Overexpression of chloroplastic Zea mays NADP-malic enzyme (ZmNADP-ME) confers tolerance to salt stress in Arabidopsis thaliana

The C4 plants photosynthesize better than C3 plants especially in arid environment. As an attempt to genetically convert C3 plant to C4, the cDNA of decarboxylating C4 type NADP-malic enzyme from Zea mays (ZmNADP-ME) that has lower Km for malate and NADP than its C3 isoforms, was overexpressed in Arabidopsis thaliana under the control of 35S promoter. Due to increased activity of NADP-ME in the transgenics the malate decarboxylation increased that resulted in loss of carbon skeletons needed for amino acid and protein synthesis. Consequently, amino acid and protein content of the transgenics declined. Therefore, the Chl content, photosynthetic efficiency (Fv/Fm), electron transport rate (ETR), the quantum yield of photosynthetic CO2 assimilation, rosette diameter, and biomass were lower in the transgenics. However, in salt stress (150 mM NaCl), the overexpressers had higher Chl, protein content, Fv/Fm, ETR, and biomass than the vector control. NADPH generated in the transgenics due to increased malate decarboxylation, contributed to augmented synthesis of proline, the osmoprotectant required to alleviate the reactive oxygen species-mediated membrane damage and oxidative stress. Consequently, the glutathione peroxidase activity increased and H2O2 content decreased in the salt-stressed transgenics. The reduced membrane lipid peroxidation and lower malondialdehyde production resulted in better preservation, of thylakoid integrity and membrane architecture in the transgenics under saline environment. Our results clearly demonstrate that overexpression of C4 chloroplastic ZmNADP-ME in the C3 Arabidopsis thaliana, although decrease their photosynthetic efficiency, protects the transgenics from salinity stress.


Introduction
Plants with C4 photosynthesis have a higher potential energy conversion efficiency compared to those with C3 photosynthesis; this is because of the existence of a CO 2 concentrating mechanism in C4 plants, that largely eliminates photorespiration (Zhu et al. 2008).Further, as compared to C3, C4 plants have higher water and nitrogen use efficiencies (Sage 2004).Introducing C4 photosynthesis genes into C3 crops is an important strategy to improve their overall photosynthetic efficiency (Hibberd et al. 2008;Von Caemmerer et al. 2012;Bräutigam et al. 2014;Kandoi et al. 2016Kandoi et al. , 2018Kandoi et al. , 2022;;Lin et al. 2020;Ermakova et al. 2021;Zhang et al. 2021;Zhao et al. 2022).
Plants performing C4 photosynthesis have two known decarboxylating enzymes, i.e., NADP-malic enzyme (NADP-ME), and phosphoenolpyruvate carboxykinase (PEPCK).In C4 plants, CO 2 is initially fixed in the mesophyll cells by phosphoenolpyruvate carboxylase (PEPC), leading to the formation of oxaloacetate (OAA).In NADP-ME type C4 species, OAA is then predominantly reduced to malate and transported into the bundle sheath cells (BSCs), where it is decarboxylated by NADP-ME to generate pyruvate, NADPH and release CO 2 for its re-fixation through ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) (Hatch 1987;Gerald and Carlos 1992).In PEPCK type C4 species OAA is decarboxylated to phosphoenolpyruvate and CO 2 via the help of PEPCK (Kanai and Edwards 1999;Wingler et al. 1999).In this process, CO 2 concentration in the BSCs rises to levels up to ~ 2000 μL L −1 that successfully competes with O 2 to reduce photorespiration and increase photosynthesis (Furbank and Hatch 1987;Sage et al. 2012).Furthermore, malate decarboxylation product NADPH acts not only as a reducing agent for carbon assimilation but also plays an important role in proline biosynthesis and in the detoxification of ROS by ascorbate-glutathione cycle and thioredoxindependent network (Chen et al. 2019a, b;Kalemba et al. 2021).In our present study, the chloroplastic C4 Zea mays NADP-ME was chosen for its overexpression in C3 plants Arabidopsis thaliana since major our food crops, such as maize, and sorghum, as well as key bioenergy crops, such as Miscanthus giganteus and Saccharum officinarum, are all of NADP-ME subtype.
Several researchers have worked on overexpression of C4 NADP-ME in C3 plants and reported variable or contradictory results in the transgenics.Overexpression of C4 NADP-ME in C3 plants is shown to cause various aberrations such as abnormal chloroplasts, bleaching of leaf colour, growth hindrance and even senescence in the dark (Takeuchi et al. 2000;Tsuchida et al. 2001;Chi et al. 2004;Fahnenstich et al. 2007).ZmNADP-ME overexpressing Arabidopsis thaliana grown under short day (8 h L/16 h D) condition had lower photosynthetic efficiency and reduced biomass whereas when they were grown in long day (16 h L/8 h D) did not have substantial changes in photosynthesis and biomass (Zell et al. 2010).Transgenic rice plants expressing C4 NADP-ME showed high photo-inhibition under high light intensity but no significant change was observed in the carbon assimilation rate (Chi et al. 2004).Similarly, maize NADP-ME, overexpressed in transgenic tobacco lines had reduced stomatal behavior and gained more fresh mass per unit of water consumed due to alteration of malate metabolism in guard cells (Laporte et al. 2002).In contrast, overexpression of maize NADP-malic enzyme (NADP-ME) in the guard and vascular companion cells of Nicotiana tabacum, resulted in increased net carbon assimilation rate, reduced stomatal aperture, higher biomass, better water use efficiency, early flowering, and shorter life cycle (Müller et al. 2018).
In C3 plants including rice, bean and wheat, the gene expression of NADP-ME isoforms increases in response to environmental stresses, drought, low temperature, salinity, osmotic stress, heavy metals, UV and hypoxia (Aragao et al. 1997;Emami et al. 2016;Fushimi et al. 1994;Walter Michael et al. 1994;Pinto et al. 1999;Cheng and Long 2007;Liu et al. 2007;Detarsio et al. 2008;Fu et al. 2009Fu et al. , 2011;;Guo et al. 2009;Nguyen et al. 2009;Doubnerová and Ryšlavá 2011;Shao et al. 2011;Alvarez et al. 2013;Doubnerová et al. 2014;Sicher et al. 2015;Sun et al. 2019;Swain et al. 2021).However, the impact of salt stress in C3 plants overexpressing C4 NADP-ME having 4-6 fold low Km for malate and NADP has not been studied.Therefore, in the present study, we have overexpressed the maize C4 NADP-ME under the control of 35S CaMV promoter in Arabidopsis thaliana to study its impact on photosynthesis, stomatal conductance, biomass and tolerance to salt stress.Although, transgnics had reduced photosynthesis and biomass under optimal growth conditions, they had higher photosynthetic rate, biomass as they were tolerant to salt stress mainly due to the overproduction of the osmoprotectant proline, reduced ROS generation and MDA production.

