Effect of Nutrient Solution pH on the Growth, Yield and Quality of Taraxacum ofﬁcinale and Reichardia picroides in a Floating Hydroponic System

: Given the important medicinal and nutritional value of wild edible greens, the last few years there is an increasing interest for their domestication and commercial exploitation. However, information concerning their adaptation to environmental conditions and their response to modern agricultural systems are scarce. In the present study, the effect of nutrient solution pH (4.0, 5.5 and 7.0) on the growth, chemical composition and inorganic nutrition of Taraxacum ofﬁcinale and Reichardia picroides plants grown indoors in a ﬂoating hydroponic system was evaluated. Both species performed better at pH 5.5 and were slightly affected by pH 7.0, whereas pH 4.0 was not prohibitive for growth for both species, although R. picroides was less tolerant than T. ofﬁcinale at low pH. Moreover, pH 4.0 did not severely affect nutrients uptake and transport within the plant tissues, suggesting that R. picroides susceptibility to low pH should not be attributed to nutrients imbalance. Nevertheless, low pH positively enhanced the content in total soluble solids, total phenolics, chlorophylls (a, b and total) and carotenoids, and decreased nitrates in both species. In conclusion, the studied species could be successfully grown in soilless systems with nutrient solutions of varied pH. Moreover, low pH levels (pH = 4.0) seemed to be beneﬁcial to nutritional and dietary value in both species highlighting the potential of commercial cultivation under adverse conditions, especially in sustainable farming systems.


Introduction
The Mediterranean basin is thriving in wild species which have been used in human diet for centuries and have great importance for the agrobiodiversity of this region [1,2]. Among the various wild species, dandelion (Taraxacum officinale (L.) Weber ex F.H. Wigg.) and common brighteyes (Reichardia picroides (L.) Roth.) are two herbs of the Asteraceae family that are traditionally consumed raw or boiled in green salads or used as medicinal plants [3,4]. They are rich in nutrients, antioxidants and numerous bioactive compounds such as organic acids, sugars, phenolics, tocopherols, fatty acids etc. [5][6][7][8][9][10][11], and considered basic ingredients of the Mediterranean diet [12,13]. Currently, consumers are seeking food products with health beneficial effects and of known origin, a trend which has rekindled the interest in wild edible species and is starting to increase the adherence to the

