Ectopic Expression of CrPIP2;3, a Plasma Membrane Intrinsic Protein Gene from the Halophyte Canavalia rosea, Enhances Drought and Salt-Alkali Stress Tolerance in Arabidopsis

Aquaporins are channel proteins that facilitate the transmembrane transport of water and other small neutral molecules, thereby playing vital roles in maintaining water and nutrition homeostasis in the life activities of all organisms. Canavalia rosea, a seashore and mangrove-accompanied halophyte with strong adaptability to adversity in tropical and subtropical regions, is a good model for studying the molecular mechanisms underlying extreme saline-alkaline and drought stress tolerance in leguminous plants. In this study, a PIP2 gene (CrPIP2;3) was cloned from C. rosea, and its expression patterns and physiological roles in yeast and Arabidopsis thaliana heterologous expression systems under high salt-alkali and high osmotic stress conditions were examined. The expression of CrPIP2;3 at the transcriptional level in C. rosea was affected by high salinity and alkali, high osmotic stress, and abscisic acid treatment. In yeast, the expression of CrPIP2;3 enhanced salt/osmotic and oxidative sensitivity under high salt/osmotic and H2O2 stress. The overexpression of CrPIP2;3 in A. thaliana could enhance the survival and recovery of transgenic plants under drought stress, and the seed germination and seedling growth of the CrPIP2;3 OX (over-expression) lines showed slightly stronger tolerance to high salt/alkali than the wild-type. The transgenic plants also showed a higher response level to high-salinity and dehydration than the wild-type, mostly based on the up-regulated expression of salt/dehydration marker genes in A. thaliana plants. The reactive oxygen species (ROS) staining results indicated that the transgenic lines did not possess stronger ROS scavenging ability and stress tolerance than the wild-type under multiple stresses. The results confirmed that CrPIP2;3 is involved in the response of C. rosea to salt and drought, and primarily acts by mediating water homeostasis rather than by acting as an ROS transporter, thereby influencing physiological processes under various abiotic stresses in plants.


Introduction
Canavalia rosea (Sw.) DC., also known as bay bean, is mainly distributed in coastal and island sandy soils in tropical and subtropical areas. It is a perennial halophyte belonging drought stress, when heterologously overexpressed in Arabidopsis thaliana [33]. Another A. canescens nodulin-like plasma membrane intrinsic protein gene, AcNIP5;1, showed a totally opposite phenotype when expressed in A. thaliana [34]. Sesuvium portulacastrum is a perennial halophyte that typically grows in coastal and inland sandy soils and exhibits excellent tolerance to salt. SpAQP1 (a PIP2 member) enhanced the salt tolerance of yeast strains and tobacco plants and promoted seed germination and root growth under salt stress in transgenic plants [35]. Selaginella moellendorffii is a desiccation-tolerant plant, and a total of 19 AQPs were detected in the S. moellendorffii-expressed sequence tag (EST) database, while only three conserved PIPs were detected [36]. SmPIP1;1 and SmPIP2;1 exhibited different water-channel activities, and their co-expression showed a synergistic effect on the water membrane in an oocyte system, while no synergistic effect was observed in yeast [37]. Stipa purpurea showed great adaptability to stresses such as drought and a changing environment in the meadows of the Tibetan Plateau [38]. The transcription of SpPIP1 increased significantly under drought, and the ectopic expression of SpPIP1 in A. thaliana elevated the drought tolerance of the plants [38]. Jojoba (Simmondsia chinensis) is a typical desert plant with strong tolerance to drought, salinity, and nutrient-poor soils, and ScPIP1 could improve salt and drought resistance when expressed in A. thaliana [39]. Thellungiella salsuginea displays strong resistance to high salinity, drought, and chilling stress. The overexpression of TsPIP1;1 in rice could also enhance the salt tolerance of rice by maintaining the osmotic potential and promoting photosynthesis [40]. Both NnPIP1-2 and NnPIP2-1, from the hydrophyte salt-tolerant lotus Nelumbo nucifera, could enhance the abiotic stress tolerance of transgenic plants when overexpressed in A. thaliana [41]. These findings indicate that PIPs from plants growing in specialized habitats can be excavated as potential target genes for enhancing plant resistance to abiotic stresses though genetic engineering.
To further understand the mechanism of salinity tolerance in C. rosea, an entire cDNA library was constructed from C. rosea seedlings for yeast functional screening. A series of ESTs were previously characterized, including CrPIP2;3. In this study, this AQP gene CrPIP2;3 was further functionally characterized based on an over-expression assay in yeast and A. thaliana. Our results suggested that CrPIP2;3 was significantly induced by high osmotic stress and could enhance the salt-alkali and drought tolerance of transgenic plants by facilitating water transport and increasing the drought tolerance of the plants.

