Root system architecture, copper uptake and tissue distribution in soybean (Glycine max cv. Kowsar) grown in copper oxide nanoparticles (CuONPs) amended soil and implications to human nutrition

Department of Agronomy, Faculty of Agriculture, Shahrekord University, Shahrekord, Iran 2 Department of Electroceramics and Electrical Engineering, Malek Ashtar University of Technology, Iran. Email: rostamnejadi@mut-es.ac.ir. Department of Public Health, The Brody School of Medicine, Department of Health Education and Promotion, College of Health and Human Performance, East Carolina University, Greenville, NC, USA.


Introduction
Soybean (Glycine max) is an important and economical legume cultivated worldwide for food and feed products (Singh et al., 2007;Mataveli et al., 2010). Soybean seeds contain 20% oil (Lee et al., 2019), about 35-40% protein, and a complete set of essential amino acids critical for improving human (Xu et al., 2020). Soybean is considered the best source of plant protein and a standard for other plant protein sources (Blair, 2008). It is also an excellent source of carbohydrates (35%) and necessary elements including, copper, zinc, calcium, magnesium, iron, manganese, and phosphorus (Mataveli et al., 2010). Further, it contains metabolites such as isoflavone, saponins, phytic acids, and oligosaccharides (Ososki and Kennelly, 2003;Sakai and Kogiso, 2008). Soybean, like other legumes, enables nitrogen fixation by establishing a symbiotic association with specific rhizobium bacterium (Bradyrhizobium japonicum) (Kanchana et al., 2016).
The world's population is growing rapidly and is forecasted to reach 9.6 billion by 2050 (UN, 2019), thereby increasing demand for agricultural production (70% for grain production by 2050) (FAO, 2019). On the other hand, due to limited arable lands and usable water resources, dramatically increasing the use of chemical fertilizers is a conventional approach to achieve increased food production and meet population demands, globally (Liu and Lal, 2015). Crops such as soybean need nutrients to grow and improve yield, and soil nutrient deficiency can significantly reduce N2-fixation, growth, and performance in plants (Miransari, 2016). Macronutrients (N, P, K, Ca, Mg, and S) and micronutrients (Fe, Zn, Mn, and Cu) promote different morphological physiological functions in plants, including enzyme activities and oxidation-reduction processes (Miransari, 2016). Micronutrients can be supplemented to plants through chemical fertilizers (Welch and Graham, 2004). Interest in the use of nanofertilizers have recently increased due to their unique physicochemical characteristics not found in bulk or ionic counterparts (Servin et al., 2015;Raliya et al., 2017;Pokhrel et al., 2017;Kah et al., 2018). Micronutrients in nano-form may improve both yield and nutrient quality of crops compared to common micronutrients (ionic or bulk) conventionally delivered . The response of plants to nanoparticles (NPs) depends on the chemistry of soil and nanoparticle, exposure dose, and species of crop Dubey, 2013, 2015;Dimkpa et al., 2019). Nanofertilizers may come in various forms: 1) nanofertilizers made of macro-and micro-nutrients; and 2) application of nanomaterials as nutrient carriers (Kah et al., 2018). Accordingly, the first group could supplement one or more nutrients for plants, and the second group could enhance the efficiency of conventional fertilizer delivery but does not directly supply nutrients to plants (Liu and Lal, 2015). Upon soil application, NPs can enter root, then penetrate the cell wall/plasma membrane, reaching the root cortex and enter the xylem vessels thereby moving up through the stele to aerial plant tissues (Ma et al., 2010;Rajput et al., 2019).
