• Open Access

Enhanced salt stress tolerance of rice plants expressing a vacuolar H+-ATPase subunit c1 (SaVHAc1) gene from the halophyte grass Spartina alterniflora Löisel

Authors


(fax +1 225 5781403; email nbaisakh@agcenter.lsu.edu)

Summary

The physiological role of a vacuolar ATPase subunit c1 (SaVHAc1) from a halophyte grass Spartina alterniflora was studied through its expression in rice. The SaVHAc1-expressing plants showed enhanced tolerance to salt stress than the wild-type plants, mainly through adjustments in early stage and preparatory physiological responses. In addition to the increased accumulation of its own transcript, SaVHAc1 expression led to increased accumulation of messages of other native genes in rice, especially those involved in cation transport and ABA signalling. The SaVHAc1-expressing plants maintained higher relative water content under salt stress through early stage closure of the leaf stoma and reduced stomata density. The increased K+/Na+ ratio and other cations established an ion homoeostasis in SaVHAc1-expressing plants to protect the cytosol from toxic Na+ and thereby maintained higher chlorophyll retention than the WT plants under salt stress. Besides, the role of SaVHAc1 in cell wall expansion and maintenance of net photosynthesis was implicated by comparatively higher root and leaf growth and yield of rice expressing SaVHAc1 over WT under salt stress. The study indicated that the genes contributing toward natural variation in grass halophytes could be effectively manipulated for improving salt tolerance of field crops within related taxa.

Introduction

Soil salinity is a major threat to agricultural productivity worldwide. At high concentrations of salts in the soil, plants experience a physiological drought because of the inability of roots to extract water, and high concentrations of salts within the plant can be toxic (Munns and Tester, 2008). Because salt tolerance is a quantitative trait governed by multiple genes, success with classical breeding has been low in developing salt-tolerant crops (Winicov, 1998). Genetic manipulation provides an alternative yet sustainable approach to engineering rice plants for salinity tolerance. Recently, however, four markers from candidate genes including SKC1 within the ‘Saltol’ QTL have been identified in rice and are being used in marker-assisted selective breeding of salt-tolerant rice (Thomson et al., 2010).

Halophyte models have been used as a source of genes for engineering salt tolerance in heterologous systems. Considering the differences in the anatomical features between dicots and monocots and also in their adaptation regulation networking, it is necessary to explore halophytic models among grass species as a source for superior alleles/regulation machinery through genetic engineering onto food crops, which are mostly grasses, such as rice (Tester and Bacic, 2005). Earlier we reported that Spartina alterniflora (smooth cordgrass), a monocot halophyte with reportedly all possible mechanisms of salt tolerance, shares 80–90% similarity with rice with regard to their DNA and protein sequences (Baisakh et al., 2008). Although there are a handful of reports documenting transgenic overexpressers showing some degree of salt tolerance, actual production of transgenic plants with demonstrably improved salt stress tolerance is few and slow (Flowers, 2004).

Among various mechanisms, control of ion movement across tonoplast (and plasma membrane) to maintain low Na+ concentration in the cytoplasm is the key cellular factor in salinity tolerance. Salt-tolerant plants differ from salt-sensitive ones in having a low rate of Na+ and Cl transport to leaves and the ability to compartmentalize these ions in vacuoles to avoid salt toxicity by preventing their build-up in cytoplasm or cell walls. In halophytes such as common ice plant (Mesembryathemum crystallinum), vacuolar sodium sequestration under salinity is mediated by an active Na+/H+ antiporter energized by the proton motive force, which is generated and maintained by plant V-ATPase (EC3.61.34), which is one of the several members of the ATP-dependent protein pumps, through primary active H+ transport at the tonoplast (Barkla et al., 1995). The most abundant subunit of the V0 complex of vacuolar ATPase is subunit c, which is encoded by the largest multigene family. It is present in six copies that form the part of the proton-conducting pore responsible for proton translocation (Sze et al., 1999), although the biological significance of this is unknown. Subunit c was the first multigene family reported in eukaryote V-ATPases (Sze et al., 1992). H+-ATPase acts as a primary transporter that pumps protons out of the cytoplasm, thus creating a pH and electric potential gradient across the vacuole that activates many secondary transporters involved in ion and metabolite uptake (reviewed in Serrano, 1989; Sussman, 1994; Michelet and Boutry, 1995; Palmgren, 1998). Subunit c is a highly hydrophobic protein in the V0 domain. It is essential for the production of an active V-ATPase holoenzyme and is likely to be directly involved in H+-transport. Transcriptional changes of subunits of the vacuolar ATPase in response to salinity stress have been reported from a number of plants. For example, salt-induced transcriptional activation of V-ATPase subunit c has been observed in common ice plant (Dietz and Arbinger, 1996; Löw et al., 1996; Tsiantis et al., 1996), which included a considerable and fast increase in vacuolar-type H+-ATPase activity in tonoplast vesicles, irrigated with high NaCl concentrations (Ratajczak et al., 1994). This demonstrates the prime importance of V-ATPase in the adaptation of common ice plant to high sodium concentrations. In the halotolerant sugar beet, transcripts of the V-ATPase subunits A and c were found in root and leaf tissue, and NaCl treatment caused an increase in the transcript levels in leaves, but not in the roots (Kirsch et al., 1996; Lehr et al., 1999). In Porteresia coarctata, roots showed immediate pronounced (two-fold) increase in V-ATPase c transcript, whereas in the leaves, there was no significant increase until or after 5 h and it achieved twofold and threefold accumulation only after 24 and 48 h following salt stress (Senthilkumar et al., 2005). The subunit c transcript level declined in both leaf and root after 10 h of salt withdrawal. Recently, we demonstrated that an expressed sequence tag (EST) showing similarity to vacuolar H+-ATPase subunit c1 (SaVHAc1) was highly induced in both leaf and root tissues of smooth cordgrass (Spartina alterniflora) when plants were grown under hypersaline (500 mm NaCl) condition (Baisakh et al., 2008). These studies indicate that coordinated enhanced steady-state transcript levels of specific V-ATPase subunits in root and/or shoot are a characteristic for halotolerant plants, whereas salinity-induced up-regulation of V-ATPase subunits has also been demonstrated in non-halophytes (Tyagi et al., 2005) including resurrection plants (Chen et al., 2002).

