNP) that is mostly expressed in rice roots and is strongly responsive to drought stress. Immunocytochemistry showed that OsPIP1;3 majorly accumulated on the proximal end of the endodermis and the cell surface around the xylem. Expression of GFP-OsPIP1;3 alone in Xenopus oocytes or rice protoplasts showed OsPIP1;3 mislocalization in the endoplasmic reticulum (ER)-like neighborhood, whereas co-expression of OsPIP2;2 recruited OsPIP1;3 to the plasma membrane and led to a significant enhancement of water permeability in oocytes. Moreover, reconstitution of 10×His-OsPIP1;3 in liposomes demonstrated water channel activity, as revealed by stopped-flow light scattering. Intriguingly, by patch-clamp technique, we detected significant NO3− conductance of OsPIP1;3 in mammalian cells. To investigate the physiological functions of OsPIP1;3, we ectopically expressed the OsPIP1;3 gene in Nicotiana benthamiana (tobacco). The transgenic tobacco plants exhibited higher photosynthesis rates, root hydraulic conductivity (Lpr) and water-use efficiency, resulting in a greater biomass and a higher resistance to water deficit than the wild-type did. Further experiments suggested that heterologous expression of OsPIP1;3 in cyanobacterium altered bacterial growth under different conditions of CO2 gas supply. Overall, besides shedding light on the multiple functions played by OsPIP1;3, this work provides insights into the translational value of plant AQPs. Introduction Membrane intrinsic proteins (MIPs) constitute an ancient family of proteinaceous channels widely distributed in all three kingdoms of life (eukarya, bacteria and archaea). MIPs are small proteins with six transmembrane (TM) α-helices connected by five loops as well as cytosol-localized N and C termini. Aquaporins (AQPs) belong to the MIP superfamily and function as channel proteins known for their abilities to conduct water and/or small molecules across biological membranes. Since the first AQP was discovered (Preston et al., 1992) numerous homologs have been identified and characterized from microorganisms, plants and animals, including humans. Orthodox AQPs act as filters that stringently exclude the passage of charged molecules and solutes larger than water. As a result of long-term evolution, however, different organisms have evolved and now possess different cell organization and physiological features. For example, photoautotrophic plants, some photoautotrophic microorganisms and heterotrophic animals possess distinct AQPs, with different substrate specificities and channel activities (Ishibashi et al., 2017). Aquaporin substrate specificity, expression and localization varies among the different AQP members from different species. Human AQP1 was the first cloned AQP to transport water (Preston et al., 1992) and CO2 (Nakhoul et al., 1998) in Xenopus oocytes across the cell membrane, and human AQP3 was discovered to be permeable to glycerol and urea (Echevarría et al., 1996). The first work addressing plant AQPs led to the cloning of γ-TIP, a water-specific channel found in the vacuolar membrane of Arabidopsis (Maurel et al., 1993). Thereafter, more plant water channel proteins were identified and characterized (reviewed by Kaldenhoff and Fischer, 2006; reviewed by Maurel et al., 2008). In addition, many plant AQPs were found to have additional functions in transporting other small molecules, such as glycerol (Dean et al., 1999), urea (Gaspar et al., 2003), boron (Takano et al., 2006), H2O2 (Rodrigues et al., 2017), ammonia (Hwang et al., 2010), CO2 (Uehlein et al., 2003) and O2 (Zwiazek et al., 2017). No NO3− permeability in plant AQPs except mammalian Aquaporin-6 (Ikeda et al., 2002) has been reported, until now. With the rapid development of DNA sequencing technology, many plant genomes have now been sequenced. Genomics data reveal that plants have greatly evolved and diversified, possessing many more AQP homologs than are found in other organisms, mostly because of their sessile nature in the face of ever-changing environmental conditions (Abascal et al., 2014). For example, Oryza sativa (rice) has 34, Arabidopsis has 35, Zea mays (maize) has 43, Glycine max (soybean) has 72, and Brassica oleracea has 67 homologs (Deshmukh et al., 2016). Based on their sequences and structures, plant AQPs are generally grouped into five subfamilies: NIP (similar to aquaglyceroporin of nitrogen-fixing bacteria of legume), PIP (plasma membrane intrinsic protein), SIP (small basic intrinsic protein), TIP (tonoplast intrinsic protein) and XIP (AQPs of unknown function). These AQPs localize in plasma membranes (PIPs and some NIPs), tonoplast (TIPs), endoplasmic reticulum (SIPs) and/or other subcellular compartments (Maurel et al., 2015). The permeability of the different isoforms is directly or indirectly related to their specific physiological functions. Until now, huge progress has been made in the characterization of this family, allowing insights into their roles in the control of plant–environment relationships (Fox et al., 2017). For example, ectopic expression of the Malus domestica (apple) MdPIP1;3 gene in Solanum lycopersicum (tomato) enlarged fruit size and enhanced drought tolerance (Wang et al., 2017a). Arabidopsis PIP1b overexpression in tobacco significantly increased plant growth, transpiration rate, stomatal density and photosynthetic efficiency (Aharon et al., 2003), whereas overexpression of GsPIP2;1 from wild soybean in Arabidopsis attenuated the resistance to salt and dehydration (Wang et al., 2015). Moreover, ectopic expression of a fungal aquaglyceroporin gene in tobacco improved water-use efficiency (WUE) and drought tolerance (Vieira et al., 2017). On the other hand, tobacco NtAQP1 (NtPIP1) (Uehlein et al., 2003) and cyanobacterial SsAqpZ (Ding et al., 2013) were proven to help permeate CO2 and thus enhanced plant and cyanobacterial photosynthesis. A recent study reported that the overexpression of rice aquaporin OsPIP1;2 improved yield by enhancing mesophyll CO2 conductance and phloem sucrose transport (Xu et al., 2019); however, the arguments about this aspect have continued over decades (Missner et al., 2008). Water deficit and drought stress can induce a series of plant responses, including leaf wilting, decreased photosynthesis and metabolic reactions, and oxidative damage in chloroplasts (Almeida-Rodriguez et al., 2010). To tolerate these stresses, plants evolve various strategies, such as the adjustment of stomatal closure and hydraulic conductivity by altering turgor pressure in leaf and root cells, where AQPs may play important roles in regulating the trans-cellular transport of water (Maurel et al., 2008). Rice has 11 PIP isoforms, including three OsPIP1 and eight OsPIP2 members. Although systematic expression analyses have been conducted for the AQP family in rice (Nguyen et al., 2013), their precise functions remain mostly elusive. Generally, the plant PIP2 group shows higher water permeability than the PIP1 group in Xenopus oocytes (Chaumont et al., 2000). We have functionally characterized rice OsPIP1;1 and OsPIP1;3, the overexpression of which enhanced plant salt and chilling tolerances, respectively (Matsumoto et al., 2009; Liu et al., 2013). Lian et al. (2004, 2006) observed that OsPIP1;3 was strongly responsive to drought challenge and played pivotal role in drought avoidance in upland rice; however, the molecular mechanism of how OsPIP1;3 functions in the plant physiological processes is largely unknown. In this work, we characterized a single-nucleotide polymorphism (SNP) of OsPIP1;3 from a drought-resistant rice cultivar and unveiled its functions in promoting plant tolerance to water deficit and growth. Although the water permeability of OsPIP1;3 from a chilling-tolerant but drought-sensitive O.sativa L. spp. japonica cultivar has been characterized in Xenopus oocytes (Matsumoto et al., 2009), the OsPIP1;3 gene described in this work was isolated from a drought-tolerant O.sativa L. spp. indica cultivar. The alignment of OsPIP1;3 sequences from the two subspecies indicates that the fragments in their N-terminal tails have an eight-amino-acid difference, and that one glycine residue in the fifth TM domain of OsPIP1;3 from the indica cultivar is missing (Figure S1), suggesting that SNPs of AQP genes are widely present in different species and ecological types of rice and may potentially contribute to functional diversity. Total RNA from roots, shoots and leaves was extracted from 1-month-old rice seedlings (O.sativa L. spp. indica cv. Zhonghan 3) under favorable growth conditions and used for quantitative real-time PCR (qPCR) assays using the primers listed in Table S1. The data show that OsPIP1;2 was ubiquitously expressed in the whole plants at lower levels, whereas OsPIP1;1 had higher expression level in leaves than in roots. OsPIP1;3 was mostly expressed in both leaves and roots, however, and OsPIP2;2 was dominantly expressed in roots (Figure 1a). To investigate the response of OsPIP1;3 to different environmental stimuli further, the rice seedlings were treated with 500mm NaCl, 20% polyethylene glycol (PEG) or 4°C for 2h and the total RNA from whole plants was extracted for qPCR assays. The results indicated that OsPIP1;1 was significantly and specifically induced by PEG, chilling and salt, whereas OsPIP1;3 was induced by drought and chilling but not by salt (Figure 1b). By using an anti-OsPIP1;3-specific antibody, the protein levels of endogenous OsPIP1;3 in different rice tissues could be determined. The quantified protein levels were basically consistent with the transcript levels. Taken together, it is reasonable to state that OsPIP1;3 is dominantly expressed in rice roots (Figure 1c, d). Expression patterns of rice PIP1 genes. Total RNA and protein samples were extracted from Oryza sativa (rice) tissues for RT-qPCR and Western blotting analyses. (a) Analyses of the spatial expression of PIP1 genes in rice tissues by RT-qPCR. (b) Responsive expression patterns of PIP1 genes in roots to different environmental stimuli. The relative values were normalized by ubiquitin gene abundance. (c) Western blotting using anti-OsPIP1;3-specific antibody against protein lysates extracted from whole plant, stems, leaves and roots of the wild-type (WT) rice. OX is protein lysate from the OsPIP1;3-overexpression plant. (d) Relative abundance of OsPIP1;3 protein in rice tissues. Bars with different lowercase letters were significantly different (Duncan’s multiple range test: P 0.05). To assess the in situ localization of OsPIP1;3, we used an anti-OsPIP1;3-specific antibody to immunostain the endogenous OsPIP1;3 protein in rice root tips. The immunoelectron microscopy data show that OsPIP1;3 was mainly localized at the proximal end of the endodermis and the cell surface around the xylem (Figure S2a–c), suggesting that OsPIP1;3 may play important roles in transporting water and other nutrients. As a negative control, no significant signal was detected from the root tip cells by staining with rabbit sera (Figure S2d). To investigate the subcellular localization of OsPIP1;3 further, OsPIP1;3 and OsPIP2;2 open