*2.6. Statistical Analysis*

The effects of the biofortification process were evaluated using one-way analysis of variance (ANOVA) followed by means separation with Fisher's protected least significant difference (LSD) at *p* ≤ 0.05. In the bioaccessibility parameter analysis, the effects of treatments and species were estimated using a two-way analysis of variance (ANOVA) followed by means separation with Fisher's protected least significant difference (LSD) at *p* ≤ 0.05. The software Statistica 10.0 (StatSoft, Tulsa, OK, USA) was used.

#### **3. Results and Discussion**

## *3.1. Mineral Analysis*

Analysis of the accuracy of the analytical measurements of macro and trace elements in the edible parts and in digested samples, from biofortified and non-biofortified baby leaf vegetables, was performed. The mineral elements Al, B, Ca, Fe, K, Mg, Mn, Sr, and Zn were detected and measured. The limits of detection (LOD) and the limit of quantification (LOQ) of the methods were calculated as suggested by D'Imperio et al. [34]. Tomato leaves (NIST-1535a) were used as CRM to evaluate the accuracy of the measurements in the plants and in the digested samples, as reported in Tables 1 and 2. The recovery of mineral elements in the vegetable samples ranged from 90 to 107%. After the in vitro digestion of the CRM, some trace elements, such as Al, Fe, and K, showed the lowest recovery values (%), whereas B, Ca, Mg, Mn, Sr, and Zn showed higher recovery values, as reported in Table 2.


**Table 1.** Mineral content recovered from certified reference materials (NIST tomato leaf 1535a), LOD, and LOQ of methods.

Results are reported as mean ± standard error. Magnesium and strontium: non-certified value. Insufficient information is available to assess the uncertainty associated with the value, and, therefore, no uncertainty is provided (NIST).


**Table 2.** Mineral content recovered from bioaccessibility assays of certified reference materials (NIST tomato leaf 1535a).

Results are reported as mean ± standard error. Magnesium and strontium: non-certified value. Information available is not sufficient to assess the uncertainty associated with the value, and, therefore, no uncertainty is provided (NIST). BF: bioaccessible fraction = concentration of element release from plant material during in vitro digestion process. Residue: residual concentration of the element in digested samples. MB: mass balance = BF + Residue. Certified: the certified value from the National Institute of Standards and Technology (NIST). BF%: bioaccessibility = (BF/Certified) × 100. Recovery = (MB/certified) × 100.

#### *3.2. Mineral Profile of Biofortified and Non-Biofortified Rocket and Purslane*

The biofortification process aims to improve the nutritional value of crops without altering the performance of the crops. In both species, the agronomic protocol applied in this study did not cause any toxic effect in the vegetables nor alteration of the crop performances (data not shown).

Using 5.2 mg/L of Zn in the NS, the tissue content of Zn in the edible parts of rocket and purslane increased, respectively, by 1.76 and 3.97-fold compared with the nonbiofortified counterpart (0.13 mg/L of Zn), as reported in Figure 1. According to our results, the level of Zn used in the biofortification treatment favored its absorption. In fact, zinc is absorbed by plants from the soil as an ionic element or bound to an organic acid and transported through the xylem to the aerial parts (shoots and leaves) [39]. Similar increases in Zn content were found in lettuce [40], cabbage [41], soybean sprouts [42], and in three different types of microgreens that were produced in soilless systems using different levels of Zn in the NS [43].

**Figure 1.** Zinc content in non-biofortified and biofortified rocket (**A**) and purslane (**B**), harvested at the phenological stage of "baby leaf vegetables". Results are reported as mean ± standard error of treatment (*n* = 3). Means separation within columns by LSD (α = 0.05). Significance: \*\* *p* < 0.01. Nonbiofortified (0.13 mg/L of Zn in nutrient solution), Biofortified (5.2 mg/L of Zn in nutrient solution).

The content of Al, B, Ca, Fe, K, Mg, Mn, and Sr measured in rocket and purslane did not reveal significant differences imputable to biofortification (Table 3). The overall mean contents (mg/kg of FW) were 3.32 (Al), 2.69 (B), 3364 (Ca), 6.68 (Fe), 7371 (K), 520 (Mg), 2.24 (Mn), and 6.17 (Sr) in rocket and 0.89 (Al), 2.36 (B), 940 (Ca), 4.07 (Fe), 4279 (K), 856 (Mg), 8.55 (Mn), and 2.77 (Sr) in purslane. In our study, no antagonistic effects were found between Zn and other mineral elements, such as K, Ca, and Fe, although this kind of antagonism has been reported in other studies and is related to the fact that these mineral elements share the same transporters on the plasma membrane [44]. However, our result could be related to the low Zn level used in this study (5.2 mg/L of Zn in NS). Di Gioia et al. [43] reported antagonistic effects between Zn and the other mineral elements using higher levels of Zn in the NS (10 and 20 mg/L) than the level used in this study.

