Fe and Zn stress induced gene expression analysis unraveled mechanisms of mineral homeostasis in common bean (Phaseolus vulgaris L.)

Iron (Fe) and zinc (Zn) stress significantly affects fundamental metabolic and physiological processes in plants that results in reduction of plant growth and development. In the present study, common bean variety; Shalimar French Bean-1 (SFB-1) was used as an experimental material. Four different MGRL media i.e. normal MGRL medium (Control), media without Fe (0-Fe), media without Zn (0-Zn) and media with excess Zn (300-Zn) were used for growing seeds of SFB-1 under in vitro condition for three weeks under optimum conditions. Three week old shoot and root tissues were harvested from the plants grown in these four different in vitro conditions and were, subjected to Fe and Zn estimation. Further, extraction of total RNA for differential gene expression of ten candidate genes selected based on our in silico investigation and their classification, phylogeny and expression pattern was unraveled. Expression analysis of three candidate genes (OPT3, NRAMP2 and NRAMP3) in roots revealed possible cross talk among Fe/Zn stress that was further confirmed by observing less accumulation of Fe in roots under both these conditions. However, we observed, higher accumulation of Fe in shoots under 0-Fe condition compared to control that suggests precise sensing for priority based compartmentalization and partitioning leading to higher accumulation of Fe in shoots. Furthermore, the expression analysis of IRT1, FRO1 and Ferritin 1 genes under Fe/Zn stress suggested their role in uptake/transport and signaling of Fe and Zn, whereas the expression of ZIP2, NRAMP1, HA2 and GLP1 genes were highly responsive to Zn in Phaseolus vulgaris. The identified genes highly responsive to Fe and Zn stress condition can be potential candidates for overcoming mineral stress in dicot crop plants.

www.nature.com/scientificreports/ formatted CDS and genomic DNA sequences 14 . MEME analysis (http:// meme-suite. org/) was performed with default parameters to identify conserved motifs in the identified Fe/Zn responsive protein sequences.
Phylogenetic analysis. The  Physical mapping of Fe/Zn responsive genes in Phaseolus vulgaris. MapInspect 1.0 (https:// mapin spect. softw are. infor mer. com) was used to physically map the genes onto individual chromosomes of common bean.
Protein Interaction Network analysis. The Fe/Zn responsive protein interaction network was examined using the STRING online server (https:// string-db. org/ cgi accessed) and network generated was formatted in Cytoscape 3.8.2.
In vitro growth of Phaseolus vulgaris L. plants. In the present study the common bean variety SFB-1 (Shalimar French Bean-1) used, was collected by fulfilling the institutional/national/international guidelines and legislation. This variety is released by Sher-e-Kashmir University of Agricultural Sciences and Technology of Kashmir-Shalimar, Srinagar, which is our parent University. The seed material was procured from Division of Genetics and Plant Breeding and the material was identified by Sajad M Zargar. The seeds of this variety are available in SKUAST-K and can be used by farmers for cultivation or by the researchers for research purpose. Seeds of Shalimar French Bean-1 (SFB-1) variety were surface-sterilized and were germinated on 4 different sterile MGRL media i.e. normal MGRL medium (Control), without Fe (0-Fe), without Zinc (0-Zn) and with 300 μM ZnSO 4 (300-Zn). For in vitro growth of plants we have used our standardized method as detailed in Urwat et al. 16 . After 3 weeks growth, Fe and Zn was estimated and total RNA was extracted from the shoots and roots of the plants grown under all 4 conditions (normal, 0-Fe, 0-Zn and Excess Zn) as detailed above. RNA extraction and quantification. Total RNA was extracted from both shoot and root tissues of SFB-1 grown in vitro under 4 different conditions as detailed above, using Trizol method as per the manufacturer's instructions 16 . The quantity and quality of isolated RNA was checked at 260 and 280 nm with Nanodroplite (ThermoScientific) and UV-visible spectrophotometer (Thermo Scientific). Prior to reverse transcription, isolated total RNA samples were run on a 1% agarose gel. To rule out DNA contamination DNase treatment was given by using DNase kit (Sigma Aldrich, USA) 16 .

