The H+-Translocating Inorganic Pyrophosphatase From Arabidopsis thaliana Is More Sensitive to Sodium Than Its Na+-Translocating Counterpart From Methanosarcina mazei

The H + -Translocating Inorganic Pyrophosphatase From Arabidopsis thaliana Is More Sensitive to Sodium Than Its Na + - Translocating Counterpart From Methanosarcina mazei Jose ´R. Pe ´rez-Castiñeira*and Aurelio Serrano Instituto de Bioquı ´mica Vegetal y Fotosı ´ntesis, Universidad de Sevilla-CSIC, Sevilla, Spain Overexpression of membrane-bound K + -dependent H + -translocating inorganic pyrophosphatases (H + -PPases) from higher plants has been widely used to alleviate the sensitivity toward NaCl in these organisms, a strategy that had been previously tested in Saccharomyces cerevisiae. On the other hand, H + -PPases have been reported to functionally complement the yeast cytosolic soluble pyrophosphatase (IPP1). Here, the efficiency of the K + -dependent Na + -PPase from the archaeon Methanosarcina mazei (MVP) to functionally complement IPP1 has been compared to that of its H + -pumping counterpart from Arabidopsis thaliana (AVP1). Both membrane-bound integral PPases (mPPases) supported yeast growth equally well under normal conditions, however, cells expressing MVP grew significantly better than those expressing AVP1 under salt stress. The subcellular distribution of the heterologously-expressed mPPases was crucial in order to observe the phenotypes associated with the complementation. In vitro studies showed that the PPase activity of MVP was less sensitive to Na + than that of AVP1. Consistently, when yeast cells expressing MVP were grown in the presence of NaCl only a marginal increase in their internal PPi levels was observed with respect to control cells. By contrast, yeast cells that expressed AVP1 had significantly higher levels of this metabolite under the same conditions. The H + -pumping activity of AVP1 was also markedly inhibited by Na + . Our results suggest that mPPases primarily act by hydrolysing the PPi generated in the cytosol when expressed in yeast, and that AVP1 is more susceptible to Na + inhibition than MVP both in vivo and in vitro. Based on this experimental evidence, we propose Na + - PPases as biotechnological tools to generate salt-tolerant plants. Keywords: Arabidopsis thaliana, heterologous expression, membrane-bound ion (H + -orNa + )-pumping inorganic pyrophosphatases, Methanosarcina mazei,Saccharomyces cerevisiae, salt stress, soluble inorganic pyrophosphatase Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12401 Edited by: Gabino Rı ´os, Instituto Valenciano de Investigaciones Agrarias, Spain Reviewed by: Ali Ferjani, Tokyo Gakugei University, Japan Roberto Adrian Gaxiola, Arizona State University, United States Anssi Malinen, University of Turku, Finland *Correspondence: Jose ´R. Pe ´rez-Castiñeira [email protected] Specialty section: This article was submitted to Plant Metabolism and Chemodiversity, a section of the journal Frontiers in Plant Science Received: 17 March 2020 Accepted: 28 July 2020 Published: 12 August 2020 Citation: Pe ´rez-Castiñeira JR and Serrano A (2020) The H + -Translocating Inorganic Pyrophosphatase From Arabidopsis thaliana Is More Sensitive to Sodium Than Its Na + -Translocating Counterpart From Methanosarcina mazei. Front. Plant Sci. 11:1240. doi: 10.3389/fpls.2020.01240 ORIGINAL RESEARCH published: 12 August 2020 doi: 10.3389/fpls.2020.01240 INTRODUCTION The membrane-bound K + -dependent H + -translocating inorganic pyrophosphatase (H + -PPase), encoded by AVP1, has been reported to be the protein responsible for removing most of the inorganic pyrophosphate (PPi) generated in the cytosol of Arabidopsis thaliana cells (Ferjani et al., 2011;Kriegel et al., 2015; Segami et al., 2018b). The removal of PPi, a by-product of many biosynthetic reactions, is essential in all living cells because its accumulation causes the collapse of anabolism, eventually leading to cell death (Lahti, 1983;Serrano-Bueno et al., 2013). AVP1 couples the energy released by the hydrolysis of PPi to the translocation of protons into the vacuole, thus acting as a second proton pump in the plant tonoplast along with the V-type H + - ATPase (Sarafian et al., 1992;Graus et al., 2018). By contrast, in the cytosol of Saccharomyces cerevisiae cells only one protein seems to have the capacity to hydrolyse PPi: the extremely efficient soluble inorganic pyrophosphatase (sPPase), encoded by IPP1 (Kolakowski et al., 1988). In a seminal paper published in 2001, Gaxiola and co-workers reported that overexpression of AVP1 alleviated the sensitivity of A. thaliana to NaCl (Gaxiola et al., 2001). Thereafter, several groups obtained other plants with higher tolerance to salinity and other stresses based on this approach (Park et al., 2005;Gao et al., 2006;Li et al., 2008;Lv et al., 2008;Pasapula et al., 2011; Gaxiola et al., 2012;Schilling et al., 2014;Yang et al., 2015;Lv et al., 2016;Ahire et al., 2018). This strategy had been originally tested in S. cerevisiae (Gaxiola et al., 1999), thus demonstrating the suitability of this yeast to study and test solutions to salt stress in plants. Although AVP1 and its orthologs in higher plants have been used as tools to develop salt-tolerant plants, membrane-bound PPases with different biochemical properties have been characterized from other organisms. A K + -independent H + - PPase was identified in 1966 in the photosynthetic alphaproteobacterium Rhodospirillum rubrum (RVP) (Baltscheffsky and Baltscheffsky, 1995). The corresponding gene was cloned and sequenced in the 1990s (Baltscheffsky et al., 1998) and expressed both in E. coli (Schultz and Baltscheffsky, 2004) and S. cerevisiae (Drake et al., 2010). Two ortologs of RVP have been identified in A. thaliana (AVP2 and AVP3) and reported to be located in the Golgi membranes, however, these