Alliance Bioversity-CIAT Research Online Accepted Manuscript Differential expressions and enzymatic properties of malate dehydrogenases in response to nutrient and metal stresses in Stylosanthes guianensis The Alliance of Bioversity International and the International Center for Tropical Agriculture believes that open access contributes to its mission of reducing hunger and poverty, and improving human nutrition in the tropics through research aimed at increasing the eco-efficiency of agriculture. The Alliance is committed to creating and sharing knowledge and information openly and globally. We do this through collaborative research as well as through the open sharing of our data, tools, and publications. Citation: Song, J.; Zou, X.; Liu, P.; Cardoso, J.A.; Schultze-Kraft, R.; Liu, G.; Luo, L.; Chen, Z., (2022) Differential expressions and enzymatic properties of malate dehydrogenases in response to nutrient and metal stresses in Stylosanthes guianensis. Plant Physiology and Biochemistry 170 p. 325-337. ISSN: 0981-9428 Publisher’s DOI: https://doi.org/10.1016/j.plaphy.2021.12.012 Access through CIAT Research Online: https://hdl.handle.net/10568/117447 Terms: © 2022. The Alliance has provided you with this accepted manuscript in line with Alliance’s open access policy and in accordance with the Publisher’s policy on self-archiving. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. You may re-use or share this manuscript as long as you acknowledge the authors by citing the version of the record listed above. You may not change this manuscript in any way or use it commercially. For more information, please contact Alliance Bioversity-CIAT - Library Alliancebioversityciat-Library@cgiar.org 1 Differential expressions and enzymatic properties of malate 2 dehydrogenases in response to nutrient and metal stresses in 3 Stylosanthes guianensis 4 Jianling Songa, Xiaoyan Zoua, Pandao Liub, Juan Andres Cardosoc, Rainer 5 Schultze-Kraftc, Guodao Liub, Lijuan Luoa,*, Zhijian Chenb,* 6 7 a Hainan Key Laboratory for Sustainable Utilization of Tropical Bioresources, College of 8 Tropical Crops, Hainan University, Haikou, 570110, China 9 b Institute of Tropical Crop Genetic Resources, Chinese Academy of Tropical Agricultural 10 Sciences, Haikou, 571101, China 11 c Alliance of Bioversity International and International Center for Tropical Agriculture, 12 Cali, A.A.6713, Colombia 13 14 * Correspondence: 15 Zhijian Chen 16 Email: jchen@scau.edu.cn 17 Lijuan Luo 18 Email: luoljd@126.com 19 20 Highlights (no more than 85 characters including spaces) 21 • At least seven SgMDH genes existed in stylo. 22 • SgMDH genes were regulated by nutrient and metal stresses in stylo root. 23 • SgMDH proteins displayed higher catalytic efficiency towards OAA than malate. 24 • Activities of the recombinant SgMDH proteins were affected by metal ions. 25 26 1 27 Abstract (no more than 250 words) 28 Malate dehydrogenase (MDH, EC 1.1.1.37) is a key enzyme that catalyzes a reversible 29 NAD-dependent dehydrogenase reaction from oxaloacetate (OAA) to malate. Although 30 MDH has been documented to participate in cellular metabolism and redox homeostasis 31 in plants, the roles of MDH members in the tropical legume Stylosanthes guianensis 32 (stylo) remain less definitive. In this study, except SgMDH1 that had been previously 33 characterized, six novel MDH genes were isolated from stylo, which were then 34 designated as SgMDH2 to SgMDH7. All of the SgMDH proteins possessed the common 35 features of NAD binding, dimerization interface and substrate binding sites. Expression 36 analysis showed that three SgMDHs exhibited preferential expressions in leaf, and one 37 SgMDH was mainly expressed in root. Furthermore, SgMDHs were regulated by nutrient 38 deficiencies in stylo roots, especially for phosphorus (-P) and potassium (-K) deficiencies. 39 Differential responses of SgMDHs to trace metal stress and heavy metal toxicity were 40 observed in stylo roots, suggesting the involvements of SgMDHs in stylo response to 41 metal stresses. The six novel SgMDHs were subsequently expressed and purified from 42 Escherichia coli to analyze their biochemical properties. Although SgMDHs exhibited 43 variations in subcellular localizations, each SgMDH protein displayed a high level of 44 catalytic efficiency towards OAA and NADH, but a low level of catalytic efficiency 45 towards malate and NAD+. In addition, the activities of recombinant SgMDH proteins 46 were pH-dependent and temperature-sensitive, and exhibited differential regulations by 47 various metal ions. These results together suggest the potential roles of SgMDHs in stylo 48 coped with nutrient and metal stresses. 49 50 Keywords: Malate dehydrogenase, Malate, Gene expression, Enzymatic properties, 51 Nutrient deficiency, Metal stress, Stylosanthes guianensis 52 53 Abbreviations: 54 GST, glutathione S-transferase; IPTG, isopropylthio-D-galactoside; MDH, malate 55 dehydrogenase; OAA, oxaloacetate; ORF, open reading frame; PVDF, polyvinylidene 56 difluoride; qRT-PCR, quantitative real-time polymerase chain reaction; SDS-PAGE, 57 sodium dodecyl sulfate polyacrylamide gel electrophoresis. 2 58 1. Introduction 59 In plants, malate dehydrogenase (MDH, EC 1.1.1.37) is a key enzyme of the 60 malate-oxaloacetate (OAA) shuttle and participates in various metabolic pathways, 61 including respiration, photosynthesis and energy generation (Nunes-Nesi et al., 2005; 62 Imran et al., 2016; Ma et al., 2018). MDH catalyzes a reversible reaction from OAA to 63 malate, accompanied by a redox reaction between NADH and NAD+, which are the 64 essential components in electron transport chain (Gietl, 1992; Hadži-Tašković Šukalović 65 et al., 2011; Imran et al., 2016). The reaction direction catalyzed by MDH is mainly 66 depended on the ratio of substrate and product, redox status and the changing 67 environment of plant growth (Tomaz et al., 2010; Chen et al., 2015). Furthermore, MDH 68 and its catalytic product, malate, have been documented to participate in the regulation of 69 root growth (van der Merwe et al., 2009), leaf respiration (Tomaz et al., 2010) and 70 embryo development (Beeler et al., 2014). In addition, malate also plays central roles in 71 increasing phosphorus (P) acquisition, symbiotic nitrogen (N) fixation and aluminum (Al) 72 toxicity tolerance (Schulze et al., 2002). 73 To date, a variety of MDH homologues have been identified from different plant 74 species, such as Arabidopsis (Tomaz et al., 2010; Beeler et al., 2014), soybean (Glycine 75 max) (Chen et al., 2011; Zhu et al., 2021), alfalfa (Medicago sativa) (Miller et al., 1998; 76 Tesfaye et al., 2001), cotton (Gossypium hirsutum) (Wang et al., 2015; Imran et al., 2016) 77 and apple (Malus domestica) (Yao et al., 2011a; Ma et al., 2018). MDHs exhibit various 78 subcellular localizations, including cytoplasm, mitochondria, glyoxisome and peroxisome, 79 serving various roles in plant metabolisms (Gietl, 1992). For example, nine putative 80 MDH members have been identified in Arabidopsis, including eight NAD-MDHs and 81 one NADP-MDH. Among the eight NAD-MDHs, two members are mitochondrial MDHs 82 (mMDHs), two members are peroxisomal MDHs (pMDHs), one member is plastidial 83 MDH (pdNAD-MDH), and the remaining three members are thought to be cytosolic 84 MDHs (cyMDHs) (Beeler et al., 2014). The mitochondrial MDHs in Arabidopsis are 85 involved in leaf respiration and photorespiration, thereby affecting plant growth (Tomaz 86 et al., 2010). Arabidopsis peroxisomal MDHs are demonstrated to participate in 87 -oxidation of fatty acids and CO2 release in the photorespiratory pathway 88 (Pracharoenwattana et al., 2007; Cousins et al., 2008). The plastidial NAD-dependent 3 89 MDH is critical for redox homeostasis and is involved in embryo development in 90 Arabidopsis (Beeler et al., 2014). Arabidopsis cytosolic MDH is found to catalyze malate 91 synthesis, which subsequently contributes to Al detoxification and facilitates P 92 acquisition (Wang et al., 2010). In addition, cell wall-localized MDH in maize (Zea mays) 93 is proposed to provide NADH for the catalytic reaction of cell wall-bound peroxidases 94 (PODs) as reducing equivalent, which further participate in phenolic metabolism 95 (Hadži-Tašković Šukalović et al., 2011). Recently, it has been documented that MDHs in 96 nodules are probably involved in regulation of symbiotic N fixation and nodule growth in 97 soybean (Zhu et al., 2021). 