Identification and Bioinformatics Analysis of CrPIP2;3
The full-length cDNA of CrPIP2;3 was isolated from the cDNA library constructed using C. rosea seedlings. The length of the CrPIP2;3 cDNA is 1254 bp, with a 51-bp 5 untranslated region (UTR) and a 333-bp 3 UTR (GenBank Accession No.: MT787666). The open reading frame (ORF) is 870 bp, encoding a protein with a predicted molecular weight of 31.07 kDa and a theoretical isoelectric point of 7.66. CrPIP2;3 was characterized as an aquaporin with an MIP (major intrinsic protein) domain (33 aa to 269 aa, PF00230) ( Figure 1A). The instability index (II) of CrPIP2;3 is 36.11, and the grand average of hydropathicity (GRAVY) is 0.383, while the aliphatic index (AI) showed a relatively high-level value of 99.31, which indicated that CrPIP2;3 encoded a stable hydrophilic protein. The three-dimensional structure predication by Phyre2 showed that CrPIP2;3 has six transmembrane α-helices (TM1 to TM6) ( Figure 1B). Sequence analyses also indicated that CrPIP2;3 contained two conserved Asn-Pro-Ala (NPA) motifs (LB and LE loops) and a putative MIP signal sequence (SGxHxNPAVT) in loop B, with several relatively conserved amino acid residues, including aromatic/arginine (ar/R) selectivity filters (H2, H5, LE1, and LE2) and Froger's positions (FPs) (P1, P2, P3, P4, and P5) ( Figure 1A), which are associated with the substrate selectivity of AQPs [42,43]. Phylogenetic analysis of CrPIP2;3 with other plant species PIPs (including AtPIPs, CaPIPs, and GmPIPs) [44][45][46] showed that CrPIP2;3 is closely related to CaPIP2;2 or GmPIP2;7 and GmPIP2;8, which could be classified into a PIP2 subgroup ( Figure 1C). We also predicted the localization patterns of CrPIP2;3 with different online programs, and the prediction results from WoLF PSORT showed that CrPIP2;3 localization at the plasma membrane and Golgi apparatus scored 13 and 1, respectively, while Plant-Ploc predicted CrPIP2;3 to be mainly distributed in the cell membrane. In general, CrPIP2;3 was predicted as a plasma-membrane-localized protein.
ters (H2, H5, LE1, and LE2) and Froger's positions (FPs) (P1, P2, P3, P4, and P5) ( Figure  1A), which are associated with the substrate selectivity of AQPs [42,43]. Phylogenetic analysis of CrPIP2;3 with other plant species PIPs (including AtPIPs, CaPIPs, and GmPIPs) [44][45][46] showed that CrPIP2;3 is closely related to CaPIP2;2 or GmPIP2;7 and GmPIP2;8, which could be classified into a PIP2 subgroup ( Figure 1C). We also predicted the localization patterns of CrPIP2;3 with different online programs, and the prediction results from WoLF PSORT showed that CrPIP2;3 localization at the plasma membrane and Golgi apparatus scored 13 and 1, respectively, while Plant-Ploc predicted CrPIP2;3 to be mainly distributed in the cell membrane. In general, CrPIP2;3 was predicted as a plasma-membrane-localized protein.  The conserved amino acid residues in all proteins are highlighted in black and gray. The MIP domain (PF00230) is marked with a green frame. Six transmembrane helices domains (TM1-TM6) and two conserved Asn-Pro-Ala ("NPA") motifs (LB and LE) are marked with black and red solid lines. The aromatic/arginine (ar/R) selectivity filters (H2, H5, LE1, and LE2) and Froger's positions (FPs) (P1, P2, P3, P4, and P5) are marked with blue, brown, and green solid lines, respectively. The MIP signal sequence between TM2 and TM3 is also lined in a gray color. (B) Predicted 3D structure of CrPIP2;3 generated using the Phyre2 server. (C) The phylogenic tree of CrPIP2;3 and other PIP proteins from A. thaliana and soybean. The black triangle symbol showed the position of CrPIP2;3. (A) Sequence alignment analysis of CrPIP2;3 with other known PIP proteins. The conserved amino acid residues in all proteins are highlighted in black and gray. The MIP domain (PF00230) is marked with a green frame. Six transmembrane helices domains (TM1-TM6) and two conserved Asn-Pro-Ala ("NPA") motifs (LB and LE) are marked with black and red solid lines. The aromatic/arginine (ar/R) selectivity filters (H2, H5, LE1, and LE2) and Froger's positions (FPs) (P1, P2, P3, P4, and P5) are marked with blue, brown, and green solid lines, respectively. The MIP signal sequence between TM2 and TM3 is also lined in a gray color. (B) Predicted 3D structure of CrPIP2;3 generated using the Phyre2 server. (C) The phylogenic tree of CrPIP2;3 and other PIP proteins from A. thaliana and soybean. The black triangle symbol showed the position of CrPIP2;3.
Analysis of the upstream 2000-bp sequence from the transcription start site (TSS) of CrPIP2;3 (promoter region) using PlantCARE indicated that the putative cis-acting regulatory elements were classified into three types: the transcription factor-binding site involved in drought inducibility, the hormone-responsive element, and the lightresponsive element (Table S1). The sequence analysis of the CrPIP2;3 promoter and the putative cis-acting element location and function was showed in Figure S1. Generally, the basic promoter elements, TATA-Box and CAAT-Box, were located at −11 and −132 bp, respectively ( Figure 2). The CrPIP2;3 promoter contained three drought-related elements, namely, MYB, MYC, and CCAAT-box, and it harbored CGTCA or TGACG (response to MeJA), the WUN motif (wounding inducible), and several light-responsive elements (AT1-motif, Box-4, GATA motif, I-box, TCCC-motif, and TCT-motif) ( Figure 2). This result indicated that CrPIP2;3 was related to a drought-resistant pathway, with other possible roles in light or biotic stress responses. Analysis of the upstream 2000-bp sequence from the transcription start site (TSS) of CrPIP2;3 (promoter region) using PlantCARE indicated that the putative cis-acting regulatory elements were classified into three types: the transcription factor-binding site involved in drought inducibility, the hormone-responsive element, and the light-responsive element (Table S1). The sequence analysis of the CrPIP2;3 promoter and the putative cis-acting element location and function was showed in Figure S1. Generally, the basic promoter elements, TATA-Box and CAAT-Box, were located at −11 and −132 bp, respectively ( Figure 2). The CrPIP2;3 promoter contained three drought-related elements, namely, MYB, MYC, and CCAAT-box, and it harbored CGTCA or TGACG (response to MeJA), the WUN motif (wounding inducible), and several light-responsive elements (AT1-motif, Box-4, GATA motif, I-box, TCCC-motif, and TCT-motif) ( Figure 2). This result indicated that CrPIP2;3 was related to a drought-resistant pathway, with other possible roles in light or biotic stress responses.

Expression Patterns of CrPIP2;3 in C. rosea
To investigate and predict more detailed results regarding the stress responses of CrPIP2;3, qRT-PCR was performed to explore the expression patterns under stress or ABA treatments and normal growth conditions using the total RNA extracted from the various tissues. We checked the expression of CrPIP2;3 in the whole C. rosea plant. The result indicated that CrPIP2;3 was expressed extensively in all tested tissues/organs of C. rosea and showed a relatively high transcriptional level with respect to the reference gene CrEF-α ( Figure S2). In Figure 3, our results revealed that under high-salt (600 mM NaCl) treatment, the expression of CrPIP2;3 decreased immediately (2 and 12 h) in all three tested tissues when compared with CK samples (the beginnings of each treatment) (Figure 3A). The expression level of CrPIP2;3 in C. rosea roots gradually recovered with the increase in high-salt stress duration (24 and 48 h), while in the vine and leaf, the expression level of CrPIP2;3 remained obviously decreased until 48 h. Under alkali toxicity (150 mM NaHCO3), the expression level of CrPIP2;3 exhibited correspondingly low levels, with no major changes observed ( Figure 3B). The high osmotic stress, achieved by immersing the roots in 300 mM mannitol, significantly induced the expression of CrPIP2;3 in the root and vine, while the expression of CrPIP2;3 in the leaf was inhibited ( Figure  3C). The ABA treatment also affected CrPIP2;3 expression ( Figure 3D). In brief, in contrast to the other stresses, spraying ABA first repressed the expression of CrPIP2;3 immediately in the entire C. rosea seedlings, following which the expression level recovered slowly both in the root and in the leaf, while in the vine, the expression of CrPIP2;3 was obviously elevated following ABA treatment at 48 h. We also performed qRT-PCR to assess the expression differences of CrPIP2;3 in different C. rosea organs. Our results showed that the expression level of CrPIP2;3 in the various parts of C. rosea was obviously different ( Figure S2). CrPIP2;3 expression showed differentially regulated patterns in the different tissues of C. rosea, which indicated a complex response mechanism and implicated CrPIP2;3 in the response of C. rosea to these abiotic stresses.

