Biofortification of Silage Maize with Zinc, Iron and Selenium as Affected by Nitrogen Fertilization

Agronomic biofortification is one of the main strategies for alleviation of micronutrient deficiencies in human populations and promoting sustainable production of food and feed. The aim of this study was to investigate the effect of nitrogen (N)fertilization on biofortification of maize crop (Zea mays L.) with zinc (Zn), iron (Fe) and selenium (Se) grown on a micronutrient deficient soil under greenhouse conditions. Factorial design experiment was set under greenhouse conditions. The experiment consisted of two levels of each N, Zn, Fe and Se. The levels for N were 125 and 250 mg N kg−1 soil; Zn were 1 and 5 mg Zn kg−1 soil; levels of Fe were 0 and 10 mg Fe kg−1 soil; levels of Se were 0 and 0.02 mg Se kg−1 soil. An additional experiment was also conducted to study the effect of the Zn form applied as a ZnO or ZnSO4 on shoot growth, shoot Zn concentration and total shoot Zn uptake per plant. Shoot Zn concentrations increased by increasing soil Zn application both with ZnSO4 and ZnO treatments, but the shoot Zn concentration and total Zn uptake were much greater with ZnSO4 than the ZnO application. Under given experimental conditions, increasing soil N supply improved shoot N concentration; but had little effect on shoot dry matter production. The concentrations of Zn and Fe in shoots were significantly increased by increasing N application. In case of total uptake of Zn and Fe, the positive effect of N nutrition was more pronounced. Although Se soil treatment had significant effect, N application showed no effect on Se concentration and accumulation in maize shoots. The obtained results show that N fertilization is an effective tool in improving the Zn and Fe status of silage maize and contribute to the better-quality feed.


Introduction
Agronomic biofortification of field crops with micronutrients is one of the main strategies in sustainable production of healthy and nutrient rich food and feed [1]. Human health problems are associated with micronutrient deficiency worldwide, especially in developing countries, but also the productivity of farm animals grown in these countries is negatively affected [2][3][4]. Low amount of phytoavailable micronutrients in cultivated soils and commonly consumed food and feed crops are main reason of the high prevalence of micronutrient deficiencies in humans [5,6]. According to Hill and Shannon (2019) [7], grazing animals are often exposed to high risk of reduced Zn intake because the pastures usually contain inadequate Zn concentration for a proper animal nutrition that is associated with low amount of phytoavailable Zn in soils. Consequently, there is an increasing trend for biofortification of feed crops with Zn to contribute to better Zn nutrition of livestock. Similarly, enrichment of feed crops with Se is of great importance for animal nutrition and health [8]. Selenium and selenoproteins, such as selenomethionine and selenocysteine, play a role in several critical biological functions in human and animal body and prevent development of various important diseases [9,10]. In essence, micronutrient deficiency affects all phases of food and feed production chain, from field to the final consumer.
Fe on grain Fe was not adequately high compared to Zn, Se and iodine. In previous studies, it has been suggested that N nutritional status of plants has an important role in increasing root uptake, shoot transport and grain deposition of Fe as well as Zn [1,37]. It was interesting to notice that increasing rate of Fe application had little effect on grain Fe; but at a given Fe application rate increasing N application increased grain Fe [32]. Kutman et al. (2010) and Erenoglu et al. (2011) [38,39] showed that increasing N application promotes root uptake of Zn and Fe and increases shoot and grain concentration of these nutrients. Literature reports also suggest that, in case of maize, N application positively affects maize shoot and grain micronutrient concentrations to certain extent under different field conditions [40][41][42][43]. Agronomic biofortification, a widely accepted approach in preventing micronutrient deficiency in several food and feed crops, is not a well-known practice in Western Balkan countries [15].
Considering that, maize silage is one of basic feed component for the dairy cattle in many Balkan countries [44], it is important to investigate the effect of N fertilization on uptake and concentration of Zn, Fe and Se in maize plant. Since N, fertilization is shown to stimulate the uptake and concentration of Zn and Fe and perhaps Se in plants, this may help to produce silage of higher nutritional value, with respect to daily needs of dairy cattle. Practically this fortified silage could lead to reduce the use of different micronutrients supplements.
