Vacuolar H⁺-pyrophosphatase HVP10 enhances salt tolerance via promoting Na⁺ translocation into root vacuoles

Abstract

Vacuolar H⁺-pumping pyrophosphatases (VPs) provide a proton gradient for Na⁺ sequestration in the tonoplast; however, the regulatory mechanisms of VPs in developing salt tolerance have not been fully elucidated. Here, we cloned a barley (Hordeum vulgare) VP gene (HVP10) that was identified previously as the HvNax3 gene. Homology analysis showed VP10 in plants had conserved structure and sequence and likely originated from the ancestors of the Ceramiales order of Rhodophyta (Cyanidioschyzon merolae). HVP10 was mainly expressed in roots and upregulated in response to salt stress. After salt treatment for 3 weeks, HVP10 knockdown (RNA interference) and knockout (CRISPR/Cas9 gene editing) barley plants showed greatly inhibited growth and higher shoot Na⁺ concentration, Na⁺ transportation rate and xylem Na⁺ loading relative to wild-type (WT) plants. Reverse transcription quantitative polymerase chain reaction and microelectronic Ion Flux Estimation results indicated that HVP10 likely modulates Na⁺ sequestration into the root vacuole by acting synergistically with Na⁺/H⁺ antiporters (HvNHX1 and HvNHX4) to enhance H⁺ efflux and K⁺ maintenance in roots. Moreover, transgenic rice (Oryza sativa) lines overexpressing HVP10 also showed higher salt tolerance than the WT at both seedling and adult stages with less Na⁺ translocation to shoots and higher grain yields under salt stress. This study reveals the molecular mechanism of HVP10 underlying salt tolerance and highlights its potential in improving crop salt tolerance.

A barley vacuolar H⁺-pyrophosphatase plays a critical role in developing salt tolerance via the modulation of Na⁺ sequestration into root vacuoles.

Introduction

Soil salinity is one of the major abiotic constraints for global food security and crop productivity (Munns and Tester, 2008; Ismail and Horie, 2017). Currently, salinity affects at least 800 million ha land, almost one-fourth of the world’s arable land (Rengasamy, 2010; Mokrani et al., 2020). Salt stress is a two-phase process to plants with a rapid osmotic stress and a slow ionic stress (Ismail and Horie, 2017). For most plants, soluble sodium (Na⁺) is the widespread dominant toxic ion, causing a dramatic loss of biomass and grain yield (Munns and Tester, 2008; Shen et al., 2016). Almost all cereal crops are glycophytes, causing more than 50% yield reduction under salt condition over 20 dS·m⁻¹ (∼200-mM sodium; Maas and Hoffman, 1977; Ayers and Westcot, 1985; Zhu, 2001; Munns and Tester, 2008; Horie et al., 2012). Thus, it is quite imperative to develop cereal crop cultivars with much improved salt tolerance (Deinlein et al., 2014).

After evolution for approximately 500 million years, green land plants have developed remarkably morphological, ecological, and genetic divergence in response to salt stress (Flowers et al., 2010; Zhao et al., 2019). Correspondingly, plants modulate multiple membrane-localized ionic transporters to eliminate excess toxic Na⁺. As suggested, the process of Na⁺ exclusion in roots was dominated by Salt Overly Sensitive 1 (mainly SOS1; Shi et al., 2000; Zhu, 2016), Na⁺ compartmentation into vacuoles by Na⁺/H⁺ antiporters (NHXs; Apse et al., 1999; Leidi et al., 2010), and K⁺/Na⁺ homeostasis by high-affinity potassium transporters (HKTs; Kobayashi et al., 2017; Huang et al., 2020). Because vacuoles can account up to 99% of cellular volume in some cell types, excessive Na⁺ could be gradually sequestered into the vacuole, resulting in alleviation of cytosol toxicity (Kim et al., 2020). The positive function of some NHX members on salt tolerance has already been verified for the exchange of Na⁺ and H⁺ across the tonoplast (Apse et al., 1999; Leidi et al., 2010). Meanwhile, Na⁺ compartmentation into vacuoles is facilitated through proton-pumping system, which is believed to be an effective and contributing pathway for salt tolerance (Apse et al., 1999; Siao et al., 2020). However, the exact roles and functions of plant vacuolar proton-pumping systems in this process are still unclear.

Na⁺ pumping activity of vacuolar Na⁺/H⁺ antiporters is driven by two tonoplast proton pumps: H⁺-pumping adenosine triphosphatase (H⁺-ATPase) and H⁺-pumping pyrophosphatase (H⁺-PPase; Pasapula et al., 2011; Baisakh et al., 2012). Both act as the primary energy forces pumping protons out of the cytoplasm to generate a pH electric potential gradient, thus facilitating Na⁺ uptake across the vacuole and providing energy fore for salt stress adaption (Gaxiola et al., 2001, 2012; Shavrukov et al., 2013; Graus et al., 2018). Compared with H⁺-ATPase, vacuolar H⁺-pumping pyrophosphatase (VP) contains a simplified structure with only one peptide of nearly 80 kDa, which makes it easier to perform genetic engineering (Segami et al., 2018). Some VPs play positive roles in salt tolerance of plants (Brini et al., 2007; Chen et al., 2007; Pasapula et al., 2011; Graus et al., 2018). For example, the overexpression of Arabidopsis (Arabidopsis thaliana) vacuolar H⁺-PPase (AVP1) enhanced cell differentiation, and auxin transport and accumulation, contributing to higher salt tolerance (Li et al., 2005; Cheng et al., 2018; Chen et al., 2007). However, the relationship between candidate VP and NHX in barley (Hordeum vulgare) has not been studied. In addition, it is imperative to make selection of suitable genes editing VPs in order to improve salt tolerance of crops.