Biological activities including metabolism of cell wall and ethylene, photosynthesis, mitochondrial respiration, and protection against free radicals and oxidative stress, are disrupted with Cu deficiency (Puig et al., 2007;Burkhead et al., 2009;Yuan et al., 2010). Moreover, Cu deficiency could affect overall plant growth, and fruit and seed yield (Rai et al., 2018). Exposure to toxic levels of Cu could increase chlorosis, necrosis, root growth inhibition and an increase in lignin content, leading to reduced cell expansion and nutrient uptake (Lequeux et al., 2010;Finger-Teixeira et al., 2010;Shaw and Hossain, 2013;Miransari, 2016). Mixture of ionic and nanoparticles of Cu, Zn, and B were investigated, and mixed responses were observed for growth, yield and nutrient uptake in soybean . Likewise, several studies have reported mixed results for Cu-NPs in lettuce, wheat and mung bean (Lee et al., 2008;Shah and Belozerova, 2009;Stampoulis et al., 2009).
Cu is naturally found in several foods and is available over the counter as a dietary supplement. Cu serves as a cofactor for many enzymes called cuproenzymes that play key roles in energy production, iron metabolism, neuropeptide activation, and synthesis of connective tissues and neurotransmitters in humans (IOM, 2001;Prohaska, 2012;Collins, 2014). Ceruloplasmin (CP), a cuproenzyme that constitutes over 95% of the total Cu in human plasma, plays a key role in iron metabolism (Hellman and Gitlin, 2002). Several physiologic processes including neurohormone homeostasis, angiogenesis, brain development, pigmentation, and regulation of gene expression and immune functioning are influenced by Cu (Collins, 2014). Cu-containing superoxide dismutases also play a major role in defense against oxidative damage (Owen, 1982;Allen and Klevay, 1994). According to the National Health and Nutrition Survey (NHANES) of data from 2009-2012, 6% to 15% of adults aged 19 and older who do not take dietary supplements containing Cu have Cu intake below the Estimated Average Requirement (EAR; an average daily intake estimated to meet the requirements of 50% of healthy individuals) (Blumberg et al., 2017). For adults using supplements, 2.2% to 7.2% of adults had intake below the copper EAR (Blumberg et al., 2017). Albeit uncommon, Cu deficiency could lead to anemia, hypopigmentation, hypercholesterolemia, connective tissue disorders, osteoporosis and other bone defects, abnormal lipid metabolism, ataxia, and increased risk of infection (Fairweather-Tait et al., 2011;Collins, 2014;Prohaska, 2014). It is therefore important to find ways to improve food Cu levels in edibles that are inexpensive and consumed widely.
Because soil-applied NPs will first come in contact with the root surfaces, following which biouptake and biodistribution of the elements (in pristine and/or modified form) could occur within the plant tissues, it is important to investigate potential effects of NPs on the root system architecture, soil-root interface, and potential accumulation in different tissues/organs in plants relative to its soluble ions, while building knowledge on ways to improve nutritional elements such as Cu in edible plant parts. Thus, in the present study, we assessed potential influence of particle size (25 nm, 50 nm, and 250 nm) and concentration (0, 50, 100, 200, 500 mg/kg-soil) of CuONPs on the root system architecture, and the physicochemical attributes of soil at the soil-root interface, leading to Cu transport and accumulation in root, stem, leaf and seeds in soybean, and compared with soluble Cu 2+ ions (positive control) and water-only control (negative control). A comparative assessment of total seed Cu levels in soybean with chickpeas and other valuable food sources for Cu intake and its implications to human health are also presented.