Expression Patterns of CrPIP2;3 in C. rosea
To investigate and predict more detailed results regarding the stress responses of CrPIP2;3, qRT-PCR was performed to explore the expression patterns under stress or ABA treatments and normal growth conditions using the total RNA extracted from the various tissues. We checked the expression of CrPIP2;3 in the whole C. rosea plant. The result indicated that CrPIP2;3 was expressed extensively in all tested tissues/organs of C. rosea and showed a relatively high transcriptional level with respect to the reference gene CrEF-α ( Figure S2). In Figure 3, our results revealed that under high-salt (600 mM NaCl) treatment, the expression of CrPIP2;3 decreased immediately (2 and 12 h) in all three tested tissues when compared with CK samples (the beginnings of each treatment) ( Figure 3A). The expression level of CrPIP2;3 in C. rosea roots gradually recovered with the increase in high-salt stress duration (24 and 48 h), while in the vine and leaf, the expression level of CrPIP2;3 remained obviously decreased until 48 h. Under alkali toxicity (150 mM NaHCO 3 ), the expression level of CrPIP2;3 exhibited correspondingly low levels, with no major changes observed ( Figure 3B). The high osmotic stress, achieved by immersing the roots in 300 mM mannitol, significantly induced the expression of CrPIP2;3 in the root and vine, while the expression of CrPIP2;3 in the leaf was inhibited ( Figure 3C). The ABA treatment also affected CrPIP2;3 expression ( Figure 3D). In brief, in contrast to the other stresses, spraying ABA first repressed the expression of CrPIP2;3 immediately in the entire C. rosea seedlings, following which the expression level recovered slowly both in the root and in the leaf, while in the vine, the expression of CrPIP2;3 was obviously elevated following ABA treatment at 48 h. We also performed qRT-PCR to assess the expression differences of CrPIP2;3 in different C. rosea organs. Our results showed that the expression level of CrPIP2;3 in the various parts of C. rosea was obviously different ( Figure S2). CrPIP2;3 expression showed differentially regulated patterns in the different tissues of C. rosea, which indicated a complex response mechanism and implicated CrPIP2;3 in the response of C. rosea to these abiotic stresses.

Heterologous Expression of CrPIP2;3 Confers Osmotic Stress and H2O2 Sensitivity in Yeast
The functional identification of CrPIP2;3 was first performed with a yeast expression system. The cell culture was adjusted to an OD600 value of 1 and then gradient-diluted (to 1:10, 1:100, and 1:1000). Two microliters of yeast solutions were spotted onto SDG agar plates with or without NaCl, sorbitol, or H2O2. As the results showed, W303 transformed with either CrPIP2;3 or pYES2 grew normally and did not show growth differences on the SDG control plate. However, with increased NaCl, W303 transformed with CrPIP2;3 showed an obvious growth lag compared with the yeast containing the pYES2 control ( Figure 4A). Similarly, the yeast strain with the pYES2 control also displayed better growth performance on the hyperosmotic plate (plus PEG8000 or sorbitol), which indicated that CrPIP2;3 could accelerate water loss when the yeast cells survived under water deprivation ( Figure 4B,C). We also assessed the H2O2 transport activity with the yeast expression system. CrPIP2;3 increased the H2O2 sensitivity of the yeast when grown on SDG medium containing different concentrations of H2O2, while both the BY4741 and H2O2 sensitive mutant strain skn7Δ yeast cells showed similar growth performance on the SDG control plate ( Figure 4E,F).

Heterologous Expression of CrPIP2;3 Confers Osmotic Stress and H 2 O 2 Sensitivity in Yeast
The functional identification of CrPIP2;3 was first performed with a yeast expression system. The cell culture was adjusted to an OD600 value of 1 and then gradient-diluted (to 1:10, 1:100, and 1:1000). Two microliters of yeast solutions were spotted onto SDG agar plates with or without NaCl, sorbitol, or H 2 O 2 . As the results showed, W303 transformed with either CrPIP2;3 or pYES2 grew normally and did not show growth differences on the SDG control plate. However, with increased NaCl, W303 transformed with CrPIP2;3 showed an obvious growth lag compared with the yeast containing the pYES2 control ( Figure 4A). Similarly, the yeast strain with the pYES2 control also displayed better growth performance on the hyperosmotic plate (plus PEG8000 or sorbitol), which indicated that CrPIP2;3 could accelerate water loss when the yeast cells survived under water deprivation ( Figure 4B,C). We also assessed the H 2 O 2 transport activity with the yeast expression system. CrPIP2;3 increased the H 2 O 2 sensitivity of the yeast when grown on SDG medium containing different concentrations of H 2 O 2 , while both the BY4741 and H 2 O 2 sensitive mutant strain skn7∆ yeast cells showed similar growth performance on the SDG control plate ( Figure 4E,F).

CrPIP2 Exhibits Apparent Tolerance to Drought Stress in Transgenic A. thaliana Plants
To further evaluate the effects of CrPIP2;3, A. thaliana transgenic plants ectopically expressing CrPIP2;3 under the control of the 35S promoter were generated. After confirmation with genomic PCR ( Figure 5A) and qRT-PCR ( Figure 5B), three homozygous T3 line plants (OX 3#, OX 6#, and OX 10#) were selected and applied in the following tests. Briefly, about 30 sterilized seeds were spotted on the MS medium with the addition of salt, saline-alkaline, or mannitol stress, following which the germination rate was calculated. As indicated in Figure 5C, a difference in the germination of the WT and CrPIP2;3 OX lines could be observed, and the germination of the CrPIP2;3 OX lines was higher than that of the WT when the seeds were grown in MS medium containing 200 mM NaCl or 5, 7.5, and 10 mM NaHCO3, and the corresponding statistical analysis of the seed germination rates also showed consistent results with the germination status, with better salt and alkali tolerance ( Figure 5D,E). There was little difference in the seed germination rates between the WT and transgenic plants under osmotic stress (200, 300, and 400 mM mannitol) ( Figure 5C). Although the CrPIP2;3 OX lines produced a larger radicle and hypocotyl than WT under 400 mM mannitol stress, the statistical analysis of the seed germination rates did not indicate obvious differences between WT and the three CrPIP2;3 OX lines ( Figure 5F).