The phenotypes caused by ectopic expression of a vacuolar pyrrophosphatase (AVP1) in Arabidopsis suggested that manipulation of vacuolar proton pumps in economically important crops holds promise for the reclamation of farmlands lost to salinization and lack of rainfall (Gaxiola et al., 2002). Little, however, is known about the effect and biological role of orthologous expression of genes encoding V-ATPase subunits on the plants’ ability to cope with salt stress, although its role in salt stress response has been indirectly shown in RNAi mutants (Padmanaban et al., 2004) or, as described earlier, by its increased transcript accumulation under salt stress. The present work was undertaken with an objective to understand the role that SaVHAc1 plays in plant’s physiological response to salinity through its functional expression in transgenic rice.

Results

Sequence analysis of SaVHAc1

A cDNA clone (897 bp) containing the entire coding sequence of SaVHAc1 was isolated from the cDNA library of Spartina alterniflora, which contained a 498-bp-long open reading form (ORF). The SaVHAc1 was predicted to be a membrane protein of 165 (16.62 kDa) deduced amino acid residues under transport and binding function GO category with subcellular localization in the inner tonoplast (Figure S1). The SaVHAc1 protein consisted of two hydrophilic and two hydrophobic alpha-helices with an average hydrophobicity of 0.979. The protein consisted of three transmembrane helices (each of 23 aa length) and a primary signal peptide (30 aa).

Comparison of the genomic sequences, isolated through standard genome walking procedure using primer specific to 5′- and 3′- UTR and ACP primer (Seegene Inc.), and cDNA- sequences of SaVHAc1 revealed five exons ranging from 28 to 289 nucleotides in length separated by four introns ranging from 14 to 1034 nucleotides in length with GT-AG conformation expected for eukaryotic nuclear genome (Figure 1a). Comparative analysis of the V-ATPase subunit c sequences from other plants showed that SaVHAc1 formed an independent group within the cluster that comprised of Eleusine glauca, Arabidopsis thaliana, Gossypium hirsutum and Mesembryathemum crystallinum (Figure 1b). It shared 87% and 99% identity with rice at the DNA and protein level, respectively. However, at the protein level, it showed 100% identity with the V-type proton ATPase (Asvat-P1) from Avena sativa. Restriction mapping (with six different restriction endonucleases viz., HindIII, BamHI, EcoRI, XhoI, XbaI, and SacI) showed the presence of more than one copy of the SaVHAc1 in the genome of S. alterniflora (Figure 1c). The complexity of the restriction patterns detected by full length SaVHAc1 suggested that a multigene family encodes the 16-kD proteolipid SaVHAc1 in S. alterniflora, which may account for more than the fact that S. alterniflora in itself is a allohexaploid (2n = 6x = 62).

Figure 1.

 Characterization of SaVHAc1 from Spartina alterniflora. Genomic organization (a) showing 5 exons (E1…E5). DNA sequence similarity of SaVHAc1 with V-ATPase gene from other plants (b). Copy number analysis of SaVHAc1 in Spartina alterniflora (c). Size fragments of λ/Hind III marker in kb are shown next to the horizontal bars. Sa, Spartina alterniflora; Eg, Eleucine glauca; At, Arabidopsis thaliana; Gh, Gossypium herbaceum; Mc, Mesembryathemum crystallinum; Zm, Zea mays; As, Avena sativa; Pc, Porteresia coarctata; Os, Oryza sativa; Te, Triticum aestivum; Pg, Pennisetum galucum; HIII, HindIII; BHI, BamHI; EI, EcoRI; Xh, XhoI; Xb, XbaI; SI, SacI.

Stable integration and inheritance of SaVHAc1 in transgenic rice

The integration of SaVHAc1 expression cassette (Figure 2a) in the rice genome was initially identified by positive PCR signals using gene-specific primers and subsequently confirmed by Southern blot analysis. The copy number and integration was further validated by T-DNA-rice genomic DNA flanking sequence analysis. In CLA9 and CLA20, SaVHAc1 was integrated as a single copy (Figure 2b), whereas in CLA19, four copies were integrated. SaVHAc1 was mapped in the long arm of chromosome 7 of CLA20 (Figure 2c). The segregation of SaVHAc1 in the first selfing generation (T1) progenies of CLA9 and CLA20 followed a single gene (3:1) Mendelian inheritance (Figure 2d). The SaVHAc1 locus was fixed to homozygosity in T2 generation (Figure 2e).

Figure 2.

 Generation and characterization of SaVHAc1 plants. Partial linear map of the plant transformation binary vector p35S:SaVHAc1 (a), Restriction analysis showing copy number of three independent SaVHAc1-rice (CLA9, CLA20 and CLA19), WT, wild type; M, λ/Hind III DNA size marker (b). Genome mapping of SaVHAc1 showing its integration (arrow marked) in Chromosome seven of rice (c). Single gene Mendelian segregation of SaVHAc1 in T1 generation (d) and homozyosity of SaVHAc1-rice in T2 generation (e). PC, plasmid positive control.