The present study was planned to examine the effect of N and Zn application on: (1) maize shoot dry matter yield; (2) Zn, Fe and Se concentration and uptake in maize; and (3) the relationship between the concentration of N and that of Zn and Fe in maize plants. Furthermore, the effect of the Zn form applied as a ZnO or ZnSO 4 on shoot growth, shoot Zn concentration and total shoot Zn uptake was also investigated.
The soil used for this experiment originated from the Zn-deficient region of Central Anatolia and has clay loam texture with pH 7.6 (H 2 O), 1.5% organic matter, and 18% CaCO 3 . The diethylentriamine pentacetic acid (DTPA)-extractable Zn and Fe concentration was 0.1 mg kg −1 and 2.1 mg kg −1 soil, respectively, and 0.1 M KH 2 PO 4 extractable Se was 0.002 mg kg −1 soil.
Two experiments were established. In the first one, the experiment consisted of two levels of each N and Zn. The levels for N were 125 and 250 mg N kg −1 soil and that of Zn were 1 and 5 mg Zn kg −1 soil. The sources of N and Zn were Ca (NO 3 ) 2 4H 2 O and ZnSO 4 7H 2 O, respectively. Similarly, there were two levels of Fe, i.e., 0 and 10 mg Fe kg −1 soil and two levels of Se, i.e., 0 and 0.02 mg kg −1 . Iron and Se fertilizers were applied in form of Fe-sequestrene and Na 2 SeO 4 , respectively.
Basic fertilizers added were phosphorus (P) 100 mg P kg −1 soil in the form of KH 2 PO 4 , sulfur (S) 50 mg S kg −1 in the form of K 2 SO 4 , and potassium (K) in the form of KH 2 PO 4 and K 2 SO 4 . All these nutrients and basic fertilizers were homogeneously mixed with 3 kg soil prior to putting the mixture into plastic pots. The experiment was laid out as factorial randomized design with 4 replicates. Twelve seeds of maize (Zea mays L.cv. Shemal) were sown in each plastic pot. Shortly after emergence, number of plants was reduced to 6 plants/pot. The pots were irrigated daily with deionized water. When plants were 25 days old, their shoot parts were harvested.
In the 2nd experiment, two different forms of Zn (i.e., ZnSO 4 and ZnO) were used to study how these Zn sources affect the shoot growth and shoot concentration of Zn of maize plants. Zinc was applied at the rates of 0, 0.5, 1, 2.5 and 7.5 mg kg −1 soil in the forms of ZnO and ZnSO 4 at the sowing time. When the plants were 35 days old, plants were harvested and analyzed for shoot production and shoot Zn concentration.

Plant Sampling and Chemical Analysis
Plant shoot materials, after washing with deionized water, were dried at 70 • C for the dry mater determination of shoot weight and analysis of micronutrients. Dried and ground plant samples (0.2 g) were digested with acid [a mixture containing 2 mL of 30% (v/v) H 2 O 2 and 5 mL of 65% (v/v) HNO 3 ] in a closed-vessel microwave system (Mars Express; CEM Corp., Matthews, NC, USA). The digested solution was diluted with DI water. For the determination of Zn, Fe and Se in the digested solution, inductively coupled plasma optical emission spectrometry (ICP-OES) (Vista-Pro Axial, Varian Pty Ltd., Mulgrave, Australia) was used. A dry combustion method (950 • C) using a LECO Tru-Spec C/N Analyzer (Leco Corp., St Joseph, MI, USA) was used for the determination of N concentration in the samples. The certified standard reference materials, obtained from the National Institute of Standards and Technology (Gaithersburg, MD, USA), were used for checking the precision in mineral nutrient analysis.
As standard sample, the SRM 1547 peach leaves, were used and the deviation was below 2%. The total uptake per plant was calculated by multiplying the concentration with dry mater yield.

Statistical Analysis
The R commander program was used for statistical analysis of the results obtained. Analysis of variance (ANOVA) was used to assess the significance level of the effects of treatments and their interactions on the reported traits. Tukey test at 5% level (p ≤ 0.05) was used for significant difference among means, whenever ANOVA (general linear model) indicated significant effect of treatments. The relationship between the treatments was assessed by linear regression model.