Estimation of Fe and Zn content.
cDNA synthesis and amplification of genes. cDNA synthesis was performed with equal concentration of RNA (1 μg) in all the samples using Thermo Scientific Revert Aid First Strand cDNA Synthesis Kit using oligo dT primers as per manufacturer instruction. The integrity of cDNA synthesis was checked by conventional PCR using cDNA as template and 2 house keeping genes (actin and tubulin) as primers that amplify 175 bp and 120 bp gene fragments respectively. For the validation of genes, cDNA as template and the forward and reverse primers of nine target genes given in Table 1 were used to amplify by PCR. The amplified PCR products were electrophoretically separated on 2% agarose gel. Details of primer sequences and amplicon size of all genes are given in Table 1. Reported primers were used for Actin 17 and Ferritin gene 18 and primers used for the rest of the genes were designed by using PRIMER 3 Plus program. The amplicons were sequenced and submitted to NCBI and their assigned accession no. are given in Table 1.
Relative quantification by qRT-PCR. Total RNA was extracted from both shoot and root tissues of plants grown under 4 different conditions (control, 0-Fe, 0-Zn, 300Zn). cDNA was synthesized as mentioned above. Actin and Tubulin was used as internal control for normalization of all the reactions. Primers used for normalization (ACT-F, ACT-R and TUB-F, TUB-R) and expression analysis of ten target genes (IRT1F-IRT1R,  FER1F-FER1R, ZIP2F-ZIP2R, GLP1F-GLP1R, FRO1F-FRO1R, OPT3F-OPT3R, NRAMP1F-NRAMP1R, NRAMP2F-NRAMP2R, NRAMP3F-NRAMP3R, AHA2F-AHA2R and GLP1F-GLP1R) are given in Table 1. Criteria for selection of genes were based on their major role/involvement of genes for transport and accumulation followed in Strategy I by dicot plants 19 . qRT-PCR reactions were performed on a Light Cycler 480 (Roche) using KAPA SYBR® FAST qPCR Master Mix kit as per manufactures instruction. Amplicon dissociation curve was also recorded at the end of the PCR cycles. All reactions were run in triplicate and repeated twice. Relative expression of genes was analyzed by using Livak and Schmittgen 20 method. Zn estimation and expression analysis respectively each with three technical replicates for all four treatments. The data was analyzed using one-way analysis of variance (ANOVA) followed by post hoc test (Multiple Comparisons) using SAS software (statistical analysis software institute, Cary, NC, USA) to investigate significant differences among multiple samples at 0.05 levels 16 .

Results
Identification and annotation of Fe/Zn responsive genes. A total of 10 Fe/Zn responsive genes/ proteins were identified in Phaseolus vulgaris genome through blastp search by using sequences available in the NCBI as queries in Phytozome. These retrieved sequences were then examined by other databases to examine their protein domain for identifying their family /superfamily. Databases like pfam, InterPro, PANTHER and Gene Ontology were employed to identify the family /superfamily to which the predicted Fe/Zn responsive proteins belong and their profile IDs under which they are classified into family /superfamily are given in (Supplementary Table1). The protein characteristics such as coding sequence (CDS) lengths, protein lengths, isoelectric point, and molecular weight of identified Fe/Zn responsive proteins in Phaseolus vulgaris were evaluated, listed in Table 2. The sub-cellular localization of these 10 Fe/Zn responsive proteins was predicted using four different prediction tools. In most of the Fe/Zn responsive proteins, the results were the same with all the four tools. The determination of the subcellular localization of Fe/Zn responsive proteins will help understand the molecular function. Details of same are provided in Table 2.
Phylogenetic analysis. To unravel the phylogenetic relationships and functional divergence of Fe/Zn responsive genes in Phaseolus vulgaris with Fe/Zn responsive homologs from other dicot plants and three monocots (used as outgroup) like Oryza sativa, Hordeum vulgare and Zea mays, the multiple sequence alignments and phylogenetic relationship analysis was carried out (Fig. 1). An unrooted maximum-likelihood phylogenetic tree based on Poisson correction model with the amino acid sequences of 20 plants using MEGA 7.0 software was constructed with default parameters and the reliability of interior branches was assessed with 1000 bootstrap repetitions (BT values ≥ 70%) and final tree was visualized in Figtree v1.4.2. The tree constructed was combined multigene phylogenetic tree of 10 identified Fe/Zn responsive proteins from 20 plant species. The tree was divided into two main clusters-cluster I and cluster II. Cluster I (green) included sequences of monocot plant species and cluster II (red) contained Fe/Zn responsive homologs of dicot plants. A clear monocot/dicot separation was observed in the phylogenetic tree. This tree also indicates Phaseolus vulgaris and Glycine max as very similar to each other and closely related (Fig. 1).