SaVHAc1 expression conferred salt tolerance in rice plants

Upon imposition of salt stress at 100 and 200 mm NaCl, the SaVHAc1 plants showed greater tolerance than the WT plants with respect to leaf chlorophyll bleaching in leaf-floating assay, leaf rolling, leaf withering, tip burning and other physiological and agronomic traits (Figure 3). The SaVHAc1-rice plants exposed to seedling stage salt stress grew normally after recovery and set seeds in the greenhouse, whereas the WT plants had very stunted growth, abnormal/incomplete panicle exsertion and were very highly sterile (Figure 3c). Tobacco transgenics expressing SaVHAc1 also showed enhanced tolerance to salinity as compared to the WT plants (Figure S2). No apparent difference was observed between WT and SaVHAc1-rice plants with regard to their growth and development under normal (unstressed) conditions (data not shown).

Figure 3.

 Salt tolerance of SaVHAc1-rice vis-à-vis WT rice under salt stress (S1 = 100 mm NaCl, S2 = 200 mm NaCl) in leaf disc assay (a), whole plant assay under hydroponics at 150 mm NaCl (b). The SaVHAc1-rice plants showed better growth and yielded more compared to WT rice under salt stress at reproductive stage (c).

After 72 h of salt stress, the SaVHAc1-rice plants maintained shoot growth as was evident by 22% reduction of shoot length compared to approximately 43% reduction of the WT plants over their respective unstressed control (Figure 4a). Similarly, the SaVHAc1 plants had a better root growth with a root length reduction of 9% under salinity than the WT (22%; Figure 4a). The reduction in the shoot dry weight of SaVHAc1-rice associated with stress was substantially less (16%) than the WT rice (34%; Figure 4b). The SaVHAc1 plants did not show any significant reduction (0.8%) in root dry weight relative to the unstressed control plants, which was much lower than the 8% reduction of WT plants under stress (Figure 4b).

Figure 4.

 Reduction in shoot and root length (a) and dry weight (b) of SaVHAc1-rice vis-à-vis WT one week after salt stress (150 mm NaCl). Error bars represent standard error of means.

SaVHAc1-rice retained high chlorophyll, relative water content, and maintained higher yield under salinity

Rice plants expressing SaVHAc1 maintained higher chlorophyll concentration over WT under saline conditions (Figure 5a). The loss of total chlorophyll because of salt stress (150 mm NaCl) was less (32%) in SaVHAc1-rice compared to WT plants (48%) when averaged over three time points (Figure 5a). The SaVHAc1-rice plants showed much less reduction (13%) in chlorophyll a after 36 h of salt stress compared to the WT plants (35%). The loss of the chlorophyll was directly proportional to the duration of salt stress (Figure 5a). Chlorophyll concentration derived from the SPAD reading and estimated by acetone extraction method was comparable (data not shown).

Figure 5.

 Chlorophyll a (a), relative water content (b) and grain yield per panicle of SAVHAC1- rice vis-à-vis WT plants under salt stress (150 mm NaCl) (c). Note that their yield was comparable under no-stress control condition. Error bars represent standard errors of means.

Relative water content (RWC) is considered an appropriate measure of plant water status as well as osmotic adjustment under stress. RWC, which also reflects, in part, the transpirational loss of water, was estimated from the leaves of the SaVHAc1- and WT rice plants subject to salt stress. After a week under stress, the SaVHAc1-rice plants maintained higher RWC (93%) as compared with the WT (77%; Figure 5b).

Both SaVHAc1- and WT rice plants showed a reduction in grain yield per panicle upon salt stress during 2 weeks of the critical reproductive stage (i.e. panicle initiation). However, the SaVHAc1-rice produced nearly six times higher grain yield (1.42 g) per panicle than WT plants (0.24 g) under reproductive stage salt stress (Figure 5c).

SaVHAc1 expression led to accumulation of higher K+ under salt stress

Only one SaVHAc1-rice line (CLA20) along with WT was included in the elemental analysis. The root Na+ concentration of the SaVHAc1-rice was less than the WT plants, while the leaf Na+ was high in both genotypes. In contrast, leaf K+ concentration in SaVHAc1-rice was much higher than in WT, which was clear from the high K+/Na+ values in the transgenic lines (Figure 6a). Concomitantly, the SaVHAc1-rice lines also accumulated higher levels of Ca2+ and Mg2+ in their leaf and root tissues with or without stress (Figure 6b).

Figure 6.

 Inductive coupled plasma (ICP) analysis showing higher K+:Na+ (a), and Ca2+ and Mg2+ concentration (b) in leaf and root tissues of SaVHAc1-rice (CLA20) in comparison with WT rice under control (C) and salinity (S). LC, leaf control; RC, root control; LS, leaf stress; RS, root stress. Error bars represent standard error of means.

SaVHAc1 expression altered the expression of native rice genes upon salt stress

From the microarray experiment, many of the 43 311 probes were significantly different: 4287 (9.9%) between genotypes (WT and SaVHAc1-rice), 6761 (15.6%) between environments (saline or non-saline conditions) and 705 (1.6%) responded differently to the environment depending upon the genotype (Table S1). Although a large number of genes showed up-regulation in SaVHAc1-rice relative to WT (Table S1; Gene Expression Omnibus Accession no. GSE34724), eighteen genes that showed more than twofold increase or decrease in transcription in SaVHAc1-rice were considered significantly up- or down-regulated, respectively. Fourteen genes encoding proteins with either ion transport or metal binding function were significantly up-regulated (Table 1). The four genes that showed significant down-regulation in SaVHAc1-rice did not have any functional annotation (hypothetical or expressed proteins). Interestingly, none of the other subunits of V-ATPase were observed as being affected by SaVHAc1.

