**4. Discussion**

In this study, we produced microgreens of some Brassicaceae genotypes by using a hydroponic system to evaluate the effects of element concentration and chemical form of nitrogen in the nutrient solution on yield and some quality traits. We conducted an exploratory experiment by using a NS type-like Hoagland and Arnon [29] but at three different strengths (1/2 strength, 1/4 strength and 1/8 strength). This, we started from the fact that some authors reported the use of a quarter-strength Hoagland nutrient solution [19,20], while other authors reported the use of a half-strength Hoagland nutrient solution [15,16] as well as three different strengths of nutrient solution [22]. Therefore, considering the short growth cycle of microgreens, we decided to also evaluate if nutrient concentration lower than half strength may satisfy seedling needs, without negatively affecting yield and other important parameters. In this context, it is important to highlight that the optimal choice of element concentration in the NS may allow one to reduce production costs and environmental impact. In the first experiment, we observed that growing parameters were not affected by NS strength (Table 4). In addition, yield was not affected by the NS strength except for broccoli raab, which showed a lower yield when 1/8 NS was used (Figure 2) and for this cultivar, the growth rate was faster than for broccoli and cauliflower (Table 4). On average, we found that seedling height significantly decreased when passing from NS at 1/2 strength to NS at 1/8 strength (Table 3). Considering that the harvesting of microgreens is usually done manually, the higher the seedling height, the easier the harvesting can be made. Therefore, for the second experiment, we decided to use a NS at 1/2 strength but with three different NH4:NO3 molar ratios to evaluate the effect of another aspect of fertigation on physiological behaviour and some quality traits of different Brassicaceae microgreens. The choice of NS at 1/2 strength instead of other ones was also made by considering the higher temperature and photosynthetic photon flux (PPF) forecasted for the second experiment than the first one. Effectively, the rate of nutrient uptake was related to current seedling nutrient demand, positively correlated with PPF and air temperature [31].

By changing the NH4:NO3 molar ratio, no differences were found on yield and growing parameters (Table 5), while significant differences were found in regards to dry matter and content of inorganic cations, proteins and β-carotene (Tables 6 and 7). For dry matter, nitrates, sodium and proteins, we observed important interactions between genotypes and the molar ratio between the chemical forms

of nitrogen. The most abundant cation in all the microgreens samples was K<sup>+</sup>, followed by Ca2+, Mg<sup>2</sup>+ and Na<sup>+</sup>, while, in regards to anion content, NO3− was followed by SO4<sup>2</sup>− and Cl− (Table 6). A similar mineral composition was observed in previous studies [17,32]. In regards to the differences in nitrates content (Figure 3), Santamaria [23] reported that the large variation in nitrate accumulation among plant species could be associated with genetic factors. At the same time, different genotypes may show different nitrate uptake, translocation and accumulation in the vacuoles of mesophyll cells [33]. In agreement, we observed that by using a NS with the NH4:NO3 molar ratio of 5:95, broccoli raab showed a nitrate content lower than other NH4:NO3 molar ratios, while broccoli showed the lowest nitrates content when the NS with the NH4:NO3 molar ratio of 25:75 was used (Figure 3). At same time, no differences in nitrates content were found by changing the NH4:NO3 molar ratio in cauliflower (Figure 3). These results sugges<sup>t</sup> that the nitrate content in different *Brassica* microgreens can be affected by the interaction between genotypes and the NH4:NO3 molar ratio in the NS. This is in agreemen<sup>t</sup> with Dikson and Fisher [34], who observed that genotypes had a central role in anion and cation uptake by varying root zone pH. In the same way, during this study, changing the NH4:NO3 molar ratio and substrate/root zone pH changes influenced cation and anion (nitrates) uptake differently for each genotype.

From a commercial point of view, it could be interesting to evaluate the nitrate content in microgreens observed in our study in relation to the tolerable levels of nitrates in foodstuffs. On average, we found a content of 5051, 4816 and 6249 mg NO3− kg−<sup>1</sup> FW, respectively for broccoli raab, broccoli and cauliflower (processed data from Table 6). It is important to note that for Brassicaceae species the European Regulation (EU) No 1258/2011 [35] reports maximum levels of nitrate only for the "rucola" group (*Eruca sativa*, *Diplotaxis* spp, *Brassica tenuifolia*, *Sisymbrium tenuifolium*). European Regulation fixed a maximum level of 7000 mg NO3 kg−<sup>1</sup> FW for "rucola" vegetables harvested from 1st of October to 31st of March (the period of our study), and a maximum level of 6000 mg NO3 kg−<sup>1</sup> FW in the other year period. Considering these maximum levels, our results sugges<sup>t</sup> that by changing the NH4:NO3 molar ratio in the NS, it is possible to produce microgreens of broccoli raab, broccoli and cauliflower without negatively affecting an important commercial characteristic such as the nitrate content.