Conserved motif analysis.
Motif analysis was performed by using the MEME motif search tool to identify most conserved six motif types in 10 Fe/Zn stress responsive proteins of Phaseolus vulgaris and other dicot plant species. These motifs for each proteins are present in almost all respective sequences proposing that motif structures for Fe/Zn responsive proteins are well conserved in dicot plant species and is given for: FER (Fig. 2a), FRO (Fig. 2b), GLP (Fig. 2c) HA2 (Fig. 2d) IRT (Fig. 3a), NRAMP (Fig. 3b), OPT (Fig. 3c) and ZIP (Fig. 3d).     Table 1.
Fe and Zn concentrations. Fe concentration in shoots of plants grown under 0-Zn was significantly higher (P ≤ 0.05) and was followed by Fe concentration in shoots of plants grown under 300-Zn, 0-Fe and control. A significant (P ≤ 0.05) difference in Fe content among shoots of plants grown under 4 different treatments was observed (Table 3 and Fig. 7a). Concentration of Fe in roots was significantly (P ≤ 0.05) higher in 0-Zn and lower in 0-Fe. The differences in Fe concentration of roots were significant (P ≤ 0.05) among treatments. The Fe concentration differences between shoot and root under different treatments were also significant (P ≤ 0.05). Concentration of Fe in shoots were significantly (P ≤ 0.05) higher in 0-Zn, 300-Zn and 0-Fe than roots of the respective treatments where as concentration of Fe in roots was significantly (P ≤ 0.05) higher in control compared to control in case of shoots. www.nature.com/scientificreports/ In shoots the average concentration of Zn was significantly (P ≤ 0.05) much higher in 300-Zn and lower in control. There was significant (P ≤ 0.05) difference among treatments (Table 3 and Fig. 7b). In roots concentration of Zn was significantly (P ≤ 0.05) higher in 300-Zn as compared to other treatments and lower in 0-Zn. The difference in concentration of Zn was non significant (P-value ≥ 0.05) between control and 0-Fe treatments and were significantly lower than other treatments whereas Zn concentration in 0-Zn and 300-Zn were significantly (P ≤ 0.05) different in comparison to other treatments. The difference in concentrations of Zn between shoot and root of different treatments were significant. The concentrations of Zn in roots of all the treatments were significantly (P ≤ 0.05) higher than the concentration of Zn of shoots.
Differential expression of Fe /Zn responsive genes. The relative expression of the Fe/Zn responsive genes in terms of mean fold expression (2 −ΔΔCt ) of both tissues (shoot and root of Phaseolus vulgaris L.) under four different treatments is shown as in numerical data and graphical representation for fold expression of nine Fe/Zn responsive genes. The differences in fold expression levels of most of the genes in shoots are non significant (p ≥ 0.05) whereas in roots are highly significant (responsive).  Table 2). Whereas mRNA expression levels of IRT1 gene in root was significantly (p < 0.05) down regulated in 0-Zn (0.066 fold), 0-Fe (0.266 fold) and 300-Zn (0.533 fold) with respect to control (1.000 fold) and was also significantly different among all treatments.

Protein-Protein Networks Analysis of the Fe/Zn responsive Genes.
Using the STRING database ( Fig. 9), we here framed protein-protein interactions among 10 Fe/Zn responsive proteins, accordingly we observed that except GLP1, other nine Fe/Zn responsive proteins interact with one another. OPT3 was observed interacting with 5 Fe/Zn responsive proteins such as NRAMP1, NRAMP2, NRAMP3 IRT1and FRO1, while FRO1 was found interacting with FER1, HA2, OPT3 and NRAMP1. IRT1 was shown interacting with ZIP2. The interaction among these Fe/Zn responsive proteins indicate their possible function in the uptake and homeostasis of Fe and Zn in Phaseolus vulgaris L.