Plant Material and Experimental Treatments
The experiment was conducted in an unheated glasshouse at the University of the Peloponnese (Kalamata, Messinia, Southern Greece, 37 • 3 22" N, 22 • 1 43" E). Two wild edible species were studied, namely Taraxacum officinale (L.) Weber ex F.H. Wigg. and Reichardia picroides (L.) Roth. Seeds of both species are part of the seed collection of the Laboratory of Agronomy, University of the Peloponnese, produced at the experimental field of the University in Kalamata in spring of 2017. Harvested seeds were stored at 4-7 • C until the start of experiment. The starting material (seeds) was collected from wild plants in the Vicinity of Kalamata, Greece. Prior to seed collections, plants were located and identified based on flower and plant morphology and with the guidance of experienced local people, since wild species usually have various common names and different species may have similar names. Sowing of seeds was performed on 10/1/2019 in polystyrene seed trays (cell dimension 5 × 5 × 5 cm) filled with white peat (pH 5.5-6.5, without fertilizers-Base Substrate, Klasmann-Deilmann GmbH, Geeste, Germany). Trays were placed in a walk-in growth chamber at 20 • C with 16 h photoperiod (fluorescent lamps, light intensity of 55 µmol m −2 s −1 ), and were regularly fertigated using a watersoluble 20-20-20 plus microelements fertilizer (Nutrileaf, Miller Chemical & Fertliliser LLC, Hanover, PE, USA). After one month (30 days after sowing, DAS) and when seedlings reached the 2-3 true-leaf stage, seed trays were transferred to the experimental glasshouse of the University of the Peloponnese. On 6/3/2019 (55 DAS) seed trays with plants at the 6-8 leaves stage were placed into containers (length × width × height: 1 × 1 × 0. 25  Nutrient solution was prepared from two stock solution (stock solution A and stock solution B) following the procedure applied in hydroponic systems to avoid precipitation of nutrients. The composition of stock solutions is described in supplementary material (Table S1). Water composition is provided in supplementary material (Table S2). The growing conditions (temperature, relative humidity and light conditions are described in Figures 1 and 2. Agronomy 2021, 11, x FOR PEER REVIEW 3 of 16 dimension 5 × 5 × 5 cm) filled with white peat (pH 5.5-6.5, without fertilizers-Base Substrate, Klasmann-Deilmann GmbH, Geeste, Germany). Trays were placed in a walk-in growth chamber at 20 °C with 16h photoperiod (fluorescent lamps, light intensity of 55 μmol m −2 s −1 ), and were regularly fertigated using a water-soluble 20-20-20 plus microelements fertilizer (  Nutrient solution was prepared from two stock solution (stock solution A and stock solution B) following the procedure applied in hydroponic systems to avoid precipitation of nutrients. The composition of stock solutions is described in supplementary material (Table S1). Water composition is provided in supplementary material (Table S2). The growing conditions (temperature, relative humidity and light conditions are described in Figures 1 and 2.     Nutrient solution was prepared from two stock solution (stock solution A and stock solution B) following the procedure applied in hydroponic systems to avoid precipitation of nutrients. The composition of stock solutions is described in supplementary material (Table S1). Water composition is provided in supplementary material (Table S2). The growing conditions (temperature, relative humidity and light conditions are described in Figures 1 and 2.   Two harvests were performed in each species, the first one when plants were reached the harvesting stage when collected in the wild, and the second just before the appearance of the flower stem. In dandelion, first harvest (H1) was carried out at 22/3/2019 (16 days after the start of pH treatment-DAST) and second harvest (H2) at 4/4/2019 (29-DAST), whereas for R. picroides the respective dates were 28/3/2019 (22-DAST) for H1 and 16/4/2019 (41-DAST) for H2. Four containers (n = 4) for each pH treatment, each harvest and each species were used totaling to 48 containers. In each container 44 plants were grown, with 10 plants (from the center of the container) being used for each harvest.
For each plant species, a bifactorial experiment was carried out following a completely randomized experimental design (factor 1: nutrient solution pH, factor 2: growth stage at harvest-harvest date). In each treatment combination, four replications (n = 4) were used comprising 10 plants each.
Throughout the experimental period, nutrient solution EC remained constant at 2.2 ± 0.1 in all the treatments without any need for readjustment. Similarly, nutrient solution pH remained stable in all three treatments. Mean temperature of the nutrient solution during the cultivation period of T. officinale was 16.2 • C (ranging from 14.0 to 17.2 • C) and that of R. picroides was 16.0 • C (ranging from 13.4 to 18.2 • C).

Plant Growth Parameters
At each harvesting date and pH treatment, ten plants per replication were sampled for each species. Subsequently, plant leaf number, rosette diameter, percentage of nonmarketable leaves (not green, dried or injured), leaf SPAD index values, total plant fresh weight (FW), root FW and marketable leaf FW were assessed. Leaves from all ten plants per replication were grouped in one batch sample, and subsequently divided in two identical samples for further analyses. One sample was used for the determination of dry matter content (%DMC, by drying samples at 72 • C to constant weight), and macro and micro-elements concentration, whereas the other sample was kept at −80 • C for chemical composition analyses. Similarly, roots from all ten plants per each replication were grouped in one sample, which after drying was used for the determination of dry matter content and macro and micro-elements concentration.

Leaves and Roots Macro-and Micro-Elements Concentrations
For the determination of total N, P, K, Ca, Mg, Fe, Mn, Zn, Cu and B concentrations in leaves and roots, dried leaves and roots samples were ground to fine powder and dryashed in a furnace at 500 • C for 6 h. 500 g of the dry-ashed material was dissolved with 10.0 mL 1.0 N HCl solution and quantitatively transferred into a 50-mL volumetric flask, diluted to volume with deionized water for further minerals analysis. P concentration was determined by the vanado-molybdo-phosphate yellow color method [37], B by azomethin-H [38], K, Ca, Mg, Fe, Mn, Zn and Cu concentrations were determined by atomic absorption spectrometry (SpectrAA, 240 Atomic Absorption FS; Varian, Palo Alto, CA, USA) in the dry digest [38], while N was determined by the indophenol-blue method in the wet digest [39]. N-NO 3 − concentration in leaves was determined in 200 mg of dried material suspended in 20 mL of deionized water using the colorimetric method of Cataldo et al. [40] based on the nitration of salicylic acid. All measurements were carried out in triplicate (n = 3) and expressed on a dry weight (DW) basis, except the leaf nitrate concentration, which was expressed in leaf fresh weight (FW).