98 Nutrient stress, such as nutrient deficiency and trace element excess, together with 99 heavy metal toxicity that inhibits plant growth and yield, has been found to regulate the 100 expressions of MDH genes (Wang et al., 2000; Uhde-Stone et al., 2003; Armengaud et al., 101 2009; Abd El-Moneim et al., 2015). For example, deficiencies of a set of macronutrients, 102 such as N, P and potassium (K), regulate the expressions of various MDH homologues in 103 Arabidopsis, soybean, cotton and white lupin (Lupinus albus) (Uhde-Stone et al., 2003; 104 Armengaud et al., 2009; Wang et al., 2015; Vengavasi et al., 2016; Zhu et al., 2021). 105 Furthermore, the expressions of MDH homologues are also regulated by excess copper 106 (Cu) and manganese (Mn) stresses and Al toxicity (Yang et al., 2012; Abd El-Moneim et 107 al., 2015; Chen et al., 2015). In addition, changes in enzyme activities of MDH under 108 phosphate (Pi) starvation as well as metal stresses are observed in soybean, Lupinus 109 angustifolius, maize and rape (Brassica napus) (Ligaba et al., 2004; Le Roux et al., 2006, 110 2008; Hadži-Tašković Šukalović et al., 2011; Vengavasi et al., 2016). Therefore, studying 111 the comprehensive changes in expression profiles and enzyme properties of MDHs 112 during stress conditions will help to provide new insights into the roles of MDHs in 113 plants coped with nutrient and metal stresses. 114 Stylosanthes guianensis (stylo), an important forage legume with high quality and high 115 yield, is commonly used for livestock nutrition and soil improvement in tropical and 116 subtropical areas (Chandra, 2009; Guo et al., 2019). As it originates from the tropics, 117 stylo exhibits great adaptability to acid soil-based nutrient and metal stresses (e.g., P 118 deficiency, Al and Mn toxicity), which is possibly attributed to the fine modification of 119 root characteristics through involvements of a variety of candidate genes (Sun et al., 2014; 4 120 Chen et al., 2015; Liu et al., 2016; Jiang et al., 2018; Chen et al., 2021). Among those 121 genes, a gene encoding MDH (SgMDH1) was found to participate in malate synthesis and 122 exudation in stylo roots, alleviating Mn toxicity by decreasing Mn accumulation (Chen et 123 al., 2015). Furthermore, several MDH homologues in stylo had also been found to be 124 regulated by low P availability, and Al and Mn toxicity through transcriptomic analyses 125 (Jiang et al., 2018; Jia et al., 2020; Chen et al., 2021). Due to the critical roles of MDHs 126 in plant metabolic processes, it is of considerable importance to investigate the SgMDHs 127 expression profiles and biochemical properties in response to nutrient and metal stresses 128 in stylo. Thus, in this study, six novel SgMDHs were isolated from stylo, and 129 transcriptional profiles of SgMDHs in response to nutrient and metal stresses were 130 investigated by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. 131 Enzymatic properties of the six novel SgMDH proteins were further investigated. 132 133 2. Materials and methods 134 2.1. Isolation of S. guianensis MDHs 135 Except for SgMDH1 (GenBank accession no. KJ123727), which has been previously 136 characterized from stylo (Chen et al., 2015), six novel genes encoding MDHs were 137 obtained from the sequencing data of transcriptome analyses (Jiang et al., 2018; Jia et al., 138 2020; Chen et al., 2021). The coding sequence of each MDH gene was amplified from a 139 full-length cDNA library constructed using P-deficient roots of stylo (Sun et al., 2013). 140 The six novel MDHs were named from SgMDH2 to SgMDH7. The sequence data of the 141 six novel SgMDHs were deposited in GenBank under accession numbers OK188912, 142 OK188913, OK188914, OK188915, OK188916 and OK188917 for SgMDH2 to 143 SgMDH7, respectively. All of the SgMDH members contained the conserved Ldh_1_N 144 (PF00056) and Ldh_1_C (PF02866) domains according to the Pfam database 145 (http://pfam.xfam.org/search/sequence). Phylogenetic analysis was performed by MEGA 146 4.1. 147 2.2. Plant growth and treatments 148 In this study, the elite stylo cultivar ‘RY2’ that is commonly cultivated in South China 149 (Tang et al., 2009) was used. After seed germination for 3 d, seedlings were transferred to 150 Hoagland solution containing 3 mM KNO3, 2 mM Ca(NO3)2, 0.25 mΜ KH2PO4, 0.5 mM 5 151 MgSO4, 5 μM MnSO4, 0.5 μM ZnSO4, 1.5 μM CuSO4, 0.09 μM (NH4)6Mo7O24, 23 μM 152 NaB4O7 and 80 μM Fe-Na-EDTA as previously described (Chen et al., 2021). The pH of 153 the nutrient solution was adjusted to 5.8 every 2 d. Plants were grown in a greenhouse 154 with a photoperiod of about 13 h and temperatures ranging from 25 to 35 ℃, under 155 natural sunlight condition. After 21 d of growth, leaf and root were harvested for RNA 156 extraction and gene expression analysis. 157 To analyze the transcripts of SgMDHs under nutrient deficient conditions, 158 fourteen-day-old seedlings grown in half strength Hoagland solution were separately 159 transplanted into fresh nutrient solution without N, P and K application according to Qin 160 et al. (2012). For -N treatment, KNO3, Ca(NO3)2 and (NH4)6Mo7O24 were replaced by 161 K2SO4, CaSO4 and Na2MoO4, respectively. For -P treatment, KH2PO4 was replaced by 162 K2SO4. For -K treatment, KNO3 and KH2PO4 was replaced by Ca(NO3)2 and NaH2PO4, 163 respectively. Plants grown in full-strength nutrient solution were used as the control 164 (CON). After 7 d treatments, roots were harvested for RNA extraction and gene 165 expression analysis. 166 To detect the expressions of SgMDHs in stylo in response to excess trace elements, 167 fourteen-day-old seedlings were separately transplanted into fresh solution (pH 5.0) 168 containing 800 μM Fe-Na-EDTA, 400 μM MnSO4, 10 μM CuSO4 or 20 μM ZnSO4 for 7 169 d, which were set as individual Fe, Mn, Cu or Zn stresses, respectively. To assay the 170 transcripts of SgMDHs in stylo exposed to heavy metal treatments, fourteen-day-old 171 seedlings were transplanted into a 0.5 mM CaCl2 solution (pH 4.5) individually supplied 172 with 0, 100 μM AlCl3, 40 μM CdCl2 or 20 μM LaCl3 for 2 d, which were set as the 173 control (CON), Al, Cd or La stresses, respectively. Roots with the above treatments were 174 separately harvested for RNA extraction and gene expression analysis. For all 175 experiments, a hydroponic box containing three seedlings was set as one biological 176 replicate, and three biological replicates were included. 177 2.3. RNA extraction and qRT-PCR analysis 178 Total RNA was extracted using Trizol reagent according to the manufacturer’s 179 instructions (TIANGEN Biotech, China). First strand cDNA was synthesized using the 180 HiScript III cDNA synthesis kit (Vazyme, China) from 2 µg total RNA after treated with 181 DNase I. qRT-PCR reaction was performed using SYBR Green Master mix (Vazyme, 6 182 China) and monitored by QuantStudio™ 6 Flex Real-Time system (Thermo Fisher, USA). 183 The primers for qRT-PCR analysis are listed in Supplementary Table S1. Gene expression 184 was calculated by comparison with a standard curve using cycle threshold value. Relative 185 expression level was calculated by the ratio of expression of each SgMDH to that of the 186 housekeeping gene, SgEF-1a, according to Chen et al. (2021). Gene expression analysis 187 included three biological replications. 