Plasmid constructs, transformation and selection of transgenic lines
Zea mays malic enzyme (Accession Number-J05130) has 1911 bp of coding region was amplified by PCR using two gene specific primers: F, 5′-GAA TTC ATG CTG TCC ACG CGC ACC G-3′; R, 5′-GAA TTC CTA CCG GTA GTT GCG GTA GAC GGG A-3′.In both cases, EcoRI restriction sites were introduced (underlined).The pGEMT-Easy recombinant plasmid containing the full length ZmNADP-ME cDNA was digested by EcoRI and further cloned into the binary vector pCAMBIA1304 under the control of CaMV 35S promoter and Ω enhancer (Pattanayak and Tripathy 2002;Pattanayak et al. 2005).The recombinant plasmid (pCAMBIA1304::ZmNADP-ME) was transformed into Agrobacterium tumefaciens strain (GV1301) and introduced into 6-week old Arabidopsis thaliana plants (cv.columbia) via agrobacterium mediated floral dip method (Clough and Bent 1998).Vector control (VC) plants containing the null vector, pCAMBIA1304 (binary vector without ZmNADP-ME cDNA), were also generated and used for comparison with the transgenics since no significant differences in growth parameters were seen between the wild type (WT) and the VC plants (Kandoi et al. 2022).Seeds of the transformed plants were screened on half-strength Murashige and Skoog (MS) agar medium containing 50 μg/ml kanamycin and were grown to T3 generation.

Genomic DNA PCR
Genomic DNA was isolated by cetyl trimethyl ammonium bromide (CTAB) method (Nickrent 1994) from 4-week old plants of the T1 generation.The presence of trans-gene in the plants was confirmed by PCR using 35S forward internal primer (5'-CCC ACT ATC CTT CGC AAG AC-3′) and ZmNADP-ME reverse primer (5′-CTA CCG GTA GTT GCG GTA GAC GGG A-3′) to ensure the incorporation of the whole cassette in sense orientation.

Southern blot
The presence of NADP-ME transgene was checked by Southern blot analysis.The genomic DNA from the leaves of the T3 generation of the VC and ZmNADP-ME overexpresser plants was digested with the restriction enzyme XbaI.Thirty µg of DNA was loaded and resolved on 1% agarose gel and blotted onto Nylon 66 membrane (MDI) (Sambrook and Russell 2001).The nptII coding sequence was used for probe preparation and labeled with (α 32 P) dCTP, using a radioactive random primer labeling kit (Amersham-GE, UK).Southern blot was developed as described by Sambrook and Russell (2001).

Western blot analysis
Total protein was extracted from 4-week-old leaves of VC and transgenic plants for immunoblot analysis, in a protein extraction buffer (Jilani et al. 1996).Protein concentrations of the supernatants were measured, as described by Bradford (1976).Electrophoresis was carried out on a 15% (w/v) SDS-PAGE gel using a total of 25 µg of soluble plant protein per lane.Separated polypeptides were blotted on nitrocellulose membrane.Proteins were probed with anti-maize ME antibody (Imgenex, India).The rabbit anti-mouse IgG (1:25,000) was used as a secondary antibody, conjugated to alkaline phosphatase.Blots were stained for alkaline phosphatase, using 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitro blue tetrazolium (NBT) and quantified using an Alpha Imager 3400.

Malic enzyme activity
For screening of the primary transgenic plants, the malic enzyme activity was measured in leaves of VC and ZmNADP-ME overexpresser plants.One hundred mg of leaf discs were quickly ground in 1 ml of extraction buffer containing 100 mM Tris-HCl (pH 7.3), 1.0 mM EDTA, 1.0 mM PMSF, 2% glycerol, 100 µM β-ME, 25 mM MgCl 2 , 20 mM K 2 HPO4, 2 mM phenylmethylsulfonyl fluoride (PMSF) and 2% (w/v) insoluble polyvinylpyrrolidone (PVP).The homogenate was centrifuged at 14,000g for 5 min and the supernatant was used immediately for spectrophotometric assay of malic enzyme at room temperature.The assay buffer contained 25 mM HEPES-KOH (pH 8.0), 0.1 mM EDTA, 2 mM MgCl 2 and 0.5 mM NADP, 5 mM malate (pH 7.0 with NaOH) and 20 μl crude extract in 1 ml reaction (Tsuchida et al. 2001).This reaction was initiated by adding malate, and the reduction of NADP was monitored by absorbance at 340 nm.Protein concentration in the enzyme extracts was determined by the method of Bradford (1976); enzyme specific activity was expressed in terms of μmol NADP + reduced (mg protein) -1 min -1 .

Estimation of pigment, amino acids and protein
Leaf total chlorophyll was extracted in 90% ammonical acetone (acetone:1 N NH 4 OH: 9:1) and estimated as described by Porra et al. (1989).Free amino acids were analyzed by the ninhydrin colorimetric method, using leucine as a standard (Kandoi et al. 2016).The leaf soluble protein was measured according to Bradford (1976).