All cereal crops belong to the grass family, including the four largest grain crops, wheat (Triticum spp.), maize (Zea mays), rice (Oryza sativa), and barley. Among them, rice is a diploid crop with a relatively small genome (430 Mb), whereas barley (5.1 Gb) and other diploid Triticeae species (4–16 Gb) contain much larger genomes (Dubcovsky et al., 2001; Yu et al., 2002; Mascher et al., 2017). More importantly, barley shows much higher salt tolerance than rice and other cereal crops, which makes it particularly useful in revealing the salt-tolerant mechanisms and exploiting the relevant genes (Ligaba and Katsuhara, 2010; Fu et al., 2018; Jayakodi et al., 2020). The attempt to enhance rice salt tolerance by introducing barley genes has been proven as a possible approach (Xu et al., 1996; Flowers, 2004; Pan et al., 2018). In rice, there are six genes encoding VPs with relatively conserved relationship (Muto et al., 2011). Recently, one rice VP (OVP1) has been identified to positively regulate root growth and biomass under salt stress (Kim et al., 2020). Compared with rice, barley contains a wider genetic variation (Ellis et al., 2000; Nevo and Chen, 2010; Dai et al., 2018; Morris et al., 2018;Shen et al., 2018), thus possibly being rich in elite alleles of HVPs. A previous study found that the Nax3 locus controls sodium exclusion in shoots and estimated one barley HVP gene (HVP10) as the candidate of this locus, but its exact function has not been reported (Shavrukov et al., 2010, 2013). Therefore, the questions arise, does the difference in the composition and expression of VPs account for the difference of salt tolerance between barley and rice? Could the expression of HVP10 in rice enhance its salt tolerance?

To answer these questions, in this study, we cloned and characterized the gene encoding HVP protein in barley, namely HVP10 (Gene ID: HORVU7Hr1G028910). We performed evolutionary analysis of VPs in the representative plants and explored the expression and function of HVP10 in barley responding to salt stress. Knockdown (RNA interference) and knockout (CRISPR/Cas9 gene editing) barley plants of HVP10 were generated to measure physiological response and ionic changes under salt conditions. Furthermore, we also generated the transgenic rice lines overexpressing HVP10 to evaluate its salt tolerance in terms of plant biomass and grain yield.

Results

VP10s are evolutionarily conserved in green plants

We cloned the full-length DNA of HVP10 from barley cultivar “Golden Promise” (GP), which consists of 4,369 bp with eight exons and seven introns (Supplemental Figure S1A). The full-length coding region of HVP10 was 2,289 bp, and encoded 762 amino acid residues containing a typical H⁺-PPase domain (Supplemental Figure S1A). Phylogenetic analysis showed HVP10 is relatively close to OVP2 in rice, sharing 74%–92% sequence identity with six OVPs, as well as conserved topology structure (Supplemental Figure S1B). However, HVP10 has the unique amino acid residues in the third conserved segment (CS3), N- and C-terminals compared with OVPs (Supplemental Figure S1C), indicating their possible differences in loop folding or function.

Evolutionary analysis suggests VP10 and proteins with the closest identities (defined as the closest one by blast-p) are relatively conserved in 19 angiosperm plant species, sharing an averaged 86% amino acid identity, and averagely 56% identity between the plants and rhodophyta algae (Figure 1A;Supplemental Table S1), indicating that VP10s probably evolved earlier than the origin of land plants. Moreover, VP10s of green plants and rhodophyta (e.g. Cyanidioschyzon merolae) showed distinctive differences in amino acid residues of three conserved segments (CS1, CS2, and CS3; Figure 1B). The results indicated that VP10 has evolved from the order of Ceramiales in rhodophyta algae. Topology prediction and evolutionary conservation analysis also showed that VP10s are evolutionarily conserved in green plants (Figure 1, A–C).

Figure 1.

Evolution analysis of HVP10 proteins. A, Phylogenetic tree of VP10s in 34 species of major lineage of plants and algae. The scale bar shows 0.1 substitutions per site. B, Sequence alignments of three conserved segments in amino acids of VP10s from eight representative species with “*” in (A). C, Predicted 3D structures of eight VPs by the Swiss Model. Hv, Horduem vulgare; At, Arabidopsis thaliana; Pt, Pinus taeda; Af, Azolla filiculoides; Mp, Marchantia polymorpha; Sf, Sphagnum fallax; Vc, Volvox carteri; Cm, Cyanidioschyzon merolae.

HVP10 is a salt-induced and tonoplast-localized protein, and mainly expresses in barley roots

Reverse transcription quantitative polymerase chain reaction (RT-qPCR) analysis showed that HVP10 is mainly expressed in barley roots under normal condition (Figure 2A). Meanwhile, the HVP10 transcripts were significantly higher in root tips than the elongation and hair zones (Figure 2B). After salt treatment of high concentrations (i.e. 400- and 500-mM NaCl), the expression levels of HVP10 in roots were greatly increased by 1.63–1.81 folds compared to the control or low salt concentrations (Figure 2C). Moreover, the expression of HVP10 was significantly induced over salt-exposure time and reached a maximum of 2.21-fold greater expression at 24 h before gradually decreasing to a similar level as control at 72 h (Figure 2D).

Figure 2.

Expression pattern of HVP10 by RT-qPCR analysis. A, Relative expression levels of HVP10 in root, stem, leaf sheath, and leaf blade of cv GP under control condition. B, Relative expression of HVP10 in root tips (0–5 mm), root elongation (5–10 mm), and root hair zones (10–20 mm) of cv GP under control condition. C, Root HVP10 expression under different salt concentrations (100-, 200-, 300-, 400-, and 500-mM NaCl) for 2 h. D, Root HVP10 expression under 200-mM NaCl treatment for 0, 2, 4, 8, 24, 48, 72, and 168 h. Data in (A)–(D) indicate means (n = 4, ±se). Different letters indicate significant difference at P < 0.05 using Tukey’s test.

To determine subcellular localization of HVP10, we performed transient expression of fluorescent proteins in onion (Allium cepa) epidermis cells and found GFP signal of HVP10-sGFP completely overlapped with tonoplast-localized cyan fluorescent protein (CFP) signal (Figure 3A), indicating that HVP10 is localized on the tonoplast. Moreover, transient expression of HVP10-sGFP in Nicotiana benthamiana leaf cells containing nucleus-localized red fluorescence protein (RFP) also showed it was not in the nucleus (Figure 3B). In the leaves of wild-type (WT), HVP10 transcripts examined by in situ PCR were predominately found in phloem vessel and mesophyll cells (Figure 3, C and D). In roots, HVP10 of cross sections were mainly expressed in phloem and xylem parenchyma cells of the stele, and to a lesser extent in cortex cells (Figure 3, F and G); while its abundance of longitudinal sections was detected in root tips and stele sections, in consistence with the results of RT-qPCR analysis (Figures 3, I, J, and 2, B). The transcripts of HVP10 in RNAi lines was obviously decreased in mesophyll, phloem vessel and stele cells, compared to the WT (Figure 3, E, H, K). In short, HVP10 is a tonoplast-localized transporter, which may participate in the regulation of salt stress response.