CuONPs synthesis and characterization
CuONPs with three different particle sizes (25 nm, 50 nm and 250 nm) were prepared by sol-gel method. The details of the synthesis protocol have been reported previously by our group (Yusefi-Tanha et al., 2020). Phase formation, crystal structure, microstructure and particle size distribution of the samples were characterized by X-ray diffraction (XRD) pattern analysis, and field emission scanning electron microscopy (FE-SEM). FE-SEM was also used for NP localization in the soybean seed. Dynamic light scattering (DLS) was used to estimate hydrodynamic diameter (HDD) and zeta potential of the CuONPs synthesized.

Experimental set up
Our experiment was carried out at the research farm of the Shahrekord University (50° 49 E, 32° 21 ́ N), Iran. The experiment was conducted using a factorial arrangement based on completely randomized design (CRD) with three replications. The treatments were CuCl2 (Cu 2+ ; positive control) and three different sizes (CuONP; 25, 50, and 250 nm) and five concentrations (0, 50, 100, 200 and 500 mg CuONP or Cu 2+ /kg). Each experimental unit consisted of two plants.

Soil preparation and CuONPs exposure conditions
Soil was collected from 0-30 cm depth and air dried for 7 days. It was sieved with a 2 mm sieve to separate any larger soil aggregates, wood chips and rocks. The percentage of sand, silt, and clay were 16%, 58%, and 26%, respectively. Additional properties of the soil are presented in Table   S1. Based on soil test, fertilizers of urea (86 kg/ha, 46% N; as a starter) and triple superphosphate (100 kg/ha, 44% P2O5) were added before planting ensued. Soil pH and EC were continuously measured until harvest (at intervals of 30 days) following Garcí a- Gómez et al. (2018) in soil:water (1:5 suspensions). For soil amendment, copper compounds (Cu 2+ and CuONP 25 nm, 50 nm, and 250 nm) were suspended in 100 mL of distilled water to achieve desired concentrations (0, 50, 100, 200 and 500 mg CuONP or Cu 2+ /kg-soil). Untreated soil was used as the negative control for each compound, and CuCl2 was added to soil as the positive control. Each NP suspension was shaken (30 min, 25°C) before adding to the soil, then mixed with soil using a hand-mixer before sowing.

Planting and crop management
The present study was done under outdoor microcosm conditions for better simulation of the experimental conditions in the natural environment. Seeds of soybean (Kowsar cultivar) were procured from the Seed and Plant Improvement Institute, Karaj, Iran. Cultivation was done in polyethylene (PE) pots. Each pot contained 4 kg soil (control and amendment). For easier plant removal from pot at harvest, each pot had an inner liner of PE mesh (with 50 holes of 5 mm for drainage), which was filled with a layer washed gravel (500 g). After inoculation of soil by symbiotic bacteria Rhizobium japonicum, two seeds were planted at 2.5 cm depth about 24 h after amendment of soil. Irrigation was based on field capacity. A water sub-sample was evaluated for total Cu concentration using inductively coupled plasma-optical emission spectroscopy (ICP-OES) in each irrigation water. Upon maturity, the plants were harvested, collecting the whole plant.
Aerial and roots tissues were then separated, oven dried (70°C for 48 h) in paper bags, weighed separately, and stored in plastic bags until analysis. Seeds were air-dried and stored.

Measurement of root parameters
To characterize the root system, root length (RL), root volume (RV), root area (RA), and root density (RD) were measured. After perfect washing the roots, a 1000 ml graduated cylinder was used to determine the root volume. So that a certain amount of water was poured into the cylinder, then the whole root was immersed in water and the volume added was equal to the root volume.
The root length was measured by a ruler. Dry weight of root was taken and expressed in gram per plant. The root density was expressed by dividing the mean root dry weight by the root volume (De Baets et al., 2007): where RDW (g) is mean root dry weight and RV (cm 3 ) is root volume.
The root area was calculated following the equation (De Baets et al., 2007): where RA (cm 2 ) is mean root area and RL (cm) is root length.

ICP-OES analysis of total copper
For measurement of total Cu accumulation in different plant tissues (root, stem, leaf, and seed), tissue samples (0.3 g) were washed several times with Milli-Q water and dried at 70°C for 48 h.
Subsequently, they were digested with 10 mL HNO3 (150°C for 1 h), then 2 mL HClO4 at 215°C for 2 h (5:1 v/v). The digests were further diluted up to 10 mL using deionized water. The extracts were filtered before being analyzed using by ICP-OES (Ghasemi Siani et al., 2017).

Seed copper concentration comparison with Recommended Dietary Allowance (RDA) and
Daily Value (DV) We compared soybean seed Cu levels with that in chickpeas seeds and other food sources. 100 g of soybean seed is assumed to be equivalent to 100 g of chickpeas seeds per serving of ½ cup, which is equivalent to 3.5 ounces.