Measurement of growth parameters
For monitoring growth parameters, VC and ZmNADP-ME overexpresser plants were grown vertically under coolwhite-fluorescent light (100 µmol photons m −2 s −1 ), under 14 h light/10 h dark photoperiods, at 21 ± 1 °C for 3 weeks in Murashige and Skoog medium in petri-dishes.The position of plants was randomized and the position of the trays rotated frequently under the light.
For the measurement of fresh weight, plants were taken from the petri dishes, and their weight was measured.For dry weight, whole plants were kept in an oven at 80 °C for 72 h, before the measurement.

Pulse amplitude modulation (PAM) measurements
Chlorophyll a fluorescence was measured simultaneously by PAM-2100 fluorometer (Walz, Germany), using the automated "Light Curve" program provided by a software from Walz.Prior to measurements, plants were placed in dark for 20 min (Demmig et al. 1987;Demmig et al. 2014).First, the initial minimal fluorescence (Fo) and then the maximum fluorescence (Fm) were measured by the 'Saturation Pulse' method.Optimum quantum yield of Photosystem II (PSII) was calculated as Fv/Fm = (Fm − Fo)/Fm, where Fv is the variable chlorophyll a fluorescence.Electron transport rate (ETR) of PSII in the light was calculated according to a formula described by Schreiber et al. (1995): ETRII = Yield II (YII) × PAR × 0.5 × 0.84, where the effective PSII quantum yield (YII) is calculated according to Genty et al. (1989) by the formula: YII = (Fm′ − Ft)/Fm′, where Ft represents the measured Chl a fluorescence yield at any given time (t) and Fm' is defined as the maximal Chl a fluorescence yield in a pulse of saturating light, when the sample is pre-illuminated, PAR is flux density of incident photosynthetically active radiation (μmol photons m −2 s −1 ), the 0.5 factor is used since transport of one electron requires absorption of two quanta by the two photosystems, i.e., it assumes that PSII: PSI ratio is 1:1, and the use of 0.84 assumes that 84% of the incident quanta are absorbed by the leaf.Non-photochemical quenching (NPQ) of the excited state of Chl a was calculated from the formula: NPQ = (Fm − Fm′)/Fm′ (Schreiber 2004).

Photosynthesis: light response curve
Light response curves of photosynthesis of six-week old vector control and transgenic plants, grown under 14 h L/10 h D, in soil, were measured with an Infrared gas analyzer (Portable Gas Exchange Fluorescence System GFS3000, Walz), using a standard head having LED light source 3040-L with an LED array providing 90% red and10% blue light.Sample chamber CO 2 concentration was maintained at 400 μmol mol −1 , whereas the air temperature was kept at 25 °C, and the relative humidity at 60%.For the light curve, different light intensities (500,400,300,200,100,50, 10 and 0 μmol photons m −2 s −1 ) were used.For measurements of stomatal conductance (gs), transpiration rate (E) and water use efficiency (WUE), the light intensity was 400 μmol photons m −2 s −1 .

Salt treatment (salt stress)
To measure the effect of salt stress on photosynthetic efficiency of plants, vector control, and transgenic plants were plated on the Murashige and Skoog medium (MS) medium for 2 weeks, and then transferred to the same medium with or without 150 mM NaCl for 9 days.Growth was measured during this period to evaluate salt tolerance.

Transmission electron microscopy (TEM)
Arabidopsis leaves were vacuum infiltrated with 2.5% glutaraldehyde solution for 30 min and kept overnight in the same solution (Karnovsky 1965).This solution was then replaced by 0.1 M sodium-phosphate buffer (pH 7.0); after this, we followed the procedure of Jiang et al. (2011).Sections of samples were then viewed in a transmission electron microscope (JEOL 2100F) at the Advanced Instrumentation Research Facility of Jawaharlal Nehru University, New Delhi, India.
Proline content Free proline was estimated from leaf samples using ninhydrin, as described by Bates et al. (1973).Leaf samples were homogenized in 3% sulfosalicylic acid and centrifuged for 30 min at 14,000×g.Then, 2 ml of acid ninhydrin and 2 ml of glacial acetic acid were added to the supernatant.The mixture was then boiled for 1 h.In this reaction mixture, 4 ml of toluene was added and vigorously mixed for 15-20 s.The upper layer of toluene containing chromophore was used to measure absorbance at 520 nm; toluene was used as a blank.

Peroxidase activity
The peroxidase activity was assayed spectrophotometrically at 436 nm, using Guaiacol, a hydrogen donor (Putter 1974).The reaction mixture consisted of 2.5 ml of a mixture containing 20 mM Guaiacol in 0.01 M Sodium phosphate buffer, pH 6.0 and 0.1 M Hydrogen Peroxide.Enzyme extract (0.1 ml) was added to initiate the reaction.The absorbance change was recorded at 436 nm.The boiled enzyme extract served as blank.
H 2 O 2 content H 2 O 2 content was quantified in leaf extracts.The extract was diluted accordingly and then used for H 2 O 2 determination with an Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Invitrogen, USA), which was done spectrophotometrically at 560 nm.

Data processing and analysis
Excel was used for statistical analyses.After the calculation of averages, standard deviations and standard errors for each of the parameters were determined.ANOVA test were used to determine statistical significance (P < 0.05) between the VC and the transgenic plants for all the parameters.