reading frames (ORFs) were fused with genes coding for different fluorescent proteins, respectively. The plant transient expression constructs obtained were transformed into rice O.c. protoplasts. Single expression of EGFP-OsPIP1;3 in the rice protoplasts resulted in the abnormal retention of GFP signal in ER-like membranes around the nucleus (Figure 2a, b), but the fluorescence of EGFP-OsPIP2;2 was solely observed in the plasma membrane of rice protoplasts (Figure 2c,d). To examine whether the abnormal localization of EGFP-OsPIP1;3 is altered by the interaction of other PIP isoforms, we co-expressed EGFP-OsPIP1;3 (green) with a small quantity of mDsRed-OsPIP2;2 (red) in rice protoplasts. The resulting fluorescence patterns suggest that OsPIP1;3 could be recruited to the plasma membrane by OsPIP2;2 (Figure 2e–g). Subcellular localization of OsPIP1;3 in Oryza sativa (rice) protoplasts and Xenopus oocytes. Rice O.c. protoplasts were prepared and transformed with the following plasmids alone or plasmid combinations by the polyethylene glycol method: (a, b) EGFP-OsPIP1;3 only; (c, d) EGFP-OsPIP1;3 plus OsPIP2;2; and (e–g) EGFP-OsPIP1;3 plus mDsRed-OsPIP2;2. At 16h after DNA transformation, the images were taken by a laser scanning confocal microscope. Xenopus oocytes were prepared and injected with the following plasmids alone or plasmid combinations: (h) 25ng of PIP1;3-EGFP cRNA; (i) 25ng of PIP2;2-EGFP cRNA; (j) 25ng of PIP1;3-EGFP cRNA and 3ng of PIP2;2 cRNA; (k) 25ng of EGFP cRNA; (l) water. At 48h after injection, the images were taken by a laser scanning confocal microscope. In the animal Xenopus oocyte system, we expressed an equal quantity of EGFP-OsPIP1;3 or EGFP-OsPIP2;2 in oocytes and observed that the eggs injected with EGFP-OsPIP1;3 showed a very weak GFP signal (Figure 2h), whereas the eggs injected with EGFP-OsPIP2;2 showed a strong signal at the plasma membranes (Figure 2i). When co-expressing EGFP-OsPIP1;3 with a small quantity of OsPIP2;2, however, we observed that the GFP signal was strongly re-localized to the plasma membrane (Figure 2j), implying that OsPIP2;2 can recruit OsPIP1;3 to the plasma membrane system, probably through a physical interaction. In order to determine the biophysical functions of OsPIP1;3 from the rice indica cultivar, we expressed OsPIP1;3 and/or OsPIP2;2 in Xenopus oocytes to measure water permeability across the cell membrane. As shown in Figure 3(a), OsPIP1;3 demonstrated almost negligible water channel activity no matter how much plasmid DNA was transformed, whereas OsPIP2;2 demonstrated significant activity with a linear dose effect. OsPIP1;3 exhibited significant channel activity in the presence of a small quantity of OsPIP2;2 in Xenopus oocytes (Figure 3a), however, suggesting that both physical interaction and heterotetramerization among different PIP1 proteins are required for the synergistic effect on water permeability (Fetter et al., 2004; Bienert et al., 2018). The phenomenon of most PIP1 proteins failing to target to the plasma membrane and residing in intracellular membranes results from either missing plasma membrane trafficking signals or existing ER retention motifs (Chevalier and Chaumont, 2015). Functional characterization of water channel activity of OsPIP1;3. (a) Measurement of water permeability of OsPIP1;3 in the Xenopus oocyte system. The coefficient of osmotic water permeability (y-axis) is expressed as Pf=μmsec−1. Vertical bars represent means of 10 replications±SEs and the different lowercase letters were significantly different (Duncan’s multiple range test: P 0.05). (b) The traces of stopped-flow light-scattering data. Purified 10×His-OsPIP1;3 protein was functionally reconstituted into liposomes to obtain proteoliposomes. A portion of proteoliposomes was incubated with 300µm of the AQP blocker HgCl2 (+Hg2+) for 5min, and a portion of Hg2+-treated proteoliposomes was incubated with 10mm β-mercaptoethanol (+Hg2++β-ME) for 15min. The water permeability of proteoliposomes or empty liposomes with or without the above treatments were measured by stopped-flow light scattering; (c) Statistical analyses of Pf values of the coefficient of osmotic water permeabilities. All Pf values are expressed as means±SEs (n=10). To overcome the complexity of the oocyte system and to determine OsPIP1;3 activity, independently from other factors, we produced and purified recombinant OsPIP1;3 by an Escherichia coli system and reconstituted it into liposomes to measure the water permeability by a stopped-flow light-scattering approach (Liu et al., 2013). The OsPIP1;3 ORF with a 10×His tag sequence at its 5′ terminus was inserted into a prokaryotic expression vector. Owing to the failure of the target protein expression, the OsPIP1;3 gene codons were optimized to maximize its expression in E.coli. The clear supernatants containing membrane fragments from the E.coli lysates were ultracentrifuged to pellet the cell membranes. As it is difficult to efficiently extract the TM proteins from the membrane pellets, we screened a series of distinct detergents to solubilize the recombinant OsPIP1;3 protein out of the E.coli crude membrane fraction and found that a solubilization buffer containing 1.5% Fos-Choline12 (FC12) or 4% perfluorooctanoic acid (PFO) detergent (lanes8 and 9 in Figure S3a) could generate higher yields of soluble OsPIP1;3 protein from membrane pellets. Subsequently, the 10×His-OsPIP1;3 recombinant protein was then purified using nickel–nitrilotriacetic acid (Ni-NTA) columns. The protein samples obtained were resolved in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels. Coomassie brilliant blue staining and Western blotting analyses using antibody against OsPIP1;3 indicated that the recombinant OsPIP1;3 was successfully expressed and purified with clear monomeric and dimeric bands (Figure S3b). A final quantity of 1.7mg of purified 10×His-OsPIP1;3 protein was obtained from 13g of fresh E.coli crude proteins harvested from 10L of E.coli culture. The purified 10×His-OsPIP1;3 protein was reconstituted into liposome to generate proteoliposomes for the measurement of its water conducting activity. 10×His-OsPIP1;3 activity was evaluated by stopped-flow light scattering after quickly exposing the proteoliposomes (versus control empty liposomes) to an osmotic upshock of 140mOsm with the impermeant solute mannitol at 20°C. The coefficient of osmotic water permeability (Pf) of the 10×His-OsPIP1;3 proteoliposomes was 62µmsec−1, significantly higher than that of the empty liposomes (8.6µmsec−1) (Figure 3b,c), indicating that 10×His-OsPIP1;3 has functional water channel activity in vitro. To further confirm these results, we incubated the proteoliposomes with 300µm Hg2+ for 5min and observed a pronounced reduction of water permeability. This inhibition could be partially alleviated by 15-min exposure to 10mm β-mercaptoethanol, however, a reducing agent removing the Hg2+ bound to threonines and/or cysteines (Figure 3b,c), thereby confirming that OsPIP1;3 is an active and bona fide water channel in vitro. To determine whether OsPIP1;3 has other permeable substrates besides water, we expressed OsPIP1;3 in mammalian cells and detected its ion permeability by patch-clamp electrophysiological technique. The fluorescence data indicated that EGFP-OsPIP1;3 was typically localized at the plasma membrane in mammalian cells, whereas EGFP was found throughout the cells (Figure S4a,b). Western blotting data indicated that both EGFP and EGFP-OsPIP1;3 fusion protein was properly expressed within the cells by using anti-GFP antibody (Figure S4c). When the cells expressing EGFP-OsPIP1;3 were exposed to 0 or 150mm NO3− solution, the electrode voltage–current curves increasingly diverged from each other (upper right panel in Figure 4a), whereas little divergence occurred in the cells expressing EGFP exposed to 0 or 150mm NO3− solution (upper left panel in Figure 4a). Surprisingly, the expression of OsPIP1;3 did not lead to significant divergence of the electrode voltage–current curves in 0 or 150mm HCO3−/CO32− solutions (lower panels in Figure 4a). The statistical data indicated that OsPIP1;3 has significant NO3− permeability but not HCO3−/CO32− exchange activity (Figure 4b), implying that OsPIP1;3 can specifically transport the NO3− anion across membranes besides water in mammalian cells. Measurement of nitrate permeability of OsPIP1;3 by patch-clamp technique. (a) EGFP or EGFP-OsPIP1;3 was expressed in mammalian HEK293 cells. At 48h after transformation, the cells were trypsinized and exposed to HCO3−/CO32−, NO3− or the control solution. The electrode voltage–current curves were detected by patch-clamp technique. (b) Statistical analyses of relative nitrate permeability. All electrophysiological recordings were repeated at least five times using different patches. Bars with different lowercase letters were significantly different (Duncan’s multiple range test: P 0.05). To investigate the physiological functions of OsPIP1;3 in planta, we inserted the OsPIP1;3 gene into a binary vector (Figure S5a) in order to ectopically overexpress it in tobacco. Expression levels of endogenous NtAQP1 and overexpressed OsPIP1;3 genes were detected by qPCR from the wild type and four representative T2 progeny. The endogenous NtAQP1 gene had similar background expressions in both wild-type and transgenic tobacco lines but OsPIP1;3 was only expressed in the transgenic plants (Figure S5b). The OsPIP1;3 protein expression levels in homozygous T2 lines transformed with FLAG-OsPIP1;3 or the empty vector were detected by anti-FLAG antibody (Figure 5a). All transgenic and wild-type plants were grown under the same conditions. The plant growth rates (plant heights) were monitored at every point in the interval and the results showed that the transgenic plants grew faster with greater height and thicker canopy than the wild-type plants at approximately 1month after germination (Figure 5c). Figure 5(b) shows the different growth vigor of 2-month-old wild-type and transgenic plants that were re-watered and recovered from a 6-day drought treatment (Figure 6a). Moreover, the transgenic plants had greater biomass accumulation with greater fresh and dry weights than the wild type (WT) (Figure 5d). Growth patterns and physiological indexes of the transgenic and wild-type (WT) Nicotiana benthamiana (tobacco) plants. (a) Western blotting of Flag-OsPIP1;3 from the transgenic and WT tobacco lines. (b) Phenotypes of 50-day-old transgenic and WT plants. (c) Plant growth rates. (d) Fresh and dry weights per plant at 2months old. (e) Photosynthesis rates (AN). (f) Leaf mesophyll CO2 conductance (gm). (g) Chloroplast CO2 concentration (Cc). Bars with different lowercase letters were significantly different (Duncan’s multiple range test: P 0.05). Analyses of water-deficit resistance and root hydraulic conductivity. (a) The phenotypes of plants after water withdrawal for 6days. (b) Relative water contents (RWCs) in leaves. The fresh leaves on transgenic and wild-type (WT) plants subjected to a 6-day drought treatment were picked and weighed, and their dry weights were measured after