Results
Zinc deficiency symptoms (i.e., development of yellow or yellowish-white stripes along the midrib of younger leaves) appeared in plants grown under low supply of Zn ( Figure 1). The symptoms started to develop first following 2 weeks of growth without Zn application under given conditions. Expression of these symptoms was more severe at low N than at adequate N supply. In case of Fe deficiency, uniform chlorosis on younger leaves appeared. Under low N supply, older leaves turn uniformly pale green and then whole shoot look slightly yellowish. The effect of soil N and Zn applications on maize shoot dry matter (DM) yield, shoot Zn and Fe concentrations, and their total uptake per plant was significant.
By contrast, soil N and Zn applications did not significantly affect shoot Se concentration and total Se uptake (Table 1). Variation in N supply did not show a significant impact on the shoot dry matter yield of the plants at both low and adequate Zn treatment ( Table 2), indicating that low N supply was still enough at this growth phase of maize under given experimental conditions.
On the other hand, variation in Zn soil application both under low and adequate N supply significantly affected shoot dry mater production ( Table 2). Table 2 shows that the shoot dry matter of plants at low Zn and adequate N supply was 10% higher than the shoot dry matter of plants produced under low Zn and low N treatments. The shoot dry matter of plants under adequate Zn and N supply was 12% higher than those plants grown under adequate Zn but low N supply.
Both Zn concentration and total Zn uptake by shoots were increased by increasing soil Zn application (Table 3). At the low Zn application rate, the plants grown under adequate N supply showed significantly higher concentration and shoot uptake of Zn than the plants grown under low N supply. The increased soil N supply showed positive effect on the shoot Zn concentration and uptake in plants grown with adequate Zn soil application (Table 3).  Variation in N supply did not show a significant impact on the shoot dry matter yield of the plants at both low and adequate Zn treatment ( Table 2), indicating that low N supply was still enough at this growth phase of maize under given experimental conditions.   Selenium soil treatment had significant effect on Se concentration and shoot Se uptake in maize shoot (Table 4), while both N and Zn application did not affect plant Se status (Table 3). Although, not statistically significant, a decrease of the Se concentration could be observed with increase of both N and Zn soil application (Table 3).  Shoot Fe concentration and uptake were greatly affected by N and Zn supply (Table 3). Both Zn and Fe uptake were positively affected by increasing soil N supply (Tables 3 and 4). Shoot Fe concentration and uptake at adequate N application were significantly higher compared to low N application, while the shoot Fe concentration and uptake at adequate Zn supply significantly decreased compared to the low Zn treatment (Table 3). Although increase in soil N supply promoted the Zn and Fe concentration in maize shoot, the ratio between Zn and Fe was mainly affected by the level of soil Zn application ( Figure 2). Shoot Fe concentration and uptake were greatly affected by N and Zn supply ( Table  3). Both Zn and Fe uptake were positively affected by increasing soil N supply (Tables 3  and 4). Shoot Fe concentration and uptake at adequate N application were significantly higher compared to low N application, while the shoot Fe concentration and uptake at adequate Zn supply significantly decreased compared to the low Zn treatment (Table 3). Although increase in soil N supply promoted the Zn and Fe concentration in maize shoot, the ratio between Zn and Fe was mainly affected by the level of soil Zn application ( Figure  2). The relationship between N-Zn and N-Fe concentrations in shoot is shown in Figure  3. The positive impact of N nutrition on Zn and Fe concentrations in shoot in relation to N shoot concentration is shown both under low and adequate Zn supply.
The relation between N-Zn in shoot is highly dependent on availability of N in the soil irrespective of Zn soil availability. Although Zn application decreased shoot Fe concentration (Figure 2), still increase in N supply significantly affected N-Fe relationship, The well-documented effects of Zn nutrition on growth of plants and shoot Zn concentrations and uptake were more distinct with the soil ZnSO4 application when compared to the soil ZnO application (Table 5). A progressive increase in shoot dry matter production was observed with increasing soil Zn application, and this increase was stronger and more pronounced in case of soil ZnSO4 application. These much better effects of ZnSO4 application on shoot growth than the ZnO application are also shown in Figure 4a,b.