188 Subcellular Localization of SgMDH proteins 189 To assay the subcellular localization of SgMDH proteins, the open reading frame (ORF) 190 of each SgMDH gene was amplified using SgMDHs-GFP-F/R primers (Supplementary 191 Table S1). The amplified product was subcloned into the N-terminal of the green 192 fluorescent protein (GFP) of the binary vector pBWA(V)HS-ccdb-GLosgfp as described 193 by Li et al (2021). Arabidopsis mesophyll protoplasts from 4-week-old seedlings were 194 prepared, and then used to analyze the subcellular localization of each SgMDH protein 195 according to Yoo et al. (2007). Each 35S:SgMDH-GFP construct was transiently 196 expressed in protoplasts, and empty vector was used as a control. The GFP fluorescence 197 in Arabidopsis mesophyll protoplasts was imaged using a Nikon C2 confocal microscopy 198 (Nikon, Japan) at 488 nm for GFP and at 640 nm for chloroplast autofluorescence. 199 2.4. Expression, purification and biochemical analyses of SgMDHs 200 The ORF of each SgMDH gene was amplified using SgMDHs-GST-F/R primers 201 (Supplementary Table S1). The resulting amplified product was digested with the 202 corresponding restriction endonuclease (Takara, Japan), and then subcloned into the 203 pGEX6P-3 vector containing a glutathione S-transferase (GST) tag (GE Healthcare, 204 USA). Six GST:SgMDH constructs and the empty vector were separately transformed 205 into the Escherichia coli strain BL21. The recombinant proteins were induced by 206 supplying 0.5 mM isopropylthio-D-galactoside (IPTG) for 4-6 h, and then purified from 207 the E. coli extracts using BeaverBeadsTM (Beaver, China) according to the manufacturer’s 208 instructions. The resulting recombinant SgMDH proteins were analyzed by the sodium 209 dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (12% acrylamide) 210 combined with Coomassie Blue R-250 staining and western blot analysis. Protein 211 concentrations were measured by the Coomassie Brilliant Blue method (Bradford, 1976). 212 MDH activity of each recombinant protein was determined according to Chen et al. 7 213 (2015) with some modifications. To detect MDH activity in OAA reduction, 214 approximately 0.2 μg recombinant MDH protein was incubated in reaction buffer 215 containing 100 mM Tris-HCl (pH 8.0), 5 mM ethylenediaminetetraacetic acid (EDTA), 216 0.025-0.3 mM NADH and 0.02-3 mM OAA. The reaction was started by protein addition. 217 MDH activity towards OAA reduction was spectrophotometrically detected at 340 nm by 218 measuring the levels of NADH within 1 min with a spectrophotometer (UV-2010, Hitachi, 219 Japan), and is expressed as the amount of enzyme required to catalyze the oxidation of 220 NADH (extinction coefficient 6.22 mM-1 cm-1) per minute. To assay MDH activity in 221 malate oxidation, 0.2 μg recombinant MDH protein was incubated in reaction buffer 222 containing 100 mM Tris-HCl (pH 9.0), 5 mM EDTA, 0.125-3 mM NAD+ and 0.1-10 mM 223 L-malate. MDH activity towards malate oxidation was detected at 340 nm by measuring 224 the levels of NADH within 1 min and calculated as the amount of enzyme required to 225 catalyze the reduction of NADH per minute. 226 To detect the activities of SgMDH proteins in catalyzing the reversible reaction using 227 NADPH and NADP+, the recombinant SgMDH proteins were incubated in 100 mM 228 Tris-HCl (pH 8.0) containing 25 mM dithiothreitol (DTT), 10 μM thioredoxin and 2 mM 229 EDTA for 30 min at 30°C according to Ferte et al. (1986) with some modifications. The 230 DTT-thioredoxin-activated enzyme was then used to analyze MDH activity in OAA 231 reduction using NADPH. Approximately 0.4 μg recombinant MDH protein was 232 incubated in reaction buffer containing 100 mM Tris-HCl (pH 8.0), 5 mM EDTA, 0.3 233 mM NADPH and 3 mM OAA. For MDH activity in malate oxidation using NADP+, 0.4 234 μg recombinant MDH protein was incubated in reaction buffer containing 100 mM 235 Tris-HCl (pH 8.0), 5 mM EDTA, 1 mM NADP+, 10 mM L-malate, 20 mM glutamate and 236 1 unit glutamate:oxaloacetate aininotransferase. MDH activity was detected at 340 nm by 237 measuring the levels of NADH within 1 min. 238 The Km and Vmax values were determined by Lineweaver-Burke plot analysis. To 239 determine the optimal reaction pH, 0.2 μg recombinant MDH protein was incubated in 240 100 mM Tris/HCl buffer (pH 7.0-9.5) using 1 mM OAA as a substrate. For temperature 241 optimization, MDH activity was detected at different temperatures ranging from 30 to 242 60 °C. To detect the effects of various metal ions on the catalyzed activity, MDH activity 243 was analyzed at the reaction buffer supplied with 1 mM FeSO4, MnCl2, CuSO4, ZnSO4, 8 244 MgCl2, or AlCl3. Analyses of MDH activity were conducted in triplicate. As the control, 245 the GST protein from the empty vector displayed no detectable MDH activity. 246 2.5. Western blot analysis 247 After the purified SgMDH proteins were resolved by 12% SDS-PAGE, they were further 248 electrophoretically transferred to a polyvinylidene difluoride (PVDF) membrane (Merck 249 Millipore, USA) according to Chen et al. (2015). The hybridization of the PVDF 250 membrane was started by addition of the anti-GST antibody (1:5,000 dilution) in blotting 251 buffer containing 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.1% (v/v) Tween 20 and 3% 252 (w/v) nonfat dry milk for 1 h. After that, the PVDF membrane was hybridized with the 253 alkaline phosphatase-tagged secondary antibody (1:2,000 dilution) in the fresh blotting 254 buffer for 1 h. The signal of target protein in the PVDF membrane was observed after 255 alkaline phosphatase reaction according to Chen et al. (2015). 256 2.6. Statistical analyses 257 Data analysis was performed by Microsoft Excel 2010 (Microsoft Company, USA). 258 Assays with one-way ANOVA and Student’s t-test were performed by the SPSS program 259 v13.0 (SPSS Institute, USA). 260 261 262 3. Results 263 3.1. Identification and characterization of MDHs in S. guianensis 264 In this study, six novel MDH genes were isolated from stylo, which were designated as 265 SgMDH2 to SgMDH7. Together with SgMDH1 that had been cloned in a previous study 266 (Chen et al., 2015), the ORF of the seven SgMDHs ranged from 999 to 1317 nucleotides. 267 The deduced amino acid sequences of SgMDHs contained 330 to 438 amino acid 268 residues, with calculated molecular masses differing from 33.6 to 48.9 kDa 269 (Supplementary Table S2). The common features, including NAD binding, dimerization 270 interface and substrate binding sites, were observed among the amino acid sequences of 271 SgMDHs; the SgMDHs possessed the conserved amino acid sequences of MDH 272 homologues from Arabidopsis, soybean, alfalfa and cotton (Fig. 1 and Supplementary Fig. 273 S1). Furthermore, the deduced amino acid sequences of SgMDH1 to SgMDH7 shared 274 15.7 to 90.7% homology identities with AtcMDH1 in Arabidopsis, 13.2 to 57.2% 9 275 homology identities with MsneMDH in alfalfa, 16.8 to 90.7% homology identities with 276 GmMDH12 in soybean and 16.9 to 83.4% homology identities with GhmMDH1 in 277 cotton (Supplementary Fig. S2). In addition, the seven SgMDHs shared 13.0 to 97.3% 278 homology identities with each other (Supplementary Fig. S2). According to WoLF 279 PSORT program, most SgMDHs were predicted to be localized to the cytoplasm, except 280 SgMDH4 and SgMDH5 that were localized to the chloroplast (Supplementary Table S2). 281 A phylogenetic tree was constructed using plants MDH homologues, including 282 SgMDHs from stylo and all MDH members from Arabidopsis. Results showed that plant 283 MDH members could be classified into three major groups (Fig. 1B). Among them, two 284 SgMDHs (SgMDH2 and 5) together with three MDH members from Arabidopsis, four 285 MDH members from rice, MsAAB99757 from alfalfa, PsAAC28106 from pea (Pisum 286 sativum), SlNM001247072 from tomato (Solanum lycopersicum) and ZmNP001148518 287 from maize were classified into group I. Group II included only SgMDH1 with high 288 similarity with GmNP001341093 from soybean and MsAAB99755 from alfalfa. Group 289 III contained four SgMDHs (SgMDH3, 4, 6 and 7), three MDH members from 290 Arabidopsis, two MDH members from rice, two MDHs from white lupin, MdDQ221207 291 from apple and other MDH members from alfalfa, cotton, maize, Prunus persica, 292 Plantago major and Ananas comosus (Fig. 1B). 