CrPIP2 Exhibits Apparent Tolerance to Drought Stress in Transgenic A. thaliana Plants
To further evaluate the effects of CrPIP2;3, A. thaliana transgenic plants ectopically expressing CrPIP2;3 under the control of the 35S promoter were generated. After confirmation with genomic PCR ( Figure 5A) and qRT-PCR ( Figure 5B), three homozygous T3 line plants (OX 3#, OX 6#, and OX 10#) were selected and applied in the following tests. Briefly, about 30 sterilized seeds were spotted on the MS medium with the addition of salt, saline-alkaline, or mannitol stress, following which the germination rate was calculated. As indicated in Figure 5C, a difference in the germination of the WT and CrPIP2;3 OX lines could be observed, and the germination of the CrPIP2;3 OX lines was higher than that of the WT when the seeds were grown in MS medium containing 200 mM NaCl or 5, 7.5, and 10 mM NaHCO 3 , and the corresponding statistical analysis of the seed germination rates also showed consistent results with the germination status, with better salt and alkali tolerance ( Figure 5D,E). There was little difference in the seed germination rates between the WT and transgenic plants under osmotic stress (200, 300, and 400 mM mannitol) ( Figure 5C). Although the CrPIP2;3 OX lines produced a larger radicle and hypocotyl than WT under 400 mM mannitol stress, the statistical analysis of the seed germination rates did not indicate obvious differences between WT and the three CrPIP2;3 OX lines ( Figure 5F). Uniformly growing seedlings (WT and CrPIP2;3 OX lines) cultured on MS medium for 7 d were transplanted to MS plates with different stresses to evaluate their tolerance at the seedling growth stage. Basically, the main root length of the WT and CrPIP2;3 OX lines was suppressed with the increase in the concentration of NaCl, NaHCO3, or mannitol. Although there were no clear differences between the WT and CrPIP2;3 OX lines under 150 or 200 mM NaCl, about half of the WT seedlings died under 200 mM NaCl ( Figure 6A,B). In the presence of NaCl and NaHCO3 (salt-alkali stress), the length of the WT root was slightly shorter than that of the CrPIP2;3 OX lines ( Figure 6A,C). In accordance with the results at germination, the high osmotic stress (200, 300, and 400 mM mannitol) did not cause obvious differences between the WT and three CrPIP2;3 OX lines ( Figure 6A,D). These results indicated that the effects of CrPIP2;3 under the different stresses varied, and the overexpression of CrPIP2;3 enhanced the salt/alkali tolerance but did not respond greatly to osmotic stress at the germination stage and seedling stage. . Asterisks indicate significant differences from the control (Student' s t-test, * P < 0.05 and ** P < 0.01).
Uniformly growing seedlings (WT and CrPIP2;3 OX lines) cultured on MS medium for 7 d were transplanted to MS plates with different stresses to evaluate their tolerance at the seedling growth stage. Basically, the main root length of the WT and CrPIP2;3 OX lines was suppressed with the increase in the concentration of NaCl, NaHCO 3 , or mannitol. Although there were no clear differences between the WT and CrPIP2;3 OX lines under 150 or 200 mM NaCl, about half of the WT seedlings died under 200 mM NaCl ( Figure 6A,B). In the presence of NaCl and NaHCO 3 (salt-alkali stress), the length of the WT root was slightly shorter than that of the CrPIP2;3 OX lines ( Figure 6A,C). In accordance with the results at germination, the high osmotic stress (200, 300, and 400 mM mannitol) did not cause obvious differences between the WT and three CrPIP2;3 OX lines ( Figure 6A,D). These results indicated that the effects of CrPIP2;3 under the different stresses varied, and the overexpression of CrPIP2;3 enhanced the salt/alkali tolerance but did not respond greatly to osmotic stress at the germination stage and seedling stage. Figure 6. Salt, salt-alkali, and high osmotic stress analyses of transgenic plants with CrPIP2;3 based on seedling root length. Four-day-old seedlings were transplanted into MS medium containing NaCl, NaCl plus NaHCO3 (pH 8.2) or mannitol and were then grown for 7 d before measuring the root length. (A) Photographs of transgenic lines (CrPIP2;3 OX 3#, OX 6#, and OX 10#) and WT seedlings on MS medium or MS medium with NaCl, NaCl plus NaHCO3 (pH8.2), or mannitol; (B-D) seedling root length (mm) of WT and transgenic lines under NaCl (B), NaCl plus NaHCO3 (pH 8.2) (C), or mannitol (D) stresses after 7 d. Error bars indicate the SD based on at least three replicates (n ≥ 3). Asterisks indicate significant differences from the control (Student' s t-test, * P < 0.05 and ** P < 0.01).
Seeds of the WT and CrPIP2;3 OX lines were grown under well-watered conditions for 30 d, and before salt, drought, and alkali stress treatment, the growth status of the adult plants (WT and three CrPIP2;3 OX lines) was relatively consistent. There was no difference in the tolerance of the adult plants between the WT and transgenic lines (OX 3#, OX 6#, and OX 10#) under salt (200 mM NaCl) or salt-alkali (100 mM NaHCO3, pH 8.2) stresses (results not shown). Apparently, CrPIP2;3 strongly improved the drought tolerance during the growth of the adult plants ( Figure 7A). After 13 d of withholding water, the plants were wilted to different degrees in both the WT and three CrPIP2;3 OX lines. After re-watering and growth for another 7 d, the WT plants did not recover and appeared to exhibit a lethal phenotype, while most of the CrPIP2;3 OX plants in all three lines recovered and remained alive, with an obviously higher survival rate than that of the WT ( Figure 7B). This result indicates that CrPIP2;3 overexpression increased plant resistance to drought. Figure 6. Salt, salt-alkali, and high osmotic stress analyses of transgenic plants with CrPIP2;3 based on seedling root length. Four-day-old seedlings were transplanted into MS medium containing NaCl, NaCl plus NaHCO 3 (pH 8.2) or mannitol and were then grown for 7 d before measuring the root length. (A) Photographs of transgenic lines (CrPIP2;3 OX 3#, OX 6#, and OX 10#) and WT seedlings on MS medium or MS medium with NaCl, NaCl plus NaHCO 3 (pH8.2), or mannitol; (B-D) seedling root length (mm) of WT and transgenic lines under NaCl (B), NaCl plus NaHCO 3 (pH 8.2) (C), or mannitol (D) stresses after 7 d. Error bars indicate the SD based on at least three replicates (n ≥ 3). Asterisks indicate significant differences from the control (Student' s t-test, * P < 0.05 and ** P < 0.01).
Seeds of the WT and CrPIP2;3 OX lines were grown under well-watered conditions for 30 d, and before salt, drought, and alkali stress treatment, the growth status of the adult plants (WT and three CrPIP2;3 OX lines) was relatively consistent. There was no difference in the tolerance of the adult plants between the WT and transgenic lines (OX 3#, OX 6#, and OX 10#) under salt (200 mM NaCl) or salt-alkali (100 mM NaHCO 3 , pH 8.2) stresses (results not shown). Apparently, CrPIP2;3 strongly improved the drought tolerance during the growth of the adult plants ( Figure 7A). After 13 d of withholding water, the plants were wilted to different degrees in both the WT and three CrPIP2;3 OX lines. After re-watering and growth for another 7 d, the WT plants did not recover and appeared to exhibit a lethal phenotype, while most of the CrPIP2;3 OX plants in all three lines recovered and remained alive, with an obviously higher survival rate than that of the WT ( Figure 7B). This result indicates that CrPIP2;3 overexpression increased plant resistance to drought.