Discussion
In silico identification of Fe and Zn responsive genes. Identification of Fe/Zn stress responsive genes has a great significance in improving accumulation of these nutrients in crop plants. Identification and characterization of Fe/Zn responsive genes in common bean through genetic screening and direct cloning is difficult due to the non-availability of specific genomic loci information data and the tissue/time specific expressions of some genes 21 . Computational approach is a powerful tool that has simplified the identification and characterization of potential candidate genes for abiotic stresses 21 . The current study was performed for identification of www.nature.com/scientificreports/ mineral (Fe/Zn) stress responsive genes in P. vulgaris through in silico tools. Similar studies have been carried out in a wide variety of plant species and a large number of stress responsive genes have been identified in these species 22 . Genes responding under various abiotic stresses like; cold, heat, drought and salt stresses were identified as cumulative abiotic stress responding genes by using BLASTx and BLASTn tools in potato 22 . The in silico research using bioinformatics tools are a rational approach to find interesting findings [23][24][25] . In silico searches of many plant genomes or proteomes was conducted to find abiotic stress responsive genes like NRAMP genes in Glycine max 26 , glutathione S-transferase genes in Vigna radiata L. 27 , proteases in Cicer arietinum 28 and so on.
In the present study, we unraveled most of the topological features and physicochemical properties of identified Fe/Zn stress responsive transporter/gene sequences that were showing consistency with the current literature.
Overall, we observed that information about gene and protein features of Fe/Zn stress transporter proteins could be used in identification of Fe/Zn stress responsive transporter homologs in different plant genomes. The transmembrane helices and subcellular localization was also predicted by using various online bioinformatic tools. Some of identified Fe/Zn responsive proteins are localized to cell membrane, mitochondria, chloroplast and vacuole providing support for their function as metal transporter or tolerance against stresses. Those localized to nucleus function as transcription factors 29 . Such compartmentalization is likely related to specific functions, namely, uptake of metals on the plasma membrane, or release from tonoplast as previously reported in Arabidopsis and rice 26 . Protein motifs for Fe/Zn stress responsive protein were generated by using MEME suite. Motif analysis was performed for the identification of most conserved six motif types for each Fe/Zn stress responsive protein. MEME analysis suggested that motifs found were present in most of the sequences of each protein taken into consideration. This analysis suggests that motif structures of Fe/Zn stress responsive proteins are well conserved in dicot plants and may be used as a base for protein classification.
The multigene phylogenetic tree generated by MEGA 7.0v was grouped into two clusters-dicots and monocots and all the dicot plants used in the study got clustered into one group. This suggested differentiation among these proteins following the divergence of monocots and dicots from a common ancestor. This phylogenetic analysis of Fe/Zn responsive genes among the 20 plants showed that these proteins have high level of conservation in dicot plants. These results suggested that phylogenetic analysis implied that Arabidopsis (model dicot crop) or any other dicot plant, Fe/Zn responsive transporter/gene sequences could be used as references/benchmarks to identify the corresponding homologs of Fe/Zn transporters in various other dicot plant species. This analysis also showed Glycine max and Phaseolus vulgaris are closely related as depicted in results. Fe and Zn partitioning during early growth stage. Different response to micronutrient shortage was observed in both tissues of common bean grown under in vitro conditions. We observed higher accumulation of Fe in shoots under 0-Fe condition compared to control. However, the total concentration of Fe in plant under 0-Fe was less compared to the concentration of Fe in plants grown under control condition. Similarly concentration of Zn was less in roots and more in shoots of 0-Zn compared