Statistical analysis
A two-way analysis of variance (ANOVA) was performed using SAS (SAS Inc., ver. 9.4) to determine significant differences in crop responses to different treatments following a completely randomized experimental design. A Fisher LSD test at the 0.05 probability level was performed to further compare the means between the treatment groups. To determine if the concentrationresponse curves were linear (monotonic) or nonlinear (nonmonotonic), we coupled visual inspection of the curves with a simple decision rule: if the co-efficient of determination (R-squared) value for the linear regression line is 65% or higher the concentration-response curves were deemed linear, suggesting that the plant response changes linearly with the concentration applied following the relationship: y = ax + b; where y denotes dependent variable, x denotes independent variable, and a and b are model parameters. The computed R-squared values are presented in Table   4.

Nanoparticle characterization
Despite apparent particle aggregation, the lognormal fitting of the particle size distribution using the FE-SEM micrographs showed that the mean particle size for S1, S2 and S3 samples are 25 nm, 50 nm and 250 nm, respectively (Fig. 1). The Reitveld analyses of the XRD patterns are shown in

Root dry weight
Our findings showed significant effects of copper compound type (Cutype; p<0.0001), concentration (C; p<0.0001) and their interactions (Cutype × C) on root dry weight in soil-grown soybean (p≤0.01) ( Table 2). Particle size-and concentration-dependent inhibition in root weight was observed upon exposure to CuONPs, and the two forms of Cu (CuONPs and Cu 2+ ) at all concentrations led to decreased root dry weight. However, the effect of CuONP-25 nm was significantly higher compared to the larger sized CuONPs or Cu 2+ ions treatments (Table 3). Root dry weight decreased linearly in a concentration-dependent fashion for the larger size CuONPs (CuONP-50, CuONP-250) and Cu 2+ ions whereas for CuONP-25 the relationships appeared nonlinear ( Table 4). Although the lowest root dry weight was at 500 mg/kg CuONP-25 nm, it was not significantly different from 200 mg/kg CuONP-25 and 500 mg/kg CuONP-50 treatments.
Although at low concentrations, Cu is a necessary micronutrient for plant growth and development, higher concentrations in soil-root interface could lead into harmful effects on plant growth (Nair and Chung, 2014a). Cu concentration in different plant tissues is typically in the range 2.0-50 µg/g dry weight (Barker and Pilbeam, 2015). Our results indicate particle size-and concentration-dependent toxicity of CuONPs in root weight of soybean, and that CuONPs toxicity may not be related to Cu 2+ ions released because Cu 2+ ions alone treatments resulted in significantly lower toxicity compared to CuONPs treatments at all concentrations (50-500 mg/kg) ( Table 3). Previously, it was documented that exposure to CuONPs (<50 nm) could decrease stem and root growth in rice (Shaw and Hossain, 2013), barley (Shaw et al., 2014), and wheat (Dimkpa et al., 2012). Likewise, particle size-and concentration-dependent toxicity of CuONPs in Arabidopsis showed reduced root growth, roots lignification, and plant biomass (Nair and Chung, 2014b). Additionally, root lignification and growth modification in Glycine max (Lin et al., 2005) and A. thaliana (Lequeux et al., 2010) were reported with Cu 2+ exposure (0-5 µM), suggesting that absorbed dissolved Cu ions can lead to decreased root growth in soybean.
Our results for root length are consistent with the earlier studies that reported decreased root length in mustard (Nair and Chung, 2015), soybean (Nair and Chung, 2014a), and mung bean upon exposure to CuONPs (Nair et al., 2014).

Root volume
Root volume in soybean exposed to different Cu compounds are presented in Table 5. The results show that the effect of Cu compound type (Cutype; p<0.0001), concentration (C; p<0.0001) and the interaction term (Cutype × C; p≤0.05) were statistically significant for root volume in soil-grown soybean ( Table 2). Our findings clearly show changes related to particle size-and concentration in root volume upon exposure to CuONPs. Amongst all the treatments tested, root volume was significantly lowest at 500 mg/kg for CuONP-25 nm, but not significantly different with 200 mg/kg CuONP-25 nm and 500 mg/kg CuONP-50 nm, similar to root dry weight (p<0.05; Table   5). Further, root volume for CuONP-250 nm was similar to Cu 2+ ions treatment at all concentrations. At 50 mg/kg Cu 2+ ions, root volume was similar to control. For all CuONPs types, root volume was not significantly different at two lower concentrations (50 and 100 mg/kg), unlike two higher concentrations (200 and 500 mg/kg) ( Table 5). The observed decrease in root volume upon exposure to small-sized CuONPs at higher concentrations could be attributed to decreased cell division, lateral root count, and root elongation ( Table 3). Consistent to our study, a previous study also found reduced cell division and cell elongation, leading to reduced root elongation in sand-grown wheat upon treatment with CuONPs (>10 mg Cu/kg) (Adams et al., 2016).