Figure 3.

Subcellular and tissue specific localization of HVP10. Subcellular localization of HVP10 in onion epidermis cells (A, scale bar = 50 μm) and N. benthamiana leaves (B, scale bar = 20 μm) with tonoplast- and nucleus-localized markers, respectively. CFP, vacuolar-localized marker; RFP, nucleus-localized marker; sGFP, empty GFP as positive control; HVP10-GFP: HVP10 CDS without terminal code with GFP. In situ PCR with HVP10 primers with or without the reverse transcription (RT) step in (C–E) leaf cross section, (F–H) root crosssection, and (I–K) root longitudinal section of WT positive control (WT + RT), Rhvp (Rhvp + RT), and WT negative (WT-RT) control, respectively. Rhvp, L3. Scale bar in (C–H) = 100 μm.

Silencing of HVP10 reduces salt tolerance due to more Na⁺ transport to shoot

RNA interference-mediated gene silencing of HVP10 in GP showed nearly 70% (0.32-fold) decrease in the expression of three RNAi lines (L3, L5, L7), in comparison with the WT (Figure 4, A and B). Salt stress (200-mM NaCl for 3 weeks) showed greater effects on all barley genotypes compared to the WT (Figure 4, C and D). The three RNAi lines showed 52%, 30%, and 11% greater reduction in shoot fresh weight (FW), dry weight (DW), and relative water content (RWC), respectively, in comparison with WT plants under salt stress (Figure 4, E–G). In roots, such a difference was also found in FW and DW, but not in RWC (Figure 4, E–G). However, no significant growth difference was found between the RNAi lines and WT plants under normal condition (Figure 4, E–G).

Figure 4.

Phenotypic analysis of HVP10 RNAi barley and WT seedlings under salt and normal conditions. A, A map of pANDA-HVP10 plasmid used for generating RNAi lines. B, Relative expression of HVP10 in the WT and RNAi lines (n = 4, ±sd). Two-week-old barley seedlings were exposed to normal (C) and 200-mM NaCl (D) hydroponic solutions for 3 weeks. Scale bar = 10 cm. (E) FW, (F) DW, and (G) RWC of root and shoot samples were measured (n = 6, ±SE) under salt and normal conditions, respectively. Different letters represent significant difference at P < 0.05 analyzed by one-way ANOVA with SPSS 20.0. “**” represents a high significance at P < 0.01. RNAi lines of HVP10, L3, L5, L7; Rhvp, the averaged value of the three RNAi lines.

Moreover, after 3 weeks of salt treatment, the three RNAi lines (95.0 ± 7.6 mg·g⁻¹ DW in average) showed significantly higher shoot Na⁺ concentration than WT seedlings (54.3 ± 8.1 mg·g⁻¹ DW), but in roots, the opposite is true, with WT seedlings having higher Na⁺ concentration (Figure 5A). Under salt stress, Na⁺ content in barley seedlings was not significantly different between RNAi and WT seedlings (Figure 5B), while RNAi lines showed a larger Na⁺ transportation rate than WT (Figure 5C). Xylem sap analysis further confirmed the role of HVP10 in root-to-shoot Na⁺ uploading, with RNAi lines (47.8 ± 2.3 mM) showing much higher (30%) Na⁺ concentration in xylem sap than WT seedlings (37.1 ± 1.3 mM) under salt condition (Figure 5D). Meanwhile, the RNAi lines showed lower root K⁺ concentration and whole-plant K⁺ content than the WT seedlings, while such a difference was not detected in shoot K⁺ concentration (Figure 5, E and F). In addition, K⁺ transportation rate of RNAi lines was slightly higher than the WT (Figure 5G), but such a difference might not be caused by xylem K⁺ uploading, as suggested by nearly the same level of K⁺ concentration in xylem sap between them (Figure 5H). The current results indicate that HVP10 probably plays roles in Na⁺ transport and sequestration, and/or K⁺ maintenance in barley roots.

Figure 5.

Na and K changes of HVP10-RNAi and WT barley seedlings under salt stress. A, Na⁺ concentration in roots and shoots (mg·g⁻¹ DW). B, Na⁺ content in total plant (mg·plant⁻¹). C, Na⁺ transportation rate from roots to shoots (%). D, Na⁺ concentration in xylem sap (mM). E, K⁺ concentration in roots and shoots (mg·g⁻¹ DW). F, K⁺ content in total plant (mg·plant⁻¹). G, K⁺ transportation rate from roots to shoots (%). H, K⁺ concentration in xylem sap (mM). Different letters represent significant difference at P < 0.05 analyzed by one-way ANOVA with SPSS 20.0 (n = 6, ±se). **P < 0.01; ns, no significant difference; RNA interference lines of HVP10, L3, L5, L7; Rhvp, averaged values of three RNAi lines.

HVP10 regulates Na⁺ sequestration in barley roots by acting synergistically with HvNHXs

To reveal the effect of HVP10 on Na⁺ transport, we determined the transcript levels of Na⁺/K⁺/H⁺ transporters in roots of HVP10 silencing lines (L3 and L5) and the WT under salt stress. The HKT, NHX, and SOS1 family genes were chosen as candidate genes according to previous studies (Fu et al., 2018; Shen et al., 2020). Correspondingly, the expression of HVP10 increased rapidly under salt stress at 1 d, and then recovered to the control level after 7 d, while its expression level was consistently lower in roots of RNAi lines relative to the WT (Figure 6A).

Figure 6.

Expression analysis of Na⁺ transporters under salt stress. Relative gene expression levels were determined in roots of HVP10 RNAi lines (L3, L5) and WT seedlings when exposed to 200-mM NaCl solution for 0, 1, 7, 14 d. (A) HVP10, (B) HvHKT1;1, (C) HvHKT1;3, (D) HvHKT1;4, (E) HvHKT1;5, (F) HvHKT2;1, (G) HvNHX1, (H) HvNHX2, (I) HvNHX3, (J) HvNHX4, (K) HvNHX5, (L) HvNHX6, (M) HvSOS1. Data in (A–M) indicate means (n = 4, ±se).