Genomic DNA PCR
Using an Agrobacterium tumefaciens mediated gene transfer system, we have transformed Arabidopsis thaliana, with cDNA of Zea mays malic enzyme (ZmNADP-ME) under the control of CaMV 35S promoter (Fig. 1a).Several transgenic (T1) Arabidopsis plants, that were resistant to kanamycin, were obtained from this transformation.We then isolated genomic DNA from the kanamycin resistant ZmNADP-ME over-expression (MEx) (T1) lines.The ZmNADP-ME was amplified by PCR, using the 35S internal forward primer and gene specific reverse primer (ZmNADP-ME R), from genomic DNA of several individual overexpression lines (MEx7, MEx9, MEx12 and MEx13) of T1 generation.PCR yielded a fragment of ~ 2 Kb from transformants that contained ZmNADP-ME (Fig. 1b).The gene was not amplified in vector control containing the null vector pCAMBIA1304 without ZmNADP-ME cDNA (Fig. 1b).These individual transgenic lines were grown to harvest seeds.Seeds collected from the above plants were again grown in kanamycin plates to select T2 transgenic lines.Transgenic seeds were grown for T3 and T4 generations to obtain homozygous transgenic plants.

Southern blot analysis
The number of integration of the T-DNA cassette containing ZmNADP-ME cDNA into the Arabidopsis thaliana host genome was checked by Southern blot analysis.The genomic DNA was extracted from MEx12 and MEx13 as well as from the vector control plants.Genomic DNA was digested by XbaI restriction enzyme that specifically cuts the T-DNA cassette introduced into host genome.Southern blot analysis of XbaI digested genomic DNA, using nptII probe, revealed single band(s) in MEx12 and MEx13 confirming single integration of the T-DNA cassette in the Arabidopsis genome (Fig. 1c).

qRT-PCR
Expression of ZmNADP-ME genes among different transgenics of T3 generation was analyzed by qRT-PCR, using genespecific primers.The transcript abundance of ZmNADP-ME in MEx9, MEx12, and MEx13 was 4.5, 5.2, and 6.4 fold higher, respectively, than that in MEx7 (Fig. 1d).An increase in the expression of ZmNADP-ME was calculated using, as a reference, MEx7, a transgenic line that had the lowest expression of ZmNADP-ME.

Western blot analysis
Figure 1e shows the coomassie brilliant blue R250 stained SDS-PAGE for the visualization of proteins separation.Further, Fig. 1f shows that increased gene expression of malic enzyme resulted in increased protein abundance.We note the presence of a 62 kD protein band in the VC as well as in transgenic lines.As compared to the VC, ME protein abundance was higher in all the transgenic lines.

ME activity
The NADP-ME activity of the transgenics were assayed spectrophotometrically by the reduction of NADP as described in material and methods.ME activity in VC was 16 nmol NADP reduced mg protein −1 min −1 ; however, this activity, in different transgenic lines were higher (~ 56-118 nmol NADP reduced mg protein −1 min −1 ) i.e., 3.0-7.0-foldhigher than in the VCs (Fig. 1g).Based on their higher transgene expression, protein abundance and enzymatic activity, MEx12, and MEx13 transgenic lines were selected for further studies (Fig. 1g).

Plant morphology
Figure 2a shows a photograph of the VC and the ZmNADP-ME overexpressed plants, used in our measurements.They were grown for 4 weeks in a controlled chamber under a 14 h day (light) and 10 h night (dark) cycle at 21 ± 1 °C, and at 100 µmol photons m −2 s −1 cool-white-fluorescent light.Visually, plants of MEx12 and MEx13transgenic lines were smaller in size than the vector control (see Fig. 2a).Their rosette diameters were reduced by ~ 12-25% (Fig. 2b).

Pigment, amino acids and total protein content
Total chlorophyll (Chl), Chl a, Chl b, and Chl a/b ratios were measured in four-week-old plants after extraction in 90% ammonical acetone.The transgenic lines MEx12 and MEx13 had 11-14% lower total Chl, as well as Chl a and Chl b, than in the VC (Fig. 2c, d, e); however, there was no significant change in the Chl a/b ratio (Fig. 2f).The protein content of transgenic plants decreased by 17-24% in MEx12 and MEx13 (Fig. 2g); further, the free amino acids content of transgenic plants also decreased by 11-15% than in the vector control (Fig. 2h).

Biomass of ZmNADP-ME overexpressing transgenic plants
After 4 weeks of growth, the fresh weight of the MEx12 and MEx13 transgenic plants decreased by 35-49% (Fig. 2i) and their dry weights by 30-44% than that of the VCs (Fig. 2j).

Chlorophyll a fluorescence measurement
To check if decreased Chl content modulates the function of photosynthetic apparatus and primary processes of photosynthesis, we used Chl a fluorescence as its noninvasive signature (Govindjee 1995; Nellaepalli et al 2012).For a background on the basics and use of chlorophyll a fluorescence for measuring different reactions in photosynthesis, see Govindjee et al. (1986), Papageorgiou and Govindjee (2004) and Padhi et al. (2021).Various Chl a fluorescence parameters were measured as described under Material and methods; our results are described below.

Fo
The minimum Chl a fluorescence (Fo), measured in darkadapted leaves of MEx12 and MEx13 plants, had 10-12% higher Fo values than in the VC (Fig. 3a).We note that although the transgenic plants had lower Chl content, their Fo was higher, probably due to the presence of non-Q B centers; these non-reducing centers are incapable of transferring electrons from the reduced primary plastoquinone Q A (Q - A ) to the secondary plastoquinone Q B (for details see Schansker and Strasser (2005).

Fm
The maximum Chl a fluorescence (Fm) was measured during the first saturation pulse after the leaf had been dark-adapted; it decreased in MEx12 and MEx13 plants by 9%-15% than that in the VC plants (Fig. 3b).

Fv
In view of the above, the variable fluorescence (Fv) had decreased by 15%-23% in MEx12 and MEx13 plants than that in the VC plants (Fig. 3c).

Fv/Fm
This ratio is an estimate of the maximum portion of absorbed quanta used in PSII reaction centers by darkadapted leaves (Guidi et al. 2019).In MEx12 and MEx13 plants, the Fv/Fm ratio was lower by 8%-10% than that in the VC plants (Fig. 3d).