Root area
ANOVA indicated that root area was affected by Cu compound type (Cutype; p<0.0001), concentration (C; p<0.0001) and the interaction term (Cutype × C; p<0.001) ( Table 2). Our results generally show particle size-and concentration-dependent root area in soybean upon exposure to CuONPs, and similar to other root parameters results (Table 3, 5), the decreasing effects of CuONP-25 nm was significantly higher compared to the larger sized CuONPs or Cu 2+ ions treatments for most concentrations ( Table 5). The lowest root area was recorded at 500 mg/kg CuONP-25 nm that was not significantly different with CuONP-25 at 200 mg/kg (Table 5).
Further, at 50 mg/kg Cu 2+ ions, root area was not significantly different from control, akin to the root length and root volume ( Table 3, 5). Also, there was no significant difference in root area between CuONP-250 nm and Cu 2+ ions at all concentrations tested (p<0.05). Small-sized CuONPs at higher concentrations decreased root area significantly due to reduced root length ( Table 3) and root volume ( Table 5). For CuONP-250 nm, the trend of root area change was relatively minimal across different concentrations tested.

Root density
Root density in soybean exposed to different copper compounds are presented in Fig. 3A. The effects of copper compound type (Cutype; p≤0.0001) and concentration (C; p<0.0001) was significant on root density, but the interaction term (Cutype × C) was not (p>0.5; Table 2). Plants exposed to CuONP-25 nm showed the lowest root density compared to the larger sized CuONPs (50 and 250 nm) or Cu 2+ ions. Also, root density was similar in larger sized CuONPs (50 and 250) and Cu 2+ ions treatments (Fig. 3A). At the highest concentration of copper compounds (500 mg/kg), the root density was significantly reduced in comparison with the lower concentrations and control ( Fig. 3B). In our study, CuONP-25 nm showed a significant decrease in root density with concomitant decrease in root dry weight (Table 3) and root volume ( Table 5). In addition, with increasing concentration, the decrease in root density was greater compared to control (Fig. 3B).

Root copper uptake
The results indicated that the effects of Cu compound type (Cutype), concentration (C) and the interaction term (Cutype × C) were significant for Cu concentration in soybean root (p<0.0001) ( were not statistically significant (p<0.05). As well as, there were no significant difference between 50 mg/kg larger sizes of CuONP (50 nm and 250 nm) and 100 mg/kg CuONP-250 and Cu 2+ ions (p<0.05) (Fig. 4A). Generally, root Cu uptake was similar for Cu 2+ ions and largest size CuONPs (250 nm), unlike the smallest size CuONPs (25 nm) that had highest root Cu uptake responses in soybean (Fig. 4A).
Several factors, including plant species, concentration used, root morphology, and soil properties, can influence Cu uptake and bioaccumulation in plants ( Monica and Cremonini, 2009).
Phytochelatins and metallothionein are momentous organic complexes, presumably formed in root cells, can promote Cu retention in the soil-root interface (Rawat et al., 2017). Cu ions from NPs have been shown to decrease root length, water content and dry biomass in lettuce. A significant accumulation of Cu in root exposed to Cu/CuONPs (20-30 nm) was documented compared to mg/kg of CuONPs as their roots contained up to ~479 mg Cu/kg dry weight (p ≤ 0.05).

