Overview

Iron (Fe) and zinc (Zn) are essential trace elements for human health[1]. However, iron and zinc deficiency is widespread worldwide, affecting approximately 2 billion people, particularly infants, young children[2], and pregnant women[3]. The risk of iron and zinc deficiency is particularly high in sub-Saharan Africa, where the diet relies heavily on maize, which provides at least 30% of total caloric intake[4]. However, the iron and zinc concentrations in maize endosperm are extremely low, and long-term maize-based diets can easily lead to iron and zinc deficiency symptoms. While iron and zinc supplements can improve human iron and zinc nutrition, these measures have limited success in developing countries due to economic constraints and low consumer acceptance[5]. Biofortification through genetically modified crops, by contrast, shows greater potential.

In recent years, genetic engineering of areas related to iron absorption and metabolism has successfully increased the iron content of edible crop parts[6]. For example, the coordinated expression of AtNAS1 (nicotinamide synthase 1), PvFERRITIN (ferritin), and AfPHYTASE (phytase) increased iron concentration in rice endosperm[7]. These studies demonstrate the broad potential of genetic engineering for crop iron fortification. Nicotinamide synthase is a key enzyme in the synthesis of nicotinamide (NA) in plants[8]. Nicotinamide is an important metal chelator that can form stable complexes with divalent metal ions such as iron (Fe) and zinc (Zn). The NAS gene directly influences the absorption, transport, and distribution of iron and zinc in plants by regulating nicotinamide synthesis, thus playing a central role in plant iron and zinc nutrition[9].

However, some literature reports confirmed the existence of an autoinhibitory signal at the C-terminus of the AtNAS1 protein through in vitro truncation experiments, thereby affecting enzyme activity[8]. This may be the reason why the effect of increasing iron and zinc in transgenic rice overexpressing AtNAS1 is not obvious[8]. Therefore, based on the above research background, this project intends to use the key amino acid sites at the C-terminus of the AtNAS1 protein predicted in published literature to affect enzyme activity, and mutate the key sites of the full-length protein. After purifying the AtNAS1 mutant protein through prokaryotic expression in Escherichia coli, the enzyme activity mechanics analysis was performed. Compared with the wild-type AtNAS1 protein, and using the truncated AtNAS1 protein reported in the literature as a positive control for increased enzyme activity, the point mutation combination with stronger activity of the AtNAS1 full-length protein was screened for Bacillus subtilis expression vector construction, and the iron and zinc content in tobacco leaves and roots was measured after tobacco soil culture. In addition, the mutant AtNAS1 protein was verified by tobacco transient expression experiments, and the transcription level of the known genes regulating iron and zinc absorption and metabolism was detected by qPCR to confirm the mutant AtNAS1 Whether the mutation affects the intracellular distribution and localization characteristics of AtNAS1 can be determined by fusing it with fluorescent protein and observing its subcellular localization under a confocal microscope. This study uses tobacco as a model organism and provides new ideas and theoretical support for solving the "hidden hunger" problem caused by the lack of trace elements such as iron and zinc in crops.

Results of our work

1. Molecular Docking and Prediction of AtNAS1 Mutated Proteins with Substrates

Based on key amino acid sites predicted in the literature, we constructed AtNAS1 mutants and conducted molecular docking analysis using its natural substrate S-adenosylmethionine (SAM) as a ligand. By comparing the differences in the binding patterns of wild-type and mutant AtNAS1 proteins with SAM, we can assess at the molecular level whether the mutations may improve substrate binding affinity and enzyme activity.

First, we searched for the protein sequence based on the gene name on the Uniprotkb website. The searched sequence was imported into the SWISS-MODLE website, where the optimal structure was found and saved in PDB format. The small molecule was then searched in the Small Molecule Database and downloaded in sdf format for subsequent operations. The first step was energy optimization, minimizing the energy of the selected ligand to reduce the energy of the docked system and achieve a stable state. The ligand was then optimized using DS software, and the optimized structure was saved in PDB format for future use.

The second step was protein preprocessing. The protein file was imported into PYMOL software, where redundant ligands and water molecules were deleted and saved in PDB format for future use. The protein was hydrogenated and charged, and then configured as a receptor and saved as a pdbqt file for future use. Finally, the small molecule was hydrogenated and charged, configured as a ligand, and torsion angles and bonds were checked and saved in pdbqt format for future use.