Non-photochemical quenching (NPQ) of the excited state of chlorophyll
The NPQ represents the fastest process of rapid and reversible thermal dissipation of absorbed light energy in the PSII antenna (Niyogi 1999;Müller et al. 2001); at 420 μmol photons m −2 s −1 , it was higher (40-64%) in the transgenic plants than that in the VC plants.We emphasize that the NPQ values of the transgenics were higher in low as well as saturating light intensities than that of the VC plants (Fig. 3f).

Light response curve of net CO 2 assimilation
To check if the decreased electron transport rates, that provide lower reducing power in the form of NADPH, as well as ATP, for CO 2 reduction, indeed leads to decreased net carbon assimilation in the antisense plants, we monitored CO 2 uptake rates of intact leaves.The photosynthetic light response curves of the attached leaves of the VC and the transgenic plants were monitored by Infra-Red Gas Analyzer (see "Materials and methods" section).As compared to the VC plants, the transgenics had lower (~ 18-28%) net CO 2 fixation; this saturated almost at 450 µmol photons m −2 s −1 (Fig. 4a).The relative quantum yield of CO 2 fixation, measured at limiting light intensities (upto 80 μmol photons m −2 s −1 ), of transgenics was lower (~ 8-12%) than that in the VC plants (Fig. 4b).The rate of respiration measured in the dark was lower (41%) in the transgenics than in the VC.The light compensation point in the VC was ~ 20 µmol photons m −2 s −1 , but it was lower (~ 15 µmol photons m −2 s −1 ) in the transgenics (Fig. 4b).

Stomatal conductance (gs) and water use efficiency (WUE)
The CO 2 assimilation rate (An) decreased by 18% in MEx12 and MEx13 at 400 µmol photon m −2 s −1 (Fig. 4c).It is important to note that these decreases in photosynthetic rates were associated with decreases in stomatal conductance and transpiration in the transgenic lines.For example, stomatal conductance decreased by 9%-22% in MEx12 and MEx13 than in the VC (Fig. 4d); in addition, transpiration rate decreased by 10%-21% in MEx12 and MEx13 plants (Fig. 4e).No significant changes were observed in water use efficiency of transgenic and vector control (Fig. 4f).

The NADP-ME overexpressers were tolerant to salt stress
To check the tolerance to salt stress, VC, MEx12 and MEx13 plants were grown in Murashige and Skoog (MS) medium for 14 days and, subsequently, transferred to MS medium or MS + 150 mM NaCl for 9 days.The VC plants almost bleached due to salt stress; however, the transgenic plants were greener than VC under identical conditions (Fig. 5a).

Biomass, Pigment and protein contents in salt stress
The fresh weight decreased by 77%, 63% and 59% in VC, MEx12 and MEx13 (Fig. 5b) As compared to that of the VC, the Chl content was higher in the transgenics under saline environment (Fig. 5c).Under salt (150 mM NaCl) treatment, the Chl content declined by 75% in the VC plants, whereas in the overexpressers, the loss of Chl was lower (51%-62%) (Fig. 5c).In the presence of 150 mM NaCl, the protein content in the VC plants declined by 65%, but in the transgenics, the decrease was somewhat less, i.e., by 46-54% in the transgenics (Fig. 5d).

Chlorophyll a fluorescence and photosynthetic efficiency in salt stress
Salt stress significantly affected chlorophyll a fluorescence parameter.In our experiments, the parameters most affected, by salt treatment, were minimal Chl fluorescence, Fo, ratio of variable to maximum fluorescence, Fv/Fm, electron transport rate (ETR) and non-photochemical quenching (NPQ); for further detailed definition of these parameters, see "Materials and methods" section.
We list below the information obtained on the various fluorescence parameters:

Fo
Fo was measured in the dark-adapted leaves as initial minimum fluorescence.Under salt stress (150 mM NaCl), the Fo of the VC decreased by 40%, but to a lesser extent (25-30%) in the two transgenic lines (Fig. 6a).Reduced Fo was mostly due to low Chl content in salt stressed plant.

Fv/Fo ratio
Variable to minimum fluorescence (Fv/Fo), which is considered to be proportional to the activity of the water-splitting complex on the donor side of the PSII; declined by 13-18% in transgenics compare to VC in salt-stressed plants.

Fv/Fm ratio
The Fv/Fm ratio, a measure of the maximum quantum efficiency of PSII in dark-adapted leaves, declined in VC and transgenic plants when grown in 150 mM NaCl.The Fv/ Fm ratio in the VC decreased by 51% upon salt (150 mM NaCl) treatment.However, its decline in the overexpressers was slightly smaller (36-40%) that demonstrates a better salt tolerance by the transgenic plants (Fig. 6b).

Electron transport rate II (ETRII)
Under control conditions, the ETRII of VC was higher (~ 20%) than that of the ZmNADP-ME overexpresser plants under limiting as well as saturating light intensities (see Figs. 3e & 6c).Salt treatment (150 mM NaCl) resulted in decreased ETRII in the VC and in the overexpresser plants under limiting as well as under saturating light intensities.However, at saturating light intensity, the ETRII under salt stress (150 mM NaCl) declined by ~ 76% in the VC, but in MEx13 and MEx12 it decreased by 50-67% (Fig. 6c).

Non photochemical quenching (NPQ) of the excited state of Chlorophyll
Salinity increased coefficient of photochemical quenching (qP) but markedly elevated coefficient of nonphotochemical quenching (qN) in the light-adapted state.As expected, we observed an increase in NPQ in response to increased light intensity (Ruban & Murchie 2012).However, at saturating light intensity (344 μmol photons m −2 s −1 ), the NPQ in MEx12 and MEx13 transgenic lines was almost similar to the VC under control conditions (Fig. 6d).Due to reduced light utilization under salt stress condition, the NPQ of VC was higher than that in the transgenics (Fig. 6d).