proteins seem to have much lower expression levels than AVP1 and their physiological importance remains to be established (Drozdowicz et al., 2000;Ferjani et al., 2011). An important breakthrough in the field of PPases was the discovery of K + -dependent Na + -translocating mPPases (Na + -PPases) in eubacteria and archaea (Malinen et al., 2007;Baykov et al., 2013). Nowadays, genome projects of a wide variety of organisms as well as metagenomics and biochemical studies are significantly broadening the Na + -PPases field (Luoto et al., 2015; Serrano A., unpublished). Our group has previously reported the generation of several S. cerevisiae mutant strains whose IPP1 genes are under the control of the glucose-repressible promoter of the GAL1 gene. These strains cannot grow on glucose, thus demonstrating the essentiality of the sPPase encoded by IPP1 (Serrano-Bueno et al., 2013), however, the capacity to grow on this carbon source is recovered when they are transformed with plasmids bearing genes coding for H + -PPases under the control of constitutive promoters (Perez-Castiñeira et al., 2002;Drake et al., 2010;Perez-Castiñeira et al., 2011;Hernandez et al., 2015). This system avoids the competition for the substrate between the heterologous H + -PPase and the yeast cytosolic sPPase in vivo, a situation that resembles that of a plant cell cytosol. In this communication, the capacity of two different mPPases to complement IPP1 under salt stress has been studied. In order to accomplish this task, several yeast mutants with different sensitivities to NaCl and conditional expression of IPP1 were generated. These mutants were transformed with plasmids bearing genes coding for AVP1 or the K + -dependent Na + - PPase from the salt-tolerant archaeon Methanosarcina mazei (MVP) (Malinen et al., 2007). MVP was more efficient than AVP1 at supporting growth of the different yeast mutants in the presence of NaCl. The addition of N-terminal signal peptides that modify the subcellular distribution of the mPPases (Drake et al., 2010) was essential for efficient complementation of IPP1. In vitro assays showed that AVP1 is significantly more sensitive to Na + than MVP. Our results suggest that mPPases may be direct targets of Na + in vivo, a situation that might lead to a toxic accumulation of PPi in the cytosol of plant cells subjected to high concentrations of salt. Based on this evidence, we propose the heterologous expression of chimeric versions of Na + -PPases with appropriate N-terminal signal peptides as a more efficient way to improve salt-tolerance in plants than the current approach, based on the use of untargeted H + -PPases. In connection with this, in silico analyses of currently available genomes suggest that Na + -PPases, originally described in bacteria and archaea (Malinen et al., 2007;Luoto et al., 2011;Luoto et al., 2015), are present in other organisms, including some of the photosynthetic lineage like certain marine microalgal groups (Prasinophyceae and diatoms). Our findings suggest new biotechnological applications for microalgae in order to improve plants of agronomical interest. MATERIALS AND METHODS Yeast Strains The strains used in this work are shown in Table 1. YPC3 mutant strain was generated from S. cerevisiae haploid strain W303-1A (MATa, ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1,ura3-1)by the single-step transplacement procedure as previously described (Drake et al., 2010). This mutant has IPP1, the gene that codes for the cytosolic soluble inorganic PPase, under the control of the Abbreviations: AVP1, K + -dependent H + -translocating inorganic pyrophosphatase from Arabidopsis thaliana; mPPase, membrane-bound integral ion (H + -orNa + ) -pumping inorganic pyrophosphatase; MVP, K + -dependent Na + - translocating inorganic pyrophosphatase from Methanosarcina mazei; PPase, Inorganic pyrophosphatase; PPi, Inorganic pyrophosphate; sPPase, Soluble inorganic pyrophosphatase. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12402 yeast galactokinase gene (GAL1) promoter (Drake et al., 2010). YPC5 and YPC6 mutant strains were obtained from S. cerevisiae strains G19 and B31, respectively, by following the same strategy except that the TRP7 cassette was used as a selection marker instead of HIS3. Mutant G19 had been derived from strain W303-1B by disrupting its ENA1-4 genes with the yeast HIS3 cassette (Quintero et al., 1996), whereas B31 was obtained from G19 by disrupting NHA1 with the LEU2 cassette (Bañuelos et al., 1998). Both mutants were generously provided by Professor Alonso Rodrı guez Navarro (Universidad Politecnica de Madrid, Spain). Plasmids Construction and Yeast Transformation YPC3, YPC5, and YPC6 cells were transformed with the plasmids shown in Table 2 as previously described (Schiestl and Gietz, 1989). All plasmids are 2 micron-based derivatives of the URA3-containing E. coli/S. cerevisiae shuttle plasmid pRS699, that bears the yeast PMA1 promoter for constitutive expression of inserts (Serrano and Villalba, 1995). The latter were made by in-frame fusion of the N-terminal signal peptides of S. cerevisiae SUC2 invertase (Inv) or the K + -dependent H + -PPase from Trypanosoma cruzi (Tc), the sequence coding for the yeastenhanced green fluorescent protein (yEGFP) and the coding sequence of the K + -dependent H + -PPase from A. thaliana (AVP1) (Drake et al., 2010). An identical approach was followed with the coding sequence of the K + -dependent Na + - PPase from M. mazei (MVP) (Malinen et al., 2007), generously provided by Professor Reijo Lahti, University of Turku (Finland). For the transformation procedure, YPC3, YPC5 and YPC6 cells were initially grown at 30°C in YPGal liquid medium [1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) galactose], while transformants were selected on 2% agar plates made in galactosecontaining