**Table 3.** Mineral content in non-biofortified and biofortified rocket and purslane harvested at the phenological stage of "baby leaf vegetables".


Results are reported as mean ± standard error of treatment (*n* = 3). Significance: ns = not significant. Means separation within columns by LSD (α = 0.05). Non-biofortified (0.13 mg/L of Zn in nutrient solution), Biofortified (5.2 mg/L of Zn in nutrient solution).

#### *3.3. Mineral Bioaccessibility in Rocket and Purslane after the Biofortification Process*

The BF is the concentration of a nutrient or a bioactive compound (mineral or organic) that is extracted from the plant matrix during the digestion process and which, potentially, becomes bioavailable in the intestinal tract. The number of mineral elements released by plant materials is related to various factors such as species, food processing (raw or cooked food), texture, nutrient concentration, and interaction with other nutrients or antinutrients [17,32,42,45,46]. In our study, after in vitro gastrointestinal digestion, Zn BF% was 98% in biofortified plants and 73% in non-biofortified plants compared to the non-digested control plants. Similar results were reported for Si-biofortified green bean pods [17]. Conversely, no differences in BF% values (72%) were found in rocket (biofortified and non-biofortified), although an increase in Zn was found in the edible parts (Figure 1). Therefore, the in vitro digestion protocol allows similar BF% values to be obtained in both biofortified and non-biofortified plants. This result was also reported in our previous study [32,47], showing that increasing the concentration of mineral elements in the edible parts of biofortified plants does not always give an increase in BF%, as reported for calcium and silicon [32,47]. However, in both rocket and purslane, after the in vitro gastrointestinal digestion (bioaccessible fraction), we measured a significant release of Zn (mg/kg) in biofortified plants compared to non-biofortified ones (76% and 298%, respectively, for rocket and purslane), as shown in Figure 2. Biofortified purslane was found to be the species with the highest amount of bioaccessible Zn released during the digestion process (16.9 mg/kg). The quantity of Zn released by biofortified rocket was 7.43 mg/kg. The quantity of bioaccessible Zn released by non-biofortified purslane and rocket was 3.75 mg/kg (on average).

**Figure 2.** Bioaccessible fraction (mg/kg) of Zn in non-biofortified and biofortified rocket and purslane after in vitro digestion process. Results are reported as mean ± standard error of treatment (*n* = 3). Different letters indicate that mean values are significantly different (means separation by LSD; α = 0.05). Non-biofortified (0.13 mg/L of Zn in nutrient solution), Biofortified (5.2 mg/L of Zn in nutrient solution).

As previously reported also in soybean sprouts [42], the BF of Zn, measured after in vitro gastrointestinal digestion, is affected by the initial content of Zn in the edible parts of the plants. The increase in the amount of Zn released during the digestion process and found in this study is a significant result, considering that this is the amount of Zn that could be potentially absorbed in the intestinal tract [48].

The BF of the mineral elements is correlated to the different compositions of the tested species and to the interaction of the plants with the intestinal juices (pancreatic enzymes and bile salts). As reported in Table 4, all mineral elements analyzed showed significant differences (*p* < 0.001) in relation to the plant species, but they were not affected by the Zn biofortification protocol used. The influence of the plant species on BF values has also been found in other studies analyzing various mineral elements, such as Si [47], Ca [32,49], K [45,49], Fe [6], Mg [49], and other trace elements [49]. In our study, the average quantities of mineral elements released in the digestion process were 0.53 mg/kg for Al, 2.36 mg/kg for B, and 7522 mg/kg for K, and these quantities were higher in rocket than in purslane. Conversely, the measured mean amounts of Fe (2.12 mg/kg) and Mg (880 mg/kg) were higher in purslane than in rocket (Table 4).