293 3.2. Expressions of SgMDHs under nutrient deficiency 294 In this study, we first detected the expressions of SgMDHs in leaf and root of 21-day-old 295 stylo seedlings by qRT-PCR analysis. Variations in SgMDHs expressions were observed 296 in leaf and root of stylo. Among them, three SgMDHs (SgMDH1, 4 and 5) showed higher 297 expressions in leaf, while SgMDH6 were mainly expressed in root (Fig. 2). In addition, 298 the transcripts of SgMDH2, SgMDH3 and SgMDH7 were constitutively expressed in leaf 299 and root (Fig. 2). 300 Subsequently, we analyzed the expression profiles of SgMDHs in response to nutrient 301 deficiencies, including N, P and K deficiencies. Results showed that SgMDHs exhibited 302 differential regulations by nutrient deficiencies in stylo roots, and all SgMDHs responded 303 to at least one nutrient deficiency treatment (Fig. 3). Among them, the transcript of 304 SgMDH3 was increased by -N treatment, while the remaining SgMDHs did not respond 305 to N deficiency (Fig. 3). Four SgMDHs (SgMDH1, 3, 6 and 7) and two SgMDHs 10 306 (SgMDH4 and SgMDH5) were increased and decreased by -P treatment, respectively (Fig. 307 3). In addition, six out of seven SgMDHs responded to K deficiency, including three 308 up-regulated SgMDHs (SgMDH4, 5 and 7) and three down-regulated SgMDHs (SgMDH1, 309 2 and 3). Interestingly, SgMDH2 and SgMDH6 exhibited specific responses to -K and -P 310 treatments, respectively (Fig. 3). 311 3.3. Responses of SgMDHs to metal stress 312 To examine the potential regulation of SgMDHs by exposure to trace metals, transcript 313 levels of these genes were assessed in roots of stylo grown in nutrient solution supplied 314 with excess Fe, Mn, Cu and Zn for 7 days. Diverse responses of SgMDHs were observed: 315 under excess Fe treatment, transcripts of five SgMDHs (SgMDH1, 2, 3, 4 and 7) were 316 increased in roots compared to the control (CON) (Fig. 4). Similarly, four SgMDHs 317 (SgMDH2, 3, 5 and 7) were also enhanced by excess Zn treatment. In contrast, three 318 SgMDHs (SgMDH3, 4 and 7) and two SgMDHs (SgMDH5 and SgMDH7) were 319 suppressed by excess Mn and Cu treatments, respectively, while the remaining SgMDHs 320 did not respond to these two trace metals. Interestingly, the transcripts of SgMDH6 were 321 not affected by any of the trace metals (Fig. 4). 322 We further investigated the responses of SgMDHs in stylo roots to heavy metal toxicity, 323 including Al, Cd (cadmium) and La (lanthanum) treatments. As shown in Fig. 5, 324 transcripts of SgMDHs were regulated by at least one of the heavy metal treatments. 325 Among them, five SgMDHs (SgMDH1, 2, 3, 4 and 6) were up-regulated by Al treatment 326 in roots, whereas SgMDH5 and SgMDH7 were not affected by Al. In contrast to the 327 regulation by Al toxicity, Cd treatment mainly resulted in inhibiting the transcripts of six 328 SgMDHs (SgMDH1, 3, 4, 5, 6 and 7), except SgMDH2. Besides, four SgMDHs (SgMDH4, 329 5, 6 and 7) were suppressed and only SgMDH3 was enhanced by La treatment. 330 Interestingly, SgMDH2 exhibited specific up-regulation by Al but not by Cd and La 331 treatments (Fig. 5). 332 3.4. Subcellular localization of SgMDH proteins 333 As the subcellular localization of SgMDH1 had been previously studied (Chen et al., 334 2015), the coding sequence of the six novel SgMDHs was separately cloned and fused 335 with the N-terminal of the GFP reporter gene and transiently expressed in Arabidopsis 336 mesophyll protoplasts. Results showed that the fluorescence of GFP in cells transformed 11 337 with the empty vector was found in many areas in protoplast cells, particularly in the 338 cytoplasm. GPF signals of SgMDH2 and SgMDH 5 were detected in the organelles, 339 probably in the peroxisome. SgMDH3 was mainly localized in the cytosol, while 340 SgMDH4, SgMDH6 and SgMDH7 were observed in both the cytosol and plasma 341 membrane. (Fig. 6). 342 3.5. Enzymatic properties of SgMDHs 343 Subsequently, the biochemical properties of six novel SgMDH proteins were detected by 344 fused with a GST tag. After expressed in E. coli, the recombinant SgMDH proteins were 345 purified from the lysates. After verification by SDS-PAGE and western-blot analyses 346 (Supplementary Fig. S3), the purified SgMDH proteins were further used to test their 347 biochemical properties in vitro, including catalytic properties and pH-dependent, 348 temperature-sensitive and metal ion-sensitive activities. The catalytic properties were 349 investigated by calculating Km, Vmax and catalytic efficiency (Kcat/Km) values. The Km 350 values of SgMDHs ranged from 0.022 to 0.121 mM for OAA and from 0.168 to 1.144 351 mM for malate, and the Vmax values of SgMDHs differed from 455.2 to 939.4 nmol -1min-1 352 for OAA and 65.8 to 135.1 nmol-1min-1 for malate (Table 1). Furthermore, the Vmax values 353 of SgMDHs for OAA and NADH were higher than those for malate and NAD+, 354 respectively. Similarly, all purified SgMDH proteins showed a higher level of catalytic 355 efficiency (Kcat/Km) towards OAA and NADH, and a lower catalytic efficiency towards 356 malate and NAD+. For example, the catalytic efficiency (Kcat/Km) of SgMDH2 was 357 475,000 for OAA and 126,017 for NADH, which was significantly higher than that for 358 malate and NAD+ (Table 1). In addition, we also detected the activities of SgMDH 359 proteins in catalyzing the reversible reaction using NADPH and NADP+. Results showed 360 that compared to NADH and NAD+, each SgMDH protein exhibited a very small activity 361 by using NADPH and NADP+ to catalyze OAA reduction and malate oxidation, 362 respectively (Supplementary Fig. S4). 363 As shown in Figs. 7 and 8, using OAA as the substrate, the enzyme activities of 364 recombinant SgMDH proteins were pH-dependent and temperature-sensitive. Increasing 365 activities of recombinant proteins were observed with increasing pH of the reaction buffer, 366 then the MDH activities decreased with increasing pH value (Fig. 7). The optimum pH 367 for activities of SgMDH2 and SgMDH6 was 7.5, while the optimum pH for SgMDH3, 12 368 SgMDH4 and SgMDH7 activities was 8.5. SgMDH5 exhibited the highest level of 369 activity at pH of 8.0 (Fig. 7). Similar to the trends of MDH activities under different pH 370 conditions, MDH activities increased and then decreased with increasing temperature of 371 the reaction buffer (Fig. 8). The optimum temperature for activities of SgMDH2, 372 SgMDH4 and SgMDH6 was 40, 50 and 50 ℃, respectively, while SgMDH3, SgMDH5 373 and SgMDH7 exhibited the highest level of activity at 45 ℃ (Fig. 8). 374 In addition, SgMDH activities were differentially influenced by metal ions, including 375 Fe2+, Mn2+, Cu2+, Zn2+, Mg2+ and Al3+, when using OAA as the substrate (Table 2). 376 Among them, SgMDH2 activities were increased by Mn2+ and Zn2+, but were inhibited 377 by Cu2+ and Al3+, while SgMDH3 activities were decreased by more than 20% when Fe2+, 378 Cu2+, Zn2+ and Al3+ were applied. SgMDH4 activities were inhibited by all tested ions, 379 except Cu2+. SgMDH5 activities were increased by Mn2+, but were inhibited by Fe2+ and 380 Cu2+, while SgMDH6 activities were decreased by all tested ions, except Mg2+ (Table 2). 381 SgMDH7 activities were more than 20% increased by Mn2+ and Mg2+, but more than 382 40% inhibited by Fe2+, Cu2+ and Zn2+. Interestingly, the activities of all SgMDHs were 383 inhibited by Cu2+ and Fe2+, except SgMDH2 and SgMDH4, respectively. Furthermore, 384 the activities of most SgMDHs were not affected by Mg2+, except SgMDH7 (Table 2). 