For most HKT genes, there was no substantial difference in expression level among all lines before salt treatment, except for HvHKT2;1 (Figure 6, B–F). At 1 d after salt treatment, the expression of HvHKT1;1 and HvHKT1;4 in roots was much more increased in the WT than RNAi lines (Figure 6, B and D). The transcripts of HvHKT1;3 and HvHKT1;5 were higher in the WT than RNAi lines at 7-d and 14-d salt treatment (Figure 6, C and E). On the contrary, while HvHKT2;1 transcription levels were lower in RNAi lines than the WT under control conditions, such a difference was not found under salt stress (Figure 6F).

For NHX genes, four members, namely HvNHX1, HvNHX2, HvNHX4, and HvNHX6, had higher transcription levels in the roots of WT than those of RNAi lines under control (Figure 6, G–L). Interestingly, HvNHX1 and HvNHX4 showed a similar expression pattern to HVP10 with highly significant correlations (r = 0.902 and 0.949, P < 0.01, respectively) in the roots of WT and RNAi lines under salt stress (Figure 6, G and J), suggesting the two genes act synergistically with HVP10. Meanwhile, expressions of HvNHX2 and HvNHX6 in all lines were decreased under salt stress and had no relation with the expression of HVP10 (Figure 6, H and L). RT-qPCR analysis also showed consistently upregulated expression patterns of HvNHX3 and HvNHX5 during salt treatment (Figure 6, I and K). Overall, the transcription levels of HvNHXs were higher in the roots of the WT than RNAi lines (Figure 6, G–L).

Transcription level of HvSOS1 was basically the same for the three RNAi lines before salt treatment (Figure 6M). After salt treatment, its expression was consistently increased by five-fold in two RNAi lines within 14 d. Meanwhile, HvSOS1 in the WT had steady expression during the first 7 d after treatment and showed an 8.8-fold increase compared with RNAi lines at 14 d after salt stress (Figure 6M).

Knockout of HVP10 reduced root H⁺ efflux and K⁺ maintenance

As mentioned above, knockdown of HVP10 significantly reduced K⁺ maintenance in barley (Figure 5, E and F). We then examined the dynamic net H⁺ and K⁺ fluxes in root elongation region of two HVP10 knockout lines (CR1 and CR2; Figure 7, A and B) using noninvasive Microelectronic Ion Flux Estimation (MIFE). Before MIFE analysis, we examined the growth performance of HVP10 knockout lines. They showed more sensitivity to salt stress than WT plants, as reflected by shoot biomass, root Na⁺ concentration, and shoot RWC (Figure 7C; Supplemental Figure S2), and were basically consistent with RNAi lines (Figures 4 and 5A). After 80-mM salt addition, H⁺ efflux dramatically decreased in the two knockout lines compared to the WT (Figure 7, D and E), indicating HVP10 is a proton pump. Moreover, knockout of HVP10 increased K⁺ efflux under salt treatment (Figure 7, F and G), thus resulting in reduced K⁺ concentration in barley tissues.

Figure 7.

Dynamic H⁺ and K⁺ fluxes from barley roots under salt conditions. A, A map of pTaU6-sgRNA-Cas9 plasmid. B, The target design and mutant type of two HVP10-knockout lines. CR1, T insertion of SG1 target; CR2, A insertion of SG2 target. C, Growth performance of two knockout lines and WT seedlings under 200-mM NaCl condition for 3 weeks. Scale bar = 10 cm. Net H⁺ (D) and K⁺ (F) fluxes were measured before and after addition of 80-mM NaCl solution for more than 40 min in root elongation region. Steady-state H⁺ (E) and K⁺ (G) fluxes at the initial, magnitude and final peaks were calculated, respectively. Different letters represent significant difference at P < 0.05 analyzed by one-way ANOVA with SPSS 20.0 (n = 6–8, ±se).

The transgenic rice lines overexpressing HVP10 had better growth and higher grain yield under salt stress

In order to further confirm the function of HVP10 in developing salt tolerance, we generated the transgenic rice lines overexpressing HVP10 (the OE lines: OE1–3) using Nipponbare (NP) as a donor (WT; Figure 8, A–C). In the hydroponic experiment, no significant difference was observed in plant growth between the OE and WT plants under normal condition (Figure 8, D and F; Supplemental Figure S3, A and B). However, after 10 d of 75-mM NaCl treatment, the OE lines showed less growth inhibition, reflected by significantly larger tissue DW and FW, and relative higher water content than the WT (Figure 8, E and F; Supplemental Figure S3, A and B). Under normal condition, there was no significant difference in Na⁺ and K⁺ concentrations of both roots and shoots (Supplemental Figure S3, C and F). After salt treatment, Na⁺ concentration in the shoots of OE lines was only half of that in the WT, while no significant difference was found in root Na⁺ concentration between them (Figure 8G). Meanwhile, the WT had higher Na⁺ transportation rate from roots to shoots than OE lines under salt stress (Figure 8, H and I); however, such a difference was not observed under normal condition (Supplemental Figure S3, D and E). Moreover, Na⁺ concentration in xylem sap was significantly higher in WT plants than the OE lines under salt stress (Figure 8I). In addition, there were no significant differences in root and shoot K⁺ concentrations, root K⁺ content, root-to-shoot transportation rate, xylem sap K⁺ concentration between OE lines and wild plants under salt conditions, except for shoot K⁺ content (Supplemental Figure S3, F–I). Obviously, HVP10-overexpressing transgenic rice lines showed higher salt tolerance than WT plants, which could be attributed to the decreased root-to-shoot Na⁺ translocation via xylem.

Figure 8.