Chloroplast ultrastructure
All the plants, used in this study, showed a marked change in the phenotype under salt stress and the effect was very prominent in the leaves.The ultrastructure of the chloroplasts, by transmission electron microscopy, showed that the VC plants had salt-induced structural distortions such as de-stacking of granal organization, as well as swelling of thylakoids (Fig. 7a, lower panel).However, the de-stacking of grana and thylakoid swelling was less pronounced in the transgenics than in the VC (Fig. 7a, lower panel).

Proline content
The amino acid proline that usually increases under stress condition is known to play a critical role in protecting plants from abiotic stresses (Hayat et al. 2012).Under control condition, the proline content of both the VC and the overexpresser plants was almost similar.In saline environment (150 mM NaCl), the proline content increased by ~ 45% in the VC plants.However, under identical conditions, the proline content was nearly ~ 122-163% higher in the overexpressers (Fig. 7b).Proline is an osmoprotecting molecule that is known to accumulate in response to environmental stress.Therefore, synthesis of excess proline is a key step involved in tolerance of overexpresser plants to environmental stress.

Peroxidase
Peroxidase is involved in the scavenging of reactive oxygen species (ROS) by deposing H 2 O 2 into water (Asada 1999).In response to salt treatment (150 mM NaCl), guaiacol peroxidase increased by 30% in the vector controls, but much more, by 81-98%, in the transgenics MEx12 and MEx13 (Fig. 7c).

H 2 O 2 content
Reactive oxygen species production is known to increase under environmental stress (Das and Roychoudhury 2014); H 2 O 2 is known to be the most stable form of the ROS (Huang et al. 2019).Under control conditions, the H 2 O 2 content of VC and overexpressers was almost similar (Fig. 7d).In response to salt treatment (150 mM NaCl), H 2 O 2 content increased to the same level (~ 96%) in both VC and overexpressers plants.However, the transgenics had 36%-50% higher H 2 O 2 than VC in saline environment (Fig. 7d).

Lipid peroxidation
In control condition, malondialdehyde (MDA) content, an index for lipid peroxidation (Kandoi et al. 2016), was quite similar in the VC and the transgenics.Under stress condition, the MDA content was found to increase by 169% in the vector controls, but by 47-78%, in MEx12 and MEx13 (Fig. 7e).

Discussion
The chloroplastic NADP-malic enzyme (NADP-ME) is a major de-carboxylating enzyme of the C4 photosynthesis pathway in NADP-ME type plants (Wang et al. 2014;Chen et al. 2019c).Therefore, introduction of de-carboxylating enzyme C4 NADP-ME (Wang et al. 2014;Chen et al. 2019c) into C3 plants is essential to genetically convert a C3 plant to C4.Our characterization of several transgenic lines of A. thaliana overexpressing ZmNADP-ME under the control of 35S CaMV promoter reveals a 4-7 fold higher gene expression and protein abundance of NADP-ME in overexpressers as compared to the vector control (VC).Similarly, the NADP-dependent malate decarboxylating activities of the enzyme in the transgenic lines were 3-7 fold higher than the VC (Fig. 1d, f, g).Further, C4 ZmNADP-ME has 5-20 fold lower Km for malate and 4-6 fold lower Km for NADP compared to its non-photosynthetic C3 isoforms (Casati et al. 1997;Drincovich et al. 2001).Because of the higher affinity for maize NADP-ME for malate, it decarboxylated most of the chloroplastic malic acid that were not metabolized by its plastidic C3 isoform (AtNADP-ME4) to pyruvate to generate more NADPH and CO 2 .Therefore, ZmNADP-ME overexpressers had higher rate of conversion of endogenous malate to pyruvate and increased NADPH.Due to increased NADP-ME activity in transgenics, the C4 carboxylic acid pool that serve as the carbon backbone of several amino acids, i.e. aspartate, asparagine and other amino acids would have declined.This resulted in reduced amino acids and protein content in transgenics (Fig. 2g, h).Proteins bind to Chl to make light-harvesting photosynthetic complexes that insert themselves into the thylakoid membrane.Due to reduced synthesis of protein, incorporation of Chls in thylakoid membrane declined, resulting in lower Chl content in transgenic plants.(Fig. 2c).