dehydration at 85℃ for 1h. The relative water contents of intact leaves (RWCi) or detached leaves (RWCd) were calculated. (c) Root hydraulic conductivities (Lpr) before and after 20% PEG treatment. Bars with different lowercase letters were significantly different (Duncan’s multiple range test; P 0.05). To decipher the mechanisms of how OsPIP1;3 promoted plant growth, we determined and compared the photosynthetic physiological indexes of young WT and OsPIP1;3 transgenic tobacco plants of the same age. We measured and calculated the leaf photosynthesis rate (AN), transpiration rate (Tr), stomatal conductance (gs), intercellular CO2 concentration (Ci), leaf mesophyll conductance to CO2 (gm) and chloroplast CO2 concentration (Cc). The resulting data indicate that the transgenic plants had higher AN, gm and Cc values than the WT plants (Figure 5e–g), whereas their gs, Tr and Ci values were similar (Figures S6–S8), suggesting that OsPIP1;3 as well as NtAQP1 (Uehlein et al., 2003; Flexas et al., 2006) promoted leaf mesophyll conductance of CO2 gas and could enhance photosynthesis efficiency. In addition, the total chlorophyll and nitrogen contents were measured and compared between OsPIP1;3 transgenic and WT leaves. Although OsPIP1;3 demonstrated in vitro nitrate permeability in this work (Figure 4), the average chlorophyll and total nitrogen contents in transgenic and WT plants did not significantly differ (Figure S9a,b). To further confirm the CO2 permeability of OsPIP1;3, FLAG-tagged OsPIP1;3 was heterogeneously expressed in Synechococcus elongatus PCC 7942 cyanobacterial cells according to our protocol (Ding et al., 2013). Western blot indicated that OsPIP1;3 was properly expressed in cyanobacterium (Figure S10a). On solid medium, although the transgenic and the WT cell lines looked similarly weak in CO2-free air, they began to differ in normal air, where the transgenic line grew stronger than the WT, indicating that OsPIP1;3 may mediate CO2 permeability across thylakoid membranes to enhance cyanobacterial photosynthesis and growth (Figure S10b). In liquid medium, the transgenic and the WT lines were cultured with equal initial cell densities, but the transgenic line demonstrated faster growth than the WT line in the exponential growth phases (Figure S10c). To assess the effect of OsPIP1;3 overexpression on plant tolerance to water deficit, we withheld water from 2-month-old tobacco plants growing in pot soil for a total of 6days in a growth chamber. After the 6-day drought treatment, the soil relative water content (SRWC) in all pots dropped from 85% to approximately 40% (Figure S11). The phenotypes were observed and recorded photographically. Our observation indicated that the WT plants wilted earlier than the transgenic lines after drought treatment (Figure 6a), but that all could recover soon after being re-watered (Figure 5b). To compare the water loss rates of transgenic and WT plants, we measured the relative water content of intact leaves (RWCi) on WT and OsPIP1;3 overexpression (OX) plants before and after the plants were given 6-day drought treatment, and also measured the relative water content of detached leaves (RWCd) from the WT and the OsPIP1;3 OX plants before and after the leaves were dried for 3h at room temperature. The data indicated that the RWCi values of transgenic lines were significantly greater than those of the WT plant after 6days of water depletion. The RWCd values did not differ, however (Figure 6b). Further investigation indicated that the transgenic and the WT leaves had the same stomatal densities (Figure S12), suggesting that the overexpression of OsPIP1;3 may result in greater water supply from the roots rather than differing water loss (Tr) from the leaves. To confirm the above hypothesis, we used 1-month-old transgenic and WT tobacco plantlets to measure root hydraulic conductivity (Lpr) in a pressure chamber (Figure S13). As a result, the Lpr values of transgenic and WT plants were similar when plants were grown in optimal conditions. The Lpr values significantly differed after the plantlets were treated with 20% PEG, however. The transgenic plants had higher root hydraulic conductivities than the WT plants under stress conditions (Figure 6c), further confirming that OsPIP1;3 may function as an active water channel in planta to enhance water uptake from the soil and thereby promote plant resistance to water deficit. As the transgenic plants showed more vigorous growth than the WT plants, this led us to investigating how OsPIP1;3 expression affects WUE. To this end, we grew 1-month-old WT and OX plantlets under well-irrigated conditions with the relative soil water content (RSWC) held at 85%. The pots were covered with a gas-resistant foil to minimize evaporation. Growth was assessed by monitoring leaf area, leaf number and plant height over time. Each pot was irrigated with an equal volume of water every 4days. After 32days, the leaf area, leaf number and plant height were measured and each individual plant was harvested and weighed. The data show that the transgenic plants had greater leaf number, plant height and WUE values than the WT plants (Figure 7a–c), but that leaf area did not differ (Figure S14). Overexpression of OsPIP1;3 increased the water-use efficiency (WUE) of Nicotiana benthamiana (tobacco) plants. One-month-old transgenic and wild-type (WT) tobacco plantlets were grown in pot soils with 85% relative water content. The pots were covered with foil to prevent water evaporation and were irrigated every 4days for a total of 32days. (a) Average numbers of expended leaves. (b) Average plant height. (c) Water-use efficiency (WUE). Bars with different lowercase letters were significantly different (Duncan’s multiple range test: P 0.05). Plants are sessile and cannot escape from their surroundings. Thus, they must adapt to varying environmental conditions through various molecular responses. Water is one of the most important molecules for all living organisms. The regulation of water transport is crucial for many physiological activities, e.g. absorption, transpiration and respiration. Most AQPs are tetrameric channel proteins allowing the transmembrane passage of small solutes and water into and out of living cells (Gena et al., 2011). In plants, members of the PIP AQP subfamily are important for the maintenance of plant water status through the control of cell and tissue hydraulics. We identified the SNPs of OsPIP1;3 genes from two rice subspecies. In sorghum, hundreds of SNPs were identified from AQP loci among 50 sorghum accessions and most of the SNPs are related to the expression of AQP genes in response to waterlogging stress (Kadam et al., 2017). Like other plant PIP1s, single OsPIP1;3 showed abnormal cellular localization and negligible water channel activity. Most PIP1 channels fail to be targeted to the plasma membrane and reside in intracellular membranes, typically the ER when heterologously expressed in oocytes or yeast, or even in plant cells (Katsuhara et al., 2002; Liu et al., 2013). The intracellular retention of PIP1s is either caused by missing plasma membrane trafficking signals or by existing ER retention motifs, whereas the interaction and heterotetramerization of certain pairs of PIP1s and PIP2s can significantly enhance water permeability (Bienert et al., 2018). For example, the combinations of OsPIP1;1/OsPIP2;1 (Liu et al., 2013), OsPIP1;3/OsPIP2;2 (this work) and BvPIP1;1/BvPIP2;1 (Jozefkowicz et al., 2013) could significantly enhance PIP1 activities through their physical interactions with PIP2s. The influence of AQPs on hydraulic conductance in plants is important in regulating the plant transpiration rate, particularly under conditions of developing soil water deficit stress and elevated atmospheric vapor pressure. As AQPs represent a big gene family, only some members are involved in the response to water deficit and drought stress (Lian et al., 2006; Song et al., 2018). Genetic engineering has brought insights to the physiological functions of AQPs, and has increased our understanding of their roles in water transport and abiotic stress responses. In Arabidopsis, the expression of AtPIP1;3, AtPIP1;4, AtPIP2;1 and AtPIP2;5 increased but the expression of AtPIP1;5, AtPIP2;2, AtPIP2;3 and AtPIP2;6 decreased in response to drought stress (Jang et al., 2004). Alexandersson et al. (2005) reported that the expression of AtPIP1;3 and AtPIP2;1 decreased in response to drought stress, however. In this work we demonstrated that the expression of OsPIP1;3 could be transcriptionally and translationally induced by drought stimulus (Figure 1b,c). OsPIP1;3 showed various response patterns to drought in upland and lowland rice cultivars (Lian et al., 2006), however, indicating that PIP expression induced by water deficit is a complicated response and that the molecular mechanisms underlying the regulation of PIP expression in drought conditions remain elusive. Here, we revealed experimentally that the overexpression of OsPIP1;3 enhanced root hydraulic conductance and WUE in transgenic tobacco plants compared with WT plants. Plant WUE, represented by the ratio of carbon assimilation to water consumption, describes the efficiency of a plant in optimizing carbon gain while minimizing water loss. In practical terms, high WUE is important for agriculture in water-deficient areas and improved WUE confers a competitive advantage when water is limited. Intriguingly, AQPs have been proven to be the major determinants of WUE in plants (Flexas et al., 2013a). Constitutive overexpression of the fungal ThAQGP gene enhanced the photosynthetic rate of tobacco plants under water stress and hence increased the efficiency of water use, despite the high transpiration rate. The conservation of the higher transpiration rate is an interesting mechanism that ensures not only continuous CO2 uptake but also the continued supply of nutrients, reduced leaf temperature and enhanced plant growth (Vieira et al., 2017). We previously found that OsPIP1;3 might play an important role in rice drought avoidance by the rapid rolling of leaves to minimize water transpiration (Lian et al., 2004). The conversion from drought avoidance to drought resistance is probably achieved through enhanced root hydraulic conductivity, ensuring water supply but not reducing water evaporation in OsPIP1;3 transgenic tobacco plants. It has been reported that AQPs contribute up to 80–90% of rice root hydraulic conductivity (Lpr) under well-watered and drought conditions, and that root plasma membrane aquaporin (PIP) expression and root anatomical properties correlate with hydraulic traits (Grondin et al., 2016). Matsuo et al. (2009) reported that different genotypic and ecotypic rice cultivars demonstrated different root hydraulic conductivities under hydroponic or aerobic water regimes. In flooding conditions, the root hydraulic conductance of most rice cultivars were greatly restrained and showed similar effects on shoot growth; however, in water-saving or drought conditions, different genotypic cultivars demonstrated various root hydraulic conductances, directly