Growth performance of rice HVP10 overexpressing lines and WT seedlings under hydroponic salt conditions. A, A map of pCambia1300: ZmUbi-HVP10 plasmid. B, PCR detection of hygromycin gene in DNA of test lines. M, DNA marker of DL2000 (Takara); OE1-3, the three transgenic rice lines; P, pCambia1300: ZmUbi-HVP10 plasmid. C, Expression levels of HVP10 in the WT and OE lines (n = 6, ±sd), with the comparison of OsGAPDH gene by the method of 2^−△△CT. OE: the averaged value of OE1-3. Phenotype of 3-week-old seedlings under (D) normal and (E) 75-mM NaCl treatment for 10 d. (F) Root and shoot DW under normal and salt conditions. (G) Na⁺ concentrations, (H) Na⁺ content, (I) Na⁺ translocation from roots to shoots, (J) Na⁺ concentration in xylem sap under salt stress (n = 8, ±se). Different letters and “***” indicate significant difference at P < 0.05 and P < 0.001 by SPSS 20.0, respectively. Bar = 5 cm.

The results obtained from the pot soil experiment were consistent with those in the hydroponic experiment (Figure 9). The OE lines had significantly greater shoot weight and more panicles per plant than the WT under salt stress (Figure 9, A–E), while there was no difference in the examined traits between OE lines and the WT under normal condition (Supplemental Figure S4, A–E). Moreover, WT plants had significantly higher shoot Na⁺ concentration than the OE lines under salt stress (Figure 9, F and G), while such a difference was not detected under normal condition (Supplemental Figure S4F). On the other hand, the OE lines had higher seed setting rate (i.e. fewer abortive florets) than the WT under salt stress (Figure 9H), but there was no significant difference in florets per panicle and grain weight between them (Figure 9, I and J). As a result, the OE lines had larger grain yields than the WT under salt stress (Figure 9K). Correspondingly, the relative grain yield of the OE lines was almost two-fold higher than that of the WT (Figure 9L), while there was no significant difference in these yield traits between OE lines and wild plants (Supplemental Figure S4, A–L). It may be concluded that HVP10 alleviates decreased grain yield caused by salt stress due to less reduction of panicles per plant and grains per panicle in rice.

Figure 9.

Phenotype and grain yield components of rice HVP10 overexpressing lines and WT under soil salinity. Phenotype of tested rice at grain-filling (A) and mature (B) stage under soil salinity. C, Shoot biomass, (D) panicles per plants, (E) plant height, (F) shoot ion concentration, (G) grain ion concentration, (H) seed setting rate, (I) floret number per spike, (J) thousand seed weight, (K) grain yield, and (L) relative grain yield under soil salinity (n = 6, ±se). Different letters indicate significant difference at P < 0.05 by Tukey’s test.

Discussion

Evolutional conservation and different expressions of VP10s in plants

The vacuole serves as a storage pool of plant cells, and it also links with toxic Na⁺ compartments passing through tonoplast vesicles in response to salt stress (Shimada et al., 2018; Cui et al., 2019). VP is a key bridge that not only generates H⁺ for vacuolar acidification, but also provides energy force for Na⁺ sequestration into vacuoles under salt stress (Pasapula et al., 2011; Kriegel et al., 2015; Martinoia, 2018; Kim et al., 2020). Some VP members has been proved to be involved in the enhancement of salt tolerance, mostly by use of Arabidopsis AVP1 gene (Li et al., 2005; Pasapula et al., 2011; Schilling et al., 2014), but little research has been done on cereal crops. In this study, we compared the sequence of barley HVP10 and its homologies in the representative plant species and found the evolutionary footmarks of VP10s are consistent with the plant lineage (Figure 1A). As expected, land plants originate from seawater and evolved out of the unicellular green algae with a large central vacuole, indicating the importance of plant vacuole in the evolution of environmental adaption (Morris et al., 2018). Besides, the evolutional conversation of vacuolar HVP10 protein and structure (Figure 1, B and C) mostly provides its biological functions in response to salt stress (Martinoia, 2018; Zhang et al., 2019).

Rice is relatively salt-sensitive among cereal crops, while barley is the most salt tolerant (Horie et al., 2012; Shabala et al., 2016; Jayakodi et al., 2020). In rice, there are six HVP genes, and only two of them (OVP1 and OVP2) show relatively high expression under salt stress (Muto et al., 2011). In addition, the expression of OVP1 in the WT plants was not induced when exposed to 200-mM salt stress, while its overexpression could enhance salt tolerance (Kim et al., 2020). Interestingly, we found HVP10 was mainly expressed in barley root tips, and its expression was greatly induced when plants were exposed to high Na⁺ concentration (such as 400 mM or higher; Figure 2). Although VPs have highly identical sequence for barley and rice (Supplemental Figure S1, B and C), the difference of their expression caused by different habitants and genetic variation might account for the difference in salt tolerance (Shen et al., 2020). In addition, plant VPs have three conserved segments and one of them plays a dominant role in catalytic function (Maeshima, 2000). According to phylogenetic analysis, there was one unique amino acid residue in the third conserved segment of HVP10 and OVPs (Supplemental Figure S1C). Therefore, it may be suggested that the difference in the expression patterns and amino acid residues of HVP10 could partially explain the difference in salt tolerance between barley and rice.

The development of salt tolerance caused by HVP10 is attributed to its acting synergistically with NHXs expression

The capacity of vacuolar Na⁺ compartment adjustment by the expression modulation of VP genes might be a probable approach for developing the cultivars or genetic accessions with enhanced salt tolerance (Flowers et al., 2010; Baisakh et al., 2012; Graus et al., 2018; Segami et al., 2018). However, a mechanistic explanation for this transformative event in the major cereal crops has remained elusive. We performed genetic manipulation of HVP10 in barley to reveal its roles in salt tolerance. The current study showed that knockdown and knockout of HVP10 in barley reduced salt tolerance (Figures 4 and 7;Supplemental Figure S3). Meanwhile, the decreased salt tolerance is related to the modulation of Na⁺ translocation from roots to shoots (Figure 5). In our previous studies, we found barley had less Na⁺ translocation from roots to shoots, resulting in lower shoot Na⁺ concentration, in comparison with rice, which may explain the difference in salt tolerance between barley and rice (Fu et al., 2018, 2019). Less Na⁺ translocation from roots to shoots via xylem, thereby resulting in lower shoot Na⁺ concentration in WT barley, is likely to be controlled by HVP10 gene (Figure 5, A–D), which might be attributed to 13%–21% increase of shoot FWs (Shavrukov et al., 2010, 2013). Therefore, it may be assumed that HVP10 promotes Na⁺ compartmentation into vacuoles of barley root cells under salt stress, thus resulting in the reduction of Na⁺ translocation to shoots and finally improved salt tolerance.