Content in Chlorophyll and Carotenoids + Xanthophylls
Chlorophyll content of leaves was assessed using two methods:

1.
By taking SPAD index readings with a SPAD-502 Chlorophyll Meter (Konica-Minolta Co. Ltd., Tokyo, Japan) at two spots on the upper surface, 0.5-1 cm right and left from the central vein of each leaf. Measurements were taken from the two youngest but fully expanded leaves, individually in all 40 plants (ten plants per replication) that were sampled in each harvesting date [42].

2.
By quantifying chlorophyll a, b and total chlorophyll content of leaves in acetone extracts of homogenized samples, as described in Karapanos et al. [43]. Absorbance of the extracts at 663 and 647 nm was measured in a spectrophotometer (model Lambda 1A, PerkinElmer, Waltham, MA, USA) and the chlorophyll content was calculated according to the equations referred by Lichtenthaler and Buschmann [44].
Carotenoids + xanthophylls were quantified in the extracts previously described by measuring the absorbance at 470 nm, according to the method of Lichtenthaler and Buschmann [44]. The chlorophyll and carotenoids + xanthophylls content was expressed as mg kg −1 FW.

Total Phenolics Content
Total phenolics were quantified using the Folin-Ciocalteu method [45], in methanolic extracts, as described in Karapanos et al. [43]. In brief, the homogenized samples were mixed with 80% (v/v) methanol in water, the mixtures were stirred for 2 h at room temperature and, after centrifugation, the supernatants were decanted and the pellets were resuspended in 80% methanol, following the same procedure. The combined supernatants were used for measurements according to the Folin-Ciocalteu method, and absorbance was measured in a spectrophotometer at 765 nm. Gallic acid was used as standard and the results were expressed as mg gallic acid equivalents (GAE) kg −1 FW.

Statistical Analysis
All data were analyzed by a two-way ANOVA in order to assess the significance of interaction between the experimental factors, namely pH value and harvesting date. Due to the statistically significant interaction between the two factors, the effect of each factor was analyzed separately. One-way ANOVA was carried out to determine the effect of nutrient solution pH on the studied parameters at each harvesting date and means were separated by the Least Significant Difference test at p ≤ 0.05. Accordingly, Student's t-test at p ≤ 0.05 was used to compare the effect of harvesting date on the studied parameters at each nutrient solution pH level. Statistical tests were performed using the StatGraphics Centurion-XVI statistical package (StatPoint Technologies Inc., Warrenton, VA, USA).