resulting in differing resistance to water deficiency. The OsPIP1;3 gene in an upland rice cultivar (subspecies indica cv. Zhonghan 3) was responsive to drought stress and overexpression of OsPIP1;3 in stress-sensitive lowland rice (subspecies japonica cv. Zhonghua 11) improved its water status by increasing root conductivity (Lian et al., 2004), thereby confirming our observations in tobacco. Ectopic expression of plant AQP genes like Musa (banana) MaTIP1;2, apple MdPIP1;3 and Arabidopsis PIP1b enhanced plant resistance to water deficit under osmotic stress or extreme temperatures (Almeida-Rodriguez et al., 2010; Wang et al., 2017a; Song et al., 2018). Apple MdPIP1;3 allowed for more rapid guard cell engorgement in transgenic leaves and therefore resulted in reduced water loss through stomata under drought conditions (Wang et al., 2017a). Interestingly, the overexpression of an aquaglyceroporin, ThAQGP, from the fungus Trichoderma harzianum could also show active channel functions and conferred significant drought resistance in tobacco by presenting a higher transpiration rate, stomatal conductance, photosynthetic efficiency and WUE than found in the wild-type specimen (Vieira et al., 2017). Some other AQPs conferred decreased stress tolerance, however (Aharon et al., 2003; Wang et al., 2015). It has been suggested that drought stress tolerance in aquaporin-overexpressing plants is closely related to aquaporin water transport activity in the roots, increasing the ability to retain water (Zhuo et al., 2016). Almost all OsPIP1;3 transgenic lines had higher fresh shoot and root weights than the WT under stress conditions, suggesting that plants adjust water content by modifying their ability to retain water in shoots and roots. Functional analyses using aquaporins predict their roles in plant–water relationships. When the aquaporin gene is silenced in a plant, decreased cellular osmotic water permeability and reduced water stress resistance are observed (Kaldenhoff et al., 1998). The timely closure of stomata in water-deficit conditions is a complicated physiological process. It has been reported that some AQPs directly participate in stomatal closure. Arabidopsis AtPIP2;1 is phosphorylated and activated by OST1 (SnRK2.6) in guard cells to facilitate water and H2O2 flux and stomatal closure (Rodrigues et al., 2017). The in planta and in vitro data from our study show that OsPIP1;3 can conduct water and NO3− across the membrane system, although whether OsPIP1;3 directly participates in stomatal closure or not needs to be elucidated further. The plant anion channel SLAC1 can be activated by nitrate to induce stomatal closure (Hedrich and Geiger, 2017); however, the sensitivity of monocot and dicot SLAC1s to nitrate-dependent stimulus differs fundamentally. Further evidence shows that when the motif of nitrate-insensitive AtSLAC1 was replaced by a tandem amino acid residue motif in the HvSLAC1 anion channel of guard cells in Hordeum vulgare (barley), AtSLAC1 was converted into a grass-type nitrate-sensitive channel (Schäfer et al., 2018), indicating that nitrate is an important signaling molecule in plant stomatal closure. Hence, we can postulate that the enhancement of nitrate uptake from soil and transport in OsPIP1;3 transgenic plants may be one of the triggers for guard cell closure through activating anion channels like SLAC1 and SLAH3 under water-limiting conditions, although this hypothesis must be confirmed experimentally. Plant SnRK2 kinase OST1 is a major kinase in regulating stomatal closure through phosphorylating anion channels. SnRK1s have many functions that are broadly and synergistically involved in the metabolism of nitrogen/carbon and in responses to biotic/abiotic stress (Crozet et al., 2014). In our recent work, we identified that a PIP1 of wild soybean interacts with SnRK1 kinase, and found that its S131 residue in the first inner loop of GsPIP1 was a potential phosphorylation site of SnRK1 (Song et al., 2019). This putative phosphorylation site is highly conserved in all PIP1 homologs. Like OST1 phosphorylating S121 to determine the activity of AtPIP2.1, SnRK1 may affect its substrate permeability by phosphorylating the S131 residue. For example, rice plasma membrane receptor-like kinase leaf panicle 2 (LP2) physically interacted with OsPIP1;1, OsPIP1;3 and OsPIP2;3 in the plasma membrane. Intriguingly, it negatively affected drought resistance by promoting stomatal opening and decreasing the H2O2 level in the guard cells (Wu et al., 2015), suggesting that the expression of PIPs including OsPIP1;3 may be generally altered by the upstream signals to actively participate in the maintenance of water homeostasis in plants. In our work, the OsPIP1;3 transgenic plants grew more vigorously than the WT plants. OsPIP1;3 might enhance the uptake of carbon and nitrogen, and subsequently promoted plant photosynthesis and growth. Previous reports indicated that some AQPs could enhance leaf mesophyll CO2 conductance (gm), photosynthesis (AN) and hydraulic conductivity (Nakhoul et al., 1998; review by Flexas et al., 2006; Groszmann et al., 2017). The conductance of CO2 in photosynthesizing leaves is commonly divided into three components: boundary layer, stomatal, and mesophyll conductance. Mesophyll conductance (gm) is the most restricted sense of a physical diffusion phenomenon. Any of these components, but especially mesophyll conductance with low diffusion, poses a limitation to photosynthesis whenever CO2 levels are not saturated. Similar to our work, when the OsPIP1;2 gene from Oryza sativa L. cv Nipponbare was overexpressed in rice, the transgenic rice lines had significantly higher photosynthesis rate and biomass accumulation (Xu et al., 2019). Overexpression of the NtAQP1 gene in tobacco increased gm and the net photosynthesis rate, measured using a combination of gas exchange, chlorophyll fluorescence and on-line 13C isotope discrimination techniques. It was reported that Arabidopsis had an unexpected lower gm than was reported for other annual plants, and the low gm and low capacity for carboxylation were major factors in limiting photosynthesis (AN) (Flexas et al., 2007). Here, we observed that ectopic expression of OsPIP1;3 in tobacco and heterogeneous expression of OsPIP1;3 in cyanobacteria did promote photosynthesis and growth of the transgenic plants and cyanobacteria. Nevertheless, our previous work indicates that cyanobacterium Synechococcussp. PCC7942 has only one aquaporin-like gene, and that knock-out of this gene caused significant abrogation of CO2 uptake and the inhibition of cell growth (Ding et al., 2013). Plant AQPs like tobacco NtAQP1 (Uehlein et al., 2003), AtPIP2;1 (Wang et al., 2016) and HvPIP2s (Mori et al., 2014) were reported to mediate CO2 permeability across the membranes, and to promote photosynthesis and plant growth. By mutagenesis, Mori et al. (2014) reported that I254 of HvPIP2;3 was a key residue in the sixth TM domain to determine its CO2 permeability, and the substitution of isoleucine with methionine (I245M) abolished CO2 permeability in HvPIP2;4. By sequence alignment, OsPIP1;3 also has an isoleucine residue in the corresponding site (Figure S1), suggesting that OsPIP1;3 may possess CO2 permeability, and supporting our observation that OsPIP1;3 overexpression plants grew better and accumulated more biomass than the WT plants. Because CO2 and water vapor share a common diffusion pathway from the air to the spaces inside leaves, an analysis of water vapor fluxes allows for accurate estimates of boundary layer and stomatal diffusion conductance. Plant WUE, represented by the ratio between carbon gain (photosynthesis) and water loss (transpiration), describes the efficiency of the plant in optimizing carbon gain while minimizing water loss. Improved WUE confers a competitive advantage when water is limited, and aquaporins are major determinants of WUE in plants (Flexas et al., 2013a). The importance of mesophyll conductance (gm) in optimizing water use during drought is shown by the correlated changes in mesophyll conductance and WUE (Maurel et al., 2016). In our work, overexpression of the OsPIP1;3 gene in tobacco plants seems to increase leaf WUE and could be explained by gas-exchange results and water balance. In practice, sustainable agriculture requires a major reduction in water use and in fertilizer use. It is therefore imperative to improve not only yields but also water productivity or WUE, i.e. the amount of carbon gained per unit water used, as well as nitrogen-use efficiency (NUE). Nitrate is an important nutrient mineral for plants. The uptake and transport of nitrogen nutrients are fulfilled by multiple pathways. Besides high- and low-affinity nitrate transporters like NRT2.1 and NRT2.4, some other channels may also possess these functions. In plants, the expression of MfPIP2-7 from Medicago falcate could enhance plant growth under NO3−-deficient conditions (Zhuo et al., 2016). Until now, only human AQP6 was reported to allow the transport of nitrate. The point mutation Asn60Gly converted AQP6 from an anion channel to a water channel (Ikeda et al., 2002) and Leu51Arg converted AQP5 from a water channel to an anion channel, demonstrating the anion permeability sequence I− NO3−≅NO2− Br− Cl− HCO3− gluconate (Qin and Boron, 2013), suggesting that key individual residues in these channels determine their important properties. In this work, we determined the in vitro NO3− permeability of OsPIP1;3 by patch-clamp technique, and compared root hydraulic conductivities, growth properties and WUE between OsPIP1;3 transgenic and WT plants. The data clearly indicate that OsPIP1;3 significantly improved the related physiological indexes. In Helianthus annuus (sunflower), the nitrate availability greatly and rapidly affected dynamic changes in root hydraulic resistance. The sunflower roots became more permeable to water within 20–30min after the application of nitrate and this increase in permeability continues over time, with the maximum enhancement dependent upon the nitrate concentration (Gloser et al., 2007). The experiments with Lotus japonicus also showed that significant changes in root hydraulic properties occurred following nitrate supply (Clarkson et al., 2000). Moreover, at the transcriptional level, the array assay revealed that the expression of many AQP genes could be induced by nitrate in tomato (Wang et al., 2001). Hence, NO3− is not only a nutrient for plant growth but is also a signaling molecule that affects PIP expression at both transcriptional and translational levels to govern plant root hydraulic conductivity (Lpr) (Li et al., 2016). Similarly, nitric oxide (NO) was also reported to be an elicitor to stimulate the transcription of OsPIP1;1, OsPIP1;2, OsPIP1;3 and OsPIP2.8 in rice germinating seeds (Liu et al., 2007). In our scenario, how nitrate affects AQP functions and how AQPs affect nitrate uptake and transport in planta are still matters for further investigation. Overall, we systematically characterized the water and nitrate permeability of OsPIP1;3 from a drought-resistant rice variety by