Then what is the molecular mechanism regulated by HVP10 in response to salt stress? We found the transcripts of HVP10 mainly occurred in the leaf vascular bundle and root stele by in situ PCR technique (Figure 3, C–K), indicating that HVP10 is involved in Na⁺ translocation from roots to shoots. Previous studies have illustrated the overexpression of both VP and NHX genes enhanced salt tolerance in Arabidopsis (Brini et al., 2007; Cheng et al., 2018), which is probably related to increased energy force for Na⁺ sequestration into vacuole (Graus et al., 2018). Recently, Kim et al. (2020) reported that the overexpression of OVP1 affected the expressions of NHX1 and other ion transporters in rice exposed to salt stress. Furthermore, we determined the expression levels of Na⁺ transporters (HKTs, NHXs, SOS1) that might respond to salt stress. Obviously, silencing of HVP10 affected the expressions of these Na⁺ transporters (Figure 6). For example, HvHKT1;5 had lower expression in the roots of RNAi lines than the WT after salt treatment (Figure 6E). Currently, the function of HvHKT1;5 has been reported, with two different hypotheses in its regulation on xylem Na⁺ uploading or unloading (Huang et al., 2020; Houston et al., 2020). Based on our functional verification in the previous studies, it may be suggested that HvHKT1;5 modulates Na⁺ xylem uploading, negatively regulating salt tolerance (Huang et al., 2020; Shen et al., 2020). As more Na⁺ was translocated to the shoots of HVP10-RNAi lines (Figure 5, A–D), this Na⁺ imbalance should be regulated by other transporters, such as HvHKT1;1 and HvNHX members, rather than HvHKT1;5. Furthermore, the NHXs members have function in Na⁺ sequestration from cytosol into vacuole, which could be one probable explanation for the involvement of HVP10 in root Na⁺ translocation (Brini et al., 2007; Zhu et al., 2017; Kim et al., 2020). Some NHX members has been functionally reported to accelerate Na⁺ sequestration and K⁺ retention into vacuoles for osmotic and ionic adjustment on salt tolerance (Barragan et al., 2012; Zeng et al., 2017). Additionally, HVP10 could provide energy which is beneficial for Na⁺ sequestration into root vacuole of NHXs. In this study, the decrease of Na⁺ and K⁺ content in the roots of HVP10-RNAi barley plants could be explained by NHX members (Figure 5). As a result, two NHX members (HvNHX1 and HvNHX4) showed the consistent expression with that of HVP10, and they likely participate in HVP10-mediated Na⁺ sequestration into root vacuoles for salinity tolerance (Figure 6, F and I; Supplemental Figure S3). As VP and NHX members were tonoplast-localized, the loss-of-function of HVP10 may decrease H⁺-pumping activity and energy force for salt stress adaption, thus reducing the activity or the expression of HvNHXs genes. Otherwise, knockout of HVP10 reduced the H⁺ gradients, which may explain the hypothesis and lead a massive leakage of K⁺ from the root epidermis after salt exposure (Figures 5, E, F, 7, D and E; Shabala et al., 2010). Thus, the hypothesis is confirmed that VP and NHX act synergistically in vacuolar Na⁺ compartment when plants are exposed to salt stress.

Transgenic rice overexpressing HVP10 show improved salt tolerance

In the present study, three transgenic rice lines expressing HVP10 showed less Na⁺ translocation from roots to shoots and lower shoot Na⁺ concentration under salt stress, finally resulting in improved salt tolerance compared with the WT (Figure 8). Based on the previous studies on HVP10 (Shavrukov et al., 2010, 2013), it may be deduced that less Na⁺ translocation to shoots and lower shoot Na⁺ concentration in the transgenic rice lines is, to a certain extent, attributed to the roles of HVP10, which enhances Na⁺ compartmentation into vacuoles of root cells, probably related to the function of tonoplast-localized NHXs. Furthermore, HVP10-overexpressing transgenic rice lines had similar Na⁺ content in the whole plant as the WT plants (Figure 8, H–J), although the transgenic lines contained greater root Na⁺ content than the wild plants. This indicates that HVP10 affects Na⁺ transportation from roots to shoots, but has no influence on Na⁺ influx or efflux in roots. In a previous study, transgenic rice lines expressing a barley late embryogenesis abundant protein (HVA1) gene showed better tolerance to drought and salinity than its WT (Xu et al., 1996). Similarly, the transgenic rice plants expressing AVP1, a gene in Arabidopsis encoding a vacuolar H⁺-PPase, also showed the improved salt tolerance (Zhao et al., 2006). However, these studies on salt tolerance of transgenic rice were only conducted at seedling stage.

Crop improvement in salt tolerance is not only reflected by a better growth performance under salt stress, and more importantly, should be well evaluated by final yield (Flowers, 2004). Here, we examined salt tolerance of adult plants and yield performance of three transgenic rice lines expressing HVP10 under soil salinity. As a result, adult transgenic rice grown in the NaCl-containing soil showed less growth inhibition, as reflected by tissue biomass and shoot Na⁺ concentration (Figure 9). Moreover, all transgenic rice lines had significantly higher grain yield than the WT under salt stress, mainly due to relatively more panicles and higher seed setting rate. Higher seed setting rate in transgenic rice lines could be attributed to improved nutrient or water availability. It was reported that another important role of VP is involved in inorganic pyrophosphate (PPi) hydrolysis in cell cytoplasm (Ferjani et al., 2011). High cytosolic PPi level could inhibit gluconeogenesis. Thus, the overexpression of HVP10 promotes removing of cytosolic PPi, beneficial for improving heterotrophic growth of transgenic rice under salt stress. Similarly, it was found that a transgenic rice expressing AVP1 was more efficient in transporting sucrose to sink organs relative to its WT (Gaxiola et al., 2012; Paez-Valencia et al., 2013). Interestingly, it was reported that transgenic barley lines expressing AVP had more grains per spike and higher grain weight (Schilling et al., 2014). However, in the current study, we found that there was little difference in grain weight between transgenic rice and WT under salt condition (Figure 9). Little change in grain weight of the WT under salt stress relative to the control is probably related to fewer fertile florets for salt-treated plants. Thus, it may be assumed that HVP10 plays positive roles in salt tolerance and other physiological activities, differing from VPs in Arabidopsis.