The plants receiving ZnO could not develop well when sol Zn application was <1 mg Zn kg −1 soil. However, shoot Zn concentrations showed a clear increase by increasing soil Zn application both with ZnSO4 and ZnO treatments. Since the enhancements in shoot dry matter by increasing soil Zn supply was more obvious by ZnSO4 application, the shoot total Zn uptake was accordingly much greater with ZnSO4 than the ZnO applications (Table 5). The relation between N-Zn in shoot is highly dependent on availability of N in the soil irrespective of Zn soil availability. Although Zn application decreased shoot Fe concentration (Figure 2), still increase in N supply significantly affected N-Fe relationship, and as in case of N-Zn relationship, N-Fe relationship is strengthened by the increase of N soil availability (Figure 3).
The well-documented effects of Zn nutrition on growth of plants and shoot Zn concentrations and uptake were more distinct with the soil ZnSO 4 application when compared to the soil ZnO application (Table 5). A progressive increase in shoot dry matter production was observed with increasing soil Zn application, and this increase was stronger and more pronounced in case of soil ZnSO4 application. These much better effects of ZnSO 4 application on shoot growth than the ZnO application are also shown in Figure 4a,b. Table 5. Effects of increasing soil Zn supply as ZnO and ZnSO 4 on shoot dry matter production, shoot Zn concentration and total Zn uptake of 34-days-old maize plants grown on a Zn-deficient calcareous soil. Zinc has been applied at the rates of 0, 0.5, 1, 2.5 and 7.5 mg Zn kg −1 soil in form of ZnO and ZnSO 4 .

Zn Source
Zn Supply mg/kg Dry Matter (g plant −1 ) Shoot Zn (mg kg −1 ) Shoot Zn Uptake (µg plant −1 )  The plants receiving ZnO could not develop well when sol Zn application was <1 mg Zn kg −1 soil. However, shoot Zn concentrations showed a clear increase by increasing soil Zn application both with ZnSO 4 and ZnO treatments. Since the enhancements in shoot dry matter by increasing soil Zn supply was more obvious by ZnSO 4 application, the shoot total Zn uptake was accordingly much greater with ZnSO 4 than the ZnO applications (Table 5).

Discussion
Maize is known to be highly sensitive to low Zn in soils [24]. In good agreement with this, decreases in Zn soil application showed detrimental effects on the shoot growth of maize plants (Tables 2-5; Figure 4a,b). Very positive response of maize plants to increasing soil Zn application was pronounced in case of ZnSO 4 application. This indicates clearly that use of water-soluble Zn fertilizers in high pH soils is highly desirable. The results presented here are similar those presented for maize plants grown in calcareous soils in the previous studies [45,46]. ZnO is known to be nearly water insoluble (e.g., 0.0016 g per liter) while ZnSO4 shows very high-water solubility (i.e., 580 g per liter) [47]. Use of watersoluble Zn fertilizer is, therefore, highly desirable for high pH soils. At least 40 to 50% of Zn in each granular fertilizer should be water soluble to achieve a positive agronomic impact on plant growth by using Zn-containing compound fertilizers in high pH soils [30,46,48]. Recently, Degryse et al. (2020) [48] emphasized importance and relevance ofwater-soluble Zn in each granular fertilizer and highlighted that water-soluble Zn rather than the total Zn should be considered in the fertilizer labeling regulations.
Studies dealing with the use of nanoparticle ZnO in Zn fertilization of plants is growing with controversial results and debates [49,50]. One critical debate is related to the very poor solubility of Zn existing in nanoparticulated ZnO. Published evidence shows that solubility and diffusion of Zn from a granular fertilizer which is coated by a bulk or nanoparticulated ZnO are not affected from the size of ZnO used [51]. Therefore, use of nanoparticulated ZnO-containing granular fertilizers in high pH soils may have a very minimal agronomic impact on plant growth and plant Zn uptake when compared to use of fertilizers containing higher percentage of water-soluble Zn.
The present study shows that besides the form of the Zn fertilizer used, the N nutritional level of the plants also influences plant Zn concentration. Although varied soil N supply only slightly affected the shoot dry matter production of plants under given experimental conditions, shoot Zn concentration as well as Zn accumulation (i.e., total Zn uptake by shoot) of plants were significantly increased through the increase in N fertilization of plants (Table 2). These results are like those published by LeBlanc et al. (1997) [40], and Xue et al. (2014) [43] for maize and Cakmak et al. (2010b) [52] for wheat grown under field conditions. Positive effects of increasing N fertilization on plant Zn concentrations have been also shown under greenhouse conditions [53].