Table 1.   Genes showing up-regulation (≥2-fold) by SAVHAC1-expression in rice under salt stress (analysed by microarray)
Probe IDGeneGO Slim IDTermOntologyFold increase
TR038922Pyrophosphate-energized vacuolar membrane proton pumpGO:0005773Membrane; vacuole, hydrolase activityCellular component2.5
TR055074Glutamine synthetase expressedGO:0003824Catalytic activityMolecular function2.7
TR066667Metallothionein-like protein 1GO:0046872Metal bindingMolecular function2.7
TR066578Pathogenesis-related protein 10GO:0035556Intracellular signal transductionBiological process2.6
TR057433Hypothetical protein   2.6
TR051207Germin-like protein subfamily 1 member 7 precursorGO:0005618Cell wallCellular component2.9
TR037990Cortical cell delineating protein precursorGO:0006810TransportBiological process2.9
TR068459Plastocyanin-like domain containing proteinGO:0005488Binding, membraneMolecular function3.1
TR066580Pathogenesis-related protein Bet v I family proteinGO:0035556Intracellular signal transductionBiological process4.0
TR066664Metallothionein-like protein 1GO:0046872Metal bindingMolecular function3.8
TR067039Cysteine synthaseGO:0003824Catalytic activityMolecular function3.6
TR037504Expressed protein   5.0
TR066668Metallothionein-like protein 1GO:0046872Metal bindingMolecular function5.0
TR047684SCP-like extracellular proteinGO:0005576Extracellular regionCellular component6.1

Semiquantitative expression analysis of genes (Figure 7) showed that SaVHAc1 transcript was much higher in the SaVHAc1-rice than the WT without or with salt stress imposition at all time points. Quantitatively, SaVHAc1 transcript accumulated up to 25-fold higher in the SaVHAc1-rice under salt stress than the SaVHAc1-rice without salt stress and approximately 40-fold higher over the WT rice plants (Figure 8a). The overexpression of SaVHAc1 in the transgenics also led to higher constitutive transcript accumulation of rice plasma membrane AAA-ATPase in the leaf tissues under no stress. Although there was no apparent change in its transcript in leaf tissues under salt stress, the root tissues showed comparable increase in its expression in both WT and SaVHAc1-rice under stress.

Figure 7.

 Semiquantitative RT-PCR analysis of rice native genes of rice in SaVHAc1-rice (CLA20) vis-à-vis WT under control (0 h) and at different time points under salt stress (150 mm NaCl) and 4 days after recovery (R) in both leaf and root tissues.

Figure 8.

 Quantitative RT-PCR of SaVHAc1 and rice plasmamembrane ATPase (a), genes involved in ion transport/exchange (NHX and CTP), extracellular (OsSCLP) and expressed protein (TR037504) (b), and genes involved in ABA signalling (c). CLA20 = SaVHAc1-rice, WT, wild type; LC, leaf control; RC, root control; LS, leaf stress; RS, root stress. Error bars represent standard error of means.

Other genes that were tested for their transcript accumulation showed up-regulation in SaVHAc1-rice as compared to WT plants under salt stress in a tissue and time-dependent manner (Figures 7 and 8b). One important function of the V-ATPase is to provide the driving force for H+-coupled Na+ antiporters, such as NHX1, which sequesters sodium into the vacuole (Apse et al., 1999). Cation transporter genes such as OsNHX1 and OsCTP showed little difference among SaVHAc1-rice and WT rice in leaf tissues, but their transcript was highly accumulated in the roots of SaVHAc1-rice compared to WT plants. OsSCPL1 showed a slight decline in its transcript accumulation until 24 h of salt stress in their root, but overaccumulated after 48 h through the recovery stage. Cysteine synthase (OsCS) behaved similarly. Genes such as metallothionein (OsMT1) and PR protein (BetV) showed only root-specific expression with subtle difference between SaVHAc1-rice and WT plants.

Quantitative expression analysis showed up-regulation of OsPLDα1 and OsGPA1, the two positive regulators of abscisic acid (ABA) signalling genes (Nilson and Assmann, 2010) in the SaVHAc1-rice over WT (Figure 8c) without and with stress. The leaves and roots of SaVHAc1-rice accumulated higher transcript of these two genes even without stress. Similarly, the expression of OsSORK (rice homologue of Arabidopsis GORK) was 2.8- and 1.9-fold higher in SaVHAc1-rice than WT with and without salt stress, respectively.

SaVHAc1 expression affected the density and opening of stomata

The SaVHAc1-rice had 39% fewer stomata per unit leaf area than the WT rice (Figure 9a,c). Further, the stomata were mostly closed in the SaVHAc1 lines under salt stress as compared with the WT (Figure 9b). However, there was no significant difference with respect to the stomata size as measured by their dimension. Cross-sections of leaf and root tissue did not show any apparent difference between the SaVHAc1- and WT rice (data not shown).

Figure 9.

 Scanning electron microscope picture of a leaf showing closed stomata (arrow marked and numbered) in SaVHAc1-rice (a) compared to open-type stomata under salt stress in WT rice (b). The SaVHAc1-rice has less stomata/sqcm in leaf surface than the WT under salinity (c). A 50× magnified picture of a representative stoma is shown as an inset on the top right corner of Panel a and b). Error bars represent standard error of means.