biochemical and biophysical approaches. The expression of OsPIP1;3 enhanced plant drought resistance and vegetative growth, offering the opportunity to advance a more holistic understanding of the additional functions played by plant AQPs and providing insights into their translational potential. The seeds of rice O.sativa L. indica cv. Zhonghan 3 and tobacco (N.benthamiana) were obtained from the Key Laboratory of Agricultural Biological Functional Genes, Northeast Agricultural University, China, and the Institute of Biotechnology, Akita Prefectural University, Japan. Anti-OsPIP1;3-specific antibody (cat no. AS09 504) was purchased from AgriseraAB (https://www.agrisera.com). Anti-FLAG (cat no. F3165-2MG) and anti-GFP (cat no. G1546) antibodies were purchased from Sigma-Aldrich (https://www.sigmaaldrich.com). The plasmids bearing the complete ORFs of OsPIP1;3 and OsPIP2;2 genes were isolated from a rice cDNA library. The genes flanked by BglII were inserted into the pXβG-ev1 vector to result in pXβG-OsPIP1;3 and pXβG-OsPIP2;2 for expression in Xenopus laevis oocytes or pXβG-OsPIP1;3-EGFP and pXβG-OsPIP2;2-EGFP for the visualization of expression of these genes in oocytes. For gene transient expression in rice protoplasts, the plant gene expression cassette (35Spro-MCS-NOS3ter) was inserted into the pUC18 vector. OsPIP1;3-EGFP or OsPIP2;2-mDsRed was cloned into SalI and EcoRI sites to generate pUC18-OsPIP1;3-green or pUC18-OsPIP2;2-red. To maximize the production of recombinant OsPIP1;3 protein in E.coli, the codons of the OsPIP1;3 gene were modified by referring to preferential codons of AqpZ and AqpM (Calamita et al., 1995; Kozono et al., 2003). The optimized sequence was synthesized at Blue Heron (http://www.blueheronbio.com). The plasmid pTrc10HisAqpZ (Borgnia et al., 1999) was digested with EcoRI and SalI to release AqpZ for insertion of OsPIP1;3 to generate pTrc10His-OsPIP1;3 with a 10×His tag (MGHHHHHHHHHHSSIEGRHEF) in the N terminus of OsPIP1;3. To generate the transgenic plants, the 35Spro-Flag-OsPIP1;3-NOS3ter cassette was inserted into EcoRI and SacI sites of the Ti plasmid binary vector pPBEL, derived from pCambia2300, to obtain pPBEL-Flag-OsPIP1;3. The general procedure was similar to a previously described protocol (Liu et al., 2013), with some modifications. Briefly, the capped complementary RNA (cRNA) samples were synthesized using T3 RNA polymerase of the mMESSAG mMACHINE High Yield Capped RNA Transcription Kit (Ambion, now ThermoFisher Scientific, https://www.thermofisher.com) after linearization of the pXβG-OsPIP1;3 plasmid. Fresh oocytes (1.0–1.2mm in diameter) of stageV–VI were treated with 0.1% collagenase (Clostridium histolytium, type II; Sigma-Aldrich) to exfoliate the oocyte membrane. A 50-nl (10–50ng) volume of cRNA solution or water was injected into the oocytes using a nanoject injector (Narishige, http://narishige-group.com). The injected oocytes were incubated in 2-(N-morpholine)-ethanesulphonic acid (MES)-buffered saline (MBS; 200mOsmin) for 48h at 20°C. The fluorescence generated by the chimeric OsPIP1;3-EGFP protein could be observed under a confocal laser scanning microscope. To measure water channel activity, the injected oocytes were transferred into hypotonic solution (one-third diluted MBS, 70mOsmout). The sizes of the oocytes at intervals of 1sec were captured by digital camera (Shimadzu, https://www.shimadzu.com), and the swelling rates of oocytes at each time point were calculated with motic images plus 21s (Shimadzu). Water permeability (coefficient of osmotic water permeability, Pf) was calculated as previously described (Kozono et al., 2003). The pTrc10His-OsPIP1;3 construct obtained was transformed into E.coli strain BL21-CodonPlus(DE3)-RIL (Stratagene, now Agilent, https://www.agilent.com). The cells were grown in 10-L Luria Bertani (LB) medium in a fermentation tank to an OD600 of 1.5 prior to the addition of isopropyl-β-d-thiogalactoside at a final concentration of 1mm, and then vigorously shaken at 37°C for 2h. The cells were harvested and resuspended in 120ml of ice-cold lysis buffer (100mm K2HPO4, 1mm MgSO4, 0.4mgml−1 lysozyme, 0.1mgml−1 DNase I, and 1mm phenylmethylsulfonyl fluoride) and then subjected to three French press cycles (1266 kg cm−2) at 4°C. The cell lysates were centrifuged at 6000g for 10min at 4°C. The supernatants were transferred to fresh tubes and were ultracentrifuged at 200000g for 60min at 4°C. To efficiently extract 10×His-OsPIP1;3 protein, the membrane pellets were incubated with solubilization buffer (25mm Na2HPO4, 10% (v/v) glycerol, 5mm β-mercaptoethanol, and 125mm NaCl, pH 8.0) containing 5% or 10% OG (n-octyl-β-d-glucoside), 5% DM (n-dodecyl-β-d-maltoside), 5% C12E8 (octaethylene glycol monododecyl ether), 5% OPOE (n-octylpolyoxyethylene), 3% LDAO (N,N-dimethyldodecylamine N-oxide), 1.5% FC12 (Fos-choline12) and 0.4% or 4% PFO (perfluorooctanoic acid), with gentle agitation at 4°C overnight. All of the samples were centrifuged at 100000g for 60min at 4°C to pellet the insoluble materials. Equal volumes of the supernatant were resolved in 12% SDS-PAGE gels. The protein extraction efficiency using different detergents could be evaluated by Western blotting using anti-OsPIP1;3 antibody. The purification of recombinant 10×His-OsPIP1;3 was performed by mixing with prewashed Ni-NTA agarose beads and incubated with gentle agitation at 4°C overnight. The beads were washed with cold wash buffer (2% detergent, 100mm K2HPO4, 10% glycerol, 5mm β-mercaptoethanol, 200mm NaCl, and 100mm imidazole, pH7.0) three times to remove nonspecifically bound materials. Recombinant 10×His-OsPIP1;3 was eluted with 2ml of elution buffer (0.4 % PFO, 25mm Na2HPO4, 10% glycerol, 5mm β-mercaptoethanol, 150mm NaCl, and 1m imidazole, pH7.0). The protein obtained was subsequently dialyzed and concentrated for functional characterization. The purified 10×His-OsPIP1;3 protein produced from above was reconstituted into liposomes as previously described (Borgnia et al., 1999). Briefly, reconstitution mixtures were prepared at 23–28°C room temperature by the sequential addition of the following solutions to certain concentrations: 100mm 3-(N-morpholino)propanesulfonic acid (MOPS; pH7.5), 1mm NaN3, 1mm dithiothreitol (DTT), 1.25% (w/v) N-octyl-β-d-glucopyranoside, 100μgml−1 purified 10×His-OsPIP1;3 protein and 9mgml−1 of bath-sonicated E.coli polar lipids (Avanti Polar Lipids Inc., https://avantilipids.com). The mixture was vigorously vortexed and then incubated on ice for 20min. The reconstitution mixture was rapidly passed through a 23-gauge needle into 25 volumes of reconstitution buffer [50mm MOPS, 150mm N-methyl-d-glucamine, 1mm DTT, and 0.5mm phenylmethylsulfonyl fluoride (PMSF)] to dilute the detergent. Proteoliposomes or liposomes were collected by centrifugation at 123000g, held at 4°C for 1h, and washed three times. The lipid pellets were resuspended in reconstitution buffer. The diameters of proteoliposomes and liposomes were determined by electron microscopy. To assess the osmotic water permeability of the 10×His-OsPIP1;3 reconstituted proteoliposomes, light-scattering experiments were performed in a stopped-flow light-scattering apparatus, as previously described (Calamita et al., 2006). The coefficient of osmotic water permeability (Pf) of the liposome or proteoliposome membranes was calculated from the Ki, as previously described (van Heeswijk and van Os, 1986), using the equation: Pf=(Ki⋅V0)/(Av⋅Vw⋅ΔC), where Ki is the fitted exponential rate constant, V0 is the initial mean vesicle volume, Av is the mean vesicle surface, Vw is the partial molar volume of water (18cm3mol−1) and ΔC is the osmotic gradient. The Pf unit is expressed as [µm3 (H2O)]/[µm2 (membrane area)×s (time)]=µmsec−1. The medium osmolarity was verified by a vapor-pressure osmometer (Wescor, http://water.wescor.com). In some experiments, the liposome specimens were pre-incubated for 5min with 300mm HgCl2, a sulfhydryl compound blocking the AQP channels. In other experiments, to verify the blocking action of the Hg2+ ion, the HgCl2 treatment was followed by a 15-min exposure to 10mm of the reducing agent β-mercaptoethanol (β-ME). The OsPIP1;3 gene was cloned into the pEGFP-C1 vector. The resulting construct or empty pEGFP-C1 vector was transfected into HEK293 cells grown in a six-well tissue culture dish using Lipofectamine 2000 (Life Technologies, now ThermoFisher Scientific, https://www.thermofisher.com). Forty-eight hours after transfection, the cells were dissociated by trypsin treatment and kept in complete serum-containing medium and were re-plated on 35-mm tissue culture dishes in a tissue culture incubator until recording. The patch-clamp assay was performed according to the method described by Hamill et al. (1981). The standard bath solution contained (in mm): 145 sodium methanesulfonate (Na-MS), 5 NaCl and 10 HEPES, pH7.4. NaMS and NaCl were replaced by 150mm NaHCO3 or 150mm NaNO3 to measure HCO3− and NO3− currents, respectively. The pipette solution contained (in mm): 145 Cs-MS, 2.5 MgCl2, 1 EGTA, and 10 HEPES, pH7.4. The patch pipettes were pulled from borosilicate glass (Harvard Apparatus, https://www.harvardapparatus.com) and heat polished to a resistance of 5–7MΩ. Data were acquired using an AxoPatch 200B amplifier (Molecular Devices, https://www.moleculardevices.com) and a low-pass analogue filter set to 1kHz. The current signal was sampled at a rate of 20kHz using a 1322A digitizer (Molecular Devices) and further analyzed with pclamp9 software (Molecular Devices). After the patch pipette was attached to the cell membrane, the giga seal (5–10GΩ) was formed by gentle suction. The whole-cell configuration was formed by a short electric zap or suction to rupture the membrane patch. The holding potential was set to −100mV. To obtain the current–voltage relationship curve, the membrane was ramped from the holding potential (−100mV) to 100mV for 1sec and then returned to the holding potential (She et al., 2018). All electrophysiological recordings were repeated at least five times using different patches. Most data points shown are means±SEMs (n=5 independent experiments). The rice O.c. cell line was isolated from seedling roots of the indica-type rice O.sativa L. accession C5924 (Baba et al., 1986) and was obtained from Dr Isao Shimamoto. The cells were subcultured at 25±2°C in the dark at 2 g in Murashige and Skoog medium supplemented with 1.0mgL−1 of 2,4-dichlorophenoxyacetic acid at regular intervals of 2weeks. The cells were harvested and treated in 4% (w/v) cellulase and 1% (w/v) Macerozyme solution at 30°C for 5h. Afterwards, the digested cells were filtered through a steel mesh (38μm) and the protoplasts were collected by centrifugation at 100g for 10min. The transient transformation of rice O.c. protoplasts was performed according to the method described by Shen et al. (2014). Briefly, the rice protoplasts obtained were resuspended in W5 solution (154mm NaCl, 125mm CaCl2, 5mm KCl, 2mm MES, pH5.8) at the desired cell density of 1.5–2.5×106protoplastsml−1, determined by counting the cells with a hemacytometer. Ten-forty μg of