to control but the total concentration of Zn in plants under 0-Zn was less compared to plants grown under control conditions. The differences in concentration of these micronutrients between shoot and root suggests precise sensing for priority based compartmentalization and partitioning leading to higher accumulation of these micronutrients in shoots under mineral stress. These are the mechanisms developed by the plants under mineral stress both at the cellular and systemic levels called plant metal homeostasis to balance the concentrations of micronutrients that are involved in central cellular processes of plants including compartmentalization and partitioning, daily redox oscillations, or transcriptional regulation of micronutrients. The intracellular compartmentalization and partitioning of metals seems essential for optimizing the use of micronutrients during development and in response to deficiencies 4 . We also found higher accumulation of Fe and Zn in shoot and root of plants under 0-Zn and 0-Fe respectively. As Fe and Zn, are divalent elements, they show antagonistic behavior during absorption. Therefore, a change in concentration of any one element in a nutrient solution could affect uptake of other elements. Generally some transporters are involved in uptake these elements 30 . The reasons for antagonistic effects of Fe and Zn are competition for uptake by transporters located on root cells to enter into xylem cells and disorder in metal chelating process in roots 31   www.nature.com/scientificreports/ of all 3 treatments whereas FRO1 was up regulated in roots and down regulated in shoots of all 3 treatments. Iron regulated transporter 1 (IRT 1) is a major player in the regulation of plant Fe homeostasis and is completely down regulated when Fe is completely lacking 32 . Fe deficiency in Arabidopsis up regulates the expression of IRT1, the primary transporter responsible for root Fe uptake. IRT1 also contributes to the accumulation of a broad range of divalent transition metals including Zn because of its weak substrate specificity [33][34][35] . Conversely, excess Zn causes physiological Fe deficiency. Early studies reported an absence of IRT1 protein in Arabidopsis roots from plants grown under Zn excess conditions 36,37 , suggesting ubiquitin-mediated proteasomal degradation of IRT1 protein which is the known post-translational regulation 38,39 . This is in consistent with the observation that the expression levels of the IRT1 gene in roots under Fe and Zn stress conditions were down regulated in the present study. FRO1 gene was up regulated to 1.855-, 1.772-and 2.600-fold in response to 0-Fe, 0-Zn and 300-Zn respectively in roots of Phaseolus vulgaris but was down regulated in shoots of all stressed plants that suggests its role in metal homeostasis. Ferric reductase oxidase (FRO2 in A. thaliana; LeFRO1 in tomato) is responsible for the reduction of Fe 3+ to Fe 2+ which is a crucial step for the Fe uptake in Strategy I plants [40][41][42] . LeFRO1 mRNA is also detected in shoots regardless of Fe status, suggesting that LeFRO1 may play a role in Fe mobilization and responsive to Fe status in the shoots as well. In onion epidermal cells, LeFRO1 localizes to the plasma membrane and confers Fe (III) reductase activity when expressed in yeast 43 . In root tissues, the expression of FRO1 mRNA is consistent with its proposed role in reduction of rhizosphere Fe (III). Expression of FRO1 gene was low in Fe-sufficient shoots, but was elevated in shoots of Fe-deficient plants. These results suggest a role for Fe (III) reduction in leaf tissues 44 . In the present study Fe is sufficient in shoots of stressed plants which was also confirmed by elemental analysis of shoots which supports down regulation of FRO1 in shoots compared to control conditions. In some previous experiments, FRO1 mRNA in the vascular bundles of leaves was not highly expressed, indicating that other proteins or transport mechanisms may be at work mobilizing xylem Fe (III). A Fe (III)-chelate reductase activity of FRO1 and its expression in chloroplast-containing cells has suggested to aid uptake of intracellular Fe by chloroplasts 45 . These results indicated that expression of FRO1 in shoots and roots is affected by different signals or Fe-sensing mechanisms. FRO1 regulation in shoots appears to respond to the Fe status of these tissues.