Several compounds, such as some antinutritional factors (carbonate, phytic and oxalic acids) and some healthy food components (proteins, fibers, and polyphenols), can modify the release of nutrients from the food matrix [50]. The interaction of mineral elements with these compounds generates insoluble salts and determines the reduction of BF and a reduced absorption of minerals [30,31]. Egea-Gilabert et al. [22] reported that purslane is a vegetable with a high oxalate content (2000 mg/kg of fresh weight). On the contrary, rocket is generally considered to be free of oxalate [23]. This difference in oxalate content could influence the BF of all mineral elements evaluated: in particular, Ca and Sr. Oxalate forms an insoluble salt with Ca [51] and probably also with Sr, considering the similar chemical and biological properties of these mineral elements [52]. The effects of plant species on Ca bioaccessibility and the high amount of Ca released during the digestion process were reported in our previous study [32].


**Table 4.** Bioaccessible fractions of Al, B, Ca, Fe, K, Mg, Mn, and Sr in non-biofortified and biofortified rocket and purslane after in vitro digestion process.

Results are reported as mean ± standard error of treatment (*n* = 3). FW: fresh weight. Significance: ns = not significant; \* *p* ≤ 0.05; \*\* *p* < 0.01; \*\*\* *p* ≤ 0.001. Different letters within column indicate that mean values are significantly different (means separation by LSD; α = 0.05). Non-biofortified (0.13 mg/L of Zn in nutrient solution), Biofortified (5.2 mg/L of Zn in nutrient solution).

The highest amounts of Ca and Sr in the digested liquid were found in the nonbiofortified rocket, followed by the biofortified rocket, whereas the purslane released lower amounts of Ca in the gastrointestinal digestion, and this result was not affected by the biofortification treatment with Zn. The high amounts of Ca observed in rocket could lead to the formation of low-solubility complexes that reduce the BF of Mn. Furthermore, mineral elements with similar electronic configurations (Zn2+, Ca2+, Fe2+, Mg2+, Mn2+, and Sr2+) are involved in mechanisms of mutual competition to bind antinutrient compounds [27,28,46]. Therefore, different values of BF and BF% can be attributable to different factors, including the mechanisms of competition at different levels in a plant-based food system.

#### *3.4. Daily Intake, Coverage of RDA-Zn (Male and Female), and Hazard Quotient*

The DI, the RDA-Zn coverage (for men and women), and the HQ for Zn intake through digesting 100 g of baby leaf vegetables (average servings for this type of products) are shown in Table 5. The Zn biofortification significantly increased those parameters (*p* < 0.001), and differences between the two vegetables were found (Table 5). The highest values of DI, RDA-Zn coverage, and HQ were obtained for biofortified purslane, whereas the lowest values were found for non-biofortified rocket and purslane (Table 5). After digestion of 100 g of biofortified purslane, an increase in DI (3.9-fold) and RDA-Zn coverage was found in males and females, compared to non-biofortified vegetables (Table 5).

**Table 5.** Daily intake, coverage of RDA for Zn, and HQ for Zn intake through consumption of 100 g portions of baby leaf vegetables, biofortified and non-biofortified, by adult male and female humans (70 kg body weight).


Results are reported as mean ± standard error of treatment (*n* = 3). Significance: \*\*\* *p* ≤ 0.001. Different letters within columns indicate that mean values are significantly different (means separation by LSD; α = 0.05). Daily intake, coverage of RDA for Zn, and HQ were calculated in relation to the quantity of Zn released from vegetables during the gastrointestinal digestion process. Major details are reported in Section 2.5 of Materials and Methods.

The increase of DI and RDA-Zn coverage accentuates the efficiency of the applied biofortification protocol, suggesting its use to produce Zn-biofortified baby leaf vegetables for different target consumers groups for which the increase of the DI is advisable, such as pregnan<sup>t</sup> and breastfeeding women, vegetarians/vegans, people with various diseases, and the elderly [7].

The HQ values found in rocket (biofortified and not) and in non-biofortified purslane were less than 1. However, an excessive increase of Zn in the edible portions of purslane can result in an increase of this parameter. When the HQ is higher than 1, adverse health effects are likely to occur. According to our findings, the consumption of 100 g of our biofortified products does not pose any health risk to consumers. This aspect must be taken into due consideration when approaching a biofortification process; an excessive content of Zn in the edible parts of vegetables would represent a risk for consumers (the maximum tolerable intake level is 40 mg Zn/day) since vegetables are only a relative portion of the diet and other foods and water intake can significantly contribute to the daily intake of Zn [39].