The Overexpression of CrPIP2;3 Affects the Expression of Genes Related to the ABA Signaling Pathway and Osmotic Stress in A. thaliana
To evaluate the potential impact of CrPIP2;3 overexpression in stress signal pathways, the expression of marker genes involved in certain stress responses in A. thaliana was analyzed by qRT-PCR ( Figure 8). HAI2 (At1g07430) encodes a highly ABA-induced protein phosphatase 2C member, which is responsible for the ABA-activated signaling pathway; the expression of RD29B (At5g52300) is induced in response to water deprivation, and this response is mediated by ABA. Under salt stress treatment (200 mM NaCl), both HAI2 and RD29B in the CrPIP2;3 OX plants showed obviously elevated expression patterns ( Figure 8A), and the expression of RD29B was enhanced in the CrPIP2;3 OX plants even under normal growth conditions ( Figure 8B). Salt-alkali (50 mM NaHCO3 at pH 8.2) and high osmotic (400 mM mannitol) stress both induced a higher expression of RD29B in CrPIP2;3 OX plants than in WT plants ( Figure 8B), which also indicated that in the CrPIP2;3 OX plants, the high expression of CrPIP2;3 might affect the ABA response signal pathway, and thereby improve the adaptability to water-deficit stress. Both RD26 (At4g27410) and ANAC019 (At1g52890) encode NAC transcription factors, which are involved in ABA-mediated dehydration responses. Both genes were significantly enhanced in CrPIP2;3 OX plants under salt, high osmotic, and alkali stress treatments ( Figure 8C,D), which also implied that CrPIP2;3 plays a regulatory role in the ABA-associated water imbalance response signaling pathway.

The Overexpression of CrPIP2;3 Affects the Expression of Genes Related to the ABA Signaling Pathway and Osmotic Stress in A. thaliana
To evaluate the potential impact of CrPIP2;3 overexpression in stress signal pathways, the expression of marker genes involved in certain stress responses in A. thaliana was analyzed by qRT-PCR ( Figure 8). HAI2 (At1g07430) encodes a highly ABA-induced protein phosphatase 2C member, which is responsible for the ABA-activated signaling pathway; the expression of RD29B (At5g52300) is induced in response to water deprivation, and this response is mediated by ABA. Under salt stress treatment (200 mM NaCl), both HAI2 and RD29B in the CrPIP2;3 OX plants showed obviously elevated expression patterns ( Figure 8A), and the expression of RD29B was enhanced in the CrPIP2;3 OX plants even under normal growth conditions ( Figure 8B). Salt-alkali (50 mM NaHCO 3 at pH 8.2) and high osmotic (400 mM mannitol) stress both induced a higher expression of RD29B in CrPIP2;3 OX plants than in WT plants ( Figure 8B), which also indicated that in the CrPIP2;3 OX plants, the high expression of CrPIP2;3 might affect the ABA response signal pathway, and thereby improve the adaptability to water-deficit stress. Both RD26 (At4g27410) and ANAC019 (At1g52890) encode NAC transcription factors, which are involved in ABAmediated dehydration responses. Both genes were significantly enhanced in CrPIP2;3 OX plants under salt, high osmotic, and alkali stress treatments ( Figure 8C,D), which also implied that CrPIP2;3 plays a regulatory role in the ABA-associated water imbalance response signaling pathway.
Accordingly, several A. thaliana antioxidant-enzyme genes, including CAT1 (At1g20630), CSD1 (At1g08830), APX1 (At1g07890), and FSD1 (At4g25100), were also evaluated by qRT-PCR, and the results indicated that the expression of these genes was not affected by CrPIP2;3 overexpression, even under salt, high osmotic, and alkali stress challenges ( Figure S3A). We also assessed the ROS accumulation status in the WT and three CrPIP2;3 OX seedlings under salt, high osmotic, and alkali stress challenges by NBT and DAB stain-ing. As we identified CrPIP2;3 as a H 2 O 2 transporter in yeast, we hypothesized that this protein might be involved in ROS signal transduction in plants. However, unexpectedly, the overexpression of CrPIP2;3 in A. thaliana did not alleviate the ROS accumulation caused by these above stress challenges, and in the WT and three CrPIP2;3 OX lines, the ROS accumulation seemed to be similar, although the ROS level was elevated by these stresses in both the WT and three CrPIP2;3 OX lines ( Figure S3B,C). Accordingly, several A. thaliana antioxidant-enzyme genes, including CAT1 (At1g20630), CSD1 (At1g08830), APX1 (At1g07890), and FSD1 (At4g25100), were also evaluated by qRT-PCR, and the results indicated that the expression of these genes was not affected by CrPIP2;3 overexpression, even under salt, high osmotic, and alkali stress challenges ( Figure S3A). We also assessed the ROS accumulation status in the WT and three CrPIP2;3 OX seedlings under salt, high osmotic, and alkali stress challenges by NBT and DAB staining. As we identified CrPIP2;3 as a H2O2 transporter in yeast, we hypothesized that this protein might be involved in ROS signal transduction in plants. However, unexpectedly, the overexpression of CrPIP2;3 in A. thaliana did not alleviate the ROS accumulation caused by these above stress challenges, and in the WT and three CrPIP2;3 OX lines, the ROS accumulation seemed to be similar, although the ROS level was elevated by these stresses in both the WT and three CrPIP2;3 OX lines ( Figure S3B,C).

Discussion
As a mangrove-associated species, C. rosea is well adapted to environments with high saline-alkaline levels and is subjected to drought and high osmotic stress in tropical or subtropical coastal areas or islands. It can therefore be considered an "extremophile" due to its unique tissue tolerance adaptations. C. rosea also exhibits a remarkable growth rate on coral islands with limited fresh water and nutrient resources, which suggests that this species can be used as a ground cover plant in ecological restoration or reconstruction on tropical islands and reefs to maintain species diversity and landscape characteristics [1,2]. However, the key genes involved in the molecular stress tolerance mechanisms remain unknown. Elucidating the molecular mechanisms of water utilization in this plant could be crucial to explaining its survival under extreme water deprivation. While these genes exhibiting stress resistance might share many common features with homologs from other plant species, there could also be some specific differences consid-