synthetic medium [0,17% (w/v) yeast nitrogen base without amino acids and ammonium sulphate, 2% (w/v) galactose, 50 mM TRIS-HCl pH 6, and a mixture of nucleotides and amino acids described in (Treco and Lundblad, 2001)], devoid of histidine and uracil (YPC3); histidine, tryptophan, and uracil (YPC5); or histidine, tryptophan, leucine, and uracil (YPC6). Colonies appeared after incubating the plates for 3-4 days at 30°C. Phenotype Complementation Tests Complementation studies were performed by inoculating 2 ml of galactose-containing selective medium with transformed cells from the plates and growing overnight at 30°C on an orbital shaker (150 r.p.m.). The following day, 10 ml of these cultures were used to inoculate 2 ml of YPD medium [1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) glucose] (1:200 dilution) and allowed to grow overnight as described above. This treatment is necessary to bring down the PPase activity associated with the chromosome-encoded IPP1, controlled by the glucose-repressible GAL1 promoter (Perez-Castiñeira et al., 2002). After overnight growth on glucose, ten-fold serial dilutions of the cultures were made in sterile water and 5 ml drops of each dilution were spotted onto YPD agar plates containing different NaCl concentrations and 50 mM TRISMES adjusted to pH 6. Plates were incubated at 30°C for several days. Isolation of Yeast Microsomal Membranes Preparations of membrane fractions were obtained from YPC6 cells transformed with the different plasmids by a modification of a method previously described (Serrano and Villalba, 1995): yeast colonies were collected from a plate and grown up to stationary phase in galactose-containing selective liquid medium; then, 400 ml of YPD were inoculated with 2 ml of stationary cultures. After overnight growth, cells were collected by centrifugation at 700 x gfor 10 min, washed thoroughly with water, resuspended in 5 ml of ice-cold buffer A (25 mM Tris- HCl, pH 8, 10% glycerol, 4 mM b-mercaptoethanol, 2 mM DTT, 1 mM benzamidine, 2 mM ϵ-aminocaproic acid, 1 mM PMSF), and homogenized by vigorous vortexing for five one-min bursts in the presence of glass beads (0.5 mm Ø) with one-min intervals on ice. From this point on, the whole procedure was carried out TABLE 2 | Plasmids used for yeast transformation. Plasmid Insertion Reference pIPP1 Sequence coding for the cytosolic soluble PPase from S. cerevisiae (Perez-Castiñeira et al., 2002) pInvGFPAVP1 Sequence coding for the putative N- terminal signal peptide of Suc2p followed by those of yEGFP and the H + -PPase from A. thaliana (AVP1) (Drake et al., 2010) pInvGFPMVP Sequence coding for the putative N- terminal signal peptide of Suc2p followed by those of yEGFP and the Na + -PPase from M. mazei (MVP) This study pTcGFPAVP1 Sequence coding for the putative N- terminal signal peptide of the K + - dependent H + -PPase from Trypanosoma cruzi followed by those of yEGFP and AVP1 (Drake et al., 2010) pTcGFPMVP Sequence coding for the putative N- terminal signal peptide of the K + - dependent H + -PPase from T. cruzi followed by those of yEGFP and MVP This study Original and chimeric nucleotide sequences described in the table were inserted between yeast PMA1 promoter and terminator of the URA3-containing plasmid pRS699 (Serrano and Villalba, 1995). TABLE 1 | Yeast strains used in this work. Name Relevant genotype Reference W303- 1A MATa, ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 ura3-1 (Thomas and Rothstein, 1989) G19 MATa, ade2-1 can1-100 his3-11,15 leu2- 3,112 trp1-1 ura3-1 ena1D::HIS3::ena4D (Quintero et al., 1996) B31 MATa, ade2-1 can1-100 his3-11,15 leu2- 3,112 trp1-1 ura3-1 ena1D::HIS3::ena4D nha1D::LEU2 (Bañuelos et al., 1998) YPC3 W303-1A ipp1 UAS -ipp1 TATA ::HIS3-GAL1 UAS - GAL1 TATA -IPP1 (Drake et al., 2010) YPC5 G19 ipp1 UAS -ipp1 TATA ::TRP1-GAL1 UAS - GAL1 TATA -IPP1 This study YPC6 B31 ipp1 UAS -ipp1 TATA ::TRP1-GAL1 UAS - GAL1 TATA -IPP1 This study Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12403 at 0°C (except centrifugations, that were performed at 4°C). The homogenate was diluted up to 25 ml with buffer B (10 mM Tris- HCl, pH 7.6, 10% (w/v) glycerol, 2 mM DTT) and centrifuged for 10 min at 700 x gto remove beads and debris. The resulting supernatant was centrifuged for 20 min at 20,000 x gand the pellet [the crude mitochondrial fraction, that includes mitochondria and plasma membrane vesicles among others (Serrano and Villalba, 1995)] was discarded for PPi hydrolysis assays due to the potential interference of the mitochondrial PPase (Lundin et al., 1991). The supernatant of this step was centrifuged for 30 min at 120,000 x g. The pellet thus obtained, the microsomal fraction, was taken up, homogenized in stripping buffer [60 mM Tris/HCl, pH 8, 12% (w/v) glycerol, 0.72 M KCl, and 1.2 mM CaCl 2 (Lenoir et al., 2002)] and centrifuged for 20 min at 120,000 x g. Finally, the stripped microsomal fraction was washed with buffer B and centrifuged (20 min, 120,000 x g). The final pellet was resuspended and homogenized in 0.5–1mlof buffer B. Microsomes were directly used for activity assays or divided into aliquots and kept at -80°C until use. Pyrophosphatase Activity and ACMA Quenching Assays, Western Blot, and Total Protein Estimation Pyrophosphatase (PPase) activity was assayed at 30°C by incubation of protein preparations with Mg 2 PPi at pH 7.2 with the addition of KCl, NaCl and/or NaF when indicated, and colorimetrically measuring the released ortophosphate as previously described (Rathbun and Betlach, 1969). Mg 2 PPi was obtained by adding fixed volumes of stock solutions of appropriate concentrations of MgCl 2 and Na 4 PPi to the assay mixes. Reactions were started by the addition of the latter and stopped with a mixture of trichloroacetic acid, sodium acetate, acetic acid, and formaldehyde (final concentrations: 4% (w/v), 0,5 M, 