pUC18-OsPIP1;3-green or/and pUC18-OsPIP2;2-red plasmids were added to 200μl of suspended protoplasts and gently mixed in a 2-ml centrifuge tube, and then 250μl of PEG solution [40% (w/v) PEG 4000, 0.3m mannitol, 0.1m CaCl2] was added and gently mixed. The protoplast/DNA mixture was incubated at room temperature for 20min and was diluted with 2ml of R2S medium, and was gently mixed by inverting the tubes to stop the transfection process. The protoplast suspension was transferred to a 35mm×10mm Petri dish and was incubated for 6–18h at 27°C. The fluorescent signals of transformed protoplasts were observed under a confocal microscope. The rice root tips from young seedlings were fixed in 0.1m phosphate buffer (pH7.4) with 4% paraformaldehyde and 0.05% glutaraldehyde at 4°C for 1h. The root samples were washed and dehydrated in 70% ethanol for 10min and 96% ethanol for 20min, and then were transferred into a mixture of LR-White resin (London Resin Co. Ltd, now Agar Scientific, http://www.agarscientific.com) and 96% ethanol (1:1) for 2h and then in pure LR-White for 5h. The roots were embedded in 100% LR-White gelatin capsule at 50°C for 24h. Ultra-thin sections were cut with a diamond knife (Jumdi; Electron Microscopy Sciences, Hatfield, PA, USA, https://www.emsdiasum.com/microscopy/default.aspx) and then mounted on 200 nickel mesh grids. The sections were blocked in 5% BSA solution (in PBS, pH7.2) for 2h, and then were immunostained with an anti-OsPIP1;3-specific rabbit antibody or rabbit serum, and were subsequently stained with gold colloid labeled anti-rabbit IgG antibody. Lastly, the ultra-thin sections were stained with TI blue and were observed with a JEM-1230 microscope (JEOL, https://www.jeol.co.jp). pPBEL-OsPIP1;3 plasmid harboring OsPIP1;3 under the control of the 35S promoter and the nopaline synthase terminator was introduced into Agrobacterium tumefaciens EHA101 by electroporation. Tobacco transformation was performed according to the prototcol previously described by Ding et al. (2004). Briefly, the leaves from 3-week-old tobacco plantlets growing on sterile medium were excised and cut into small pieces of approximately 0.5cm2. The leaf explants were immersed in the agrobacterial suspension for 1min and were then transferred onto sterile filter paper for a few seconds prior to MS medium for 2days in the dark. The leaf explants from the dark were transferred onto MS medium supplied with 1mgL−1 BAP, 100mgL−1 kanamycin sulfate and 200mgL−1 cefotaxime, and were cultured under light for 3–4weeks. The shoots with 1cm of height were excised from the calli and were transferred onto rooting medium. The plantlets with roots were transferred to soil in the glasshouse for propagation to generate T2 progeny. The expression levels of OsPIP1;3 in homologous progeny were identified by qPCR and Western blotting. The hydraulic conductivities (Lpr) of the transgenic and control roots were measured according to the method described by Murai-Hatano et al. (2008). Briefly, 1-month-old OsPIP1;3 and wild-type tobacco seedlings were divided into two groups. One group of plantlets were cultured in half-strength MS medium and the other group of plantlets were cultured in half-strength MS medium containing 20% PEG to mimic drought treatment for 6h. Afterwards, individual plantlets were fixed in a pressure chamber filled with nutrient solution. In order to apply static pressure to the root medium, the gap between the stem and the metal lid was sealed with dental paste and then the upper shoot was cut off with a razor blade at the root base. Immediately, the pressure in the chamber was stepped up from 50kPa to 500kPa above atmosphere. The xylem sap exudated over a period of 10min was collected. The rate of sap flow (J=V/Ar·t) was determined by measuring sap volume per unit root area and per unit time. The surface area of the root (Ar) was measured using the WinRHIZO system (Regent Instruments Inc., https://www.regentinstruments.com). The Lpr of individual samples was calculated from the slope of a plot of J/(ΔP+σΔΨs), where ΔP is the applied pressure, σ is the reflection coefficient for nutrient salts in the xylem, and ΔΨs is the difference in osmotic potential between the exudated xylem sap and the root medium (hydroponic solution). The selected homologous T2 seeds were germinated and grown under normal conditions, with a 12-h photoperiod (200µmolm−2sec−1) and with day/night temperatures of 30/22°C, respectively, in soil pots with regular irrigation in a growth chamber. To investigate the effect of the overexpression of OsPIP1;3 on the growth and development of the transgenic plants, the transgenic and WT plants were grown under equal conditions and their growth indexes, including plant size, and fresh and dry weights, were measured. One-month-old plantlets were used to measure WUE according to the method described by Wang et al. (2017b). The OX and the WT plantlets were grown in pots with SRWC maintained at 85%. The pots were covered with foil to prevent water evaporation. Each pot was irrigated with equal volumes of water to achieve 85% SRWC every 4days for a total of 32days. Each plant was weighed and the WUE was calculated as follows: WUE=(FWf−FWi)/Wu, where FWi and FWf are the fresh weights of each plantlet at 0 and 32days, and Wu is the total water used for the 32days. Stomatal density was determined essentially by light microscopy of nail polish imprints taken from the top and bottom surfaces of the leaves. Total chlorophyll from tobacco leaves was extracted in 80% aqueous acetone and the contents were measured at wavelengths of 633 and 645nm by spectrophotometry. Fifty-day-old transgenic and WT tobacco plants growing in normal conditions were selected for the measurement of gas exchange and chlorophyll fluorescence (Flexas et al., 2007, 2013b). The transpiration and photosynthesis rates of newly expanded leaves were measured using a 6400-40 Leaf Chamber Fluorometer (LICOR Biosciences, https://www.licor.com) under natural sunlight in the glasshouse. The CO2 concentration in the cuvette was set at 400μmolmol−1 with a CO2 cartridge. The temperature of the leaf chamber was maintained at 25°C, with a photosynthetically active radiation of 1500μmolm−2sec−1. After equilibration to a steady state, the fluorescence was recorded (Fs) and a 0.8-sec saturating pulse of light (approximately 8000μmolm−2sec−1) was applied to measure the maximum fluorescence (Fm). The actual photochemical efficiency of photosystemII, ΦPSII=(Fm−Fs)/Fm, was used to calculate the electron transport rate (JETR=α×PFD×ΦPSII), where α is a term including the product of leaf absorption and the partitioning of absorbed quanta between photosystemsI and II, determined as the slope of the relationship between PSII and CO2 obtained by varying light intensities under non-photorespiratory conditions in an atmosphere containing 1% O2, and PFD is the photosynthetically active photon flux density. The net photosynthetic rate (AN), stomatal conductance (gs), transpiration rate (Tr) and intercellular CO2 concentration (Ci) were measured and calculated. According to Harley et al. (1992), the leaf mesophyll conductance (gm) was estimated using the equation: gm=AN/{Ci−[Γ*×JETR+8Γ*(AN+Rd)]/[JETR−4(AN+Rd)]}, where Γ* represents a CO2 compensation point without Rd. Rd is the rate of mitochondrial respiration in the light and was calculated using a linear regression line of the relationship between AN and Ci for the range of Ci below 150μmolmol−1. CO2 concentration at the carboxylation sites in the stroma was calculated by using the gm value obtained. Based on Fick’s law of diffusion, the net photosynthetic flux (AN) can be expressed as: AN=gs(Ca–Ci)=gm(Ci–Cc), where Ca is the CO2 concentration in the atmosphere, Ci is the CO2 concentration in the substomatal cavity and Cc is the chloroplast stroma, and can be calculated by the equation. For drought tolerance treatment, young tobacco plants (2months old) were grown in a growth chamber without irrigation for 6days. Afterwards, the plants were then re-watered. The phenotypes were observed and recorded using photography. To measure the relative water content (RWC) of intact leaves on plants, representative leaves with a certain area size were collected before and after a 6-day drought treatment. To measure the RWC of the detached leaves, representative leaves of the same sizes and age as described above were detached from the plants growing in normal conditions, and were weighed before and after air drying for 3h. The water contents of all leaf samples were measured by comparing the weight differences before and after dehydration at 85°C for 1h. The relative water contents of intact leaves (RWCi) and detached leaves (RWCd) were calculated using the following equation: RWC=(WFW−DWw)/(TFW−DWt)×100%, where TFW is the fresh weight of turgid leaves, WFW is the fresh weights of wilted leaves after a 6-day drought (for intact leaves) or after 3h of air drying (for detached leaves), DWt is the dry weight of turgid leaves and DWw is the dry weight of wilted leaves after a 6-day drought (for intact leaves) or after 3h of air drying (for detached leaves). The total soil water content was measured by determining soil weights before and after drying in an oven for 12–16h at a temperature of 110°C. The SRWC was calculated using the following equation: SRWC=(Wa−Wd)/Wb−Wd)×100%, where Wb is the soil weight before drought treatment, Wa is the soil weight after drought treatment and Wd is the soil weight after drying. Wild-type Synechococcussp. PCC7942 cells (strain kindly provided by Dr Zhihan Lin, Institute of Oceanology, Chinese Academy of Sciences, China) were grown and maintained in liquid BG-11 medium (Ding et al., 2013). The pSyn_1 vector for transformation of PCC7942 strain was purchased from Invitrogen (now ThermoFisher Scientific, https://www.thermofisher.com). The target gene with FLAG tag sequence was inserted into HindIII and EcoRI sites. The resulting construct or empty vector was transformed into cyanobacterial cells. The transformants were selected on solid BG-11 medium containing 50μgml−1 spectinomycin. To compare the growth rates, equal quantities of cells were dropped onto solid BG-11 media lacking NaHCO3 and grown in normal air or CO2-free air, which was obtained by passing air through two CO2 trap filters filled with NaOH and Ca(OH)2. The air was supplied at 30mlmin−1 for 10days at 50μmolphotonsm−2sec−1 of continuous illumination at 28°C. The cell growth patterns on solid media were photographed. Moreover, the cells were subjected to liquid culture and the cell densities (OD730) were measured at the exponential phase. All experiments were carried out using biological duplications for each treatment, and were replicated on at least three occasions. Multiple-comparison tests were performed by Duncan’s test using spss16.0 (IBM, https://www.ibm.com). The significance level was set at P 0.05. We are thankful to Drs Mari Murai-Hatano and Atsuko Sakurai of the National Agricultural Research Center for Tohoku Region for providing us with anti-OsPIP1;3 antibody, and to Dr Isao Shimamoto of the Nara