In summary, the vacuolar H⁺-pyrophosphatase HVP10 in barley positively regulates salt tolerance, probably through promoting Na⁺ compartmentation into root vacuole. Knockdown of HVP10 decreased salt tolerance and increased Na⁺ accumulation in shoots. HvNHX1 and HvNHX4 act synergistically in the regulation of Na⁺ compartmentation by HVP10 under salt stress. The overexpression of HVP10 in rice improved salt tolerance at both seedlings and adult stages, reducing grain yield loss by increasing panicles per plant and seed setting rate under salt stress. The current study reveals the functions and salt-tolerant mechanism of HVP10 and highlights the potential of its application in improving crop salt tolerance by genetic manipulation.

Materials and methods

Gene cloning and evolutional analysis

HVP10 (HORVU7Hr1G028910) gene was cloned from a barley (Hordeum vulgare) cultivar GP (as the WT of barley). Phylogenetic analysis of VP10s was performed according to Feng et al. (2020). The optimal sequences of VP10 proteins from 34 representative plant species were obtained by the Blast-p of HVP10 protein in online databases of ONEKP (Leebens-Mack et al., 2019) and EnsemblPlants (http://plants.ensembl.org/index.html). Multiple alignments were performed by the MAFFT (https://mafft.cbrc.jp/alignment/software/). Phylogenetic tree was constructed by the FastTree using the maximum-likelihood method. Jalview was used to display protein sequence alignment. The 3D structures of VP proteins were predicted by online SWISS-MODEL (https://swissmodel.expasy.org/). The sequences of rice VPs were obtained according to Muto et al. (2011) and Kim et al. (2020). The primers used in this study were listed in Supplemental Table S2. Accession numbers for amino acid sequences of VP proteins were listed in Supplemental Table S1.

Genetic transformation

Barley genetic transformation was performed on GP according to Harwood (2014) with some modification by Shen et al. (2020). To generate RNAi plasmid, a 367-bp specific fragment of HVP10 transcript (primers RNAi_F/R) was cloned into pDONR-Zeo entry vector and then recombined with pANDA express vector using Gateway BP and LR kits (11789 and 11791; Invitrogen), respectively (Miki and Shimamoto, 2004; Miki et al., 2005). To generate CRISPR/Cas9 plasmid, two single-guided RNA sequences of HVP10 were designed and separately ligated to the pTaU6:Cas9 vector via the restriction enzyme site BsaI (Wang et al., 2019). Immature embryos of GP were infected by Agrobacterium tumefaciens AGL1 with target vectors. Three independent positive HVP10-RNAi (L3, L5, L7) and CRISPR/Cas9 knockout (CR1, CR2) lines were selected from T₂ generations. Only positive lines verified by PCR (primers RNAi_test_F/R) or sequencing were used.

Rice (O. sativa) HVP10-overexpressing lines were obtained using Agrobacterium EHA105-mediated transformation in a cultivar NP (as the WT of rice) by Biogle Company (China). To generate overexpressing construct, the 2,289-bp coding region of HVP10 (primers OE_F/R) was amplified and transferred into pCAMBIA1300 vector driven by maize (Zea mays) ubiquitin promoter using ClonExpress kit (C112-01; Vazyme). Three independent transgenic lines (T₂ generation: OE1, OE2, OE3) were used by PCR verification with the insertion of hygromycin genes (primers HYG_F/R).

Plant growth and salt treatments

The seeds of the tested barley were germinated on moist filter papers in a growth chamber (Shen et al., 2020). For hydroponic experiments, 7-d-old seedlings were transplanted into the pots containing one-fifth-strength Hoagland solution (pH = 6.0). Salt treatments were performed at 14-d-old seedling stage by adding 100-mM NaCl per day to reach final salt concentrations. In addition, the seedlings of knockdown (or knockout) and the WT were exposed to 0 and 200-mM NaCl for 3 weeks. Six biological replicates were used in each treatment. For the determination of spatial gene expression, 3-week-old seedlings were divided into root, stem, leaf sheath and leaf blade, and then roots were cut into 0–5, 5–10, and 10–20 mm from the root apex. For determination of gene expression in response to salt stress, roots were sampled from the plants exposed to 0, 100, 200, 300, 400, and 500-mM NaCl for 2 h, and to 200-mM NaCl for 0, 2, 4, 8, 24, 48 (2 d), 72 (3 d), 168 (7 d), and 336 h (14 d), respectively. The solutions were renewed every 3 d. Seedlings were grown in a growth chamber with 16-h light at 22°C/8-h dark at 18°C of photoperiod, 350 μmol·m⁻²·s⁻¹ of light intensity supplied by fluorescent lamps, and 60% of relative humidity.

Rice seeds were germinated according to Fu et al. (2019). For hydroponic experiment, 4-week-old seedlings were exposed to half-strength Kimura B solutions (pH = 5.6) with the addition of 0 and 75-mM NaCl for 10 d. The solutions were renewed every 3 d. There were eight biological replicates for each treatment. For soil salinity experiment, 4-week-old seedlings of three transgenic lines and the WT were transplanted into 17-L pots filled with paddy soil (from the experimental farm, Zijingang campus, Zhejiang University, China). Four seedlings were planted in each pot. Soil salt treatment was initiated at tillering stage by adding 2 L 0.75% (w/w) NaCl solution to each pot every 7 d and lasted for 4 weeks, and then normal water supply was resumed until grain maturity (Takagi et al., 2015; Wang et al., 2019 ). The plants irrigated with tap water were the control. There were six biological replicates for each treatment. Agronomic characters, including plant height, panicles per plant, florets per panicle, seed setting rate, and grain weight were determined. Shoots and grains were dried for element determination. Rice plants were grown in a glasshouse with 14-h light at 30°C/10-h dark at 22°C, 600 μmol·m⁻²·s⁻¹ of light intensity supplied by fluorescent lamps, and 70% of relative humidity.