Stem copper uptake
A significant effect of Cu compound type (Cutype; p<0.0001), concentration (C; p<0.0001) and the interaction term (Cutype × C; p<0.001) on stem Cu uptake in soybean was observed ( Table 6). Our results show particle size-and concentration-dependent Cu uptake in soybean stem upon exposure to CuONPs. The concentration-response curves for all CuONPs and Cu 2+ ions were linear ( Table   4). Exposure to CuONP-25 nm led to significantly higher Cu uptake in stem compared to the larger sized CuONPs and Cu 2+ ions treatments, and this uptake increased nearly two-fold at 500 mg/kg treatment compared to 50 mg/kg treatment for all Cu compound types (Fig. 4B). The stem Cu concentration increased to 7.3 mg/kg, on average, for CuONP-25 nm from the baseline of 3 mg/kg in control. Plants treated with CuONP-250 nm had similar stem Cu content compared to Cu 2+ ions treatment at all concentrations tested (Fig. 4B). Generally, stem Cu uptake was similar for Cu 2+ ions and larger size (50 nm and 250 nm) CuONPs, unlike the smallest size CuONPs (25 nm) that had highest stem Cu uptake responses in soybean (Fig. 4B).

Metal and metal oxide NPs can induce toxicity either by releasing toxic metal ions or by
direct interaction with the cell (Manusadžianas et al., 2012;Andreotti et al., 2015). Plant can absorb metals as dissolved or soluble ionic fractions or absorb as NPs themselves. Dimkpa et al. (2012) observed Cu bioaccumulation in wheat stem upon exposure to 500 mg/kg CuONPs (<50 nm), and that soluble Cu from CuONPs was implicated in phytotoxicity. Consistent with our results, uptake of CuONPs by leaf fronds in Landoltia punctata was found to be more toxic with CuONPs treatments compared to ionic Cu treatments (Shi et al., 2011). Exposure to CuONPs (10-100 nm) and CuNPs (100-1000 nm) showed reduction in root growth in lettuce and alfalfa, and increased stem Cu content in alfalfa (Hong et al., 2015).

Leaf copper uptake
Our results show leaf Cu concentration in soybean was significantly affected by Cu compound type (Cutype), concentration (C), and the interaction term (Cutype × C) (p<0.0001; Table 6). Further, leaf Cu uptake was particle size-and concentration-dependent. Similar to root Cu concentration, the concentration-response curves for the larger size CuONPs (CuONP-50 nm, CuONP-250 nm) and Cu 2+ ions were linear ( Fig. 4C; Table 4). Leaf Cu concentration upon treatment with CuONP-25 nm at 50, 100, 200, and 500 mg/kg concentrations increased 3.9, 4.7, 5.7, and 6.0 times, respectively, compared to control. Leaf Cu uptake that was the highest at 500 mg/kg CuONP-25 nm treatment was not significantly different with 200 mg/kg CuONP-25 nm treatment.
Additionally, the effects of larger sized CuONPs (50 nm and 250 nm) were not significantly different when compared between 50 and 500 mg/kg treatments (Fig. 4C). Generally, leaf Cu uptake was similar for Cu 2+ ions and larger size (50 nm and 250 nm) CuONPs, unlike the smallest size CuONPs (25 nm) that had highest leaf Cu uptake responses in soybean (Fig. 4C).
Nanoparticles must be absorbed by the root for uptake and accumulation by aerial plant parts. They enter vascular tissue (xylem) upon penetrating the cell walls and plasma membrane, and translocating to stem, leaf and ultimately to seed (Ma et al., 2010). A linear relationship between Cu uptake/accumulation in different tissues and exposure concentrations of Cu-NPs in the growth medium was previously reported (Ma et al., 2010). The pores in cell wall could be below 10 nm in diameter (Ma et al., 2010;Albersheim et al., 2011), which is much smaller than the size range we tested for CuONPs. It was hypothesized that smaller sized ENP aggregates can pass through the pores reaching the plasma membrane, unlike larger aggregates that would not (Navarro et al., 2008 a,b;Kim et al., 2012). It is likely for ENPs to also create new pores upon cell surface interactions, enabling larger ENP internalization into plant tissues. However, cellular payloads such as innate and foreign macromolecules (e.g., proteins, peptides) could actively transport in and out of the cell (Ross et al., 2015), and studies have documented that such larger molecules might transport via plasmodesmata, a roughly cylindrical channel reaching up to 40 nm in diameter (Heinlein and Epel, 2004;Lucas and Lee, 2004;Microbe Notes, 2019). Shi et al. (2014) found that exposure to CuONPs with diameter 34-52 nm (100, 200, 500, 1000, and 2000 mg/L) led to reduced root length, and NPs accumulation in root and leaf, in Elsholtzia splendens. Cu bioaccumulation in cowpea leaf was also enhanced by CuNPs with sizes < 25 nm and 60-80 nm; however, increasing exposure concentrations led to decreased translocation (Ogunkunle et al., 2018), suggesting a threshold for NP uptake and translocation in plants. Deng et al. (2020) also reported that leaf Cu accumulation pattern of Brassica rapa treated with nano CuO (75, 150, 300, and 600 mg Cu/kg soil) depends on particle size and plant phenotype. Our findings suggest that CuONPs with small size (25 nm) may be favored for cellular entry and translocation compared to the larger size CuONPs, promoting increase in leaf Cu uptake in soybean. Although dissolved, Cu 2+ ions showed overall lower uptake in different plant tissues compared to small size CuONPs (25 nm). In addition, the leaves of the soybean plant fall off after ripening or before harvesting. Therefore, biofortification of leaves in CuONPs treatments (Fig. 4C) may affect the microbial decomposition of soybean residues, and the positive and negative aspects of this subject need to be investigated in the future research.