Discussion

The present results identified a major role for SaVHAc1, a gene similar to vacuolar H+-ATPase, from a halophyte grass in several physiological processes particularly those related to salt stress. The vacuolar H+-ATPase is the major proton pump that establishes and maintains an electrochemical proton gradient across the tonoplast, thus providing the driving force for the secondary active transport of ions and metabolites against their concentration gradient (Gaxiola et al., 2007). The functional significance of V-ATPase in plant’s ability to adapt to unfavourable stress was provided earlier (Dietz et al., 2001). Transcriptional activation of V-ATPase at an early stage of salt stress has been shown in several instances (Baisakh et al., 2008; Golldack and Deitz, 2000). The ability of a plant to change its gene expression profile to respond to salt stress, as we observed, might be a key mechanism for salt tolerance.

SaVHAc1 and ion homoeostasis

The capacity of plants to maintain a high cytosolic K+/Na+ ratio is likely to be one of the key determinants of salt tolerance. Halophytes utilize ion homoeostasis in the cytosol as one of the most important strategies for their adaptation to salinity. The SaVHAc1 plants accumulated high level of Na+ in roots and leaves; growth, however, was little affected by the toxic Na+, which could be due to the sequestration of Na+ at the tonoplast by a secondary Na+/H+ transporter that was energized by a proton motive force created by the overexpression of SaVHAc1 (Apse et al., 1999). Further, the increase in the concentration of other cations (K+, Ca2+, and Mg2+) in the leaf and root tissues in the SaVHAc1-rice plants (Figure 6) could be due to the electrochemical gradient generated by the SaVHAc1 expression, which is used by proton-coupled antiporters to accumulate more of Ca2+ and Mg2+ inside the lumen (Hirschi et al., 1996) and exclusion of toxic Na+ from the cytosol to protect the transgenics. This led to a reestablished ion homoeostasis which was of critical importance for the adaptation of plants to salt stress (Niu et al., 1995). It is widely established that Ca2+ ameliorates Na+ toxicity in a variety of plant species (Hasegawa et al., 2000). Elevated Ca2+ levels also stimulate net uptake of K+, possibly by enhancing membrane integrity (Maathuis et al., 2003), and thus may have modulated toward higher K+/Na+ ratio in the SaVHAc1-rice. Because, Mg2+ is a central molecule in chlorophyll, the increased accumulation of Mg2+ may also have contributed to higher photosynthesis in the SaVHAc1 plants.

SaVHAc1 and vegetative growth and yield of plants under salinity

The SaVHAc1 plants showed better root and shoot growth and improved dry weight and grain yield compared to the WT plants under saline conditions. SaVHAc1-rice possessed higher leaf N concentration relative to WT rice as predicted from their higher SPAD values, which is known to be linearly correlated with leaf N concentration. Physiologically it is well established that primary proton pumps are crucial for plant growth and survival, and VHAc genes are expressed to support growth in actively growing cells and to supply increased demand for V-ATPase in cells with active exocytosis (Padmanaban et al., 2004). In addition to maintaining ion homoeostasis, the SaVHAc1 could be involved in protein sorting and membrane fusion events that are needed to promote growth, exocytosis of wall materials in certain cell types and tolerance to salt stress. Genetic evidence that V-ATPase influenced plant development and signalling came from the first V-ATPase mutant, det3, which had reduced subunit C transcript and V-ATPase activity (Schumacher et al., 1999). The mutant was deetiolated when germinated under dark conditions and the mature plant was dwarf compared to wild-type plants. From our study and several earlier studies, it is clear that VHAc1 has implications in cell expansion. Padmanaban et al. (2004) observed that high VHAc1 promoter activity was consistently seen in the elongating zone of roots, expanding cotyledons or elongating hypocotyls, but not in non-expanding organs such as cotyledons of dark-grown seedlings or hypocotyls of light-grown seedlings. Further, they demonstrated that dsRNA-mediated RNAi mutant plants of VHAc1 were more sensitive and showed reduced root and hypocotyl length relative to wild type after 4-days under salt stress (50 mm).

The expression of SaVHAc1 played a role in the protection of plant’s photosynthetic machinery from higher Na+, which was evident by higher chlorophyll retention of the SaVHAc1 plants in comparison with the WT plants under salinity (Figure 8). In principle, enhanced expression of vacuolar proton pumps can energize solute transport by increasing the availability of protons and thus increases vacuolar solute accumulation. The net increase in the cell solutes concentration of SaVHAc1-rice must have led to an increase in the uptake of water as supported by their higher leaf RWC such that these transgenic plants could maintain turgor under low soil water conditions.

SaVHAc1 and expression of transporter genes and other rice native genes under salinity

The expression of SaVHAc1 was higher in both leaf and root tissues of SaVHAc1-rice with or without salt stress, which indicated that the SaVHAc1-rice already had the proton pump activated and prepared to adapt to the salt stress.