Chlorophyll a fluorescence
Chl a fluorescence is a non-invasive probe for measurement of photosynthesis (Krause and Weis 1991;Govindjee 1995;Baker 2008).The minimal fluorescence (Fo) increased (~ 10-12%) in ZmNADP-ME overexpressers compare to VC (Fig. 3a); such an increase of Fo is often related to PSII inactivation by photoinhibition and is possibly due to increased amount of Q A − , reduced form of the first plastoquinone electron accepter and formation of non-Q B centers (Schreiber and Armond 1978;Neubauer and Schreiber 1987;Cao and Govindjee 1990;Krause and Weis 1991;Govindjee 1995;Zlatev and Yordanov 2004;Dutta et al. 2009).Further, the observed decrease (~ 9-15%) in maximal fluorescence (F m ) in transgenics (Fig. 3b) is, in all likelihood, due to inhibition of electron transport on the electron donor side of the PSII resulting in the accumulation of P 680 + , a quencher of chlorophyll a fluorescence.The Fv/Fo an indicator of the efficiency of the oxygen evolving complex decreased ~ 13-18%, further corroborating the donor side impairment of PSII in transgenics likely due to photoinhibition.The Fv/Fm that measures the maximum photochemical efficiency of PSII in dark adapted leaves (Björkman and Demmig 1987) decreased (8-10%) in the ZmNADP-ME overexpressers demonstrating that their PSII efficiency was lower than that in the VCs (Fig. 3d).Since NADP-ME catalyzes the reduction of NADP to NADPH, this enzyme decreases the NADP and increases the NADPH in the chloroplast stroma to stimulates photoinhibition (Shi et al. 2022).Thus, all our results suggest that the transformants were more susceptible to photoinhibition.This conclusion is supported by the observed bleaching of plants, the decreased Fv/Fm ratios and deteriorative effects on growth in rice plants overexpressing NADP-ME (Takeuchi et al. 2000;Tsuchida et al. 2001).In contrast, under longer-day conditions (16 h L/8 h D), the Fv/Fm ratios of the NADP-ME overexpressing Arabidopsis thaliana were similar to wild type (Fahnenstich et al. 2007).
In our experiments, the ETR decreased in the transgenics under low as well as high light intensities, demonstrating a decrease in the energy capture and its utilization (Fig. 3e).Non-photochemical processes that dissipate excitation energy by quenching the excited state of Chl a are higher in the transgenics than in the vector control demonstrating that in the transgenics, more light energy is dissipated as heat (Fig. 3f).Furthermore, the light dependent CO 2 assimilation was lower in the transgenics than in VC (Fig. 4a) due to constitutive loss of carbon pool as a consequence of decarboxylation of malate to pyruvate and CO 2 .The quantum yield of photosynthetic CO 2 assimilation, measured in limiting light intensities, decreased (8-12%) in the overexpressers as compared to the VC (Fig. 4a).Similarly, rice plants overexpressing ZmNADP-ME also had lower rate of CO 2 assimilation than the WT (Tsuchida et al. 2001).Due to lower carbon assimilation efficiency, plant growth, fresh weight and dry matter accumulation decreased in the transgenics (Fig. 2a, i, j).
Malate is known to be involved in stomatal movement through regulation of the osmotic pressure of the guard cells (Asai et al. 2000).During stomatal opening, malate is transported from the apoplastic space to the guard cell cytoplasm by the transporter AtABCB14, a member of the ABC (ATP Binding Cassette) family (Lee et al. 2008;Daloso et al. 2017).Malate uptake into guard cells results in increased turgor pressure and opening of stomata.Conversely, during stomatal closure, malate is transported out of the guard cell via the voltage gated malate channel AtQUAC1.Furthermore, the silencing of AtQUAC1 results in an impaired stomatal closure due to malate accumulation in guard cells that increases stomatal conductance and loss of water from plants (Medeiros et al. 2016).Similarly, in our transgenic plants malate present in the guard cells decarboxylated by NADP-ME to yield pyruvate and CO 2 that decreases the malate concentration.This leads to loss of guard cell turgor pressure and stomatal closure that resulted in reduced stomatal conductance and decreased transpiration rate in transgenics.However, the decreased transpiration rate in transgenics did not result in the increased water use efficiency due to decrease in rate of photosynthesis in ZmNADP-ME overexpressers (Fig. 4f).Similarly, overexpression of NADP-ME in tobacco resulted in decreased stomatal conductance (Laporte et al. 2002).Further, Arabidopsis thaliana overexpressing a plastidic ZmNADP-ME grown in short days (8 h L/16 h D) had low malate and fumarate levels with decreased biomass and lower photosynthetic performance (Zell et al. 2010).Under our experimental conditions, 14 h L/10 h D photoperiod the transgenics had smaller rosette diameter, lower photosynthetic rate and reduced biomass than the VC.Taken together, our results demonstrate that C4 type NADP-ME overexpression in the chloroplast of C3 plants decreases the rate of photosynthesis, carbon assimilation and biomass due to the unregulated loss of carbon backbone.