0.5 M, and 1% (w/v), respectively) (Schultz and Baltscheffsky, 2004). Control yeast microsomal membranes devoid of PPase activity and containing the same amount of total protein were used as blanks in every assay. ACMA quenching assays and Western blots were performed as described elsewhere (Perez-Castiñeira and Apps, 1990;Valverde et al., 1997). A polyclonal antibody raised against the purified Na + -PPasefromthemarinebacterium Thermotoga maritima (TVP) (Lopez-Marquesetal.,2005)anda commercial polyclonal rabbit antibody against yeast IPP1 protein (AP21326F_N, OriGene EU, Germany) were used for immunodetection of mPPases and sPPases, respectively, after SDSPAGE and semi-dry electroblotting transfer. Antibody against TVP was affinity-purified as previously described (Pringle et al., 1991). Protein concentration was estimated using a Coomassie-blue dye binding-based assay from Bio-Rad (München, Germany) following manufacturer’s instructions and using ovalbumin as a standard. Analysis of Enzyme Kinetics Kinetic data obtained for AVP1 and MVP were fitted to the Hill equation and to a mathematical expression based on the concerted MWC allosteric model (see Results) by weighted non-linear least squares (nls) regression using RStudio (version 1.1.456 – © 2009– 2018 RStudio, Inc.), downloaded from https://cran.cnr.berkeley.edu. This version of RStudio is an integrated development environment (IDE) for the programming language R, version 3.5.1 (2018-07-02), “Feather Spray”( © 2018 The R Foundation for Statistical Computing), and was run under Ubuntu 18.04.1 LTS (“Bionic Beaver”) and Debian GNU/Linux 10 (“buster”). Averages of three to five independent experiments were used for regression calculations and each experimental data point was weighted as 1/variance. Determination of Internal PPi Levels in Yeast Cells 100 ml cultures of YPC6 cells transformed with the plasmids shown in Table 2 and treated to minimize the chromosomeencoded IPP1 (see above) were grown in glucose-containing medium (YPD adjusted to pH 6) with agitation in the presence or absence of 100 mM NaCl. The initial optical densities at 660 nm (OD 660 ) of the cultures where individually adjusted so that all of them had an OD 660 of around 0.5 (roughly 10 9 cells) after 8– 9 h. Then, cells were collected by centrifugation, washed with icecold deionised water and broken as described for isolation of yeast membranes, except that a 4% (v/v) perchloric acid aqueous solution was used instead of buffer A. Beads, debris, and denatured proteins were removed by centrifugation (20 min, 20,000 × g,4°C) and PPi concentrations were determined colorimetrically in the supernatants after neutralization with KOH, as previously described (Heinonen et al., 1981). Molecular Phylogenetic Analysis of mPPases The amino acid sequences of selected H + - and Na + -translocating mPPases orthologs of bacteria, archaea, parasitic protists, microalgae and plants were aligned with the CLUSTAL X software tool (Larkin et al., 2007). Phylogenetic trees were constructed with the evolutionary distances (neighbor joining) and maximum parsimony methods using the SeaView v5.2 software (Gouy et al., 2010). RESULTS MVP Functionally Complements Yeast IPP1 More Efficiently Than AVP1 Under Salt Stress Three yeast strains with increasing sensitivities to NaCl were generated (Table 1) and transformed with E. coli/S. cerevisiae shuttle plasmids bearing the genes coding for chimeric versions of AVP1 and MVP under the control of the yeast PMA1 promoter (Table 2).Dropteststocheckforgrowthof transformants onto YPD plates at pH 6 containing different concentrations of NaCl were carried out. We have previously shown that addition of certain sequences in the N-termini of the coding sequences of some mPPases altered the subcellular distribution of the resulting chimeric peptide and improved their expression levels. Among these sequences were those encoding the N-terminal signal peptides from the S. cerevisiae invertase SUC2 or the K + -dependent H + - Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12404 PPase from Trypanosoma cruzi (TcVP), which targeted the chimeric mPPases mainly to the plasma membrane or the internal membrane systems, respectively. Fusion with the yEGFP sequence further enhanced mPPase expression levels, especially in the case of AVP1 (Drake et al., 2010;Perez- Castiñeira et al., 2011). This protein engineering strategy was applied to the coding sequence of the K + -dependent Na + -PPase from M. mazei with similar results (Supplementary Figure S1). Initial studies of growth in the presence of NaCl were done with YPC3, a mutant derived from the strain W303-1A, which is not especially sensitive to Na + and is able to grow in the presence of NaCl concentrations in the molar range (Rios et al., 1997). Figure 1 shows drop tests comparing growth obtained with YPC3 cells transformed with plasmids bearing different PPase genes and with the parental strain W303-1A. Cells expressing MVP grew significantly better than those expressing AVP1 in the presence of NaCl, however, no difference was observed in the absence of salt. Moreover, the chimera with the N-terminal signal peptide of yeast invertase SUC2 (InvGFPMVP) performed better than that with the N-terminal signal peptide of TcVP (TcGFPMVP). InvGFPAVP1 also supported growth more efficiently than TcGFPAVP1, albeit at lower NaCl concentrations than the MVP chimeras. None of the cells expressing mPPases grew better than those expressing their soluble counterpart from yeast (IPP1) under any of the conditions tested. The use of YPC3 mutant and its parental strain forced the addition of very high concentrations of NaCl to observe the effects exerted by this salt (above 1 M). At such high concentrations, NaCl poses two types of stress to yeast cells, namely, osmotic stress and sodium toxicity. Although in