Institute of Science and Technology for providing us with rice O.c. cells. This work was supported by the grants from National Natural Science Foundation of China (31670272) to X.D., the Natural Science Foundation of Heilongjiang province (C2017014) to X.D. and the Starting Fund of Northeast Agricultural University to X.D., and a grant from the Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN, Japan) to Y.K. We thank Dr Victor Manon for his critical reading of this article. XD, YK and GC designed this project; SL, TF, FP, QS and TM performed the gene cloning, protein expression, oocyte performance, immunoelectron microscopy, tobacco genetic transformation and characterization; PG and GC performed the stopped-flow work; MK and TK measured root hydraulic conductivity; SL, YZ, TF and QL performed the plant transgenic work and protein purification; SZ, WZ and AW performed the electrophysiology and biochemistry work; and XD, SL and GC wrote and revised the article. This article has earned an Open Data Badge for making publicly available the digitally shareable data necessary to reproduce the reported results. This article has earned an Open Materials Badge for making publicly available the components of the research methodology needed to reproduce the reported procedure and analysis. Figure S1. Alignment of OsPIP1;3 sequences from different rice varieties. OsPIP1;3in is from Oryza sativa L. indica cv. Zhonghan 3 and OsPIP1;3jp is from Oryza sativa L. japonica cv. Nipponbare. The isoleucine residues that are supposed to affect CO2 permeability of plant AQPs are marked in the red box. Figure S2. In situ localization of OsPIP1;3 in rice root. Rice root tips were fixed and subsequently embedded in 100 % LR-white. Ultra-thin sections were prepared and immunostained with an anti-OsPIP1;3 specific rabbit antibody (a–c) or rabbit sera (d) and gold colloid labeled anti-rabbit IgG antibody. The samples were stained with TI blue and observed by electron microscope. Figure S3. Expression, solubilization and purification of OsPIP1;3 in the E.coli system. (a) Western blotting analysis using anti-OsPIP1;3 polyclonal antibody to detect solubilization efficiency by different detergents. Lanes 0 and 1 show the crude membrane fraction before and after being re-suspended in solubilization buffer (25mm Na2HPO4, 10% (v/v) glycerol, 5mm β-mercaptoethanol, and 125mm NaCl, pH 8.0) without any detergent. Lanes2–9 show the profiles of the proteins solubilized in solutions with 5% OG (lane 2), 10% OG (lane 3), 5% DM (lane 4), 5% C12E8 (lane 5), 5% OPOE (lane 6), 3% LDAO (lane 7), 1.5% FC12 (lane 8) and 4% PFO (lane 9), respectively. (b) Protein purification analysis. The detergent-solubilized membrane proteins were separated by SDS-PAGE and stained with Coomassie brilliant blue (CBB) and then were subjected to Western blotting using anti-OsPIP1;3 polyclonal antibody. Loading profile: molecular weight marker (lane M), total PFO soluble membrane fraction (lane 1), the 1st passed fraction through the Ni-NTA column (lane 2), washing fraction (lane 3), and elution fraction (lane 4). Figure S4. Expression of EGFP or EGFP-OsPIP1;3 in HEK293 cells. (a) Images of EGFP or EGFP-OsPIP1;3 in HEK293 cells. The arrow indicates the localization of EGFP-OsPIP1;3 in the membrane. (b) WB to show expression of EGFP or EGFP-OsPIP1;3. Figure S5. Generation of OsPIP1;3 transgenic tobacco plants. (a) Schematic illustration of the T-DNA sequence of the pPBEL-FLAG-OsPIP1;3 construct. A FLAG tag sequence was inserted in the 5′ of the OsPIP1;3 gene to generate FLAG-OsPIP1;3 fusion protein for Western blotting. (b) Expression levels of OsPIP1;3 and endogenous NtAQP1 genes in tobacco plants. Total RNA was extracted from each T2 plant and subject to RT-qPCR analyses. Figure S6. Stomatal conductances (gs) of OsPIP1;3 overexpression and wild-type plants. 50-day-old transgenic and WT tobacco plants growing in normal conditions were selected for gas exchange measurement. The indexes were measured with a 6400-40 Leaf Chamber Fluorometer (LICOR Biosciences) under natural sunlight in the glasshouse. See detailed description in the ‘Experimental procedures’. Figure S7. Transpiration rate (Tr) of OsPIP1;3 overexpression and WT plants. See the detailed description in ‘Experimental procedures’. Figure S8. Intercellular CO2 concentration (Ci) of OsPIP1;3 overexpression and WT plants. See the detailed description in ‘Experimental procedures’. Figure S9. Total chlorophyll and nitrogen contents in tobacco leaves. (a) Total chlorophyll contents of wild-type and OX leaves. (b) Total nitrogen contents of WT and OX leaves. Figure S10. Promotion of cyanobacterial growth by expression of OsPIP1;3. (a) FLAG-OsPIP1;3 expression in cyanobacteria revealed by Western blotting. (b) Growth patterns of transgenic and wild-type cells on solid medium exposed to CO2-free or normal air. (c) The transgenic and wild-type cells were inoculated in liquid medium with equal density and were cultured in liquid medium in normal air. Their relative growth rates were measured and compared at the exponential growth phases. Figure S11. Relative water content in pot soils. The plants were grown in pot soil with equal water contents. After a 6-day drought treatment, the pot soil water contents were measured. The values were the percentages of water contents after versus before the 6-day drought treatment. Figure S12. Stomatal densities on WT and OsPIP1;3 transgenic leaves. The leaves from the top and bottom of representative WT and OsPIP1;3 transgenic plants were sampled. The stomatal densities were determined by means of nail polish imprints. Figure S13. Pressure chamber for measuring root hydraulic conductance. Diagram of pressure chamber for measurement of root hydraulic conductance. Figure S14. Effect of OsPIP1;3 overexpression on tobacco leaf area. One-month-old transgenic and WT tobacco plantlets were grown in pot soils. The pots were irrigated every 4days for a total of 32days. The first six true leaves were selected and the leaf areas were measured. Table S1. Primer sequences for real-time RT-qPCR used to quantify gene expression levels. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. Abascal, F., Irisarri, I. and Zardoya, R. (2014) Diversity and evolution of membrane intrinsic proteins. Biochim. Biophys. Acta, 1840, 1468– 1481. Aharon, R., Shahak, Y., Wininger, S., Bendov, R., Kapulnik, Y. and Galilia, G. (2003) Overexpression of a plasma membrane aquaporin in transgenic tobacco improves plant vigor under favorable growth conditions but not under drought or salt stress. Plant Cell, 15, 439– 447. Alexandersson, E., Fraysse, L., Sjövall-Larsen, S., Gustavsson, S., Fellert, M., Karlsson, M., Johanson, U. and Kjellbom, P. (2005) Whole gene family expression and drought stress regulation of aquaporins. Plant Mol. Biol. 59, 469– 484. Almeida-Rodriguez, A.M., Cooke, J.E.K., Yeh, F. and Zwiazek, J.J. (2010) Functional characterization of drought-responsive aquaporins in Populus balsamifera and Populus simonii×balsamifera clones with different drought resistance strategies. Physiol. Plant. 140, 321– 333. Wiley Online Library Baba, A., Hasezawa, S. and Syono, K. (1986) Cultivation of rice protoplasts and their transformation mediated by Agrobacterium spheroplasts. Plant Cell Physiol. 27, 463– 471. Bienert, M.D., Diehn, T.A., Richet, N., Chaumont, F. and Bienert, G.P. (2018) Heterotetramerization of plant PIP1 and PIP2 aquaporins is an evolutionary ancient feature to guide PIP1 plasma membrane localization and function. Front. Plant Sci. 9, 382. Borgnia, M.J., Kozono, D., Calamita, G., Maloney, P.C. and Agre, P. (1999) Functional reconstitution and characterization of AqpZ, the E. coli water channel protein. J. Mol. Biol. 291, 1169– 1179. Calamita, G., Bishai, W.R., Preston, G.M., Guggino, W.B. and Agre, P. (1995) Molecular cloning and characterization of AqpZ, a water channel from Escherichia coli. J. Biol. Chem. 270, 29063– 29066. Calamita, G., Gena, P., Meleleo, D., Ferri, D. and Svelto, M. (2006) Water permeability of rat liver mitochondria: a biophysical study. Biochim. Biophys. Acta-Biomembr. 1758, 1018– 1024. Chaumont, F., Barrieu, F., Jung, R. and Chrispeels, M.J. (2000) Plasma membrane intrinsic proteins from maize cluster in two sequence subgroups with differential aquaporin activity. Plant Physiol. 122, 1025– 1034. Chevalier, A.S. and Chaumont, F. (2015) Trafficking of plant plasma membrane aquaporins: multiple regulation levels and complex sorting signals. Plant Cell Physiol. 56, 819– 829. Clarkson, D.T., Carvajal, M., Henzler, T., Waterhouse, R.N., Smyth, A.J., Cooke, D.T. and Steudle, E. (2000) Root hydraulic conductance: diurnal aquaporin expression and the effects of nutrient stress. J. Exp. Bot. 51, 61– 70. Crozet, P., Margalha, L., Confraria, A., Rodrigues, A., Martinho, C., Adamo, M., Elias, C.A. and Baena-González, E. (2014) Mechanisms of regulation of SNF1/AMPK/SnRK1 protein kinases. Front. Plant Sci. 5, 190. Dean, R.M., Rivers, R.L., Zeidel, M.L. and Roberts, D.M. (1999) Purification and functional reconstitution of soybean nodulin 26. An aquaporin with water and glycerol transport properties. Biochemistry, 38, 347– 353. Deshmukh, R.K., Sonah, H. and Bélanger, R.R. (2016) Plant aquaporins: genome-wide identification, transcriptomics, proteomics, and advanced analytical tools. Front. Plant Sci. 7, 1896. Ding, X., Iwasaki, I. and Kitagawa, Y. (2004) Overexpression of a lily PIP1 gene in tobacco increased the osmotic water permeability of leaf cells. Plant Cell Environ. 27, 177– 186. Wiley Online Library Ding, X., Matsumoto, T., Gena, P. et al. (2013) Water and CO2 permeability of SsAqpZ, the cyanobacterium Synechococcus sp. PCC7942 aquaporin. Biol. Cell, 105, 118– 128. Wiley Online Library Echevarría, M., Windhager, E.E. and Frindt, G. (1996) Selectivity of the renal collecting duct water channel aquaporin-3. J. Biol. Chem. 271, 25079– 25082. Fetter, K., van Wilder, V., Moshelion, M. and Chaumont, F. (2004) Interactions between plasma membrane aquaporins modulate their water channel activity. Plant Cell, 16, 215– 228. Flexas, J., Ribas-Carbó, M., Hanson, D.T., Bota, J., Otto, B., Cifre, J., McDowell, N., Medrano, H. and Kaldenhoff, R. (2006) Tobacco aquaporin NtAQP1 is involved in mesophyll conductance to CO2 in vivo. Plant J. 48, 427– 439. Wiley Online Library Flexas, J., Ortuño, M.F., Ribas-Carbo, M., Diaz-Espejo, A., Flórez-Sarasa, I.D. and Medrano, H. (2007) Mesophyll conductance to CO2 in Arabidopsis thaliana. New Phytol. 175, 501– 511. Wiley Online Library Flexas, J., Niinemets, U., Gallé, A. et al. (2013a) Diffusional conductances to CO2 as a target for increasing photosynthesis and photosynthetic water-use efficiency. Photosynth Res. 117, 45– 59. Flexas, J., Scoffoni, C., Gago, J. and Sack, L. (2013b) Leaf mesophyll conductance and leaf hydraulic conductance: an introduction to their measurement and coordination. J. Exp. Bot. 64, 3965– 3981. Fox, A.R., Maistriaux, L. and Chaumont, F. (2017) Towards understanding of the high number of plant aquaporin isoforms and multiple regulation mechanisms. Plant Sci. 264, 179– 187. Gaspar, M., Bousser, A., Sissoeff, I., Roche, O., Hoarau, J. and Mahe, A. (2003) Cloning and characterization of ZmPIP1-5b, an aquaporin transporting water and urea. Plant Sci. 165, 21– 31. Gena, P., Pellegrini-Calace, M., Biasco, A., Svelto, M. and Calamita, G. (2011) Aquaporin membrane channels: biophysics, classification, functions, and possible biotechnological applications. Food Biophys. 