The Effect of Nutrient Solution pH on Taraxacum officinale
In the case of T. officinale, pH treatments did not affect the leaf number, the percentage of non-marketable leaves per plant, the SPAD index values and the leaf % DMC (Table 1). Plants grown in solution with pH 4.0 had smaller rosette diameter than those grown in solution with pH 5.5 or pH 7.0, lower total FW than those grown in pH 7.0 (H1) and in pH 5.5 (H2), less marketable leaves FW than those grown in pH 5.5 and pH 7.0 (H1) and less root FW per plant than those grown in pH 5.5 (H2). Moreover, roots % of DMC (ranged between 8.3%-9.8% at H1 and 8.8%-10.1% at H2) was not affected by the pH (data not shown).  The studied pH treatments did not affect leaf concentration in N, Ca, Mg, Fe and B (H1, H2), P and Mn (H1) and Cu (H2). On the other hand, K (67.3 g kg −1 DW at H2) and Cu (5.0 g kg −1 DW at H1) were the lowest in plants grown in solution with pH 4.0. P and Zn were the lowest in pH 7.0 (16.8 g kg −1 DW at H2), whereas Mn was the highest in pH 4.0 (25.8 g kg −1 DW at H2). The time of harvest also affected leaf N concentration which was higher at H1 compared to H2, regardless of pH treatments, while P, Mn and Zn concentrations were higher at H1 compared to H2, only in the case of pH 7.0. K was higher by 9.8% at H1 compared to H2 in plants grown in pH 4.0 but lower by 8.0% at H1 compared to H2 in the case of pH 5.5 (Table 2).  Root N, Ca, Mg and B concentrations were not affected by pH treatments. P was the lowest in the case of pH 7.0 (9.2 g kg −1 DW; H2), K was the lowest in pH 4.0 (38.5 g kg −1 DW at H1; 33.5 g kg −1 DW at H2) and Fe was the highest in pH 4.0 (146.6 g kg −1 DW at H1). Moreover, Mn (23.9 g kg −1 DW at H1 and 29.4 g kg −1 DW at H2) and Cu (16.7 g kg −1 DW at H1) were the highest in pH 7.0, while root P, Fe and Zn concentrations were higher at H1 compared to H2 in pH 7.0 (increased by 25.2%, 9.6% and 33.7%, respectively). On the other hand, Mn in pH 5.5 and pH 7.0, Zn in pH 5.5 and Cu in pH 4.0 and pH 5.5 showed higher values at H2 compared to H1 (Table 3). Leaf nitrate content was lower in pH 4.0 than in pH 7.0 (reduced by 37.1% and 33.7% at H1, H2, respectively) and pH 5.5 (reduced by 38.8% at H2). Leaf content in TSS, TA, total phenolics, chlorophyll a, chlorophyll b, and total chlorophylls was the highest in pH 4.0 compared to pH 7.0 (increased by 22.9%, 29.2%, 26.9%, 13.45%, 12.8% and 13.2%, respectively) ( Table 4). At H2, plants grown in pH 4.0 had higher content of total chlorophylls and chlorophyll a than in pH 7.0 (increased by 14.1% and 12.0%, respectively) and higher content of total phenolics than in pH 5.5 (increased by 16.1%). However, pH treatments did not affect the chlorophyll a:b ratio (ranged between 2.07-2.10 at H1 and 1.90-2.04 at H2, data not shown). In addition, TA was higher at H1 compared to H2 irrespective of the pH treatment, TSSC was lower at H2 compared to H1 only in the case of pH 4.0 (reduced by 17.9%) and chlorophylls a, b and total chlorophylls, as well as carotenoids + xanthophylls, were lower at H2 compared to H1 only in pH 5.5 (reduced by 11.3%, 7.1%, 10.4% and 15.5%, respectively) ( Table 4).

The Effect of Nutrient Solution pH on Reichardia picroides
In R. picroides, the percentage of non-marketable leaves per plant, the SPAD index values and the leaf % of DMC were not affected by the pH treatments (Table 5). By contrast, the leaf number, the rosette diameter, the total plant FW and the marketable leaves FW were the lowest in plants grown at pH 4.0. At H2, plants grown in pH 4.0 had lower leaf number, total FW and marketable leaves FW than those grown in other pH values. Roots FW was the lowest in pH 4.0 (H1) but roots % of DMC (ranged between 6.1%-7.2% at H1 and 6.7%-7.5% at H2, data not shown) was not affected by the pH treatment.  The studied pH treatments had no effect on leaf N, Mg and B concentrations. However, the concentrations of the rest of the macro-and micro-nutrients showed a varied response depending on the pH treatment and the harvesting time. K (51.9 g kg −1 DW at H1) and Cu (3.3 g kg −1 DW at H2) were the lowest in pH 4.0 and pH 5.5, respectively. P was the highest in pH 5.5 (9.8 g kg −1 DW at H1), and in pH 4.0 (7.8 g kg −1 DW at H2) than in pH 7.0. Ca was the highest in pH 5.5 (9.6 g kg −1 DW at H1) and in pH 4.0 and pH 5.5 at H2 (7.8 g kg −1 DW and 8.7 g kg −1 DW, respectively). Fe was higher in pH 4.0 than in pH 5.5 (increased by 30.7% at H1 and by 26.8% at H2) and pH 7.0 (increased by 20.3% at H2), Mn was higher in pH 4.0 than in pH 7.0 (increased by 28.0% at H1 and by 70.0% at H2) and pH 5.5 (increased by 33.9% at H2), and Zn content was the highest in pH 4.0 at both harvesting stages (H1, H2). The content of N, P and Mn in pH 5.5 and pH 7.0 and Ca, Mg, Fe and Zn in pH 7.0 were higher at H1 compared to H2. On the other hand, Cu and Zn in pH 4.0 and B in pH 4.0 and pH 7.0 were higher at H2 compared to H1 (Table 6).  (Table 7).  Leaf nitrate content was not affected by the pH treatment at H2, but it was the lowest in pH 4.0 at H1 (Table 8). pH treatments had no effect on the content of leaves in TSSC (H1), TA (H1), chlorophylls (a, b and total-H2), carotenoids + xanthophylls (H2) and on the chlorophyll a /b ratio (ranged between 1.91-2.01 at H1 and 2.15-2.25 at H2, data not shown). On the contrary, all the other parameters were similarly affected by the pH values, presenting higher values in pH 4.0. In addition, TA at H2 was lower by 24.2% in pH 7.0 compared to pH 4.0. Plants grown in pH 4.0 higher content in a, b and total chlorophylls at H1 compared to H2 (increased by 12.2%, 19.5% and 14.6%, respectively) ( Table 8).