glucosecontaining medium the latter contributes more to growth inhibition than the former (Rios et al., 1997), we were interested in the specific effects of Na + . It was then decided to use yeast strains that were sensitive to lower NaCl concentrations in order to minimize osmotic stress, therefore, we generated mutant strains YPC5 and YPC6. These strains have conditional expression of their respective IPP1 genes in genetic backgrounds associated with higher sensitivities to NaCl (knock-outs of genes encoding major sodium pumps and/or transporters; Table 1). YPC5 and YPC6 cells were transformed with the plasmids described in Table 2 and the results obtained are shown in Figures 2A, B, respectively. An identical pattern to that obtained in the case of transformed YPC3 cells was observed, the only difference being the concentrations of NaCl required to inhibit growth. YPC5 cells stopped growing at around 0.5 M NaCl, FIGURE 1 | Drop tests for yeast mutant YPC3 cells and parental strain W303 grown in the presence different concentrations of NaCl. Cells were transformed with plasmids shown in and grown as described in the main text. Serial dilutions of the cultures were made in sterile water and spotted on to YPD plates adjusted to pH 6 and containing the indicated concentrations of NaCl. Growth was recorded after 4 days at 30 °C. Negative controls are YPC3 cells transformed with empty plasmid. A B FIGURE 2 | Drop tests for sodium-sensitive yeast strains YPC5 (A), YPC6 (B), and their respective parental strains G19 and B31. Cells were transformed with plasmids pIPP1, pInvGFPAVP1 and pInvGFPMVP as described in the main text. The rest of conditions as described in . Negative controls are YPC5 (A) and YPC6 (B) cells transformed with empty plasmid. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12405 whereas 150 mM NaCl was sufficient to obtain a similar effect in YPC6 cells. As far as mPPases are concerned, the chimera InvGFPMVP supported growth more efficiently than InvGFPAVP1, and both performed better than TcGFPMVP and TcGFPAVP1, respectively (not shown for clarity), as in the case of YPC3. YPC5 and YPC6 cells expressing IPP1 always performed better than those expressing chimeric mPPases. YPC6 cells transformed withplasmidspInvGFPMVPand pInvGFPAVP1 were selected for subsequent studies. The results obtained with YPC3, YPC5 and YPC6 showed that the capacity of AVP1 to complement IPP1 was severely hampered by the presence of NaCl in the medium, whereas that of MVP was less sensitive to this circumstance. The fact that identical results were obtained with NaCl concentrations ranging from 1.5 M down to 150 mM suggests that the effects exerted by NaCl are actually due to Na + toxicity rather than to osmotic stress. This idea was supported by drop tests performed with transformed YPC5 and YPC6 cells in YPD plates supplemented with 1 M and 0.3 M sorbitol, that showed no difference with respect to those performed in YPD alone (not shown). Moreover, the addition of KCl instead of NaCl in the culture medium did not inhibit growth in any case (not shown), demonstrating that the phenotypes observed in the presence of NaCl were due to the toxic effects of Na + . Membrane-Associated PPase Activity and Expression Levels of the Heterologously- Expressed mPPases Do Not Change When Yeast Cells Are Subjected to Salt Stress In order to check whether the phenotypes observed could be due to differences in protein expression and/or PPase activity, different studies were done with microsomal membrane preparations and soluble extracts obtained from transformed YPC6 cells. The upper panel of Figure 3A shows fluoride-insensitive PPi hydrolysis activity associated with microsomal membranes. As expected, PPase activity could only be detected in preparations obtained from cells transformed with plasmids bearing genes coding for mPPases.Theseactivities increased6to7-foldinthepresenceof100 mMKCl invitro,afeatureof bothAVP1 andMVP(Kimet al., 1994; Malinen et al., 2007). The presence of NaCl during growth did not alter in vitro activity levels. No membrane-bound PPase activity was detected in YPC6 control cells (transformed with empty plasmid pRS699), whereas in membrane preparations of cells transformed with plasmid pIPP1 some residual activity was detected in the absence of NaF (not shown). This indicates that some molecules of the yeast cytosolic sPPase remain attached to these membranes even after treatment with stripping buffer, as sensitivity to fluoride is a characteristic of this enzyme (see below). Immunodetection carried out with microsomal membranes using an affinity-purified polyclonal antibody against the Na + - PPase from T. maritima (TVP) confirmed the results obtained with PPase assays, that is, salt stress does not significantly alter expression levels and/or protein processing patterns in any case (Figure3A,lowerpanel).Thelatterareduetopartialdegradationof the heterologously-expressed mPPases by yeast vacuolar proteases in vivo and it is a characteristic of our expression system (Perez- Castiñeira et al., 2011). A parallel set of experiments were carried out with soluble protein extracts of YPC6 cells transformed with the same plasmids as above. Fluoride-sensitive sPPase activity could be measured only in extracts from cells transformed with plasmid pIPP1, consistently, a commercial antibody against IPP1 recognized a band of the expected size (ca. 37 kDa) in this sample (Figure 3B). Similar results were obtained with cells of the parental strain B31, albeit with lowerlevelsofactivityandIPP1polypeptide (not shown). Growing cells in the presence of NaCl did not significantly alter IPP1 expression levels. A B FIGURE 3 | Levels of PPi hydrolysis activity and immunodetection of mPPases in microsomal membranes (A) and sPPases in soluble extracts (B) of YPC6 mutant cells transformed with plasmids shown in Table 2. Cells were cultured in liquid YPD up to an OD 660nm of 0.3, as described in main text, then, each culture was divided into two aliquots, one of which was subjected to a treatment with 100 mM NaCl by adding the appropriate volume of a 5 M stock solution of NaCl, whereas the same volume of deionized water was added to the other aliquot. All cultures were grown up to an OD 660nm of 0.8-1, cells were collected and microsomal membranes and soluble crude extracts were obtained as described in Materials and Methods. An affinity purified polyclonal antibody against Thermotoga maritima mPPase (A) and a commercial antibody against IPP1 (B) were used for immunodetection. 