A number of rice native genes were affected by the SaVHAc1 expression that was evident from the microarray experiment, although the number of genes that were significantly up- and/or down-regulated was low (Table S1). Temporal and spatial changes in the gene expression were not detected in the microarray because of the fact that the RNA samples were pooled from root and leaf tissues at different time points of salt stress in WT and SaVHAc1-rice lines. However, differences in their transcript accumulation were captured in the (semi)quantitative RT-PCR (Figure 7). Although the expression of cation transporter (OsCTP) and exchanger (OsNHX1) genes were immediately affected at the early stage of salt stress, their subsequent up-regulation was observed in the SaVHAc1-rice under stress, which further explained that a proton electrochemical gradient was generated by the expression of SaVHAc1. Higher accumulation of OsCTP (a novel cation transporter in rice with homology to ChaC associated with Ca2+/H+ transport in Escherichia coli) in the SaVHAc1-rice may have contributed to the increased Ca2+ transport. The high expression of SaVHAc1, even without salt stress and subsequent up-regulation of other genes including cation transporter genes, which may have been energized by the SaVHAc1 expression as a component of coordinated regulation under salinity stress, led us to the presumption that the SaVHAc1 lines maintained an anticipatory preparedness to adapt to salt stress.

SaVHAc1 and stomata closure

As an adaptation/avoidance strategy, plants are known to close the stomata under dehydration stress to save water and maintain turgor (Skirycz and Inze, 2010). The possible involvement of ABA in the regulatory pathway leading to induction of V-ATPase activity has been established (Tsiantis et al., 1996). Overaccumulation of V-ATPase in SaVHAc1-rice might play an important role in the coregulation of ABA signalling induced by salt stress. ABA treatment mimics NaCl treatment of plants, and only the c-subunit of the V-ATPase has been shown to be transcriptionally regulated by NaCl in plants (Tsiantis et al., 1996).

Stomata opening is induced by increased turgor pressure as a result of K+ and anion influx energized by H+-ATPase. Osmotic stress, at the early stage of salt stress first experienced by the roots, releases ABA signalling molecules, which under the overexpression of vacuolar ATPase could induce stomata closure (Allen et al., 2000). The loss of turgidity as a result of K+ and water efflux from the guard cell may have contributed to the stomata closure in SaVHAc1-rice plants (Figure 9). H+ is known as secondary messenger in hormone action of plants. The increase in cytoplasmic pH (by the expression of vacuolar ATPase) in the SaVHAc1 in coordination with ABA molecules may have resulted in the opening of K+ (out) channel and closing of K+ (in) channel. Further, the increase in Ca2+Cyt in the SaVHAc1-rice possibly brought about a reduction in the K+ (in) activity of the plasma membrane, leading to the reduction in guard cell turgidity and ultimately stomata closure. Although the Ca2+- and H+-mediated K+ efflux are independent of each other, it is possible both the mechanisms are in operation to maintain a balance of the Ca2+Cyt and alkalinization (Swamy, 1999). Also, the kinetics of expression changes of the ABA signalling genes (OsPLDα1 and OsGPA1) in the SaVHAc1-rice vis-à-vis WT plants provided clues to the closure of stomata in the leaves of SaVHAc1-rice. The expression of these two positive regulators of stomata ABA signalling was higher in SaVHAc1-rice as compared with the WT (Figure 8) without and with salt stress. Similarly, the expression of OsSORK (rice homologue of Arabidopsis GORK) was 2.8- and 1.9-fold higher in SaVHAc1-rice than WT with and without salt stress, respectively. Although the mechanism of reduced stomata density in SaVHAc1-rice remains to be investigated, the earlier observation that det3 mutants were defective in guard cell signalling or movement could explain the apparent anatomical adjustments of SaVHAc1-rice to adapt to the stress condition. Further, the reduced stomata density in the SaVHAc1-rice could contribute to the slow/lower rate of water loss from SaVHAc1-rice leaves for which they maintained a higher RWC under salt stress as compared to the WT leaves. Higher RWC is one of the important factors contributing to tolerance to physiological drought that is experienced by the plant at the initial stage of salt stress as a result of low osmotic potential caused by the accumulation of soluble salts.

The time point analysis of SaVHAc1 as well as other related genes further established that SaVHAc1-rice tolerated salt stress by an early stage priming and preconditioning response (Harb et al., 2010), thus maintaining a higher RWC and chlorophyll (thereby photosynthesis), protecting cytosol from ion toxicity, and early stomata closure. Present results suggested a clearly defined functional role of SaVHAc1 with implications in the modification of a signalling pathway in the salt stress response by driving the expression of other genes involved in ion homoeostasis, solute accumulation, and ABA signalling. This study including that of others as described earlier again substantiates that grass halophytes, in addition to their usefulness to understand the gene regulation mechanism of their natural salinity-adaptability, could be effectively used as a source for mining and bioprospecting superior alleles for the manipulation of salt tolerance in monocot field crops such as rice.

Experimental procedures

Cloning of SaVHAc1, sequence analysis, and construction of binary vector

An expressed sequence tag (EST#617) of Spartina alterniflora (GenBank Acc. No EH277293; Baisakh et al., 2008) showed similarity to plant vacuolar H+-ATPase subunit c1 when its nucleotide and deduced protein sequences were blasted against the non-redundant nucleotide and protein database using BLASTN and BLASTP interface, respectively. The sequence alignment of this EST (hereinafter referred to as SaVHAc1) was carried out with orthologs from other plants (see Figure 1 for details) using CLUSTALW.

The complete open reading frame of SaVHAc1 was amplified from the S. alterniflora root cDNA library (Baisakh et al., 2008) by PCR using primers SaVHAc1 Fwd: 5′- ggaagatctatgtcgtcgacgttcag -3′ and SaVHAc1 Rev: 5′- gggtwaccctaatctgcacggac -3′ containing the BglII and BstEII restriction endonuclease (RE) recognition sites (underlined), respectively. The PCR product and pCAMBIA1301 (CAMBIA, Australia) was digested using the same REs and ligated to yield the binary plasmid p35S:SaVHAc1 (Figure 2a). The identity and orientation of p35S:SaVHAc1 was confirmed by restriction digestion and sequencing. The plasmid was mobilized into Agrobacterium tumefaciens LBA4404 using the freeze-thaw method (An et al., 1998).