Since we do not know the binding pocket area, we set the grid to the entire protein area for ligand binding and run this operation. After the grid setting is completed, we perform 50 molecular Autodock dockings. The official recognition of more than 50 docking times indicates that the docking results are relatively accurate. After the docking is completed, the conformation with the highest repeatability and the best binding energy is selected as the output result conformation, and DS or PYMOL is used to display the visualization, three-dimensional structure and surface model.

Fig. 1 Two-dimensional, three-dimensional, and surface model images of the wild-type and mutant AtNAS1 proteins docked with SAM.

Two-dimensional and three-dimensional results show that wild-type AtNAS1 primarily forms multiple hydrogen bonds with SAM through residues such as GLU77, SER133, and LEU138, resulting in a relatively stable binding. The binding site of the truncated AtNAS1 shifts to residues such as GLN196 and ASP198, resulting in a more exposed binding pocket and reduced stability. Among the point mutants, T287A interacts with ASN285 and VAL284, but the binding is weak; T288A and G289A maintain moderate levels of interaction; and TRG→AAA forms multiple hydrogen bonds around residues such as GLY277, GLY278, and GLY286, resulting in a deeper and tighter binding pocket (Fig. 1).

The binding energy results were consistent with the structural observations: the wild-type binding energy was -6.0 kcal/mol, while the truncated forms decreased to -5.3 kcal/mol. T287A exhibited the weakest binding energy (-4.5 kcal/mol), while T288A and G289A exhibited -5.7 kcal/mol, and TRG→AAA exhibited the best binding energy (-6.5 kcal/mol). Overall, this suggests that the TRG→AAA mutation enhances SAM binding stability and may improve AtNAS1 enzymatic activity (Fig. 2).

Fig. 2 Binding energies between wild-type and mutant AtNAS1 proteins and SAM ligands.

2.Construction of Plasmids

2.1 Construction of E. coli Expression Plasmids

We constructed the pET28a-AtNAS1_Wildtype and pET28a-AtNAS1_Truncated plasmids using Infusion recombination. Initially, we amplified the AtNAS1_Wildtype and AtNAS1_Truncated fragments via high-fidelity PCR, resulting in fragment lengths of 963 bp and 861 bp, respectively. The right panel of Fig. 3 displays the correct bands corresponding to the target sizes of the two genes. Subsequently, we used the restriction endonuclease BamHI to linearize the pET28a vector backbone, as shown in the left panel of Fig. 3. After performing agarose gel electrophoresis and recovering the bands, we proceeded with the Infusion assemble kit to obtain the recombinant plasmids.

Fig. 3 Agarose gel electrophoresis of target genes and linearized pET28a vector

After the Infusion reaction assay, we transformed the recombinant plasmids into the cloned strain E. coli DH5α. Fig. 4A shows the single colonies on the plate. We then selected six colonies from each plate. Next, we used vector universal primers, T7-Forward and T7-Terminal, to detect the insertion of target genes using colony PCR. As shown in Fig. 4B, the AtNAS1_Wildtype and pET28a-AtNAS1_Truncated gene fragments were successfully inserted into the pET28a backbone.

Fig. 4 Verification of the transformation results for positive colonies of pET28a-AtNAS1_Wildtype and pET28a-AtNAS1_Truncated plasmids. A: Single colonies on the plate; B: Amplification results from colony PCR.

Finally, we used the universal primer T7-Forward to verify the positive colonies through Sanger sequencing. Fig. 5 displays the correct sequences of the recombinant plasmids. We selected the correct colonies for further IPTG induction.

Fig. 5 Sanger sequencing results of pET28a-AtNAS1_Wildtype and pET28a-AtNAS1_Truncated positive colony.

To further construct single-base and multi-base mutations, we linearized the pET28A-AtNAS1-Wildtype plasmid using inverse PCR to incorporate specific mutation sites. Subsequently, we constructed the mutated plasmids AtNAS1-T287A, AtNAS1-R288A, AtNAS1-G289A, and AtNAS1-TRGtoAAA via Infusion assay. Following the same construction process as for pET28A-AtNAS1-Wildtype, we also used colony PCR and Sanger sequencing to verify the positive colonies of the mutated plasmids, as shown in Fig. 6.

Fig. 6 The construction process of pET28A-AtNAS1 mutated plasmids. A. Agarose gel electrophoresis of mutated genes; B. Single colonies on the plate; C: Amplification results from colony PCR; D: Sanger sequencing results.