70 mg of total protein were loaded per lane in all cases. Positive (+) signs denote samples obtained from cells subjected to treatment with 100 mM NaCl. Negative (-) signs denote samples obtained from control cultures with no added salt. “n. d.”: not detected. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12406 Membrane-Bound PPases Exhibit Different Sensitivities to NaCl In Vitro The effect exerted by Na + on the PPase activities of AVP1 and MVP in vitro was studied by performing assays in the presence of increasing concentrations (0, 50, and 100 mM) of added NaCl. It must be borne in mind that PPi is added as Na 4 PPi (see Materials and Methods), therefore, basal concentrations of Na + four times higher than those of PPi are always present in the assays. Both AVP1 and MVP depend on K + for full activity (Kim et al., 1994; Malinen et al., 2007), therefore, 100 mM KCl was always present in the assay mix to provide the concentration of K + usually maintained in most plant cells (Nieves-Cordones et al., 2016). Mg 2 PPi was used as substrate, as previously reported for diverse mPPases (Leigh et al., 1992;Gordon-Weeks et al., 1996;Tsai et al., 2014). Results obtained are shown in Figure 4. AVP1 seemed to be quite sensitive to NaCl, showing a decrease in V m of 20% and 50% in the presence of 50 and 100 mM NaCl, respectively; by contrast, MVP showed virtually no inhibition at 50 mM NaCl and only a 20% decrease in this parameter when the salt concentration was raised up to 100 mM. The experimental data were fitted with the empirical Hill equation (Hill, 1910) by weighted non-linear least squares regression using RStudio, yielding values of the Hill coefficient n H of 2–3 for AVP1 and 3.5–4.5 for MVP. This indicates a significant degree of positive cooperativity for both proteins, MVP showing a higher deviation from the Michaelis-Menten behaviour (that is, higher sigmoidicity) than AVP1. The V m value predicted by the Hill equation for MVP decreased somewhat with increasing concentrations of NaCl, however, neither n H nor the K 0.5 values changed significantly (Table 3A). In order to model the kinetic properties of AVP1 and MVP and relate them to the structural information available for these proteins, a simplified equation was derived from the MWC concerted allosteric model (Monod et al., 1965). The following assumptions were made to obtain the equation: 1. The ratio v/V m is proportional to the saturation function (Y  ) of the allosteric model 2. AVP1 and MVP can exist in two states, R and T, in thermodynamic equilibrium driven by the allosteric constant L= [T]/[R]. R is assumed to be active and T inactive. 3. The substrate (Mg 2 PPi) binds only to the R state, thus defining a single dissociation constant K R . 4. The number of catalytic binding sites for substrate is 2 for both mPPases functional oligomers. This assumption is based on the structural information currently available, which A B FIGURE 4 | Graphical representation of PPase activity versus substrate (Mg 2 PPi) concentration for AVP1 (A) and MVP (B) in the presence of 0 mM (●), 50 mM (▲), and 100 mM (■) NaCl. Data points correspond to the averages ± standard errors (SE) of 4 independent experiments. The substrate, Mg 2 PPi, was obtained by mixing stock solutions of Na 4 PPi and MgCl 2 , as described in Materials and Methods, consequently, basal concentrations of Na + four times higher than those of PPi are present in the assays. Mg 2 PPi concentrations were estimated considering the dissociation constants for individual PPi complexes with H + ,Na + ,K + , and Mg 2+ ,as previously described (Baykov et al., 1993). Graph lines were calculated with Equation 1 using RStudio, as described in Materials and Methods. TABLE 3 | Kinetic values obtained by fitting the Hill equation (A) and Equation 1 (B) to the experimental data shown in Figure 4. (A) AVP1 [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.47 ± 0.01 0.37 ± 0.01 0.23 ± 0.01 K 0.5 (mM) 0.14 ± 0.08 0.14 ± 0.11 0.19 ± 0.17 n H 2.06 ± 0.14 2.45 ± 0.22 2.63 ± 0.34 MVP [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.41 ± 0.04 0.35 ± 0.04 0.33 ± 0.01 K 0.5 (mM) 0.27 ± 0.27 0.20 ± 0.27 0.30 ± 0.25 n H 3.74 ± 0.60 4.27 ± 1.50 3.46 ± 0.3 (B) AVP1 [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.51 ± 0.01 0.40 ± 0.01 0.24 ± 0.05 L 17 ± 5 50 ± 3 106 ± 14 K R (mM) 84 ± 9 64 ± 13 69 ± 30 MVP [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.46 ± 0.01 0.47 ± 0.09 0.34 ± 0.04 L 615 ± 45 123 ± 22 7 10 3 ± 790 K R (mM) 62 ± 8 90 ± 16 35 ± 11 Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12407 shows that AVP1 and MVP are homodimers with one substrate binding site in each subunit (Li et al., 2016). 