Seed copper uptake
ANOVA showed that the effect of Cu compound type (Cutype; p<0.0001), concentration (C; p<0.0001), and the interaction term (Cutype × C; p<0.01) were statistically significant for seed Cu concentration ( Table 6). The results generally show particle size-and concentration-dependent seed Cu uptake in soybean upon exposure to CuONPs, and plant exposed to CuONP-25 nm typically showed higher Cu uptake compared to the larger sized CuONPs (50 nm and 250 nm) or Cu 2+ ions at most concentrations tested; a result consistent with the root, stem and leaf Cu uptake (Fig. 4D) ; Fig. 4D). Generally, seed Cu uptake was similar for Cu 2+ ions and larger size (50 nm and 250 nm) CuONPs, unlike the smallest size CuONPs (25 nm) that had highest seed Cu uptake responses in soybean (Fig. 4D).
Potential aggregation of CuONP-250 nm at higher concentrations may decrease metal bioavailability, reducing metal uptake and toxicity. In our study, total Cu concentrations differed among tissue types in the order: roots > leaves > stem > seeds (Figs. 4A-D), suggesting that the tissues furthest away from the root-soil system had the lowest total Cu (i.e., seed) and the root that is in direct contact with the soil had the highest total Cu levels. Wang et al. (2012) also showed metallic NPs phloem-based translocation from leaves to other parts of plant. CuONP-25 enabled higher Cu translocation to seed, probably due to its smaller size, less accumulation and easier passage through the cell wall. It was illustrated that NPs with smaller particle size can be increasing Cu availability even at low concentration, which may be due to increasing surface area of small nanoparticles (Andreotti et al., 2015). However, CuONPs with larger particle size can form more aggregates, reduce surface area and thus decrease availability of Cu, especially in the seed. Seeds of cowpea accumulated significant Cu amounts comparison to control, and the highest content of Cu was related to CuNPs < 25 nm and 60-80 nm, respectively at 500 mg/kg and 1000 mg/kg (Ogunkunle et al., 2018). In lettuce, Cu/CuONP treatments altered nutritional quality compared to the control, because this plant had more Cu, S, and Al, but less Mg, Ca, P, and Mn ( Our findings indicate that during the growth period, pH of the soil reduced, but remained alkaline (Figs. 5A-B). At low pH, higher concentrations of NPs are more phytotoxic due to potential for more dissolution of metals (Qiu and Smolders, 2017). Accordingly, as CuONPs concentration increased, the available Cu increased with decreasing pH. Shi et al. (2014) found CuONPs dissolution promoted inside the cell because of the decreased cellular pH.
In the present study with reducing pH as Cu concentrations increased in soil, a higher Cu uptake was associated with lower root growth ( Table 3, 5), which could lead to decreased plant growth and yield in soybean (Yusefi-Tanha et al., 2020). Fig. 5B shows the pH changes for CuCl2 amended soil at different concentrations, which were similar to the changes documented for CuONPs (Fig. 5A).