Since the enzymatic activity assay of nicotinamide synthase requires the substrate SAM and two coenzymes, MtnN and AdeD, we cloned the MtnN and AdeD gene fragments from the E. coli XL-1 Blue strain. Using the same construction strategy as for the pET28A-AtNAS1 plasmid, we constructed the expression plasmids pET28a-MtnN and pET28a-AdeD. The construction process is shown in Fig. 7.

Fig. 7 The construction process of pET28A-MtnN and pET28A-AdeD plasmids. A. Agarose gel electrophoresis of MtnN and AdeD genes; B. Single colonies on the plate; C: Amplification results from colony PCR; D: Sanger sequencing results.

3. Protein Expression and Purification

3.1 Construction of E. coli Expression Strains

We transformed the correct plasmids into the E. coli expression strain BL21 and selected single colonies for culture in preparation for the next step of prokaryotic expression. The plates are shown in Fig. 8.

Fig. 8 Single colonies of recombinant plasmids in E. coli BL21 strain.

3.2 Preliminary Experiment for Protein Expression「

Based on the lac operator system in the pET28a vector backbone, we used IPTG for protein expression. Initially, we applied a final concentration of 1 mM IPTG for standard protein expression. The expected molecular weights of AtNAS1 (including the mutated proteins), AtNAS1-Truncated, MtnN, and AdeD are 35.5 kDa, 31.7 kDa, 24.4 kDa, and 63.7 kDa, respectively. As shown in Fig. 9, we can clearly observe the protein bands corresponding to the target genes.

Fig. 9 SDS-PAGE analysis of preliminary experiment for target proteins expression.

3.3 His-Tag Affinity Purification of Proteins

We utilized the principle of His-tag affinity with Ni²⁺. This allowed us to bind the crude protein extracted from the E. coli expression cells to Ni-NTA beads. After washing away impurities with a low concentration of imidazole, we eluted the protein with a high concentration of imidazole. As shown in Fig. 10, we obtained a high-purity protein with a high expression level. Subsequently, we used dialysis to remove the high concentration of imidazole from the buffer and aliquoted the protein at a concentration of 1 mg/mL for storage, to be used in subsequent functional assays.

Fig. 10 SDS-PAGE analysis of large-scale His-tag protein purification of target proteins. Total: Total Protein; FT: Flow Through.

4. Functional Tests Part1: Mutation Sites Selection

4.1 Enzymatic Reactions Detection of AtNAS1 Mutated Proteins' Catalytic Ability

We assessed the enzymatic activity of nicotinamide synthase (NAS1) mutants in Functional Tests Part 1. The enzymatic activities of the wild-type and various mutants (Truncated, T287A, R288A, G289A, TRG_to_AAA) of NAS1 were evaluated by measuring the initial reaction rates (ΔA/Δt) and calculating the enzyme activities (nkat/mg) based on the parameters of the reaction system (Fig. 11B). Compared to the slope and correlation coefficient of AtNAS1-Wildtype, the enzyme activities of the different mutants varied significantly, with some mutants, such as AtNAS1-T287A and AtNAS1-R288A, exhibiting markedly reduced activities. In contrast, AtNAS1-Truncated and AtNAS1-TRG_to_AAA demonstrated steep slopes and high enzyme catalytic efficiency (Fig. 11A). The AtNAS1-TRG_to_AAA mutant protein, unlike AtNAS1-Truncated, retains the complete C-terminal region of the NAS1 protein, thereby avoiding potential functional redundancies associated with the truncated form. Notably, the TRG_to_AAA mutant also exhibited a relatively higher level of nicotinamide synthesis activity compared to the truncated protein (Fig. 11B). Therefore, the TRG_to_AAA mutant may possess greater potential for application than the truncated variant.

Fig. 11 Kinetic and Thermodynamic Analysis of Engineered Variants of AtNAS1 protein. A. Enzyme Kinetics of AtNAS1 mutated protein; B. Enzyme Activity Analysis.