Ferritin1 (FER1) gene expression in roots was decreased to 0.012-, 0.090-and 0.082-fold for 0-Fe, 0-Zn and 300-Zn condition respectively compared to normal condition. Ferritins (FER), functioning as ferric iron binding and participating in the cellular Fe homeostasis, are essential to protect cells against oxidative damage and flowering 46 and is modulated by many abiotic and biotic stresses (nutrient deficiency, stress, and microorganism attack) in plants 47,48 . FER1 was identified to be robustly down-regulated in roots of Arabidopsis under Fe deficiency [49][50][51] and this finding is in agreement with the present study. mRNA expression of FER gene was seen in the vascular cylinder of the mature root hair zone 43 . The down regulation of FER1 gene in stressed conditions could be attributed to that Fe/Zn stress altered nutrient status of the stressed plants that leads to modulation in gene expression of FER1.These results suggests that common regulation of these three genes under Fe/Zn stress have role in uptake/transport and signaling network of Fe and Zn status in Phaseolus vulgaris. Differential expression of OPT3, NRAMP2 and NRAMP3 reveals possible cross talk between Fe-deficient and excess Zn. Among nine genes, we observed that the expression of three identified candidate genes (OPT3, NRAMP2 and NRAMP3) was higher in 0-Fe and 300-Zn media. The common regulation of these three genes in response to Fe deficiency and excess Zn induced iron deficiency indicates regulatory cross talk among these two conditions. Oligopeptide transporter 3 (OPT3) was upregulated 3.208 and 11.272 fold in roots of plants grown on 0-Fe and 300-Zn media respectively. OPT3 is involved in the transport of small peptides that may have roles in nutrition 52 . OPT3 plays a critical role in the maintenance of whole-plant Fe homeostasis and Fe nutrition in developing seeds and it plays an important role in shoot to root signaling for the regulation of Fe deficiency responses in roots 53 . Stacey et al. 54 showed that OPT3 expression was enhanced by Fe limitation and excess Zn. The high level of OPT3 expression under Fe-deficient and excess-Zn conditions shown in this study provides further support for the role of this gene in defense against Fe deficiency stress and partitioning of Fe content between root and shoot in Phaseolus vulgaris. OPT3 mediates partitioning of Fe from source to sink tissues 55 . Therefore, to determine the contribution of OPT3 to Fe partitioning, we examined concentrations of Fe in shoot (sources) and root (sinks) of the Fe/Zn stressed plants. We found higher concentration of Fe content in shoots than roots of 0-Fe and 300-Zn. Our findings suggest that OPT3 may have role in both signaling of Fe demand from shoots to roots and Fe transport to developing tissues. This data also showed an aspect of crosstalk between Fe homeostasis and partitioning that is mediated by OPT3.
Natural resistance associated macrophage protein 2 (NRAMP 2)/DCT1 (divalent cation transporter 1)/DMT1 (divalent metal transporter 1) was also highly expressed under both 0-Fe and 300-Zn growth conditions. NRAMP 2 was up regulated to 2.536-and 12.346-fold in roots of 0-Fe and 300-Zn, respectively. NRAMP2 has a role in transport of Fe/Zn and may have a role in tolerance to mineral stress. Gao et al.(2018) 56 found that NRAMP2 has role in transport of Mn, Fe and Zn in yeast and is involved in remobilization of Fe in roots. NRAMP 3 (natural resistance associated macrophage protein 3) gene expression was also increased to 1.916-and 8.017-fold in case of 0-Fe and 300-Zn respectively. AtNRAMP3, a multi-specific vacuolar metal transporter protein are up regulated at the transcriptional level by Fe deficiency and mediates the release of Fe from the vacuole and provide sufficient Fe during seed germination [57][58][59] . NRAMP genes are up-regulated under Fe deficient conditions to regulate Fe nutrition. It was found that, under stress conditions, NRAMP acts as mineral regulatory element and defends plants against stresses 60 . These findings suggest its role in crosstalk and assisting the transport of Fe/Zn under Fe-deficient and Zn-excess conditions. These genes may serve as backup systems for Fe homeostasis. However, www.nature.com/scientificreports/ further evidence is needed to confirm the mechanism underlying increased NRAMP genes expression under Fe-deficient conditions.
Zn transport responsive genes. Three candidate genes/transporters-ZIP2, NRAMP1, HA2 and GLP1 were found up and down regulated in roots of stressed plants under 300-Zn and 0-Zn conditions respectively. ZIP2 gene was up-regulated to 2.888-fold and down-regulated to 0.297-fold under excess-Zn and Zn deficient conditions respectively that indicates its role in Zn uptake in roots. Root and shoot AtZIP2 transcript abundance was decreased in response to Zn, Fe, and Mn deficiency in Arabidopsis 61 . ZIP genes revealed their role in regulating the uptake, transport and accumulation of Zn in Arabidopsis [61][62][63][64][65] . In Arabidopsis at least 10 different members of the ZIP family play a role in Zn uptake in roots, including ZIP1, 2, 3, 4, 5, 9, 10, 11, 12, and IRT3 66 . The expression of natural resistance-associated macrophage protein 1 (NRAMP1) gene, was increased to 2.383-and decreased to 0.251-fold under 300-Zn and 0-Zn conditions respectively. Members of the NRAMP family are functional divalent metal ion transporters 67 that are conserved in different species and located in the plasma membrane of root apical cells 68 69,73 . These studies are in agreement with the results of the present study that NRAMP1 is involved in transportation/ uptake of Zn. H + -ATPase 2 pump (HA2) gene expression was increased to 2.264-and decreased to 0.157-fold in response to 300-Zn and 0-Zn respectively. HA2 is responsible for the major acidification activity. It is up-regulated in Fe deficiency 32 and Zn induced Fe deficiency and is also increased in conditions requiring greater transport activity as more and more Zn is taken up under excess-Zn conditions in the present study which is confirmed by up-regulation of other Zn transporters like ZIP2 and NRAMP1. The external signal caused due to heavy metal stress results in change in HA2 gene expression to control the major transport processes in the plants such as nutrient uptake and xylem or phloem loading to play role in adaptation of plants to changing conditions of stress 74 . Excess-Zn may trigger a signal transduction pathway that up-regulates the expression of transporters to increase the efficiency of Zn transport and results in accumulation of Zn in plants under excess-Zn conditions. However, further evidence is needed to confirm the mechanism underlying increased expression of these genes under excess-Zn conditions.