385 386 4. Discussion 387 The ability to catalyze a reversible reaction from OAA to malate by the NAD-dependent 388 MDH has been well elucidated in plants, and thus MDH plays a role in various 389 physiological processes (Tomaz et al., 2010; Beeler et al., 2014; Ma et al., 2018). The 390 number of MDH members differs among various plant species, suggesting they may 391 possess diverse roles in cellular metabolic pathways. Although one cytosol-localized 392 MDH member in stylo, SgMDH1, had been previously characterized to function in 393 catalyzing malate synthesis (Chen et al., 2015), data are still limited on the potential roles 394 of different MDH members in this important tropical legume due to lack of genome 395 information. Thus, in this study, six novel SgMDH genes were further cloned and 396 functionally characterized. Including SgMDH1, all seven SgMDH members contained 397 the conserved features of plant NAD-MDH proteins (Fig. 1 and Supplementary Fig. S1), 398 suggesting that SgMDH members are putative MDH proteins and possess roles in 13 399 catalyzing the reversible reaction from OAA to malate. 400 According to the phylogenetic analysis, in group I, SgMDH2 and SgMDH5 exhibited 401 high similarity to Arabidopsis peroxisomal MDHs (AtNP1979863 and AtNP196528), 402 which were demonstrated to be involved in -oxidation of fatty acids and CO2 release in 403 the photorespiratory pathway (Pracharoenwattana et al., 2007; Cousins et al., 2008). 404 Furthermore, alfalfa neMDH (MsAAB99757) in group I was suggested to function in 405 malate synthesis, contributing to increase Al tolerance and facilitate Pi acquisition 406 (Tesfaye et al., 2001, 2003). In group II, the cytosol-localized SgMDH1 had been 407 previously characterized to be involved in malate synthesis, improving Mn tolerance by 408 decreasing Mn accumulation in stylo (Chen et al., 2015). The closest homologue of 409 SgMDH1, GmMDH12 (GmNP001341093), was proposed to catalyze malate synthesis; it 410 regulates nodule growth in soybean under P deficient condition (Zhu et al., 2021). In 411 addition, AtNP564625 (mMDH1) and AtNP001078156 (mMDH2) from Arabidopsis in 412 group II were involved in leaf respiration and photorespiration, thereby affecting plant 413 growth (Tomaz et al., 2010). Furthermore, mMDH (SlNM001247072) in tomato was 414 found to participate in altering photosynthetic activity and aeration as well as root growth 415 in tomato (Nunes-Nesi et al., 2005; van der Merwe et al., 2009). Four SgMDHs 416 (SgMDH3, 4, 6 and 7), together with three MDHs from Arabidopsis and two MDHs 417 (LaAF459645 and LaAF459646) from white lupin, were divided into group III, clustering 418 with apple MdcyMDH (MdDQ221207), which participated in plant growth and tolerance 419 to cold and salt stresses (Yao et al., 2011a,b), and OscMDH (OsNP001064860), which 420 was involved in starch synthesis and seed development in rice (Teng et al., 2019). 421 Therefore, stylo SgMDH proteins are probably involved in diverse metabolic processes. 422 Subsequent expression analysis showed that SgMDHs exhibited various responses to 423 nutrient deficiencies in stylo roots (Fig. 3). It has been observed that nutrient deficiencies 424 regulate the expressions of many metabolism-related genes, including MDH (Wang et al., 425 2000; Scheible et al., 2004; Armengaud et al., 2009). For example, a variety of MDH 426 genes were found to be regulated by P deficiency in Arabidopsis, soybean and white lupin 427 (Hammond et al., 2003; Uhde-Stone et al., 2003; Wu et al., 2003; Vengavasi et al., 2017; 428 Zhu et al., 2021). Enhanced expressions of LaMDH1 were observed in P-deficient tissues 429 of white lupin, especially in normal and proteoid roots, facilitating acclimation to low-P 14 430 stress through regulation of carbon metabolism (e.g., organic acid synthesis) (Uhde-Stone 431 et al., 2003). In addition, five out of 16 GmMDH genes were significantly enhanced by 432 low-P stress in soybean nodules, suggesting the involvement of GmMDHs in nodule 433 growth during Pi starvation (Zhu et al., 2021). Furthermore, overexpressions of MDH 434 homologues could enhance malate synthesis and exudation, thereby increasing P 435 availability in the rhizosphere (Tesfaye et al., 2003; Wang et al., 2010; Lü et al., 2012). In 436 this study, thus, the four low-P stress-enhanced SgMDHs (SgMDH1, 3, 6 and 7) were 437 probably involved in increasing P utilization of stylo. Besides, we also found that the 438 expressions of SgMDHs were regulated by N and K deficiencies in stylo roots (Fig. 3). 439 Similar regulations of MDH homologues have been reported in other plants, such as 440 Arabidopsis and cotton (Scheible et al., 2004; Armengaud et al., 2009; Wang et al., 2015). 441 In addition to catalyze malate synthesis, changes in expressions of SgMDHs might 442 regulate nitrate and carbon metabolism as well as redox equilibrium in stylo response to 443 N and K deficiencies. The roles of SgMDHs in nutrient deficiencies require further study. 444 It has been demonstrated that MDH is involved in metal tolerance by catalyzing malate 445 synthesis in roots and subsequent exudation from roots to chelate metal ions, thereby 446 improving metal tolerance in plants (Tesfaye et al., 2001; Wang et al., 2010; Chen et al., 447 2015). In the present study, transcripts of SgMDHs in stylo roots were differentially 448 regulated by exposure to trace and heavy metals. Among them, five SgMDHs (SgMDH1, 449 2, 3, 4 and 7) and four SgMDHs (SgMDH2, 3, 5 and 7) were enhanced by excess Fe and 450 Zn, respectively, while four SgMDHs (SgMDH2, 3, 4 and 6) were up-regulated by Al 451 (Figs. 4 and 5). Similarly, MDH homologues were also found to respond to various metal 452 treatments, including Cu, Mn and Al (Kumari et al., 2008; Yang et al., 2012; Abd 453 El-Moneim et al., 2015; Chen et al., 2015). For example, the transcript of mMDH but not 454 cyMDH was increased by Al treatment in Citrus, which was associated with the tolerance 455 of Citrus to Al toxicity (Yang et al., 2012). In addition, mMDH1 in roots of Arabidopsis 456 was also found to be induced by Al (Kumari et al., 2008). Furthermore, overexpression of 457 neMDH in alfalfa and amdh in tobacco led to increased organic acid synthesis and 458 secretion, conferring Al tolerance in transgenic plants (Tesfaye et al., 2001; Wang et al., 459 2010). In our previous study, we found that both the transcripts of SgMDH1 and its 460 protein levels were increased under excess Mn in roots of the Mn-tolerant stylo genotype, 15 461 contributing to increase Mn tolerance (Chen et al., 2015). Therefore, these results suggest 462 the potential roles of SgMDHs in metal tolerance. 463 We thus investigated the biochemical properties of the recombinant SgMDH proteins, 464 catalyzed OAA reduction and malate oxidation. Among them, the Km values of the 465 remaining recombinant SgMDH3 and SgMDH6 for OAA and NADH were lower than 466 those for malate and NAD+ (Table 1), as the case observed in neMDH/cyMDH from 467 alfalfa, cell wall-associated and plasma membrane-bound MDHs from maize, MDH from 468 pineapple (Ananas comosus) and MDH from Aptenia cordifolia (Miller et al., 1998; 469 Cuevas and Podesta, 2000; Tripodi and Podesta, 2003; Hadži-Tašković Šukalović et al., 470 2011). In contrast, the values of Vmax and Kcat/Km of the six recombinant MDH proteins 471 were higher for the OAA reduction than for malate oxidation (Table 1). Similar results 472 were also found in MDH members from wheat, apple, maize, pineapple and A. cordifolia 473 (Cuevas and Podesta, 2000; Tripodi and Podesta, 2003; Ding and Ma, 2004; 474 Hadži-Tašković Šukalović et al., 2011; Yao et al., 2011b). However, alfalfa neMDH 475 exhibited higher Vmax values for malate and NAD + but lower Vmax values for OAA and 476 NADH, whereas the catalytic efficiency (Kcat/Km) for malate and NAD + was lower than 477 that for OAA and NADH (Miller et al., 1998). Although values of kinetic properties 478 differed in SgMDH proteins and other plant MDH members, which were probably due to 479 the recombinant proteins or native proteins used in the analyses, our results revealed that 480 SgMDH proteins displayed higher catalytic efficiency towards OAA than malate by using 481 NADH. 482 The optimum pH for catalyzed activities of the recombinant SgMDH proteins towards 483 OAA reduction ranged from 7.5 to 8.5 (Fig. 7). A wide range of optimum pH values for 484 MDH activities have been observed from 6.5 to 8.5 in maize, A. cordifolia, pineapple and 485 spinach (Spinacia oleracea), and basic pH