4.2 Molecular Dynamics Fitting Analysis of Catalytic Efficiency in AtNAS1 Mutated Proteins

After testing the enzymatic assays, we further investigated the binding affinities of the AtNAS1-Wildtype and AtNAS1 mutant forms with the substrate S-adenosyl methionine (SAM) using molecular dynamics simulations. As shown in Figure 12, the average binding free energy (dG) between AtNAS1-Wildtype and SAM was -36.83 kcal/mol, while the dG values for AtNAS1-Truncated is -13.63 kcal/mol and the dG values for AtNAS1-TRG_to_AAA is -17.06 kcal/mol. These results indicate that the ligand (SAM) can spontaneously bind to the protein, and the resulting protein-ligand complexes are dynamically stable, suggesting they may possess biological activity. This provides a theoretical energy-based foundation for further studying the biological functions of these ligand-protein complexes. The favorable binding free energies imply that the AtNAS1-Wildtype and AtNAS1-TRG_to_AAA can effectively interact with the SAM substrate, which is a critical step in the enzymatic reactions catalyzed by this protein. 

Fig. 12 Molecular dynamics simulation of the binding sffinities between AtNAS1-Wildtype and mutant forms with the substrate S-Adenosyl Methionine (SAM)

5. Functional Tests Part2: Plant Model Validation

5.1 Construction of N. benthamiana Transient Expression Plasmids

Our previous results showed that the AtNAS1-Truncated protein and the AtNAS1 triple-mutant AtNAS1-TRG_to_AAA protein have comparable enzyme activity levels. Therefore, in the next step, we would like to validate our hypothesis using a plant model system, like N. benthamiana . First, we constructed the gene sequences of AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA into the dual expression vector pCambia1305, which has a GFP fluorescent protein tag. This will allow us to detect the subcellular localization of the mutants using confocal microscope. As showed in Fig. 11A, we first amplified the gene fragments containing the homologous arms of the pCambia1305 backbone, where the AtNAS1-Wildtype served as the control. We then linearized the pCambia1305 vector using the restriction enzyme BamHI (Fig. 13B). After that, we performed an In-Fusion assembly reaction to construct the recombinant vector (Fig. 13C). Finally, we screened for positive clones using colony PCR (Fig. 13D).

Fig. 13 The construction process of pCambia1305-GFP-AtNAS1-WT, pCambia1305-GFP-AtNAS1-Truncated, and pCambia1305-GFP-AtNAS1-TRG_to_AAA plasmids. A. Agarose gel electrophoresis of genes; B. Linearized pCambia1305 backbone; C. Single colonies on the plate; D: Amplification results from colony PCR.

5.2 Transient N. benthamiana Expression Assay

5.3.1 Detection the Effect of AtNAS1 Mutant Proteins on Subcellular Localization Using Confocal Microscopy

Given that the subcellular localization of a protein can affect its function, we utilized transient expression in N. benthamiana along with microscopy to observe whether the localization patterns of the AtNAS1 mutant proteins differ from the AtNAS1-Wildtype protein. As shown in Fig. 14A, the process of transient expression in tobacco involves Agrobacterium strain in acetosyringone buffer, selecting healthy tobacco leaves, and infiltrating the bacterial suspension into the leaf abaxial surface using a 1 mL syringe. After two days of transient expression, the leaves were analyzed. We first transformed the AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA positive plasmids into the Agrobacterium GV3101 strain and grew them on dual-antibiotic plates containing kanamycin and rifampicin, picking single colonies for further culturing (Fig. 14B). We initially used a handheld fluorescence instrument to detect the protein expression levels, which revealed that the AtNAS1-Truncated and AtNAS1-TRG_to_AAA proteins exhibited higher expression efficiency compared to the AtNAS1-Wildtype protein (Fig. 14C). Further analysis using confocal microscopy showed that the AtNAS1-Truncated and AtNAS1-TRG_to_AAA proteins had stronger fluorescence intensity and were more strongly localized to the endoplasmic reticulum, providing a potential explanation for their higher enzymatic activity (Fig. 14D).

Fig. 14 Subcellular localization of AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA proteins. A. Process of transient transfection in N. benthamiana; B. Single colonies of recombinant plasmids in Agrobacterium GV3101 strain. The plate contains Kanamycin and Rifampicin dual-antibiotics; C. Detection of GFP protein expression in N. benthamiana leaves using a handheld fluorescence instrument; D. The subcellular localization patterns of AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA in N. benthamiana leaves using confocal microscope.