In our study, we observed higher expression of Germin like protein 1 (GLP-1) gene in roots under excess Zn that reveals its role to overcome heavy metal stress. GLPs have been proposed to play a role in reactive oxygen species detoxification and to function as signaling molecules inducing a range of defense responses in a direct or indirect manner 75,76 . Germin like proteins (GLPs) include the first crystal structure of extracellular superoxide dismutase (SOD) which are implicated in the response of plants to many abiotic stresses such as exposure to heat, salt, submergence, and aluminum toxicity 77,78 , which provide evidence that GLPs represent a new family of extracellular SODs, leading to the generation of H 2 O 2 in response to abiotic stress in plants 78,79 . In the present study higher accumulation of Zn in plants under 300-Zn (excess zinc) leads to oxidative stress in the roots of Phaseolus vulgaris and induces expression of GLP1 to overcome stress. GLPs members have been confirmed to have oxalate oxidase or superoxide dismutase functions which catalyze the oxalate degradation resulting in the production of hydrogen peroxide 80 , and Zn toxicity is a source of superoxide radicals that impose oxidative stress to plant cells. Li et al. 80 studied expression of soybean germin-like gene family revealed a role of GLP7 gene in various abiotic stress tolerances. Peanut germin-like proteins (AhGLPs) plays role in plant development and defense confirmed by Wang et al. 81 and also reported the genetic connection between GLPs and ABA-mediated stress responses. We presume that pvGLP1 is a good candidate gene for increasing tolerance to heavy metal stress in plants (a new finding) but further confirmation is required.

Protein-protein networks analysis of the Fe/Zn responsive Genes.
To regulate nutrient uptake and in response to different biotic and abiotic stress, it has been observed that Fe/Zn responsive genes interact with other proteins. In the present study, STRING database predicted results that revealed OPT3 interacts with NRAMP1, NRAMP2, NRAMP3, IRT1 and FRO1; while FRO1 interacts with FER1, HA2, OPT3 and NRAMP1. Further, IRT1 interacts with ZIP2. The interactions among Fe/Zn responsive proteins indicate their possible function in the uptake and regulation of Fe and Zn in Phaseolus vulgaris. Moreover, the interactions among these proteins supports that common bean deploy strategy I for the uptake of Fe and Zn. Strategy I is a reduction strategy, used by dicots and non-grass monocots, which involves proton pumps HA2 to acidify the rhizosphere surrounding the root, increasing the solubility of Fe (III), by involving ferric reduction oxidase (FRO). Iron regulated transporter 1 (IRT1)/(NRAMP1)/(DMT1/NRAMP2) is then involved in transport of Fe (II) through the assembly of metal transporters in plasma membrane of the epidermis. The results revealed interaction of IRT1 with ZIP2 for Zn transportation. In Arabidopsis, half of the zinc-regulated transporter, iron-regulated transporter proteins (ZIPs) family is induced under Zn deficient conditions 62,82 . As per the current model, based mostly on yeast studies 83 , the ZIP member IRT1 non selectively takes up Zn (as well as cadmium). This network of Fe/Zn responsive proteins showed interaction of OPT3 (An additional protein localized in the phloem) with IRT1, FRO1, NRAMP1, NRAMP2 and NRAMP3 as oligopeptide transporter 3 (OPT3), has been identified as a