Rice transformation

Agrobacterium tumefaciens transformation of rice cultivar ‘Cocodrie’ was performed following the method described earlier (Rao et al., 2009). LBA4404/p35S:SaVHAc1 was precultured overnight at 28 °C in Luria-Bertani (LB) broth with rifampicin (20 μg/mL), spectinomycin (100 μg/mL), streptomycin (50 μg/mL) and kanamycin (50 μg/mL) under constant shaking at 200 rpm. The precultured bacteria were subcultured in fresh LB with the same antibiotics and grown for 24 h. Bacteria cells were resuspended in the MS (Murashige and Skoog, 1962) liquid medium supplemented with 2 mg/L 2,4-D and 100 μm acetosyringone (MSco) to a final titre of A600 = 1.0 for transformation.

Three–four-week-old seed-derived rice embryogenic calli were vacuum-infiltrated (0.4–0.6 atm) with the bacterial suspension for 10 min and co-cultivated for 3 days in solid MSco medium at 25 °C in the dark for 3 days. Embryogenic callus development, and selection and regeneration of the putative transgenic calli was performed following the method described earlier (Baisakh et al., 2001).

Molecular analysis of plants expressing SaVHAc1

Polymerase chain reaction

Total genomic DNA was isolated from leaf tissues using a modified CTAB method (Murray and Thompson, 1980). One hundred ng of DNA was subject to PCR analysis for selectable marker gene (hph) and target gene (SaVHAc1) using gene-specific primers (5′–3′) as follows: HPH F- tacttctacacagccatc, HPH R- tatgtcctgcgggtaaat; SaVHAc1 F- aggagggtgtaccattcgtcaatg, SaVHAc1 R- ccaggctcgtagagaataccattg.

Southern blot analysis

Southern blot analysis was performed following Baisakh et al. (2006a). Ten micrograms of genomic DNA were digested with a single cutter Sst I (for copy number analysis) and BglII and BstEII (for releasing the full length SaVHAc1), electrophoresed on a 1% (w/v) TAE agarose gel, and transferred under alkaline denaturing conditions to Hybond N+ nylon membrane (GE Healthcare, Piscataway, NJ). A PCR generated 200-bp fragment of SaVHAc1, radiolabeled using (α-32P)dCTP and the Rediprime labelling system (GE Healthcare), was used as the hybridization probe. Hybridization and follow-up membrane washing, and exposure to X-ray Hyperfilm™ MP (GE Healthcare) was performed as per Baisakh et al. (2006a).

Mapping the SaVHAc1 integration site in rice

The insertion site of SaVHAc1 was mapped by isolating the T-DNA-rice genome flanking sequences using TAIL-PCR technique as described by Liu et al. (1995). The gene-specific primers designed from 35S promoter and nosT sequences were used in combination with AD1 and AD4, and the nested primers were designed from the left and right border sequences of the T-DNA. Three nested gene-specific reverse primers and one of the four arbitrary degenerate (AD1-4) primers were used in successive rounds of TAIL-PCR cycling. The primary PCR product was diluted 40-fold and used in the secondary reaction, while the latter was diluted 5-fold for the tertiary reaction. The products of the primary, secondary and tertiary reactions were analysed on a 1.5% agarose gel. Fragments exhibiting a difference in size consistent with nested gene-specific primer positions were cloned using pGEMT-easy vector (Promega, Madison, WI) and sequenced as described earlier (Baisakh et al., 2006b). The genomic sequences flanking SaVHAc1 were blasted against and mapped to the reference rice genome using megaBLASTN interface (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Salt stress treatment

Initially, cut leaves of three independent PCR positive SaVHAc1-rice plants were floated in 40 mL of water with 100 and 200 mm NaCl in plastic deep dishes and kept under continuous light for 72 h before scoring for chlorophyll bleach (Sanan-Mishra et al., 2005).

Four-week-old seedlings of homozygous SaVHAc1-rice (CLA9, CLA19, and CLA20) and WT rice were subject to salt stress (150 mm NaCl) under hydroponics in Yoshida’s nutrient solution (Yoshida et al., 1976) following the method described earlier (Batlang et al., 2012). Twenty plants were included in each of the four replications. After 1 week of stress, shoot and root length, shoot and root fresh and dry weight, chlorophyll concentration and tissue ion concentrations were measured. After a week of stress, the SaVHAc1-rice and a few surviving WT plants (those not completely dead) were transferred to fresh nutrient solution without salt, and a week-old recovered plants were grown to maturity in the greenhouse maintained at 29/21 °C day/night temperature regime under natural day light condition. At reproductive stage, salt stress was imposed as described by Batlang et al. (2012). Ten SaVHAc1 and WT plants during panicle initiation stage (90–95 days-old) were subject to 150 mm NaCl in the irrigation water continuously for 1 week before the pots were submerged up to the soil level in a deep plastic tray with salt-free water for 2 days and then irrigated normally until maturity. Data were recorded for number of seeds per plant and single panicle (primary tiller) grain yield.

Expression analysis

Microarray analysis

Total RNA was extracted from salt stressed and unstressed WT and transgenic plants at different time points (24, 48, 72 h, and 4 days of recovery) using RNeasy plant minikit (Qiagen, Valencia, CA). For microarray experiment, total RNA quality was checked in gel as well as through Agilent 2100 bioanalyzer. The RNA samples from different time points except recovery point were pooled for WT and SaVHAc1 plants for each of the four biological replications. The four samples were named WT-Control (WT-C), WT-Stress (WT-S), SaVHAc1-rice-Control (SaVHAc1-rice-C) and SaVHAc1-rice-stress (SaVHAc1-rice-S).