In regards to the nutritional quality, we found that all three genotypes of *Brassica* microgreens showed a high content of mineral elements (Table 6). This is agreemen<sup>t</sup> with several authors [17,32,36,37] confirming that microgreens can be considered as a good source of minerals in the human diet. Apart from the content of mineral elements, microgreens can provide higher amounts of other nutrients compared to their mature leaf counterparts [1]. To this end, we found that 100 g of mature cauliflower supplies about 2 g of fibers, 1.92 g of proteins and 0.08 mg of α-tocopherol [38]. The same serving size of mature broccoli supplies 2.6 g of fibers, 2.82 g of proteins and 0.78 mg of α-tocopherol [39], while 100 g mature broccoli raab supply 2.7 g of fibers, 3.17 g of proteins, and 1.62 mg of α-tocopherol [40]. Results of the present study show a fiber content (Table 7) much lower than mature plants independently of genotypes and the NH4:NO3 molar ratio. Therefore, according to Renna et al. [9], microgreens of this study can be considered as a low content fiber food for subjects with gastrointestinal disorders, such as bowel colon syndrome. Regarding protein content, microgreens showed values similar to mature *Brassica* vegetables with the exception of micro-cauliflower fertigated by using a NS with a NH4:NO3 molar ratio of 25:75, which showed a higher protein content than mature cauliflower. This, could be due to the fact that the NH4:NO3 molar ratio of 25:75 caused an increase in dry matter content compared with other treatments and proteins are one of the major constituents of the dry matter [41].

α-Tocopherol is the most common and biologically active form of vitamin E. Effectively, although the term vitamin E can refer to different types of tocopherols and tocotrienols, it should be considered the selective degradation and excretion of other vitamin E forms and the selective retention of α-tocopherol, mediated by the hepatic α-tocopherol transfer protein (α-TTP) [42]. In our study, we observed a higher α-tocopherol content, independently of the NH4:NO3 molar ratio, in microgreens than in the mature counterparts, especially in micro cauliflower (Table 7). α-Tocopherol represents part of the fat-soluble

antioxidant system of the cell, since it terminates the chain reaction of lipid peroxidation. Vitamin E deficiency is associated with a progressive necrosis of the nervous system and muscle. In this context, it is important to note that the recommended dietary allowance (RDA) of vitamin E (α-tocopherol) for people aged 14 years and over, including pregnan<sup>t</sup> women, is 15 mg per day [42]. Therefore, 100 g of microgreens produced in this study can satisfy about 70, 34 and 13% of the RDA, respectively, for micro cauliflower, micro broccoli and micro broccoli raab.

β-Carotene is the principal pro-vitamin A carotenoid considering that its symmetrical chemical structure always provides vitamin A regardless of the metabolic process. Other forms of provitamin A are α-carotene, γ-carotene and β-cryptoxanthin. β-Carotene is the most abundant dietary carotenoid present in yellow-orange fruits and vegetables, and green leafy vegetables. In humans, it plays a potent antioxidant role known to prevent oxidative damage to biological membranes by quenching free radicals [42]. Mature cauliflower lacks β-carotene [38], while 100 g of mature broccoli and broccoli raab contain 0.36 and 1.57 mg of β-carotene, respectively [39,40]. Therefore, results of the present study show a higher β-carotene content in microgreens than the mature counterparts, especially by using a NH4:NO3 molar ratio of 25:75 (Table 7). In a study aimed to evaluate the nutrient composition of ten culinary microgreens, Ghoora et al. [43] found a β-carotene content ranging from 3.1 to 9.1 mg 100 g<sup>−</sup><sup>1</sup> FW. Our results are in agreemen<sup>t</sup> with these authors, confirming that microgreens can be considered a good source of β-carotene, although the amount can vary depending on genotype.