Discussion
As a mangrove-associated species, C. rosea is well adapted to environments with high saline-alkaline levels and is subjected to drought and high osmotic stress in tropical or subtropical coastal areas or islands. It can therefore be considered an "extremophile" due to its unique tissue tolerance adaptations. C. rosea also exhibits a remarkable growth rate on coral islands with limited fresh water and nutrient resources, which suggests that this species can be used as a ground cover plant in ecological restoration or reconstruction on tropical islands and reefs to maintain species diversity and landscape characteristics [1,2]. However, the key genes involved in the molecular stress tolerance mechanisms remain unknown. Elucidating the molecular mechanisms of water utilization in this plant could be crucial to explaining its survival under extreme water deprivation. While these genes exhibiting stress resistance might share many common features with homologs from other plant species, there could also be some specific differences considering the extreme growing conditions of this plant. In the present study, we identified and characterized an aquaporin (AQP) gene, CrPIP2;3, from C. rosea for the first time.
AQPs, belonging to a large protein family, mainly mediate the transmembrane transportation of water and other neutral molecules and are thus involved in plant responses to various water-deficit environmental stresses, including salt, drought, cold, and even desiccation [3,4]. The critical role of AQPs in regulating plant stress tolerance has been extensively addressed in multiple plant species over the past decades [6,7]. However, in recent years, there has been an increasing body of research focused on the functional identification of AQPs from plants with specialized habitats. Chickpea (Cicer arietinum), wild soybean (Glycine soja), and common vetch (Vicia sativa) are all legumes with better drought or salinity stress tolerance than standard crops. The AQP family in these three species has been systematically characterized, providing valuable information for further functional analysis to infer the roles of AQPs in adaptation to diverse environmental conditions [45,47,48]. Psammophytes jojoba (Simmondsia chinensis), date palm (Phoenix dactylifera), and jujube (Ziziphus jujuba) can all tolerate drought, salinity, and nutrient stress, and their AQP genes have been intensively studied to better understand their potential for genetic engineering to improve plant stress tolerance [39,49,50]. The halophytes Atriplex canescens, Sesuvium portulacastrum, and Thellungiella salsuginea all exhibit strong salt tolerance, and their PIP genes have been found to function in the response to salt stress and could be used in genetic engineering to improve plant growth under abiotic stress [33,35,40]. However, overall, the biological functions of these plant AQP members in stress tolerance are mainly related to their structure and biochemical function in water transport in response to physiological changes in the cellular environment, and this feature appears to have been highly temporally conserved in various species. Among them, plant plasma membrane intrinsic proteins (PIPs) exhibit the most homogeneity and high sequence identity across species and have been widely proved to be involved in abiotic stress resistance [8, 10,11]. In this study, the first identified AQP gene from the halophyte C. rosea, CrPIP2;3, was grouped into the PIP2 subfamily and showed high levels of homology to CaPIP2;2 and GmPIP2;5 (more than 80% identity in the amino acid sequence) ( Figure 1A,C), which suggests that this protein is important for water channel activity, as observed for other PIP2 subfamily members. Amino acid sequence analysis showed that CrPIP2;3 contains six putative transmembrane α-helices and a conserved MIP signal sequence (SGxHxNPAVT), which is found in all PIP members ( Figure 1A,B), and the subcellular localization prediction indicated that CrPIP2;3 was mainly distributed in the plasma membrane, which further defined this protein as a PIP2 transmembrane transport protein with strong water channel activity.
Although numerous studies have shown that plant PIP2s play considerable roles in abiotic stress tolerance, including drought, salt, and even cold, the expression pattern of AQPs should provide the most immediate evidence [51]. CrPIP2;3 is expressed in different tissues or organs at different expression levels in C. rosea plants ( Figure S2). We characterized some special cis-acting elements in the CrPIP2;3 promoter region (Figure 2), which indicated that the expression of CrPIP2;3 is controlled by specific transcription factors that interact with these elements to induce the expression of this gene in response to some abiotic stresses or specific developmental signals in C. rosea. It is still unclear if CrPIP2;3 is involved in the adaptive evolution of C. rosea to the extreme environment on tropical reefs or the coast. In the present study, we mimicked some abiotic stresses, such as high salinity (600 mM NaCl), alkali toxicity (150 mM NaHCO 3 , pH8.2), high osmotic conditions (300 mM mannitol), and ABA treatment (100 µM). The expression pattern of CrPIP2;3 varied with stress treatment duration, generally first decreasing and then increasing (Figure 3). Alkali toxicity caused a slight temporary increase in CrPIP2;3 expression in the root (2 h), following which the expression decreased. This indicated that alkali stress could temporarily enhance root hydraulic conductivity in C. rosea, and with the increase in alkali stress treatment, the water balance could be maintained even though the expression of CrPIP2;3 decreased in the root. This could be because the water imbalance caused by alkali toxicity was only slight. High osmotic stress strongly induced CrPIP2;3 expression (about 25 times) in C. rosea roots after 48 h, which may suggest that the high expression of CrPIP2;3 could help to improve water absorption from the outside by the roots, and the higher expression of CrPIP2;3 in the vine could help coordinate water transport to the aerial parts of C. rosea. In the leaf, the expression of CrPIP2;3 was obviously decreased under all challenges or ABA treatment, which might be a protective strategy to reduce water loss via the leaves by transpiration. In general, our CrPIP2;3 expression results indicated that CrPIP2;3 may participate in osmotic and salt-alkali tolerance in C. rosea and is therefore probably involved in the adaptation of C. rosea to tropical reefs or coasts.
As a unicellular eukaryote with a rapid growth rate, yeast has become a convenient protein expression system for protein functional identification. We found that inducing the expression of CrPIP2;3 in yeast ( Figure S4A) resulted in obvious sensitivity to high salt and high osmotic stress (Figure 4), which is, to some extent, opposite to that observed in other plant PIP members, such as barley PIP2;5 (HvPIP2;5), foxtail millet SiPIP3;1 and SiSIP1;1, and date palm PdPIP1;2, which improved salt and osmotic stress tolerance in yeast [49,52,53]. Similarly, mammalian AQP1 and AQP5 caused moderate growth inhibition under salt and hyperosmotic stress when expressed in yeast [54], and many AQPs have been confirmed as membrane transporters of H 2 O 2 in the plant response to stress [55]. In this study, the expression of CrPIP2;3 caused sensitivity to high salt and osmotic stress, probably due to the more rapid water loss caused by the accumulation of CrPIP2;3 in the cell membrane. This offered further evidence that CrPIP2;3 is active in water transmembrane transport in vivo. Accordingly, the sensitivity to H 2 O 2 was due to the stronger import penetration to H 2 O 2 and more toxic effects on yeast cells caused by CrPIP2;3 than the control yeast.
In order to elaborate the function of CrPIP2;3 in abiotic stress in plants, we generated transgenic A. thaliana plants overexpressing the GFP-CrPIP2;3 fusion gene under the control of the constitutive CaMV 35S promoter ( Figure S4B). In general, the accumulation of aquaporins in plants may alleviate water deprivation by improving water retention and maintaining water balance under unfavorable conditions, thereby maintaining plant survival even under serious water shortages or special circumstances. To date, it has been demonstrated in many plants that the heterologous expression of plant AQP genes can improve tolerance to salinity, drought, or oxidative stress [27,28]. In the present study, at the seed germination and seedling growth stages, CrPIP2;3 overexpression generally only showed moderate improved tolerance to salt, salt-alkali, and high osmotic stress. This finding could be due to the ability of CrPIP2;3 to facilitate water absorption, being useful but not that marked when plants are grown on MS plates. The transgenic A. thaliana plants grown under NaCl, NaHCO 3 , or mannitol stresses showed stronger growth vigor than WT in terms of seed germination, hypocotyl elongation, and root length (Figures 5 and 6). These results may suggest that the overexpression of CrPIP2;3 could facilitate radicle or young root growth by maintaining the water management system under high osmotic or alkali toxicity. We also assessed the response of soil-cultivated adult plants to high salinity, drought, and alkali toxicity. Although 200 mM NaCl and 100 mM NaHCO 3 did not cause any phenotypic differences between the WT and CrPIP2;3 OX lines, the drought-stressed CrPIP2;3 OX plants were able to recover after re-watering and showed a much higher survival rate than WT (Figure 7). This indicated that CrPIP2;3 could increase the water usage under drought, whereas high salt or alkali toxicity were not related to water deficit overmuch, but rather to an ion or pH imbalance. This result provided further evidence that CrPIP2;3 confers higher water uptake in the roots at the adult plant stage.
According to the ROS levels determined by NBT and DAB staining in the leaves of CrPIP2;3 OX A. thaliana lines, our results showed that although salt, alkali, and high osmotic stress all resulted in ROS production in the plants, the effects of ROS accumulation caused by CrPIP2;3 overexpression seemed to be negligible ( Figure S3B,C). Combined with our qRT-PCR results of A. thaliana antioxidant-enzyme genes (CAT1, CSD1, APX1, and FSD1), the overexpression of CrPIP2;3 in A. thaliana did not affect the expression of these genes in the CrPIP2;3 OX seedlings. These results implied that the ectopic expression of CrPIP2;3 in A. thaliana did not result in significant differences in the redox state in the CrPIP2;3 OX plants under abiotic stress. Furthermore, the phenotypic difference between the WT and CrPIP2;3 OX lines to stress tolerance was mainly caused by water absorption, not by antioxidant ability. Even though CrPIP2;3 is an active H 2 O 2 transporter ( Figure 4D,E), it seemed that, at least in response to water-deficit stress, CrPIP2;3 played a greater role in the transport of water rather than in ROS signaling.
In conclusion, the first aquaporin gene CrPIP2;3 from the halophyte C. rosea was isolated and characterized systematically in this study and was strongly induced under salt, high osmotic stress, and ABA treatment. We demonstrated that the induced expression of CrPIP2;3 in yeast could result in the sensitivity of yeast cells to salt, high osmotic stress, and H 2 O 2 , which indicated that CrPIP2;3 is an active H 2 O and H 2 O 2 transmembrane transporter. Further, the overexpression of CrPIP2;3 in A. thaliana conferred tolerance to salinity, drought, and salt-alkali in plants by retaining a better water status rather than by reducing ROS accumulation and membrane damage by enhancing the expression of antioxidant enzymes. This is the first study aimed at functionally characterizing an aquaporin gene from C. rosea. This study also represents the first attempt to elucidate the function of CrPIP2;3 in plant genetic improvement under salt or drought tolerance. However, further research is required to better understand the function of CrPIP2;3 in C. rosea, including its involvement in maintaining good cellular water status or interacting with other protective proteins, and thus participating in the adaptation of C. rosea to the extreme environmental habitats of tropical coasts and reefs.