5. Mg 2 PPi may also act as an allosteric positive effector by displacing the R-T equilibrium to the R state. To account for this effect in the simplest possible way, L is divided by the factor (1+S/K R ) 2 . This means that either Mg 2 PPi exerts its stabilising role by binding to the two existing catalytic sites (that is, there are no specific regulatory sites for Mg 2 PPi), or it binds with the same affinity to regulatory and catalytic sites. The final equation obtained was: v=Vm=(S=KR)·(1 + S=KR) (1 + S=KR)2+½L=(1 + S=KR)2(Equation 1) Where S is the concentration of the substrate (Mg 2 PPi), K R is the dissociation constant of the substrate toward the active R state, and L is the allosteric constant ([T]/[R]). The experimental data were fitted with this equation (Figure 4), giving plots very similar to those obtained with the empirical Hill equation (data not shown). The values of the parameters in Equation 1 calculated using RStudio illustrated the differences observed in the kinetic properties of AVP1 and MVP, thus, the allosteric constant L was one order of magnitude higher for MVP than for AVP1 (Table 3). This result is consistent with the higher values of the Hill parameter n H previously obtained (see above). The value of L for AVP1 increased with the concentration of NaCl, whereas the calculated dissociation constant K R did not change significantly (Table 3B). This suggests that NaCl displaces the allosteric equilibrium of AVP1 toward the inactive T state. By contrast, in the case of MVP, L decreases in the presence of 50 mM NaCl, while a concentration of 100 mM is necessary to increase its value one order of magnitude with respect to that obtained in the absence of salt (Table 3B). In another set of experiments, PPase activities were assayed by fixing MgCl 2 concentrations in the assays (1 and 2.5 mM) and increasing PPi(upto0.25mM)inthepresenceof0,50,and100mM NaCl. The concentrations of PPi and Mg 2+ were chosen to ensure that a significant amount of free Mg 2+ (that is, not complexed by PPi) was always present in the assay. Figure 5 shows that the kinetic behaviour of both AVP1 and MVP changed significantly in this situation with respect to the results shown in Figure 4. Fitting these data with the Hill equation produced good plots and the V m values were correctly predicted but the values of the rest of the parameters (n H and K 0.5 ) showed high values of standard errors and were not considered reliable (not shown). A B FIGURE 5 | Graphical representation of PPase activity versus PPi concentration for AVP1 (A) and MVP (B) in the presence of 0 mM (●), 50 mM (▲), and 100 mM (■) NaCl. MgCl 2 was fixed in the assays at the indicated concentrations. As in Figure 4, basal concentrations of Na + four times higher than those of PPi are present in the assays. Data points correspond to the averages ± SE of 3 independent experiments. Graph lines were calculated with Equation 1 using RStudio, as described in Materials and Methods. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12408 Equation 1 predicted the kinetic behaviour of AVP1 and MVP in the presence of excess Mg 2+ and the values obtained for the different parameters are shown in Table 4. The values of the allosteric constant L significantly decreased with respect to those shown in Table 3, especially in the case of MVP; moreover, these values were not affected by the presence of increasing concentrations of NaCl, with the only exception of AVP1 when its PPase activity was assayed at 100 mM NaCl and 1 mM MgCl 2 . Kinetic data further show that V m values for AVP1 steadily decreased with NaCl concentration, especially in the presence of 1 mM Mg 2+ , this effect being somewhat attenuated at 2.5 mM Mg 2+ . PPi concentrations of 0.25 mM and higher exerted a significant inhibitory effect on AVP1 activity at the concentrations of MgCl 2 tested. Equation 1 could not model this effect, therefore, only data obtained with PPi concentrations up to 0.2 mM were used for curve fitting. Vm values for MVP only decreased significantly at 100mMNaCland1mMMg 2+ . Inhibition of MVP by PPi concentrations up to 0.5 mM was not observed (not shown). An appropriate [K + ]/[Na + ] ratio in the cytosol has been reported to be extremely important for cell viability in both yeast and higher plants and the presence of high external sodium concentrations may alter this ratio producing toxic effects (Gomez et al., 1996;Maathuis and Amtmann, 1999). PPase activity assays of AVP1 and MVP performed with optimal substrate concentration and decreasing [K + ]/[Na + ]ratios showed that AVP1 is inhibited to a higher extent than MVP when the concentration of Na + is higher than that of K + .As observed in previous experiments, the presence of excess Mg 2+ significantly protected AVP1 against inhibition by Na + (Figure 6A). Expectedly, the H + -pumping activity of AVP1 decreased in parallel with the PPase activity (Figure 6B), whereas no H + - translocation was detected with microsomes obtained from cells expressing MVP (not shown). Intracellular PPi Levels In order to check whether the sensitivities of AVP1 and MVP to NaCl in vitro could have an effect in vivo, YPC6 cells transformed with plasmids pInvGFPAVP1 and pInvGFPMVP were subjected TABLE 4 | Values of parameters obtained by fitting Equation 1 to kinetic data corresponding to AVP1 (A) and MVP (B) in the presence of fixed Mg 2+ concentrations. (A) [MgCl 2 ] = 1 mM [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.77 ± 0.11 0.64 ± 0.05 0.43 ± 0.05 L4±2 3±1 13±5 K R (µM) 56 ± 8 48 ± 2 20 ± 3 [MgCl 2 ] = 2.5 mM [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.82 ± 0.09 0.73 ± 0.05 0.52 ± 0.04 L5±1 7±2 9±3 K R (µM) 50 ± 3 42 ± 4 27 ± 4 (B) [MgCl 2 ] = 1 mM [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 0.62 ± 0.05 0.62 ± 0.02 0.45 ± 0.07 L 4±3 2±1 0±1 K R (µM) 83 ± 28 77 ± 6 88 ± 47 [MgCl 2 ] = 2.5 mM [NaCl] (mM) Parameter 0 50 100 V m (U/mg prot.) 1.07 ± 0.45 0.69 ± 0.11 0.58 ± 0.03 L 0±1 3±1 5±2 K R (µM) 200 ± 50 59 ± 38 40 ± 4 Values of specific PPase hydrolytic activity obtained with PPi concentrations up to 0.175 mM (AVP1) and 0.25 mM (MVP) were used for curve fitting and calculation of parameters. Graphical representations of experimental data and lines calculated with Equation 1 are shown in Figure 5. A B FIGURE 6 | Sensitivity of the PPase