values, such as 8 and 8.5, were the common 486 optimum pH for MDH activity (Hadži-Tašković Šukalović et al., 1999, 2011; Cuevas and 487 Podesta, 2000; Tripodi and Podesta, 2003; Cvetić et al., 2008). Furthermore, the optimum 488 temperature for activities of the recombinant SgMDH proteins ranged from 40 to 50 ℃ 489 (Fig. 8), which was similar to that for MDH members from soybean, A. cordifolia and 490 pineapple (Cuevas and Podesta, 2000; Tripodi and Podesta, 2003; Zhu et al., 2021). In 491 addition, increased or inhibited effects of metal ions on MDH activities were observed in 16 492 stylo (Table 2), and in soybean and maize as well. For example, activities of the cell 493 wall-associated MDH from maize were inhibited by Zn2+ and Cu2+, while soybean 494 GmMDH12 activities were reduced when exposed to Al3+, Fe2+, Ag+, Cu2+, Mn2+, Mg2+ 495 and Zn2+ (Hadži-Tašković Šukalović et al., 2011; Zhu et al., 2021). Furthermore, stylo 496 SgMDH1 activities had been previously found to be increased by Mn2+ and inhibited by 497 Fe2+ (Chen et al., 2015). Interestingly, increases in activities of the SgMDH1 and 498 SgMDH2 proteins exposed to Mn2+ and Zn2+ seemed to relate to the increasing gene 499 expressions under excess Mn and Zn stresses, respectively (Fig. 4 and Table 2) (Chen et 500 al., 2015). These results together suggest that SgMDH proteins possess different 501 properties due to their potential roles in cellular ion homeostasis and microenvironment 502 adaptation. 503 Although the seven SgMDH members in stylo exhibited similar structures, variations 504 in the expression patterns, subcellular localization and biochemical properties of 505 SgMDHs were observed in this study, suggesting these SgMDHs may have distinct or 506 redundant biological functions in various physiological processes. Similarly, although 507 mMDH1 and mMDH2 were mitochondrial MDH homologues in Arabidopsis, the 508 transcripts and protein levels of mMDH1 were higher than those of mMDH2 (Millar et al., 509 2001; Hruz et al., 2008; Lee et al., 2008). Furthermore, the two single T-DNA insertion 510 mutants, mmdh1 and mmdh2, had no detectable phenotypes, but the double mutant 511 mmdh1mmdh2 displayed small and slow growth; the leaf respiration rate of this double 512 mutant was higher than that of the wild type in both dark and light conditions (Tomaz et 513 al., 2010), suggesting that mMDH1 coordinates with mMDH2 regulated leaf respiration 514 and photorespiration in Arabidopsis. Besides, different enzymatic properties of MDH 515 members in the same plant were also observed, such as nodule-enhanced and cytosolic 516 MDHs from alfalfa, and cell wall-associated and plasma membrane-bound MDHs from 517 maize (Miller et al., 1998; Hadži-Tašković Šukalović et al., 1999, 2011). Therefore, 518 SgMDHs might possess various functions in the response of stylo to nutrient and metal 519 stresses. The biological significance of SgMDHs in these processes merits further 520 investigation. 521 522 Conclusion 17 523 In this study, a total of seven SgMDHs were characterized in stylo. The expressions of 524 SgMDHs were differentially regulated by nutrient deficiencies and metal stresses in stylo 525 roots. Despite various subcellular localizations, each recombinant SgMDH protein 526 displayed a higher level of catalytic efficiency towards OAA and NADH compared to 527 that towards malate and NAD+. The activities of recombinant SgMDH proteins were 528 pH-dependent and temperature-sensitive, and were differentially regulated by various 529 metal ions. This study suggests the involvements of SgMDHs in stylo adapted to nutrient 530 and metal stresses. 531 532 Authors contributions 533 ZC and LL designed the research. JS, XZ and PL performed the experiments and 534 analyzed the data. GL prepared the plant material for this work. JS and ZC wrote the 535 manuscript. JAC, RS and LL discussed and revised the manuscript. All authors have read 536 and approved the manuscript. 537 538 Conflicts of interest statement 539 The authors declare that they have no competing interests. 540 541 Acknowledgments 542 This work is supported by the National Natural Science Foundation of China (31801951, 543 31861143013), the Central Public-interest Scientific Institution Basal Research Fund for 544 CATAS (1630032020003), the Modern Agro-industry Technology Research System, and 545 the Integrated Demonstration of Key Techniques for the Industrial Development of 546 Featured Crops in Rocky Desertification Areas of Yunnan-Guangxi-Guizhou Provinces 547 (SMH2019-2021). 548 549 550 References 551 Abd El-Moneim, D., Contreras, R., Silva-Navas, J., Gallego, F.J., Figueiras, A.M., Benito, 552 C., 2015. On the consequences of aluminium stress in rye: repression of two 553 mitochondrial malate dehydrogenase mRNAs. Plant Biol. 17, 123-133. 18 554 Armengaud, P., Sulpice, R., Miller, A.J., Stitt, M., Amtmann, A., Gibon, Y., 2009. 555 Multilevel analysis of primary metabolism provides new insights into the role of 556 potassium nutrition for glycolysis and nitrogen assimilation in Arabidopsis roots. 557 Plant Physiol. 150, 772-785. 558 Beeler, S., Liu, H.C., Stadler, M., Schreier, T., Eicke, S., Lue, W.L., Truernit, E., Zeeman, 559 S.C., Chen, J., Kötting, O., 2014. Plastidial NAD-dependent malate dehydrogenase 560 is critical for embryo development and heterotrophic metabolism in Arabidopsis. 561 Plant Physiol. 164, 1175-1190. 562 Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram 563 quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 564 72, 248-254. 565 Chandra, A., 2009. Diversity among Stylosanthes species: Habitat, edaphic and 566 agroclimatic affinities leading to cultivar development. J. Environ. Biol. 30, 567 471-478. 568 Chen, Z.J., Song, J.L., Li, X.Y., Arango, J., Cardoso, J.A., Rao, I., Schultze-Kraft, R., 569 Peters, M., Mo, X.H., Liu, G.D., 2021. Physiological responses and transcriptomic 570 changes reveal the mechanisms underlying Stylosanthes guianensis adaption to 571 phosphorus deficiency. BMC Plant Biol. 10.1186/s12870-021-03249-2. 572 Chen, Z.J., Sun, L.L., Liu, P.D., Liu, G.D., Tian, J., Liao, H., 2015. Malate synthesis and 573 secretion mediated by a manganese-enhanced malate dehydrogenase confers 574 superior manganese tolerance in Stylosanthes guianensis. Plant Physiol. 167, 575 176-188. 576 Chen, Z.J., Cui, Q.Q., Liang, C.Y., Sun, L.L., Tian, J., Liao, H., 2011. Identification of 577 differentially expressed proteins in soybean nodules under phosphorus deficiency 578 through proteomic analysis. Proteomics 11, 4648-4659. 579 Cousins, A.B., Pracharoenwattana, I., Zhou, W., Smith, S.M., Badger, M.R., 2008. 580 Peroxisomal malate dehydrogenase is not essential for photorespiration in 581 Arabidopsis but its absence causes an increase in the stoichiometry of 582 photorespiratory CO2 release. Plant Physiol. 148, 786-795. 583 Cuevas, I.C., Podesta, F.E., 2000. Purification and physical and kinetic characterization of 584 an NAD+-dependent malate dehydrogenase from leaves of pineapple (Ananas 19 585 comosus). Physiol. Plant. 108, 240-248. 586 Cvetić, T., Veljović-Jovanović, S., Vučinić, Ž., 2008. Characterization of NAD-dependent 587 malate dehydrogenase from spinach leaves. Protoplasma 232, 247-253. 588 Ding, Y., Ma, Q.H., 2004. Characterization of a cytosolic malate dehydrogenase cDNA 589 which encodes an isozyme toward oxaloacetate reduction in wheat. Biochimie 86, 590 509-518. 591 Gietl, C., 1992. Malate dehydrogenase isoenzymes cellular locations and role in the flow 592 of metabolites between the cytoplasm and cell organelles. Biochim. Biophys. Acta 593 1100, 217-234. 594 Guo, P.F., Liu, P.D., Lei, J., Chen, C.H., Qiu, H., Liu, G.D., Chen, Z.J., Luo, L.J., 2019. 595 Improvement of plant regeneration and Agrobacterium-mediated genetic 596 transformation of Stylosanthes guianensis. Trop. Grassl.-Forrajes Trop. 7, 480-492. 597 Hadži-Tašković Šukalović, V., Vuletić M., Ignjatović-Micić, D., Vučinić, Ž., 1999. 598 Plasma membrane-bound malate dehydrogenase activity in maize roots. Protoplasma 599 207, 203-212. 