Plant Materials, Growth Conditions, and Treatments
The C. rosea seeds were gathered from Yongxing Island (YX, 16 • 50 N, 112 • 20 E). The seeds were dried outdoors in summer, and the seedlings were cultivated outdoors with nutrient vermiculite or sandy soil until flowering and seed maturation from April to December (2019) in the South China Botanical Garden (SCBG, 23 • 18 N, 113 • 35 E). Seedlings of C. rosea were used for stress treatment assays to detect the expression patterns of CrPIP2;3. Several stress factors and hormone treatments, including salt (600 mM NaCl), alkali (150 mM NaHCO 3 , pH 8.2), water-deficit or drought (300 mM mannitol), and abscisic acid (ABA) treatment (100 µM), were applied to the C. rosea seedlings by irrigation or spraying to challenge the plants and to induce transcriptional changes in CrPIP2;3. The aerial parts (leaves and vines) and roots of C. rosea at different timepoints (0, 2, 12, 24, and 48 h) were collected separately and then immediately placed into liquid nitrogen for future use. The different tissues or organs of the C. rosea seedlings or adult plants present at SCBG were also gathered for further expression assays.
The wild-type (Col-0) A. thaliana was grown in a growth chamber at 22 • C with a photoperiod of 16 h light/8 h darkness. Transgenic A. thaliana plants were generated using the floral dip method. The T1 generation plants were selected by Basta (30 µL of 13.5% Basta solution added to 100 mL Murashige and Skoog (MS) medium) on MS plates, and the seedlings were identified by polymerase chain reaction (PCR) and reverse transcription (RT)-PCR to verify the overexpression of CrPIP2;3 in A. thaliana. Three T2 homozygous lines were selected for further phenotype identification or subcellular localization analysis. The house-keeping gene AtActin2 (At3g18780) was used as an internal control.

Gene Isolation and Bioinformatics Analysis
A full-length cDNA library from C. rosea was constructed using the SMART ™ cDNA Library Construction Kit (Clontech, Takara Bio USA), with the Saccharomyces cerevisiae expression vector pYES-DEST52 as the carrier of the cDNA library. Dozens of colonies were picked randomly and sequenced, and one of the cDNAs showed intact open reading frames (ORFs) and had high homology to other plant PIP2 genes. Combining the genomic DNA sequence results of C. rosea, this cDNA sequence was named CrPIP2;3. The obtained sequences of this PIP2 gene were submitted to NCBI (Accession No.: MT787666). The protein sequences of CrPIP2;3 and other homologous proteins were aligned using ClustalW, and a phylogenetic tree was generated using the MEGA6 program with the neighbor-joining method and 1000 bootstrap replicates. The sequence information of other plant PIPs in this study is listed in Table S2. The three-dimensional structures and transmembrane regions or orientation prediction of CrPIP2;3 were predicted by the Phyre2 server (http://www.sbg.bio.ic.ac. uk/phyre2/html/page.cgi?id=index). The conserved domain of CrPIP2;3 was searched using the Pfam online program (http://pfam.xfam.org/). The subcellular localization of CrPIP2;3 was also predicted using the WoLF PSORT server (https://wolfpsort.hgc.jp/) and Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/).
To predict the motifs of the CrPIP2;3 promoter region, the ATG upstream sequence (2000 bp) of the CrPIP2;3 coding region was selected from the C. rosea genomic DNA sequence data (unpublished) and analyzed with the online program PlantCARE (http: //bioinformatics.psb.ugent.be/webtools/plantcare/html/).
The transformed single colony was inoculated in SDG-Ura medium (plus 2% galactose) overnight at 30 • C, diluted with fresh pre-warmed SDG-Ura medium, and then incubated with vigorous shaking for approximately 48 h at 30 • C to reach an optical density just higher than 1.0 at OD600. Then, the cells were serially diluted in 10-fold steps, and 2 µL aliquots of each were finally spotted onto SDG-Ura medium plates with or without NaCl, sorbitol, PEG8000, or H 2 O 2 . The test plates were incubated at 30 • C for 3 to 7 d.