6, 241– 249. Gloser, V., Zwieniecki, M.A., Orians, C.M. and Holbrook, N.M. (2007) Dynamic changes in root hydraulic properties in response to nitrate availability. J. Exp. Bot. 58, 2409– 2415. Grondin, A., Mauleon, R., Vadez, V. and Henry, A. (2016) Root aquaporins contribute to whole plant water fluxes under drought stress in rice (Oryza sativa L.). Plant Cell Environ. 39, 347– 365. Wiley Online Library Groszmann, M., Osborn, H.L. and Evans, J.R. (2017) Carbon dioxide and water transport through plant aquaporins. Plant Cell Environ. 40, 938– 961. Wiley Online Library Hamill, O.P., Marty, A., Neher, E., Sakmann, B. and Sigworth, F.J. (1981) Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Arch. 391, 85– 100. Harley, P.C., Loreto, F., Marco, G.D. and Sharkey, T.D. (1992) Theoretical considerations when estimating the mesophyll conductance to CO2 flux by analysis of the response of photosynthesis to CO2. Plant Physiol. 98, 1429– 1436. Hedrich, R. and Geiger, D. (2017) Biology of SLAC1-type anion channels-from nutrient uptake to stomatal closure. New Phytol. 216, 46– 61. Wiley Online Library van Heeswijk, M.P.E. and van Os, C.H. (1986) Osmotic water permeabilities of brush border and basolateral membrane vesicles from rat renal cortex and small intestine. J. Membr. Biol. 92, 183– 193. Hwang, J.H., Ellingson, S.R. and Roberts, D.M. (2010) Ammonia permeability of the soybean nodulin 26 channel. FEBS Lett. 584, 4339– 4343. Wiley Online Library Ikeda, M., Beitz, E., Kozono, D., Guggino, W.B., Agre, P. and Yasui, M. (2002) Characterization of aquaporin-6 as a nitrate channel in mammalian cells. Requirement of pore-lining residue threonine 63. J. Biol. Chem. 277, 39873– 39879. Ishibashi, K., Morishita, Y. and Tanaka, Y. (2017) The evolutionary aspects of aquaporin family. Adv. Exp. Med. Biol. 969, 35– 50. Jang, J.Y., Kim, D.G., Kim, Y.O., Kim, J.S. and Kang, H. (2004) An expression analysis of a gene family encoding plasma membrane aquaporins in response to abiotic stresses in Arabidopsis thaliana. Plant Mol. Biol. 54, 713– 725. Jozefkowicz, C., Rosi, P., Sigaut, L., Soto, G., Pietrasanta, L., Amodeo, G. and Alleva, K. (2013) Loop A is critical for the functional interaction of two Beta vulgaris PIP aquaporins. PLoS ONE, 8, e57993. Kadam, S., Abril, A., Dhanapal, A.P., Koester, R.P., Vermerris, W., Jose, S. and Fritschi, F.B. (2017) Characterization and regulation of aquaporin genes of sorghum (Sorghum bicolor (L.) Moench) in response to waterlogging stress. Front. Plant Sci. 8, 862. Kaldenhoff, R. and Fischer, M. (2006) Functional aquaporin diversity in plants. Biochem. Biophy. Acta, 1758, 1134– 1141. Kaldenhoff, R., Grote, K., Zhu, J.J. and Zimmermann, U. (1998) Significance of plasmalemma aquaporins for water-transport in Arabidopsis thaliana. Plant J. 14, 121– 128. Wiley Online Library Katsuhara, M., Akiyama, Y., Koshio, K., Shibasaka, M. and Kasamo, K. (2002) Functional analysis of water channels in barley roots. Plant Cell Physiol. 43, 885– 893. Kozono, D., Ding, X., Iwasaki, I., Meng, X., Kamagata, Y., Agre, P. and Kitagawa, Y. (2003) Functional expression and characterization of an archaeal aquaporin. AqpM from Methanothermobacter marburgensis. J. Biol. Chem. 278, 10649– 10656. Li, G., Tillard, P., Gojon, A. and Maurel, C. (2016) Dual regulation of root hydraulic conductivity and plasma membrane aquaporins by plant nitrate accumulation and high-affinity nitrate transporter NRT2.1. Plant Cell Physiol. 57, 733– 742. Lian, H.L., Yu, X., Ye, Q., Ding, X., Kitagawa, Y., Kwak, S.S., Su, W.A. and Tang, Z.C. (2004) The role of aquaporin RWC3 in drought avoidance in rice. Plant Cell Physiol. 45, 481– 489. Lian, H.L., Yu, X., Lane, D., Sun, W.N., Tang, Z.C. and Su, W.A. (2006) Upland rice and lowland rice exhibited different PIP expression under water deficit and ABA treatment. Cell Res. 16, 651– 660. Liu, H.Y., Yu, X., Cui, D.Y., Sun, M.H., Sun, W.N., Tang, Z.C., Kwak, S.S. and Su, W.A. (2007) The role of water channel proteins and nitric oxide signaling in rice seed germination. Cell Res. 17, 638– 649. Liu, C., Fukumoto, T., Matsumoto, T. et al. (2013) Aquaporin OsPIP1;1 promotes rice salt resistance and seed germination. Plant Physiol. Biochem. 63, 151– 158. Matsumoto, T., Lian, H.L., Su, W.A., Tanaka, D., Liu, C., Iwasaki, I. and Kitagawa, Y. (2009) Role of the aquaporin PIP1 subfamily in the chilling tolerance of rice. Plant Cell Physiol. 50, 216– 229. Matsuo, N., Ozawa, K. and Mochizuki, T. (2009) Genotypic differences in root hydraulic conductance of rice (Oryza sativa L.) in response to water regimes. Plant Soil, 316, 25– 34. Maurel, C., Reizer, J., Schroeder, J.I. and Chrispeels, M.J. (1993) The vacuolar membrane protein gamma-TIP creates water specific channels in Xenopus oocytes. EMBO J. 12, 2241– 2247. Wiley Online Library Maurel, C., Verdoucq, L., Luu, D.-T. and Santoni, V. (2008) Plant aquaporins: membrane channels with multiple integrated functions. Annu. Rev. Plant Biol. 59, 595– 624. Maurel, C., Boursiac, Y., Luu, D.-T., Santoni, V., Shahzad, Z. and Verdoucq, L. (2015) Aquaporins in plants. Physiol. Rev. 95, 1321– 1358. Maurel, C., Verdoucq, L. and Rodrigues, O. (2016) Aquaporins and plant transpiration. Plant Cell Environ. 39, 2580– 2587. Wiley Online Library Missner, A., Kügler, P., Saparov, S.M., Sommer, K., Mathai, J.C., Zeidel, M.L. and Pohl, P. (2008) Carbon dioxide transport through membranes. J. Biol. Chem. 283, 25340– 25347. Mori, I.C., Rhee, J., Shibasaka, M., Sasano, S., Kaneko, T., Horie, T. and Katsuhara, M. (2014) CO2 transport by PIP2 aquaporins of barley. Plant Cell Physiol. 55, 251– 257. Murai-Hatano, M., Kuwagata, T., Sakurai, J., Nonami, H., Ahamed, A., Nagasuga, K., Matsunami, T., Fukushi, K., Maeshima, M. and Okada, M. (2008) Effect of low root temperature on hydraulic conductivity of rice plants and the possible role of aquaporins. Plant Cell Physiol. 49, 1294– 1305. Nakhoul, N.L., Davis, B.A., Romero, M.F. and Boron, W.F. (1998) Effect of expressing the water channel aquaporin-1 on the CO2 permeability of Xenopus oocytes. Am. J. Physiol. 43, C543– C548. Nguyen, M.X., Moon, S. and Jung, K.-H. (2013) Genome-wide expression analysis of rice aquaporin genes and development of a functional gene network mediated by aquaporin expression in roots. Planta, 238, 669– 681. Preston, G.M., Carroll, T.P., Guggino, W.B. and Agre, P. (1992) Appearance of water channels in Xenopus oocytes expressing red cell CHIP28 protein. Science, 256, 385– 387. Qin, X. and Boron, W.F. (2013) Mutation of a single amino acid converts the human water channel aquaporin 5 into an anion channel. Am. J. Physiol. 305, C663– C672. Rodrigues, O., Reshetnyak, G., Grondin, A., Saijo, Y., Leonhardt, N., Maurel, C. and Verdoucq, L. (2017) Aquaporins facilitate hydrogen peroxide entry into guard cells to mediate ABA- and pathogen-triggered stomatal closure. Proc. Natl Acad. Sci. USA, 114, 9200– 9205. Schäfer, N., Maierhofer, T., Herrmann, J. et al. (2018) A tandem amino acid residue motif in guard cell SLAC1 anion channel of grasses allows for the control of stomatal aperture by nitrate. Curr. Biol. 28, 1370– 1379. She, J., Guo, J., Chen, Q., Zeng, W., Jiang, Y. and Bai, X. (2018) Structural insights into the voltage and phospholipid activation of mammalian TPC1 channel. Nature, 556, 130– 134. Shen, J., Fu, J., Ma, J., Wang, X., Gao, C., Zhuang, C., Wan, J. and Jiang, L. (2014) Isolation, culture, and transient transformation of plant protoplasts. Curr. Protoc. Cell Biol. 63, 2.8.1– 17. Wiley Online Library Song, S., Xu, Y., Huang, D., Miao, H., Liu, J., Jia, C., Hu, W., Valarezo, A.V., Xu, B. and Jin, Z. (2018) Identification of a novel promoter from banana aquaporin family gene (MaTIP1;2) which responses to drought and salt-stress in transgenic Arabidopsis thaliana. Plant Physiol. Biochem. 128, 163– 169. Song, Y., Zhang, H., You, H. et al. (2019) Identification of novel interactors and potential phosphorylation substrates of GsSnRK1 from wild soybean (Glycine soja). Plant Cell Environ. 42, 145– 157. Wiley Online Library Takano, J., Wada, M., Ludewig, U., Schaaf, G., Wirén, N. and Fujiwara, T. (2006) The Arabidopsis major intrinsic protein NIP5;1 is essential for efficient boron uptake and plant development under boron limitation. Plant Cell, 18, 1498– 1509. Uehlein, N., Lovisolo, C., Siefritz, F. and Kaldenhoff, R. (2003) The tobacco aquaporin NtAQP1 is a membrane CO2 pore with physiological functions. Nature, 425, 734– 737. Vieira, P.M., Santos, M.P., Andrade, C.M., Souza-Neto, O.A., Ulhoa, C.J. and Aragão, F.J.L. (2017) Overexpression of an aquaglyceroporin gene from Trichoderma harzianum improves water-use efficiency and drought tolerance in Nicotiana tabacum. Plant Physiol. Biochem. 121, 38– 47. Wang, Y.H., Garvin, D.F. and Kochian, L.V. (2001) Nitrate-induced genes in tomato roots. Array analysis reveals novel genes that may play a role in nitrogen nutrition. Plant Physiol. 127, 345– 59. Wang, X., Cai, H., Li, Y., Zhu, Y., Ji, W., Bai, X., Zhu, D. and Sun, X. (2015) Ectopic overexpression of a novel Glycine soja stress-induced plasma membrane intrinsic protein increases sensitivity to salt and dehydration in transgenic Arabidopsis thaliana plants. J. Plant Res. 128, 103– 113. Wang, C., Hu, H., Qin, X., Zeise, B., Xu, D., Rappel, W.J., Boron, W.F. and Schroeder, J.I. (2016) Reconstitution of CO2 regulation of SLAC1 anion channel and function of CO2-permeable PIP2;1 aquaporin as CARBONIC ANHYDRASE4 interactor. Plant Cell, 28, 568– 582. Wang, L., Li, Q.-T., Lei, Q., Feng, C., Zheng, X., Zhou, F., Li, L., Liu, X. and Kong, J. (2017a) Ectopically expressing MdPIP1;3, an aquaporin gene, increased fruit size and enhanced drought tolerance of transgenic tomatoes. BMC Plant Biol. 17, 246. Wang, L., Wang, S., Chen, W., Li, H. and Deng, X. (2017b) Physiological mechanisms contributing to increased water-use efficiency in winter wheat under organic fertilization. PLoS ONE, 12, e0180205. Wu, F., Sheng, P., Tan, J. et al. (2015) Plasma membrane receptor-like kinase leaf panicle 2 acts downstream of the DROUGHT AND SALT TOLERANCE transcription factor to regulate drought sensitivity in rice. J. Exp. Bot. 66, 271– 281. Xu, F., Wang, K., Yuan, W. et al. (2019) Overexpression of rice aquaporin OsPIP1;2 improves yield by enhancing mesophyll CO2 conductance and phloem sucrose transport. J. Exp. Bot. 70, 671– 681. Zhuo, C., Wang, T., Guo, Z. and Lu, S. (2016) Overexpression of MfPIP2-7 from Medicago falcata promotes cold tolerance and growth under NO3- deficit in transgenic tobacco plants. BMC Plant Biol. 16, 138. Zwiazek, J.J., Xu, H., Tan, X., Navarro-Ródenas, A. and Morte, A. (2017) Significance of oxygen transport through aquaporins. Sci. Rep. 7, 40411. The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username