Discussion
In Taraxacum officinale, the leaf number per plant was not affected by the studied pH treatments, while in Reichardia picroides it was decreased at the lower studied values (pH 4.0). However, pH did not affect the percentage of non-marketable (stressed or injured) leaves in both species, indicating that both plants were able to tolerate pH values within the tested range without severe effects. According to the literature, pH 4.0 did not also affect the leaf number per plant in Metaxylon sagu [46] and Paeonia lactifolia plants [47]; however, it reduced plant height and total leaf area of Paeonia lactifolia [47] suggesting a possible effect of low pH on plant morphology. This is in accordance to our results, as the rosette diameter of both species was decreased at pH 4.0, indicating a reduction of leaf length, similarly to what is reported by Roosta and Rezaei in rose plants [48]. In addition, the reduction of marketable leaves and total plant FW caused by low pH value (pH 4.0) in R. picroides at both harvests (particularly at H2) also indicates low pH induced changes on plant growth and morphology. Nevertheless, our results indicate a species-dependent efficiency to overcome the low pH-induced stress. Even though in T. officinale total plant FW at pH 4.0 was lower at both harvest dates, marketable leaves FW was lower only at H1, suggesting that dandelion might have developed an adaptation mechanism to grow satisfactorily after the initial stress, whereas as stress gradually builds up the defensive mechanism cannot cope with increasing stress, thus a reduction in leaves FW is observed. In addition, DMC (%) of leaves and roots of both species was not affected by pH, indicating that the absorption and accumulation of water from plants was unaffected by the studied pH levels, as well as that the root functioning was not unbalanced due to injuries from non-optimal pH values of nutrient solution.
Several species can grow well at pH values that exceed the recommended range of 5.5 to 6.5 [20], although for most (conventional) vegetable species plant growth is impaired at pH levels above 7.0 or under 5.0 [49]. Our results showed that T. officinale growth was not restricted even at the lowest studied pH values (pH 4.0), whereas R. picroides seems to be more sensitive to changes (higher or lower than optimal) in the pH level. In comparison to the treatment of pH 5.5, R. picroides growth was negatively affected by pH 4.0 and to a lesser extent by pH 7.0, despite it is reported in the literature that this species is usually found in calcareous soils [50]. This contradiction in the results could be attributed to the fact that in our study the plants were grown having their roots directly immersed in nutrient solution which does not have the buffering capacity of soil [51]. Moreover, considering the wild nature of the species there are probably many other environmental factors that may define the optimal growing conditions apart from pH value of growing medium [52].
In both species, the lowest pH level (pH 4.0) reduced root FW, indicating that root growth is directly impaired by low pH levels, probably due to the restriction of root elongation caused by higher H + concentrations at low pH in nutrient solution [53]. Limited root growth due to the restrictions of net H + release by H + -ATPase and the inability to properly regulate the cytoplasmic pH of root cells is related with the impaired plant growth and reduced biomass production at low soil/nutrient solution pH [54,55]. Moreover, the low soil/nutrient solution pH values (e.g., the level of pH = 4.0 tested in our study) is also reported to impair plant growth and biomass accumulation by reducing photosynthetic rate due to lower transpiration rate and closure of leaf stomata in tomato seedlings [56]. However, there is no evidence that low pH negatively affects chlorophyll content or the photosynthetic apparatus of plants. On the contrary, Paenia lactifolia leaves showed higher chlorophyll content at pH 4.0, which indicates a possible defense mechanism through which plants increase chlorophyll content also evidenced by the increased activity of three protective enzymes [47]. Similarly to our study, although leaf chlorophyll content expressed as SPAD index values was not affected by nutrient solution pH in both species, extractable chlorophyll a, b and total was favored by pH 4.0 in both species at H1. This finding, in relation to the unaffected chlorophyll a:b ratio by low pH value (pH = 4.0), could indicate an enhancement of the photosynthetic apparatus which allows plants to counteract the possibly impaired photosynthetic activity at low pH due to leaf stomata closure, similarly to the findings of Zhao et al. [47] who also reported an increase in leaf total soluble solids content in Paenia lactifolia plants. Indeed, in our study, leaf total soluble solids content (TSSC) in plants grown at pH 4.0 was higher in T. officinale at H1 and in R. picroides at H2 and tended to be higher at H2 and H1, respectively in each species, indicating different adaptation mechanisms depending on the species.
In both species, leaf and root N concentrations were not affected by pH treatments, even though plants were treated with slightly more N, in the form of nitric acid, in pH 4.0 and pH 5.5 compared to pH 7.0 level, aiming to establish the desirable pH level of nutrient solutions. Similarly, Anugoolprasert et al. [46] did not find any significant effect of pH level on N concentration in Metaxylon sagu, whereas Findenegg [57] reported that pH 4.0 reduced total leaf N concentration in Helianthus annuus plants.
Leaf P concentration declined when nutrient solution pH increased to 7.0, as previously reported by Assimakopoulou et al. [58] in spinach plants grown under high pH levels in the growing medium. Moreover, the findings of our study are in accordance to the acquired knowledge that P-PO 4 availability in tomato plants decreases with increasing root zone pH [59].