Expression Analysis of CrPIP2;3 in C. rosea Plants
Total RNA was isolated from C. rosea and A. thaliana using HiPure Plant RNA Kits (Magen, Guangzhou, China), and the cDNA was synthesized using TransScript One- Step gDNA Removal and cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China) according to the manufacturer's instructions. Quantitative reverse transcription (qRT)-PCR was conducted using a LightCycler ® 480 Gene Scanning system (Roche, Switzerland) and TransStart Top Green qPCR SuperMix (TransGen Biotech, Beijing, China). Gene expression levels were normalized using the C. rosea reference gene CrEF-α as internal control. The primer pairs (CrPIP2;3RTF/CrPIP2;3RTR and CrEF-αRTF/CrEF-αRTR) used for qRT-PCR are listed in Table S3.

Generation of Transgenic A. thaliana Plants
The ORF sequence of CrPIP2;3 was PCR-amplified with the primer pair CrPIP2;3OXF/ CrPIP2;3OXR (Table S3). The PCR fragments were then inserted between the EcoRI and BamHI sites of the pGEAD plasmid to generate a recombination vector with a 35S promoterdriven overexpression cassette and Basta-resistant gene ( Figure S4B). After sequencing confirmation, the pGEAD plasmid and recombination vector containing CrPIP2;3 were transformed into Agrobacterium tumefaciens GV3101, and transgenic A. thaliana was generated using the floral dip method. Positive transgenic plants with CrPIP2;3 overexpression (CrPIP2;3 OX lines) were confirmed by genomic PCR with the primer pair Cr-PIP2;3OXF/CrPIP2;3OXR and by qRT-PCR with the primer pair CrPIP2;3RTF/CrPIP2;3RTR (Table S3). The reference gene for the qRT-PCR was AtACT2 (At3g18780) in A. thaliana using the primer pair AtACT2RTF/AtACT2RTR (Table S3). The pGEAD transgenic plants were identified by a Basta-resistant screening protocol. In brief, three T3 homozygous (OX 3#, OX 6#, OX 10#) transgenic seeds were germinated for further research.

Evaluation of the Stress Tolerance of Transgenic A. thaliana
To evaluate the functions of CrPIP2;3 in abiotic stress tolerance in plants, particularly the response to high salinity or alkali, water deficit, or drought-stress challenges, the seed germination rate, seedling root length, and the phenotype of adult plants of three CrPIP2;3 OX homozygous lines (OX 3#, OX 6#, OX 10#) under salt/alkali or drought stresses were recorded and analyzed separately.
To evaluate the seedling tolerance to salt, alkali, or hyperosmotic stress, the root lengths of the OX 3#, OX 6#, OX 10#, and WT seedlings were also measured after 7 d of different stress challenges. In short, seeds were germinated on MS plates for 3 d under suitable conditions, following which the sprouts were transferred onto MS plates supplied with 100, 150, and 200 mM NaCl (salt stress), 0.5 mM NaHCO 3 plus 99.5 mM NaCl (pH 8.2), 0.75 mM NaHCO 3 plus 99.25 mM NaCl (pH 8.2), and 1 mM NaHCO 3 plus 99 mM NaCl (pH 8.2) (alkali stress), or 200, 300, and 400 mM mannitol (hyperosmotic stress). The seedling lengths were calculated, and photographs were taken.
Salt, alkali, and drought tolerance assays were also assessed using transgenic adult A. thaliana plants. The A. thaliana seeds (OX 3#, OX 6#, OX 10#, and WT) were sown in vermiculite directly and were well-cultivated for 30 d. Subsequently, these adult plants were subjected to the following stress tolerance assays. For the salt tolerance assays, the plants were well irrigated with 200 mM NaCl solution for 21 d. For the alkali tolerance assays, the plants were watered with 100 mM NaHCO 3 (pH 8.2) solution for 35 d. For the drought tolerance assays, the plants were maintained under continuous drought for 13 d and were re-watered for another 7 d. The survival rates were counted, and photographs were taken.
The reactive oxygen species (ROS) accumulation was also assessed with a nitro-blue tetrazolium (NBT) or 3.3 -diaminobenzidine (DAB) staining assay. In brief, three-week-old seedlings (WT and CrPIP2;3 OX lines) growing in the soil were soaked in 200 mM NaCl, 400 mM mannitol, or 50 mM NaHCO 3 solutions for 24 h, with water as the control. The rosette leaves were collected before and after each treatment, and the in situ detection of H 2 O 2 and O 2 − in the leaves was determined with 1 mg/mL NBT or 1 mg/mL DAB solution, respectively, for 12 h, followed by clearing in 96% ethanol.

Expression Analysis of Stress-Responsive Marker Genes in CrPIP2;3-Overexpression A. thaliana
For detecting the expression of antioxidative system-related genes (CAT1, CSD1, APX1, and FSD1) and abiotic stress-related genes (HAI2, RD26, RD29B, and ANAC19) in threeweek-old A. thaliana (WT or transgenic lines OX 3#, OX 6#, and OX 10#) planted in soil, the total RNA was isolated from the rosette leaves at different time points (with or without stress treatments, including 200 mM NaCl, 400 mM mannitol, or 50 mM NaHCO 3 at pH 8.2), and cDNA synthesis was performed using the above procedure. The reference gene for the qRT-PCR was AtACT2 (At3g18780) in A. thaliana. The primers used for qRT-PCR are listed in Table S3.

Statistical Analysis
All of the above experiments were repeated at least three times independently, and the data shown are the mean ± SD. In this research, statistical analyses were performed using the statistical tools (Student's t-test) of Excel 2010 software (Microsoft Corp., Albuquerque, NM, USA). The significance level was defined as * (P < 0.05) and ** (P < 0.01). Funding: This research was supported by grants from the Guangdong Science and Technology Program (2019B121201005), the 'Strategic Priority Research Program' of the Chinese Academy of Sciences (XDA13020500), the National Natural Sciences Foundation of China (31570257 and U1701246), and Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou) (GML2019ZD0408). The funders had no roles in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

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