Copper is a low-mobility element in the soil, but can form very intense chelates (Rawat et al., 2017;Sekine et al., 2017). Potential toxicity of metal oxide NPs might be due to the dissolution and release of toxic ions. Dissolution can be affected by environmental conditions where the NPs are found; for example, pH and EC (Andreotti et al., 2015). Soil pH can affect metal bioavailability and phytotoxicity of NPs (Garcí a- Gómez et al., 2018). Generally, alkalinic condition promotes NPs aggregation, while acidic environment could trigger metal/oxide NPs dissolution, transforming into ionic species (Peretyazhko et al., 2014;Zhou et al., 2016). Additionally, in alkaline pH, aggregation of NPs can alter nano-specific attributes, and the dissolution propensity into ionic forms could decrease. However, with a pH change, the NPs can also disaggregate and return to a previous stable state. Likewise, NPs reactivity and toxicity can be modified with a minor change in surface charge and particle size (Silva et al., 2014), that this can also change as a function of media chemistry Pokhrel et al., 2014;Dimkpa, 2018).  Table 7 is 0.9 mg (900 ug) for adults and children age 4 years and older (NIH, 2020). FDA required manufacturers to use these new labels starting in January 2020, but companies with annual sales of less than $10 million may continue to use the old labels that list a Cu DV of 2 mg (2,000 ug) until January 2021 (US FDA, 2013. The FDA does not require food labels to list Cu content unless it has been added to the food. Foods that offers ≥20% of the DV are considered to be high sources of a nutrient, but foods providing lower percentages of the DV may also contribute to a healthful diet (NIH, 2020).
Our results of soybean seed Cu concentrations upon different sized CuONPs (25 nm,50 nm,250 nm) treatments at variable soil Cu concentrations (50-500 mg/kg soil) demonstrated the potential for significant improvement in seed Cu uptake ( Fig. 4D mirrored that of largest size CuONPs-250 nm ( Table 7).
Comparing soybean seed Cu concentrations with chickpeas seed, a legume considered good source of nutrients and phenolic compounds (e.g., polyphenols, isoflavones) with antioxidative potential to reduce oxidative effects with evidence supporting its consumption in prevention and management of diabetes and obesity (De Camargo et al., 2019), we found that our soybean seeds had 1.38-2.30 fold higher Cu concentrations than chickpeas seed per serving of ½ cup (100 g or 3.5 ounces) ( Because we found evidence of CuONPs within the cell wall, cell membrane and protein storage vacuoles in the cytoplasm of the soybean seed embryo using electron microscopy for CuONPs-25 nm treatment (Fig. 6), it is paramount to understand the potential toxicity of CuONPs in humans upon consumption of soybean seeds and oil. Future research should address what risk, if any, does CuONPs pose to humans before such Cu fortified soybean seeds are used for daily human consumption to address Cu deficiency and related illnesses, globally.

Conclusions
Utilization of CuONPs in modern agriculture as a novel fertilizer requires an understanding of their uptake, translocation and toxicity in plants. Our findings showed particle size-and concentration-dependent influence of CuONPs on Cu uptake and tissue distribution in root, stem, leaf and seed in soybean grown for full lifecycle of 120 days.        6. TEM analysis of ultrastructure of soybean seed embryo with CuONPs-25 nm treatment at 500 mg/kg-soil (C, D) and compared with control seeds (no nanoparticles) (A, B). Electron dense metal aggregates are clearly visible within cell wall (cw)/ plasma membrane (pm) including within the cytoplasm (cy) and/or protein storage vacuoles (psv) for the seeds with CuONPs-25 nm treatment at 500 mg/kg-soil (C, D). nu = nucleus, ics = intracellular space, ob = oil bodies, NPs = nanoparticles (red triangle).