Leaf and root Ca and Mg concentrations of both species were not reduced at pH 4.0, except for leaf Ca concentration at H1. It has been documented that higher H + concentration in the nutrient solution inhibits Ca and Mg uptake in basil [29] and gerbera [60], leading to instability of root cells membranes [30]. Nevertheless, those responses can be alleviated by increasing Ca in the rhizosphere, since Ca is considered a base cation which does not increase acidity in nutrient solution [29,30]. In our study, the lack of effect of pH 4.0 on Ca accumulation in roots and leaves, could be ascribed to the presence of adequate Ca in the nutrient solution and/or to the proven ability of these plant species to adapt to adverse environments either due to weather extremities or to nutrient deficiencies [61].
The decrease in leaf and root K concentrations caused by the low pH level (pH 4.0) in T. officinale and R. picroides (only at H1) plants has also been reported in other species, e.g., Trifolium repens L. and Lolium perenne L. [53]. According to Alam et al. [30], this effect could be associated with the H + -induced inhibition of the root cells plasma membrane function as K + movement is symplastic, as well as with the reduced ability of roots to uptake K or the increased loss of K from the root cells due to root injury caused by low Ca content. Although reduced, the K concentration of both species at pH 4.0 was at levels considered sufficient for plant growth, suggesting that this reduction may be mainly caused by a direct antagonism of K + -H + uptake and to a lesser extent by reduced K uptake or increased K loss due to root injury [62].
In both species, leaf and root B concentrations were not affected by the pH values in nutrient solution, whereas the level of pH 7.0 did not affect leaf and root Cu in T. officinale, but it reduced leaf and root Cu in R. picroides at H2 and H1 harvesting dates, respectively. It is reported that increased nutrient solution pH reduces the concentration of micronutrients in several plant species [63]. A reduction in root Fe concentration was found at pH 7.0 in both species, due to the more intense oxidation of iron and thus its lower availability at higher pH [30]. Although the low pH level (pH = 4.0) favored Fe root concentration in both species, R. picroides roots accumulated Fe over three times more in pH 4.0 than in pH 7.0. Enhanced accumulation of Zn in the leaves of both species and in the roots of R. picroides was also found in pH 4.0, as it has been reported for other plant species grown in low pH [64]. Similarly, pH 4.0 increased leaf Mn in T. officinale (H2) and R. picroides (both harvests), a finding which is in agreement to reports in other plant species [58,60]. According to Gillespie et al. [29], decreasing pH values increase the solubility of nutrients such as Zn, Mn and B due the exchange of the hydronium ions with cationic nutrients from negatively charged particle in the substrate. On the other hand, at high pH values these nutrients precipitate and become unavailable for uptake from plant roots [29]. However, the same authors suggest the lack of negatively charged particles within the nutrient solution that could exchange cations with hydroniums; therefore, the effect of pH values on nutrients availability in nutrients solution differs from that in soil and substrate culture due to possible antagonistic effects between released hydroniums and cations. Finally, the higher concentration of Mn and Zn in the leaves of both species grown in pH 4.0 could be related with the higher chlorophyll content, as these elements play an important role in chlorophyll synthesis [65].
Total soluble solids content of leaves was significantly increased or showed increasing trends in plants grown at pH 4.0 in both species. This finding, along with the higher titratable acidity of T. officinale leaves at the same pH at H1, could suggest better taste and higher consumers' acceptance due to improved quality of the final product. Moreover, similar trends were recorded in H2 for both species where the low and moderate pH values (4.0 and 5.5, respectively) increased significantly titratable acidity compared to the highest pH level. In contrast to our study, pH values did not seem to affect titratable acidity in root and shoot sap of tomato, lettuce, and bermudagrass [66].
Chlorophyll content increased in plants grown at pH 4.0 leading to a better product visual appearance and consequently to higher consumer acceptability. Apart from its importance in photosynthesis, leaf chlorophyll content can be used to estimate the greenness of leafy vegetables [67] and plays an important role in their acceptance by consumers [28]. In addition, the decreased leaf nitrate content in plants grown at pH 4.0 contributes to final products of better quality in terms of human safety, as nitrates are significant anti-nutritional factors in leafy vegetables [68].
Leaf content in total phenolics was higher in plants grown at pH 4.0, indicating that low pH imposed an abiotic stress in both species thus eliciting the biosynthesis of secondary metabolites such as phenolic compounds [33,36,69]. This finding in turn, results in healthier products, as the dietary value and antioxidant activity of the studied species and many other wild edible greens depend to a large extent on their content and composition in phenolic and other bioactive compounds [6,7,10,70,71]. However, pH 4.0 increased more intensively the content of phenolics in the leaves of R. picroides than in T. officinale, providing additional evidence that low pH was more stressful for R. picroides plants. This is also in agreement with the increased leaf content in carotenoids + xanthophylls of R. picroides plants grown at pH 4.0 in H1. Carotenoids, except for their main role as accessory pigments and their involvement in plant photoprotective mechanisms, have also been found in higher concentrations in plants grown under abiotic stresses since they contribute to the overall defense mechanism of plants due to their antioxidative properties [53,63].