Transgenics are tolerant to salinity stress
In general, the photosynthetic performance of plants is known to decline under saline conditions (for review see: Chaves et al. 2009).It is an established fact that photosynthetic capacity of plants is determined by several factors.Often the efficiency of light captured to drive photosynthesis is correlated with the amount of chlorophyll in the leaf (see e.g., Netondo et al. 2004).Both water-stress and saltstress are known to downregulate and limit Chl biosynthesis by downregulating the expression of genes and enzymatic activities involved in the synthesis of 5-aminolevulinic acid (ALA), the substrate for chlorophyll heme and siroheme.The subsequent anabolic processes for the conversion of ALA to Chl, i.e., the gene expression, protein abundance and enzymatic activities are impaired by salt stress (Turan and Tripathy 2015).Therefore, Arabidopsis thaliana grown in saline environment had reduced Chl (Fig. 5c).In response to salt treatment, the reduction in Chl content was much higher in VC than the ZmNADP-ME overexpressers.As a result, in saline environment, the transgenics had greener leaves.This is in agreement with previous studies where it is demonstrated that the overexpression of others C4 genes, i.e., phosphoenol pyruvate carboxylase (PEPC) and NADPmalate dehydrogenase (NADP-MDH) in C3 plants protects the transgenics from abiotic stress i.e. salinity stress, drought stress and high temperature stress (Gu et al. 2013;Kandoi et al. 2016Kandoi et al. , 2018;;Li et al. 2017;Qi et al. 2017;Jiang et al. 2018).
Salt-stress-induced inhibition of growth of plants is often due to their reduced photosynthetic performance (Akram and Ashraf 2011;Nascimento et al. 2011).In response to salt stress, the Fo was shown to decline in both VC and transgenics.However, the percent reduction of Fo in the transgenics was lower (11-15%) than in the VC plants.The lower Fo could be attributed to reduce Chl content under salt stress condition.The transgenics had higher Chl content in saline condition and as a result had higher Fo.The Fv/Fm ratio that denotes maximum photosynthesis quantum yield, in dark-adapted condition, declined due to impairment of PSII reaction in vector control.However, in stress condition, the Fv/Fm was substantially higher (23-32%) in the transgenics suggesting that their PSII photochemical reactions were substantially protected from salt-stress.
The PSII-dependent photosynthetic ETR increased with increasing light intensity, and then saturated, and, in general salt-stress is known to decrease these rates (Kandoi et al. 2016).In our experiments, we found that the ETR was higher in the transgenic lines than VC plants in saline conditions.This could be due to a reduced salt-induced PSII damage of the reaction centers in the transgenics (Fig. 6c).The NPQ of the excited state of Chl molecules, which measures the dissipation of absorbed light energy as heat, was lower in salt-stressed transgenics than in the VC (Fig. 6d).These results suggest that the absorbed light energy was better utilized in photochemical reactions in the transgenics than in the vector controls, under salt stress conditions.The NADPH that serves, in addition to being used in the Calvin-Benson cycle, as the source of reducing equivalents for the xanthophyll cycle (Goss and Hanke 2014).Thus, the redox states of NADP(H) are crucial for regulating both ROS production and its scavenging in chloroplasts.The transgenics overexpressing ZmNADP-ME had higher NADPH due to malate decarboxylation and it protected them from reactive oxygen species (ROS) induced damage in salt stress condition.
The salt treatment in plants causes osmotic stress and is accompanied by the formation of excessive ROS.Proline accumulation normally occurs in cytoplasm where it stabilizes the proteins structure.It is a compatible solute that protects plants from different abiotic stresses (Delauney and Verma 1993;Madan et al. 1995;Verslues and Sharma 2010) by stabilizing the overall redox status of the cell (Hare et al. 1999;Szabados and Savouré 2010).The accumulation of proline decreases the cytoplasmic osmotic potential, facilitating water absorption, maintain cytosolic pH and scavenges reactive oxygen species (Qureshi et al. 2013;Gharsallah et al. 2016).The proline content of plants increased in saline conditions both in the VC and transgenics.However, the proline content in the transgenics was higher than the VC in salt stress.The ZmNADP-ME overexpresers had higher NADPH that was used for proline synthesis by reducing the substrate glutamate to pyrroline-5-carboxylate (Pérez-Arellano et al. 2010).The plastidic AtNADPME4 has a higher k cat for the reverse reaction than for the forward reaction, i.e. the rate of malate decarboxylation is almost half of the reverse carboxylation reaction (Drinocovich et al. 2001).Therefore, endogenous AtNADP-ME4 does not efficiently contribute in NADPH formation.
The glutathione peroxidase is a key component of the glutathione-ascorbate cycle which reduces the accumulation of H 2 O 2 by oxidizing reduced glutathione (GSH) to its disulfide form (GSSG) (Mittova et al. 2003).In the ZmNADP-ME overexpressers, glutathione peroxidase activity increased due to greater availability of NADPH that regenerates reduced form of glutathione (GSH).The increased activity of glutathione peroxidase substantially decreased H 2 O 2 content in the transgenics (Fig. 7c, d).Consequently, malondialdehyde (MDA) production, an indicator of membrane lipid peroxidation decreased in transgenics (Fig. 7e).Our results suggest that higher activity of ROS scavenging enzymes, like peroxidase (Fig. 7c), in the transgenics may lead to efficient detoxification of active oxygen species (Fig. 7d), and reduction in malondialdehyde content (Fig. 7e) under saline environment.It is already known that NADP-ME is an important player in plant-based defence in Arabidopsis (Chen et al. 2019c).Malate levels usually decrease and activities of NAD-ME and NADP-ME increase when plants are exposed to NaCl treatment (Speer and Kaiser 1991;Aragao et al. 1997).Thus, malic enzyme not only balances the concentration of malic acid in the cells, but also reduces the production of ROS.
From our observations on the chloroplast structure (Fig. 7a; lower panels), we suggest that better photosynthetic function of the overexpressers, under salt stress conditions, may have been due to better stacking of thylakoids in grana, and preserved membrane architecture (Fig. 7a) as a consequence of increased proline content.Further, under identical conditions, organization and architecture of thylakoid membranes were grossly affected in the vector control plants, which must have limited their photosynthesis potential (Fig. 7a).
Previously, we have shown that other C4 genes, PEP carboxylase (PEPC) and NADP-MDH overexpressed in Arabidopsis resulted in generation of increased proline content, lower H 2 O 2 and membrane lipid peroxidation product MDA that protected plants from salt stress (Kandoi et al. 2016(Kandoi et al. , 2018)).Similarly, our present study with ZmNADP-ME overexpression demonstrates that increased proline content and reduced ROS generation protect transgenic plants from salinity stress.

Fig. 1
Fig. 1 Vector construction and confirmation of the ZmNADP-ME transgenic plants.a Ligation of ZmNADP-ME with linearized modified pCAMBIA1304 vector having 35S promoter and omega (Ω) translational enhancer cassette and nptII for kanamycin selection; b PCR amplification of the genomic DNA with 35S internal forward and ZmNADP-ME specific reverse primers; c Southern blot analysis using nptII probe revealed that MEx12, and MEx13 transgenic lines including the vector control had single integration of the insert; d qRT-PCR of the T3 homozygous plants used to check the relative expression of ZmNADP-ME, as a reference, MEx7 line; Adenosyl

Fig. 3
Fig. 3 Pulse amplitude modulated (PAM) Chl a fluorescence measurement in VC and ZmNADP-ME transgenic plants.Plants were grown in pots under 14 h Light and 10 h Dark regime for 5 weeks under cool-white fluorescent light (100 µmol photons m −2 s −1 ). a Minimal Chl a fluorescence (Fo); b Maximal Chl a fluorescence (Fm); c Variable fluorescence (Fv = Fm − Fo); d Maximum photochemical efficiency, as inferred from Fv/Fm; e Light response curve

Fig. 5
Fig. 5 Morphological features, fresh weight, chlorophyll and protein content, of plants in saline environment.Seedlings were grown in MS for 14 days and subsequently transferred to Murashige and Skoog (MS) medium or MS + 150 mM NaCl medium for 9 days.a A photograph of VC and ZmNADP-ME overexpressers to MS (upper panel)

Fig. 6
Fig. 6 Chlorophyll a fluorescence measurements of dark adapted salt-stressed plants.Seedlings were grown in MS for 14 days and subsequently transferred to Murashige and Skoog (MS) medium or MS + 150 mM NaCl medium for 9 days.a Minimal Chl a fluorescence (Fo); b Maximum photochemical efficiency, as inferred from Fv/Fm; c electron transport rates (ETR), ETR = Yield × PAR × 0.5 × 0.84; d Non-photochemical quenching