Subcellular localization

To construct the transiently expressing plasmid for subcellular localization, the 2,286-bp coding fragment without the stop codon of HVP10 was amplified and fused with the coding region of GFP (green fluorescence protein) into pCAMBIA1300 vector (Shen et al., 2020). The target vectors were delivered into onion (Allium cepa) epidermal cells using biolistic PDS-1000/He particle device (Bio-Rad), with tonoplast-localized CFP or plasma-membrane-localized RFP markers, respectively (Nelson et al., 2007; Huang et al., 2020). Fluorescence signals were observed at 16–18 h after particle bombardment using a LSM880 exciter confocal laser scanning microscope (Zeiss, Germany) and analyzed by Zen 3.1 Blue version (Zeiss, Germany). The experimental setup followed by: excitation wavelength and acquisition bandwidth of laser at 488/490–540 nm for GFP, 561/580–630 nm for RFP, 405/460–500 nm for CFP, respectively, bit depth was eight bit, depth of focus was 1.72 μm, digital offset was 0, digital gain was 1. The vectors were also transformed into EHA105 strain to infect leaves of 4-week-old transgenic N. benthamiana expressing nucleus-localized RFP (Martin et al., 2009). Fluorescence signals were observed after 3 d infection.

Gene expression

The transcript levels of HVP10 and Na⁺ transporter genes were determined by RT-qPCR (Fu et al., 2019). After sampling, total RNA was extracted using MiniBEST Plant RNA Extraction Kit (9769; TaKaRa) and reverse transcribed to cDNA by PrimeScript RT reagent Kit (RR037A; TaKaRa). RT-qPCR was performed using iQ SYBR Green Supermix Kit (1725124; Bio-Rad) on a Roche LightCycle 480 instrument. The comparative 2^−△△Ct method was used for relative gene expression analysis. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) genes from rice and barley were regarded as internal reference genes. All primers of transporter genes were listed in Supplemental Table S1. Four biological replicates and three technical repeats were performed in the experiment.

In situ PCR

In situ PCR analysis of the HVP10 transcripts was performed in roots and leaves of 14-d-old WT and RNAi seedlings (Athman et al., 2014; Shen et al., 2020). Fresh samples were immersed into fixative solution (63% ethanol, 5% acetic acid, and 2% formaldehyde, [v/v] for all) for 4–12 h at 4°C. The samples were embedded vertically into 5% (w/v) agarose B, and cut into 50–70 μm sections on a slicing machine. The sections were treated with DNase I enzyme (Takara), and reversed with HVP10 reverse primer. PCR amplification was performed with digoxigenin (DIG)-labeled dUTP and KOD FX procedure (Toyobo) to generate the products. The sections were incubated with anti-DIG-alkaline phosphate antibody and strained with BM purple substrate solution (Roche). Only water without reverse transcription was used as the negative control. Blue color signals reacted by alkaline phosphate and DIG-labeled antibody were determined by a microscope (Nikon Eclipse Ni).

Element measurement

The FW of roots and shoots samples were determined and then dried in an oven at 70°C for 3 d to obtain DW. RWC were analyzed as RWC(%) = (FW−DW)/FW × 100%. Dried samples were digested in HNO₃ solution by a microwave heating equipment (Multiwave 3000; Antor Paar) at 100–140°C for 3–4 h. Na⁺ and K⁺ concentrations of the digested solutions were measured by an inductively coupled plasma-optical emission spectrometer (ICP-OES; iCAP 6000 series; Thermo Fisher Scientific). The ion concentration, content, and transportation rate were calculated according to Shen et al. (2020). Four biological replicates were used in these measurements.

Xylem sap analysis

Na⁺ and K⁺ concentrations in xylem sap were determined as described by Shen et al. (2020) and Huang et al. (2020). Briefly, for barley, 4-week-old seedlings of HVP10 RNAi lines and WT plants were exposed to 50-mM NaCl for 4 d. For rice, 8-week-old seedlings of HVP10-overexpressing lines and the WT were exposed to 75-mM NaCl for 1 d. Shoots were excised with a razor by 1–2 cm above the root–shoot junction. Xylem sap exudates were collected within 1 h by a micropipette. The concentrations of Na⁺ and K⁺ in xylem sap were diluted with 2% HNO₃ (v/v) for the measurement by an ICP-OES instrument (Optima 6000 series; PerkinElmer Inc).

Non-invasive K⁺ and H⁺ flux measurements

Net K⁺ and H⁺ fluxes in roots of two homozygous mutants (HVP10 knockout lines: CR1 and CR2) were measured using MIFE technique (Chen et al., 2007; Shabala et al., 2010). One hour prior to measurement, roots of 3-d-old barley seedling were immobilized to a slide and placed horizontally in a Perspex container with 10-mL BSM solution. K⁺ and H⁺ ion-selective microelectrodes were calibrated according to Shabala et al. (2010). The electrode tips were positioned 40-μm above the surface of mature root zone (∼10 mm from the root tip). K⁺ and H⁺ fluxes at initial, magnitude, and final peaks were calculated according to Shen et al. (2017). Six biological replicates were used in these measurements.

Statistical analysis

Data analysis of significant difference was performed by Student’s t or Tukey’s test using SPSS 20.0 software (IBM SPSS Statistics). The difference levels at P < 0.05 and P < 0.01 were considered as significant and highly significant, respectively. Correction analysis of gene expressions was performed by Pearson’s method using SPSS software.

Accession numbers

All gene accession numbers for sequence alignment and expression are listed in Supplemental Table S2.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Sequence, phylogenetic and structure analysis of HVP10 and OVPs.

Supplemental Figure S2. Growth performance of HVP10 knockout lines and WT plants in response to salt stress.

Supplemental Figure S3. Phenotypic and ionic analysis of rice overexpressing HVP10 and WT plants at seedling stage.

Supplemental Figure S4. Phenotypic and ionic analysis of rice overexpressing HVP10 and WT plants at reproductive stage under normal condition.

Supplemental Table S1. The information of VP proteins used for the phylogenetic analysis.

Supplemental Table S2. The primers used in this study.

 

L.F., Q.S., D.W., and G.Z. designed the work. L.F., X.Z., Y.X., L.K., S.C., and Q.S. performed experiments and data analysis. L.F. and Q.S. drafted the manuscript. D.W. and G.Z. revised the manuscript.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Qiufang Shen (shenqf@zju.edu.cn).