In a short-term experiment, Erenoglu et al. (2011) [39] showed that Zn transport through root uptake and root to shoot was significantly promoted by N fertilization. An increased N nutrition of plants showed probably positive effects on abundance of Zn-chelating ligands (such as amino acids and amines) and transporter proteins involved in the root uptake and root to shoot transport of Zn in the plants such as nicotianamine, ZIP family proteins and YSL transporters [37,38]. The presented positive effect of higher soil N supply on Zn concentration and uptake is more emphasized under adequate soil Zn application. This result compares well with those found by Kutman et al. (2010) [38] in wheat. According to Kutman et al. (2010) [38] Zn and N are synergistic in their effects on increasing plant Zn concentrations and their levels in growth medium should be at enough levels to achieve the synergistic effect of N on root Zn uptake.
Like Zn, increasing soil N supply also positively affected shoot Fe concentrations, even at much higher level than Zn (Tables 3 and 4). Losak et al. (2011) [41] in maize and Aciksoz et al. (2011a) [32] in wheat also found the positive effects of N fertilization on plant Fe concentrations. Interestingly, increasing soil Fe fertilization in the form of FeSO 4 and FeEDTA had no clear effect on shoot and grain Fe concentrations of wheat plants; but at a given Fe dose, increasing N fertilization significantly improved shoot and grain Fe [32]. The N-dependent increases in plant Fe concentrations were ascribed to N-dependent increases in level of Fe-chelating and transporting nitrogenous compounds such nicotianamine and phytosiderohores. Accordingly, it was shown that improving plants N nutritional status increased root release of phytosiderophores from roots [54]. It is known that phytosiderophores play an important role in mobilization of Fe and Zn from sparingly soluble Fe and Zn sources in soils as well as in root uptake and shoot transport of Zn and Fe within plants [55][56][57].
In contrast to Zn and Fe, increasing N fertilization did not affect plant Se concentration (Table 4), which shows that the increasing effect of N fertilization on plant Zn and Fe seems to be specific. Even, an increasing N fertilization has been found to have an inhibitory effect on Se concentrations of vegetables [58]. Similarly, the concentration and uptake of Se in maize plants was not significantly affected by varied Zn supply (Table 1), which could be explained by different root uptake mechanisms for Se and Zn.
Plants are very responsive to soil Se fertilization and show substantial increases in shoot and grain concentrations of Se as shown in field-grown maize plants by Chilimba et al. (2012) [59] in Malawi, Mao et al. (2014) [60] in China and Ngigi et al. (2019) [61] in Kenya. The results presented in Table 4 compare well with those results from the field trials.
Considering the average concentrations of observed micronutrients in experimental maize plants and dietary requirement of the dairy cattle, we can suggest that biofortified maize grown on the deficient soils can meet the dairy cattle requirements to certain extent. Average Se maize shoot DM levels ranged between 0.04-0.88 mg kg −1 (Table 4), while the defined selenium requirement is 0.3 mg/kg of dietary dry matter for all categories of dairy cattle [62].
This indicates that maize silage could be very successfully Se biofortified after optimization of the application rates. In case of Zn, required dietary concentration for milking cows is 63 mg kg −1 DM, for heifers 31 mg kg −1 DM and dry cows 22.8 mg kg−1 DM [63], while the average maize shoot DM values ranged between 7-24 mg kg −1 . According to this comparison, Zn biofortified silage could meet the requirements of dry cows and heifers but not those of milking cows. Required dietary concentration of Fe for milking cow is 24 mg kg −1 DM [63]. Thus, average range of maize Fe shoot concentration of 36-80 mg kg −1 DM could easily meet the dietary requirements of all cattle categories.
The present study showed that N nutritional status represents a key factor in biofortification of silage maize with Zn and Fe. It is suggested that an optimum N fertilization of feed and food crops are required to contribute to better human and animal dietary intake of Zn and Fe. The results also highlighted importance of use of water-soluble Zn fertilizers in Zn biofortification of plants grown on high pH soils.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
No potential conflict of interest was reported by the authors.