Conclusions
Commercial cultivation of wild edible greens is gaining increasing interest, since the current consumers trends have created marketing niches for cultivating alternative/ complementary species with functional properties. The results of the present study showed the tested species showed promising characteristics and could be successfully cultivated in a floating hydroponic system. Both species showed the ability to adapt in nutrient solutions with acidic to neutral pH levels (4.0, 5.5 and 7.0), although the application of pH 5.5 was more favorable. Moreover, T. officinale showed higher adaptability to pH 4.0, compared to R. picroides, indicating the presence of an efficient defensive mechanism in the first case. However, despite the lower yields of plants under low pH values, this particular treatment enhanced most of the studied quality and dietary characteristics in both plants, a finding which is of great importance for consumers' acceptance as novel/complementary leafy greens for human consumption and, especially, for their use in the food industry for the design of functional food products. Considering the lack of information regarding the cultivation requirements of the species, the results of the present study would provide important information regarding agronomic practices such as the nutrient solution pH and harvesting time, aiming to introduce these underexploited species in sustainable farming systems. In conclusion, the regulation of nutrient solution pH and the application of low pH values (pH = 4.0) may be beneficial to the quality of the final product, while moderate levels of pH (pH = 5.5) did not severely affect yield components in both species highlighting the potential of commercial growing of the studied species under adverse conditions. Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/agronomy11061118/s1, Table S1: The composition of stock solution used for the preparation of nutrient solution, Table S2: Water composition used for the preparation of nutrient solution.