Target preparation, that is, first-strand synthesis, second-strand synthesis, aRNA purification, dye labelling, and hybridization and washing, was performed as per the protocol described earlier (Edwards et al., 2008; Data S1). Sixty-mer oligonucleotides designed from the rice unigenes and synthesized by Operon technologies were printed on slides in Galbraith laboratory (http://ag.arizona.edu/~dgalbrai) for use on microarray chips (45K). The GPR result file was analysed in R software (http://www.R-project.org; Data S1).

Enrichment of functional Gene Ontology categories (GO; http://www.geneontology.org) for genotype, environment and their interaction was tested using the GOEAST package (Zheng and Wang, 2008) with a Fisher’s exact test, which considered topology of the GO relationships (Alexa et al., 2006), followed by a Yekutieli adjustment for multiple comparisons (Yekutieli and Benjamini, 1999). Results were considered significant at 0.05 after adjustment. The set of oligos that responded to the environment were enriched for the biological processes related to biotic stimuli and stress responses.

(Semi)quantitative reverse transcription polymerase chain reaction (Sq/qRT-PCR)

The total RNA was extracted from 100 mg of freshly collected leaf and root tissues of SaVHAc1-rice and WT rice at 0 h (control), 2, 24, 48, 72 h and 4 days of recovery following salt stress. Two micrograms of total RNA were subject to sqRT-PCR of SaVHAc1 gene and other genes that were up/down-regulated in the microarray data, employing a single-step RT-PCR kit (Qiagen, Valencia, CA). The products were resolved in 1.5% TAE agarose gel, visualized under UV transilluminator in a Kodak 200 gel doc apparatus (Carestream Health, Inc., Rochester, NY). Rice actin gene 1 (OsAct1) was used as an internal control for the template validation. The primer sequences used in the study are provided in Table S2.

The qRT-PCR was carried out using the same RNA samples that were used for sqRT-PCR as per the method described (Baisakh et al., 2008). Essentially, 1 μg total RNA was reverse transcribed using iScript 1st strand cDNA synthesis kit (Bio-rad, Carlsbad, CA). PCR was performed in triplicate (biological replicate) with two independent cDNA preparations (technical replicate) using SYBR green master mix (Bio-rad, Hercules, CA), 2 μL cDNA and 3.25 pmol of each gene-specific primer in a MyiQ Real-Time PCR detection system (Bio-rad, Hercules, CA). The relative expression ratio was calculated using the 2ΔΔCt method (Baisakh et al., 2008) with rice elongation factor 1α gene (OsEF1α) as the reference gene.

Estimation of chlorophyll concentration

Total chlorophyll was extracted from one fully expanded leaf per plant (three plants from each of the SaVHAc1-rice and WT plants following salt stress) with 80% acetone twice. The chlorophyll a and b concentration was measured spectrophotometrically following the method described by Lichtenthaler (1987) to determine the extent of bleaching and chlorophyll loss. Total chlorophyll was also estimated from the SPAD502 meter (Konica Minolta Sensing, Inc., Ramsey, NJ) reading as described earlier (Monje and Bugbee, 1992).

Estimation of tissue ion concentration

Leaf and root tissues were harvested from 1-month-old seedlings of unstressed (control) and salt-treated SaVHAc1-rice and WT, and oven-dried at 80 °C for 48 h. Five hundred mg of dried tissues was extracted with HNO3 digestion. The Na+, K+, Ca2+ and Mg2+ concentrations were measured through inductively coupled plasma-mass spectrometry (ICP-MS, Perkin-Elmer Plasma 400 emission spectrometer) in an in-house plant and soil testing laboratory.

Growth parameters study

Data were collected on different vegetative growth parameters (root and shoot length, root and shoot dry weight) of the SaVHAc1-rice lines vis-à-vis WT plants one week after salt stress was imposed. The single panicle grain yield from the primary tiller was recorded on the salt-stressed greenhouse-grown plants at maturity. Data were analysed for anova with statistical analysis software SAS 9.1.3 (SAS Institute Inc, 2004).

Relative water content

The RWC of the leaves was determined following Slatyer (1967). Middle sections of second-youngest fully expanded leaves were collected and wrapped from three different WT and SaVHAc1-rice plants after a week of salt stress and weighed [fresh weight (FW)]. The leaf pieces were immersed in dH2O placed in dark at 4 °C overnight and weighed after brief blot-drying [turgid weight (TW)]. Then, the pieces were dried at 60 °C for 24 h and weighed [dry weight (DW)]. RWC was estimated in percentage of the water content at a given time and tissue as related to the water content at full turgor using the formula:

image

Scanning electron microscopy (SEM)

Leaf samples were collected from SaVHAc1-rice and WT before and 24 h after salt stress and were fixed with glutaraldehyde buffer followed by gradual alcohol dehydration. The leaves were then critical point dried under liquid CO2 and the probe surface sputter-coated with an electric-conducting gold layer before imaging with SEM (Cambridge S-260) at 5 kV.

Acknowledgements

The technical assistance of Greg Ford, SRRC, USDA-ARS is duly acknowledged. This work was financially supported by USDA-CSREES. This manuscript is approved for publication by the Director of Louisiana Agricultural Experiment Station as MS#2011-306-6572.

Ancillary