5.3.2 Detection the Effect of AtNAS1 Mutant Proteins on the Expression of Iron Cycling Related Genes Using Real-time Quantitative PCR In Vivo

To further investigate whether mutations in the AtNAS1 gene could affect the expression of genes related to the absorption and activation of iron and zinc micronutrients, we validated this hypothesis using a N. benthamiana -based culture transient system. First, we used the previously constructed Agrobacterium strains carrying the AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA to infiltrate tobacco leaves. After 3 days of culture, we extracted RNA from the leaves, reverse transcribed it into cDNA, and used quantitative real-time PCR to detect the expression levels of the target genes. The expression levels were normalized using the tobacco NtActin1 gene as an internal reference. As shown in Fig. 15A, compared to the non-transformed tobacco leaves, the tobacco leaves expressing the AtNAS1-Wildtype, AtNAS1-Truncated, and AtNAS1-TRG_to_AAA plasmids significantly enhanced AtNAS1 gene expression, indicating that the transient expression of AtNAS1 in tobacco was successful. Further analysis revealed that the transient overexpression of both the wild-type and mutant AtNAS1 significantly increased the expression of the iron and zinc absorption and transport genes NtIRT1 and NtMTP8 (Fig. 15B-C), while the expression of the iron activation-related gene NtFRO2 remained unchanged (Fig. 15D). Notably, the AtNAS1-Truncated, and AtNAS1-TRG_to_AAA exhibited even higher expression levels of the iron and zinc absorption and transport genes NtIRT1 and NtMTP8, suggesting their greater potential for iron biofortification (Fig. 15B-C).

Fig. 15 Overexpression of wild-type and mutant AtNAS1 affects the expression of genes related to iron and zinc homeostasis. A: Relative expression levels of the AtNAS1 gene in tobacco leaves; B: Relative expression levels of the iron transporter gene NtIRT1 in tobacco leaves; C: Relative expression levels of the zinc transporter gene NtMTP8; D: Relative expression levels of the iron activation gene NtFRO2 in tobacco leaves.

5.3 Construction of B. subtilis Expression Plasmids

To further evaluate the application potential of AtNAS1 mutants in plants, we also used N. benthamiana as a model system and used B. subtilis harboring expression plasmids as a bioinoculant for soil-grown tobacco. Using the same construction strategy as for the pET28A-AtNAS1 plasmid, we constructed the B. subtilis expression plasmids pHT43-AtNAS1-Wildtype, pHT43-AtNAS1-Truncated, and pHT43-AtNAS1-TRG_to_AAA. The construction process is shown in Fig. 16.

Fig. 16 The construction process of pHT43-AtNAS1-WT, pHT43-AtNAS1-Truncated, and p pHT43-AtNAS1-TRG_to_AAA plasmids. A. Agarose gel electrophoresis of genes; B. Linearized p pHT43 backbone; C. Single colonies of DH5alpha on the plate; D: Single colonies of B. subtilis on the plate.

5.4 Biofortification Experiment of Soil-Based N. benthamiana Cultivation Using B. subtilis Stably Expressing AtNAS1 Mutated Proteins

5.4.1 Effect of B. subtilis Expressing AtNAS1 Mutated Proteins on N. benthamiana Growth and Biomass

The previous research results have shown that compared to the AtNAS1-Wildtype, both the AtNAS1-Truncated and the AtNAS1-TRG_to_AAA mutants have higher enzyme activity and a high efficiency to activate genes related to iron cycling. Since the AtNAS1-Truncated removes the C-terminal domain of the protein, it may lead to the partial loss of other biological functions, so we are more focused on the application of the AtNAS1-TRG_to_AAA mutant. In our tobacco pot culture experiment, we used B. subtilis as a probiotic. When the tobacco plants grew at the 4-leaf stage, we applied 10 mL of B. subtilis at different concentrations to the root system for root colonization. After two weeks of growth, we measured the plant fresh biomass and then dried and ground the shoot to determine the iron and zinc micronutrient contents using ICP-MS. As shown in Figure 17, we found that after applying B. subtilis, the fresh biomass of the tobacco plants actually showed a decreasing trend, and the leaves exhibited partial curling, which may be related to the applied concentration and still requires further optimization.

Fig. 17 Phenotypic analysis of AtNAS1 transient expressing tobacco plants. A: Images showing the growth of pHT43-AtNAS1-Wildtype and pHT43-AtNAS1-TRG_to_AAA transient expressing tobacco plants under different B. subtilis optical densities (PBS Control, OD=0.5, OD=1.0); B: The fresh weight of tobacco plants.