600 Hadži-Tašković Šukalović, V., Vuletić, M., Marković, K., Vučinić, Z., 2011. Cell 601 wall-associated malate dehydrogenase activity from maize roots. Plant Sci. 181, 602 465-470. 603 Hammond, J.P., Bennett, M.J., Bowen, H.C., Broadley, M.R., Eastwood, D.C., May S.T., 604 Rahn, C., Swarup, R., Woolaway, K.E., White, P.J., 2003. Changes in gene 605 expression in Arabidopsis shoots during phosphate starvation and the potential for 606 developing smart plants. Plant Physiol. 132, 578-596. 607 Hruz, T., Laule, O., Szabo, G., Wessendorp, F., Bleuler, S., Oertle, L., Widmayer, P., 608 Gruissem, W., Zimmermann, P., 2008. Genevestigator V3: a reference expression 609 database for the meta-analysis of transcriptomes. Adv. Bioinformatics 2008, 420747. 610 Imran, M., Tang, K., Liu, J.Y., 2016. Comparative genome-wide analysis of the malate 611 dehydrogenase gene families in cotton. PLoS One 11, e0166341. 612 Jia, Y., Li, X., Liu, Q., Hu, X., Li, J., Dong, R., Liu, P., Liu, G., Luo, L., Chen, Z., 2020. 613 Physiological and transcriptomic analyses reveal the roles of secondary metabolism 614 in the adaptive responses of Stylosanthes to manganese toxicity. BMC Genomics 21, 615 861. 20 616 Jiang, C., Liu, L., Li, X., Han, R., Wei, Y., Yu, Y., 2018. Insights into aluminum-tolerance 617 pathways in Stylosanthes as revealed by RNA-Seq analysis. Sci. Rep. 8, 6072. 618 Kumari, M., Taylor, G.J., Deyholos, M.K., 2008. Transcriptomic responses to aluminum 619 stress in roots of Arabidopsis thaliana. Mol. Genet. Genomics 279, 339-357. 620 Le Roux, M.R., Khan, S., Valentine, A.J., 2008. Organic acid accumulation may inhibit 621 N2 fixation in phosphorus-stressed lupin nodules. New Phytol. 177, 956-964. 622 Le Roux, M.R., Ward, C.L., Botha, F.C., Valentine, A.J., 2006. Routes of pyruvate 623 synthesis in phosphorus-deficient lupin roots and nodules. New Phytol. 169, 624 399-408. 625 Lee, C.P., Eubel, H., O’Toole, N., Millar, A.H., 2008. Heterogeneity of the mitochondrial 626 proteome for photosynthetic and non-photosynthetic Arabidopsis metabolism. Mol. 627 Cell. Proteomics 7, 1297-1316. 628 Li, J.F., Dong, R.S., Jia, Y.D., Huang, J., Zou, X.Y., An, N., Song, J.L., Chen, Z.J., 2021. 629 Characterization of metal tolerance proteins and functional analysis of GmMTP8.1 630 involved in manganese tolerance in soybean. Front. Plant Sci. DOI, 631 10.3389/fpls.2021.683813. 632 Ligaba, A., Shen, H., Shibata, K., Yamamoto, Y., Tanakamaru, S., Matsumoto, H., 2004. 633 The role of phosphorus in aluminium-induced citrate and malate exudation from 634 rape (Brassica napus). Physiol. Plant. 120, 575-584. 635 Liu, P.D., Xue, Y.B., Chen, Z.J., Liu, G.D., Tian, J., 2016. Characterization of purple acid 636 phosphatases involved in extracellular dNTPs-P utilization in Stylosanthes. J. Exp. 637 Bot. 67, 4141-4154. 638 Lü, J., Gao, X., Dong, Z., Yi, J., An, L., 2012. Improved phosphorus acquisition by 639 tobacco through transgenic expression of mitochondrial malate dehydrogenase from 640 Penicillium oxalicum. Plant Cell Rep. 31, 49-56. 641 Ma, B., Yuan, Y., Gao, M., Xing, L., Li, C., Li, M., Ma, F., 2018. Genome-wide 642 identification, classification, molecular evolution and expression analysis of malate 643 dehydrogenases in apple. Int. J. Mol. Sci. 19, 3312. 644 Millar, A.H., Sweetlove, L.J., Giegé, P., Leaver, C.J., 2001. Analysis of the Arabidopsis 645 mitochondrial proteome. Plant Physiol. 127, 1711-1727. 646 Miller, S.S., Driscoll, B.T., Gregerson, R.G., Gantt, J.S., Vance, C.P., 1998. Alfalfa malate 21 647 dehydrogenase (MDH): molecular cloning and characterization of five different 648 forms reveals a unique nodule-enhanced MDH. Plant J. 15, 173-184. 649 Nunes-Nesi, A., Carrari, F., Lytovchenko, A., Smith, A.M., Loureiro, M.E., Ratcliffe, 650 R.G., Sweetlove, L.J., Fernie, A.R., 2005. Enhanced photosynthetic performance and 651 growth as a consequence of decreasing mitochondrial malate dehydrogenase activity 652 in transgenic tomato plants. Plant Physiol. 137, 611-622. 653 Pracharoenwattana, I., Cornah, J.E., Smith, S.M., 2007. Arabidopsis peroxisomal malate 654 dehydrogenase functions in β-oxidation but not in the glyoxylate cycle. Plant J. 50, 655 381-390. 656 Qin, L., Guo, Y., Chen, L., Liang, R., Gu, M., Xu, G., Zhao, J., Walk, T., Liao, H., 2012. 657 Functional characterization of 14 Pht1 family genes in yeast and their expressions in 658 response to nutrient starvation in soybean. PLoS One 7, e47726. 659 Scheible, W.R, Morcuende, R., Czechowski, T., Fritz, C., Osuna, D., Palacios-Rojas, N., 660 Schindelasch, D., Thimm, O., Udvardi, M.K., Stitt, M., 2004. Genome-wide 661 reprogramming of primary and secondary metabolism, protein synthesis, cellular 662 growth processes, and the regulatory infrastructure of Arabidopsis in response to 663 nitrogen. Plant Physiol. 136, 2483-2499. 664 Schulze, J., Tesfaye, M., Litjens, R.H.M.G., Bucciarelli, B., Trepp, G., Miller, S., Samac, 665 D., Allan, D., Vance, C.P., 2002. Malate plays a central role in plant nutrition. Plant 666 Soil 247, 133-139. 667 Sun, L., Chen, Z., Liu, P., Liao, H., Liu, G., Tian, J., 2013. Cloning and analysis of the 668 phosphate transporter protein SgPT1 from Stylosanthes. Acta Prataculturae Sinica 22, 669 187-196. 670 Sun, L., Liang, C., Chen, Z., Liu, P., Tian, J., Liu, G., Liao, H., 2014. Superior Al 671 tolerance of Stylosanthes is mainly achieved by malate synthesis through an 672 Al-enhanced malic enzyme, SgME1. New Phytol. 202, 209-219. 673 Tang Y.Q., Wu, Z.Y., Liu, G.D., Yi, K.X., 2009. Research advances in germplasm 674 resources of Stylosanthes. Chin. Bull. Bot. 44, 752-762. 675 Teng, X., Zhong, M., Zhu, X., Wang, C., Ren, Y., Wang, Y., Zhang, H., Jiang, L., Wang, 676 D., Hao, Y., Wu, M., Zhu, J., Zhang, X., Guo, X., Wang, Y., Wan, J., 2019. FLOURY 677 ENDOSPERM16 encoding a NAD-dependent cytosolic malate dehydrogenase plays 22 678 an important role in starch synthesis and seed development in rice. Plant Biotechnol 679 J, 17, 1914-1927. 680 Tesfaye, M., Dufault, N.S., Dornbusch, M.R., Allan, D.L., Vance, C.P., Samac, D.A., 681 2003. Influence of enhanced malate dehydrogenase expression by alfalfa on 682 diversity of rhizobacteria and soil nutrient availability. Soil Biol. Biochem. 35, 683 1103-1113. 684 Tesfaye, M., Temple, S.J., Allan, D.L., Vance, C.P., Samac, D.A., 2001. Overexpression 685 of malate dehydrogenase in transgenic alfalfa enhances organic acid synthesis and 686 confers tolerance to aluminum. Plant Physiol. 127, 1836-1844. 687 Tomaz, T., Bagard, M., Pracharoenwattana, I., Lindén, P., Lee, C.P., Carroll, A.J., Ströher, 688 E., Smith, S.M., Gardeström, P., Millar, A.H., 2010. Mitochondrial malate 689 dehydrogenase lowers leaf respiration and alters photorespiration and plant growth 690 in Arabidopsis. Plant Physiol. 154, 1143-1157. 691 Tripodi, K.E.J., Podesta, E.F., 2003. Purification and characterization of an 692 NAD-dependent malate dehydrogenase from leaves of the crassulacean acid 693 metabolism plant Aptenia cordifolia. Plant Physiol. Biochem. 41, 97-105. 694 Uhde-Stone, C., Gilbert, G., Johnson, J.M.F., Litjens, R., Zinn, K.E., Temple, S.J., Vance, 695 C.P., Allan, D.L., 2003. Acclimation of white lupin to phosphorus deficiency 696 involves enhanced expression of genes related to organic acid metabolism. Plant 697 Soil 248, 99-116. 698 van der Merwe, M.J., Osorio, S., Moritz, T., Nunes-Nesi, A., Fernie, A.R., 2009. 699 Decreased mitochondrial activities of malate dehydrogenase and fumarase in tomato 700 lead to altered root growth and architecture via diverse mechanisms. Plant Physiol. 701 149, 653-669. 702 Vengavasi, K., Kumar, A., Pandey, R., 2016. Transcript abundance, enzyme activity and 703 metabolite concentration regulates differential carboxylate efflux in soybean under 704 low phosphorus stress. Ind. J. Plant Physiol. 21, 179-188. 705 Vengavasi, K., Pandey, R., Abraham, G., Yadav, R.K., 2017. Comparative analysis of 706 soybean root proteome reveals molecular basis of differential carboxylate efflux 707 under low phosphorus stress. Genes 8, 341. 