activity of AVP1 and MVP (A) and the H + -pumping activity of AVP1 (B) to the [K + ]/[Na + ] ration in the presence of different concentrations of Mg 2+ . PPase assays were performed with 0.25 mM PPi and 0.5 mM (white bars), 1 mM (gray bars), and 2.5 mM (black bars) MgCl 2 . Activities are averages ± SE of 4 independent experiments and are expressed as percentages of PPase activity with respect to the average values obtained in the absence of NaCl (a tetrasodium salt of PPi was used, therefore, an extra 1 mM Na + was always present in the assays). These values, expressed as micromoles of PPi/min/mg of protein, were: 0.49 ± 0.05 (0.5 mM Mg 2+ ), 0.60 ± 0.11 (1 mM Mg 2+ ), 0.62 ± 0.06 (2.5 mM Mg 2+ ) for AVP1; and 0.49 ± 0.06 (0.5 mM Mg 2+ ), 0.50 ± 0.05 (1 mM Mg 2+ ), 0.60 ± 0.08 (2.5 mM Mg 2+ ) for MVP. H + -pumping activity was assayed as described in Materials and Methods in the presence of 1 mM MgCl 2 at the indicated concentrations of NaCl. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 12409 Rios, G., Ferrando, A., and Serrano, R. (1997). Mechanisms of salt tolerance conferred by overexpression of the HAL1 gene in Saccharomyces cerevisiae. Yeast Chichester Engl. 13, 515–528. doi: 10.1002/(SICI)1097-0061(199705) 13:6<515::AID-YEA102>3.0.CO;2-X Rodrı guez-Navarro, A., and Sancho, E. D. (1979). Cation exchanges of yeast in the absence of magnesium. Biochim. Biophys. Acta BBA - Biomembr. 552, 322–330. doi: 10.1016/0005-2736(79)90286-4 Sarafian, V., Kim, Y., Poole, R. J., and Rea, P. A. (1992). Molecular cloning and sequence of cDNA encoding the pyrophosphate-energized vacuolar membrane proton pump of Arabidopsis thaliana. Proc. Natl. Acad. Sci. U. S. A. 89, 1775– 1779. doi: 10.1073/pnas.89.5.1775 Schiestl, R. H., and Gietz, R. D. (1989). High efficiency transformation of intact yeast cells using single stranded nucleic acids as a carrier. Curr. Genet. 16, 339– 346. doi: 10.1007/BF00340712 Schilling, R. K., Marschner, P., Shavrukov, Y., Berger, B., Tester, M., Roy, S. J., et al. (2014). Expression of the Arabidopsis vacuolar H + -pyrophosphatase gene (AVP1) improves the shoot biomass of transgenic barley and increases grain yield in a saline field. Plant Biotechnol. J. 12, 378–386. doi: 10.1111/pbi.12145 Schultz, A., and Baltscheffsky, M. (2004). Inhibition studies on Rhodospirillum rubrum H(+)-pyrophosphatase expressed in Escherichia coli. Biochim. Biophys. Acta 1656, 156–165. doi: 10.1016/j.bbabio.2004.03.005 Segami,S.,Asaoka,M.,Kinoshita,S.,Fukuda, M., Nakanishi, Y., and Maeshima, M. (2018a). Biochemical, structural and physiological characteristics of vacuolar H +-pyrophosphatase. Plant Cell Physiol. 59 (7), 1300–1308. doi: 10.1093/pcp/pcy054 Segami, S., Tomoyama, T., Sakamoto, S., Gunji, S., Fukuda, M., Kinoshita, S., et al. (2018b). Vacuolar H + -Pyrophosphatase and Cytosolic Soluble Pyrophosphatases Cooperatively Regulate Pyrophosphate Levels in Arabidopsis thaliana.Plant Cell 30, 1040–1061. doi: 10.1105/tpc.17.00911 Serrano, R., and Villalba, J.-M. (1995). “Chapter 35 Expression and Localization of Plant Membrane Proteins in Saccharomyces,”in Methods in Cell Biology. Eds. D. W. Galbraith, D. P. Bourque and H. J. Bohnert (Academic Press), 481–496. doi: 10.1016/S0091-679X(08)61052-3 Serrano, R., Mulet, J. M., Rios, G., Marquez, J. A., de Larrinoa, I. F., Leube, M. P., et al. (1999). A glimpse of the mechanisms of ion homeostasis during salt stress. J. Exp. Bot. 50, 1023–1036. doi: 10.1093/jxb/50.Special_Issue.1023 Serrano, A., Perez-Castiñeira, J., Baltscheffsky, M., and Baltscheffsky, H. (2007). H+-PPases: yesterday, today and tomorrow. IUBMB Life 59, 76–83. doi: 10.1080/ 15216540701258132 Serrano-Bueno, G., Hernandez, A., Lopez-Lluch, G., Perez-Castiñeira, J. R., Navas, P., and Serrano, A. (2013). Inorganic pyrophosphatase defects lead to cell cycle arrest and autophagic cell death through NAD+ depletion in fermenting yeast. J. Biol. Chem. 288, 13082–13092. doi: 10.1074/jbc.M112.439349 Takeshige, K., and Hager, A. (1988). Ion Effects on the H+-Translocating Adenosine Triphosphatase and Pyrophosphatase Associated with the Tonoplast of Chora corallina.Plant Cell Physiol. 29, 649–657. doi:10.1093/oxfordjournals.pcp.a077542 Thomas,B.J.,andRothstein,R.(1989).Elevatedrecombinationratesin transcriptionally active DNA. Cell 56, 619–630. doi: 10.1016/0092-8674(89) 90584-9 Treco, D. A., and Lundblad, V. (2001). Preparation of Yeast Media. Curr. Protoc. Mol. Biol. 23, 13.1.1–13.1.7. doi: 10.1002/0471142727.mb1301s23 Tsai, J.-Y., Kellosalo, J., Sun, Y.-J., and Goldman, A. (2014). Proton/sodium pumping pyrophosphatases: the last of the primary ion pumps. Curr. Opin. Struct. Biol. 27, 38–47. doi: 10.1016/j.sbi.2014.03.007 Valverde, F., Losada, M., and Serrano, A. (1997). Functional complementation of an Escherichia coli gap mutant supports an amphibolic role for NAD(P)-dependent glyceraldehyde-3-phosphate dehydrogenase of Synechocystis sp. strain PCC 6803. J. Bacteriol. 179, 4513–4522. doi: 10.1128/jb.179.14.4513-4522.1997 Yang, Y., Tang, R. J., Li, B., Wang, H. H., Jin, Y. L., Jiang, C. M., et al. (2015). Overexpression of a Populus trichocarpa H + - pyrophosphatase gene PtVP1.1 confers salt tolerance on transgenic poplar. Tree Physiol. 35, 663–677. doi: 10.1093/treephys/tpv027 Zhen, R. G., Kim, E. J., and Rea, P. A. (1997). Acidic residues necessary for pyrophosphate-energized pumping and inhibition of the vacuolar H +-pyrophosphatase by N,N’-dicyclohexylcarbodiimide. J. Biol. Chem. 272, 22340–22348. doi: 10.1074/jbc.272.35.22340 Conflict of Interest: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Copyright © 2020 Pe rez-Castiñeira and Serrano. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Pe ´rez-Castiñeira and Serrano Na+-PPases as Potential Biotechnological Tools Frontiers in Plant Science | www.frontiersin.org August 2020 | Volume 11 | Article 124016