5.4.2 Effect of B. subtilis Expressing AtNAS1 mutated Proteins on Iron and Zinc Trace Element Contents in Tobacco Leaves Using ICP-MS

After harvesting the tobacco samples with B. subtilis root application, as shown in Figure 18A, we dried the shoot of the tobacco plants at 60°C and ground them into powder. We then performed microwave digestion of the samples using nitric acid and hydrogen peroxide, and used ICP-MS to measure the micronutrient contents. The raw data from the instrument was converted to mg/kg units using standard calibration curves. As shown in Figure 18B, compared to the PBS control, the tobacco plants inoculated with B. subtilis carrying the pHT43-AtNAS1-Wildtype and pHT43-AtNAS1-TRG_to_AAA plasmids showed increased iron (Fe) and zinc (Zn) micronutrient contents in the shoot at different B. subtilis concentrations. However, at the OD=0.5 concentration, there was no significant difference in either Fe or Zn content between the AtNAS1-Wildtype and AtNAS1-TRG_to_AAA treatments. At the OD=1 concentration, the tobacco plants inoculated with the B. subtilis carrying the pHT43-AtNAS1-TRG_to_AAA plasmid had significantly higher Fe content in the shoot compared to the pHT43-AtNAS1-Wildtype, while the Zn content did not differ. This suggests the potential application of the B. subtilis strain carrying the pHT43-AtNAS1-TRG_to_AAA plasmid for iron biofortification. Overall, our experiment has achieved preliminary success, but further optimization is still needed to fully understand the effects of the different B. subtilis treatments on plant biomass and micronutrient accumulation.

Fig. 18 A: Sample processing process, B: Fe element determination results, C: Zn element determination results

Future plans

Our preliminary results demonstrate that compared to the AtNAS1-Wildtype, the AtNAS1-Truncated and AtNAS1-TRG_to_AAA mutant exhibit higher enzymatic reaction rates. Importantly, the AtNAS1-TRG_to_AAA, which retains the complete C-terminal amino acid sequence of AtNAS1, has greater potential for practical applications. We further validated our hypotheses in pot experiments, but the application dosage of B. subtilis may not have reached the optimal conditions. We only detected higher iron content in the shoot of tobacco plants treated with the AtNAS1-TRG_to_AAA mutant strain compared to the AtNAS1-Wildtype, when the application concentration was at OD=1. This indicates the need for further optimization of the experimental conditions to achieve better results. In the future, we can explore the application of the homologous genes of AtNAS1 in other crop plants, building on the insights gained from this study. The favorable enzymatic properties of the AtNAS1 mutant forms, combined with the potential for improved micronutrient accumulation, provide a promising foundation for developing biofortification strategies using these genetic resources.

References

1. Prashanth L, Kattapagari K K, Chitturi R T, et al. A review on role of essential trace elements in health and disease[J]. Journal of Dr. YSR University of Health Sciences, 2015, 4(2): 75-85.

2. Al Mamun M A, Ghani R B A. The role of iron and zinc in cognitive development of children[J]. Asian Journal of Medical and Biological Research, 2017, 3(2): 145-151.

3. Durrani A M, Parveen H. Zinc deficiency and its consequences during pregnancy[J]. Microbial biofertilizers and micronutrient availability: The role of zinc in agriculture and human health, 2021: 69-82.

4. Galani Y J H, Orfila C, Gong Y Y. A review of micronutrient deficiencies and analysis of maize contribution to nutrient requirements of women and children in Eastern and Southern Africa[J]. Critical Reviews in Food Science and Nutrition, 2022, 62(6): 1568-1591.

5. Nguyen P, Grajeda R, Melgar P, et al. Effect of zinc on efficacy of iron supplementation in improving iron and zinc status in women[J]. Journal of Nutrition and Metabolism, 2012, 2012(1): 216179.

6. Zhu C, Sanahuja G, Yuan D, et al. Biofortification of plants with altered antioxidant content and composition: genetic engineering strategies[J]. Plant biotechnology journal, 2013, 11(2): 129-141.

7. Boonyaves K, Wu T Y, Gruissem W, et al. Enhanced grain iron levels in rice expressing an iron-regulated metal transporter, nicotianamine synthase, and ferritin gene cassette[J]. Frontiers in Plant Science, 2017, 8: 130.

8. Seebach H, Radow G, Brunek M, et al. Arabidopsis nicotianamine synthases comprise a common core-NAS domain fused to a variable autoinhibitory C terminus[J]. Journal of Biological Chemistry, 2023, 299(6).

9. Seregin I V, Kozhevnikova A D. Nicotianamine: a key player in metal homeostasis and hyperaccumulation in plants[J]. International journal of molecular sciences, 2023, 24(13): 10822.