708 Wang, Q.F., Zhao, Y., Yi, Q., Li, K.Z., Yu, Y.X., Chen, L.M., 2010. Overexpression of 23 709 malate dehydrogenase in transgenic tobacco leaves: enhanced malate synthesis and 710 augmented Al-resistance. Acta Physiol. Plant. 32, 1209-1220. 711 Wang, R., Guegler, K., LaBrie, S.T., Crawford, N.M., 2000. Genomic analysis of a 712 nutrient response in Arabidopsis reveals diverse expression patterns and novel 713 metabolic and potential regulatory genes induced by nitrate. Plant Cell 12, 714 1491-1510. 715 Wang, Z.A., Li, Q., Ge, X.Y., Yang, C.L., Luo, X.L., Zhang, A.H., Xiao, J.L., Tian, Y.C., 716 Xia, G.X., Chen, X.Y., Li, F.G., Wu, J.H., 2015. The mitochondrial malate 717 dehydrogenase 1 gene GhmMDH1 is involved in plant and root growth under 718 phosphorus deficiency conditions in cotton. Sci. Rep. 5, 10343. 719 Wu, P., Ma, L., Hou, X., Wang, M., Wu, Y., Liu, F., Deng, X.W., 2003. Phosphate 720 starvation triggers distinct alterations of genome expression in Arabidopsis roots and 721 leaves. Plant Physiol. 132, 1260-1271. 722 Yang, L.T., Jiang, H.X., Qi, Y.P., Chen, L.S., 2012. Differential expression of genes 723 involved in alternative glycolytic pathways, phosphorus scavenging and recycling in 724 response to aluminum and phosphorus interactions in Citrus roots. Mol. Biol. Rep. 725 39, 6353-6366. 726 Yao, Y.X., Dong, Q.L., Zhai, H., You, C.X., Hao, Y.J., 2011a. The functions of an apple 727 cytosolic malate dehydrogenase gene in growth and tolerance to cold and salt 728 stresses. Plant Physiol. Biochem. 49, 257-264. 729 Yao, Y.X., Li, M., Zhai, H., You, C.X., Hao, Y.J., 2011b. Isolation and characterization of 730 an apple cytosolic malate dehydrogenase gene reveal its function in malate synthesis. 731 J. Plant Physiol. 168, 474-480. 732 Yoo, S.D., Cho, Y.H., Sheen, J., 2007. Arabidopsis mesophyll protoplasts: a versatile cell 733 system for transient gene expression analysis. Nat. Protoc. 2, 1565-1572. 734 Zhu, S.N., Chen, Z.J., Xie, B.X., Guo, Q., Chen, M.H., Liang, C.Y., Bai, Z.L., Wang, 735 X.R., Wang, H.C., Liao, H., Tian, J., 2021. A phosphate starvation responsive malate 736 dehydrogenase, GmMDH12 mediates malate synthesis and nodule size in soybean 737 (Glycine max). Environ. Exp. Bot. 189, 104560. 738 24 739 Table 1. Kinetic parameters of the recombinant SgMDH proteins. Km Vmax Kcat/Km Protein Substrate (mM) (nmol-1min-1) (mM-1min-1) OAA 0.065±0.02c 492.10±7.94a 1415734 NADH 0.182±0.02b 387.20±21.70b 397837 SgMDH2 Malate 0.168±0.03bc 85.60±2.38c 95281 NAD+ 0.460±0.05a 122.98±1.33c 49994 OAA 0.032±0.01b 455.17±12.0a 2523349 NADH 0.098±0.02b 416.67±34.4a 754258 SgMDH3 Malate 0.610±0.05a 65.83±2.43b 15954 NAD+ 0.600±0.05a 66.87±4.10b 16476 OAA 0.047±0.00b 939.40±30.30a 3607122 NADH 0.194±0.01b 944.44±13.10a 1131058 SgMDH4 Malate 1.144±0.24a 135.08±13.10b 24567 NAD+ 0.256±0.17b 103.87±4.69b 45357 OAA 0.022±0.01c 626.60±22.70b 5921054 NADH 0.504±0.02a 803.00±71.60a 331220 SgMDH5 Malate 0.168±0.03b 85.60±2.38c 77822 NAD+ 0.370±0.07a 75.48±3.40c 31158 OAA 0.047±0.00c 750.90±18.30b 3607122 NADH 0.207±0.01b 1037.00±37.00a 1131058 SgMDH6 Malate 0.605±0.05a 65.83±2.43c 24567 NAD+ 0.481±0.08a 96.63±4.90c 45357 OAA 0.121±0.03c 939.39±30.30b 1303501 NADH 0.268±0.03b 1157.40±46.30a 725102 SgMDH7 Malate 0.542±0.07a 77.09±5.21c 23881 NAD+ 0.216±0.02bc 71.50±1.53c 55578 740 The Km and Vmax of the recombinant MDH proteins were calculated by Lineweaver-Burke 741 plots method. Values indicate means of three replicates with standard error. Values 742 followed with different letters indicate significant differences between various substrates 743 at P<0.05 according to least significant difference (LSD) test. 744 745 25 746 Table 2. Effects of metal ions on the activities of the recombinant SgMDH proteins. Metal Relative activity (%) ion SgMDH2 SgMDH3 SgMDH4 SgMDH5 SgMDH6 SgMDH7 Fe2+ 97.2±4.3bc 45.6±5.1c 53.0±1.4c 76.3±3.4c 61.9±2.0c 48.3±2.1cd Mn2+ 107.8±1.8a 95.6±2.6a 86.6±1.4b 111.0±5.9a 88.2±4.2ab 122.7±4.7a Cu2+ 88.1±4.2cd 14.2±2.1d 100.3±1.6a 43.1±4.3d 32.1±1.9d 55.8±4.3c Zn2+ 111.6±3.5a 53.8±4.6c 25.1±1.3d 97.5±1.3b 57.9±2.2c 44.7±2.3d Mg2+ 103.4±0.8ab 105.8±1.0a 88.7±1.5b 105.7±5.8ab 98.1±2.6a 128.5±5.1a Al3+ 84.1±3.2d 77.1±3.3b 88.9±3.0b 98.5±2.7ab 85.4±5.0b 102.6±1.5b 747 Relative activity (%) was calculated by the ratios of MDH activity with metal ion 748 treatment to its activity without application of metal ion. Values indicate means of three 749 replicates with standard error. Values followed with different letters within a column 750 indicate significant differences between various treatments at P<0.05 according to LSD 751 test. 752 753 754 26 755 Supplementary data 756 Supplementary Fig. S1. Conserved domains of SgMDHs. Conserved domains were 757 predicted by CDD program (http://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). 758 Supplementary Fig. S2. Homology identity of MDH proteins from stylo and other plants. 759 (A) Identity of SgMDHs with those closely related MDH proteins from other plants. (B) 760 Homology identity among SgMDH proteins. The first two letters of each protein 761 represent the abbreviated species name. Sg, Stylosanthes guianensis; At, Arabidopsis 762 thaliana; Gm, Glycine max; Ms, Medicago sativa; Gh, Gossypium hirsutum. The identity 763 (%) was determined by Clustalw (https://www.genome.jp/tools-bin/clustalw). 764 Supplementary Fig. S3. Purification of the recombinant SgMDH proteins. A, 765 SDS-PAGE analysis of the recombinant SgMDH proteins in E. coli. Lane 1, molecular 766 mass marker. Lane 2, total protein of E. coli transformed with the empty vector 767 pGEX-6P-3 after IPTG induction. Lane 3, purified GST protein. Lanes 4, 6, 8, 10, 12 and 768 14, total protein of E. coli transformed with SgMDH2, 3, 4, 5, 6 and 7 after IPTG 769 induction, respectively. Lanes 5, 7, 9, 11, 13 and 15, purified SgMDH2, 3, 4, 5, 6 and 7 770 proteins, respectively. B, Western blot analysis of the purified SgMDH proteins using 771 anti-GST antibody. Lane 1, molecular mass marker. Lane 2, purified GST protein. Lanes 772 3, 4, 5, 6, 7 and 8, the purified SgMDH2, 3, 4, 5, 6 and 7 proteins, respectively. 773 Supplementary Fig. S4. Effects of NADH, NADPH, NAD+ and NADP+ supplements on 774 activities of the recombinant SgMDH proteins. 0.3 mM NADH and NADPH were 775 separately added into the reaction buffer to analyze OAA reduction, while 1 mM NAD+ 776 and NADP+ were used to analyze malate oxidation. All values represent means of three 777 replicates with standard error. Different letters indicate significant differences among 778 various supplements at P<0.05 according to Duncan test. 779 Supplementary Table S1. Primers used for qRT-PCR and enzymatic properties analyses. 780 Supplementary Table S2. General information for the SgMDH genes in S. guianensis. 781 782 27 783 Figure legends 784 Fig. 1. Multiple alignment and phylogenetic analysis of SgMDH proteins. (A) SgMDHs 785 with those closely related MDH members from other plants. (B) Phylogenetic analysis of 786 MDH proteins in stylo and other plants. Except for SgMDHs, AtcyMDH1 (AtNP171936), 787 GmMDH12 (GmNP001341093) and MsneMDH (MsAAB99757), the first two letters of 788 each protein represent the abbreviated species name, followed by GenBank number. At, 789 Arabidopsis thaliana; Ac, Ananas comosus; Cm, Cucumis melo; Gh, Gossypium hirsutum; 790 Gm, Glycine max; La, Lupinus albus; Md, Malus domestica; Ms, Medicago sativa; Os, 791 Oryza sativa; Pm, Plantago major; Pp, Prunus persica; Ps, Pisum sativum; Sl, Solanum 792 lycopersicum; Zm, Zea mays. The red arrows represent SgMDH proteins. 793 Fig. 2. Expressions of SgMDHs in leaf and root of stylo. After seeds germinated for 3 d, 794 seedlings were transplanted into full-strength Hoagland solution. After 21 d of growth, 795 leaf and root were separately harvested for gene expression analysis. Each bar represents 796 means of three independent replicates with standard error. Asterisks indicate significant 797 differences between leaf and root according to Student’s t-test. **, 0.001