1. Introduction
Modern farming systems aim to obtain maximum yields to ensure the alarmingly increasing food demands [
1,
2], while at the same time, the anthropogenic activities tend to gradually reduce the available arable land [
3]. Therefore, a new approach is needed focusing not only on intensification of cropping systems, but also on maximizing the efficient use of the available arable land, including degraded soils that cannot be grown with conventional crops. In this context, the valorization of underutilized and neglected species seems to be a promising alternative to the existing crops, especially when considering the ongoing climate crisis that severely affects conventional crop production [
4], which may provide new sources of food with improved nutritional properties [
5]. Moreover, the integration of such species in farming systems ensures the preservation of agrobiodiversity heritage and reduces the risk of genetic erosion in agroecosystems due to intensified cropping systems [
1,
6,
7,
8].
Scolymus hispanicus L. (also known as common golden thistle or Spanish oyster thistle) is a wild annual or perennial herb belonging to the Asteraceae family with wide distribution in the Mediterranean basin [
9,
10,
11]. It is usually found in agricultural ecosystems and is considered a difficult-to-control and noxious weed [
12,
13]. However, although it is undesirable in commercial farms, it is highly appreciated as a wild edible green due to its high nutritional value and beneficial health effects, especially in the countries of southern Europe [
9,
14]. It is a common ingredient in various gourmet and local dishes of the so-called Mediterranean diet where it is consumed in raw, boiled or fried form [
9,
11,
15,
16]. The most commonly consumed plant parts are the flowers, midribs and petioles of leaves, as well as the cortex of the roots, which after post-harvest processing can be used in various dishes [
10,
17]. According to Disciglio et al. [
18], wild
S. hispanicus plants are rich sources of Mg and Ca and contain low amounts of nitrates, while Rubio et al. [
19,
20] identified various flavonoids and phenolic acids. Petropoulos et al. [
21] suggested luteolin and kaempferol glucuronides as the major phenolic compounds, while Vardavas et al. [
22] detected moderate amounts of vitamins (K1 and C), carotenoids (lutein and carotene), Tbatou et al. [
23] detected high dietary fiber content, and Morales et al. [
24] detected low amounts of α- and total tocopherol. Moreover, Vardavas et al. [
25] reported a balanced content of n-6 and n-3 fatty acids (a ratio of 1.06) and high amounts of palmitic, linoleic and α-linolenic acids.
Considering consumers’ awareness regarding the origin of food and the production practices implemented, especially regarding the inputs of agrochemicals, the use of sustainable means for crop production is essential for fulfilling the market demands [
26]. The introduction of alternative crops such as the various wild and underutilized species falls within this context due to their low requirements in agrochemicals and natural resources (e.g., fertilizers and water) and their efficient adaption strategies to various abiotic and biotic stressors [
27,
28]. Moreover, the commercialization of these species is pivotal for the reduction of genetic erosion risks related to irrational harvest and anthropogenic activities [
29]. During the last few years, several studies have focused on the chemical characterization and the bioactivities of various wild edible plants [
30,
31,
32], while numerous ethnopharmacological studies have highlighted their contemporary uses in modern diets and their positive health effects [
13,
22,
33,
34,
35,
36,
37,
38]. However, in order to domesticate these species, useful information regarding the best practice guides that farmers should follow in order to achieve high yields and high quality of the final product should be also provided [
9,
13]. Thus far, various species have been suggested for commercial exploitation in small-scale farming systems of the Mediterranean, including
Cichorium spinosum [
39,
40,
41],
Portulaca oleracea [
42,
43,
44],
Sanguisorba minor [
45,
46]
Crithmum maritimum [
47,
48,
49] and several others [
21,
31,
50,
51]. Among the cultivation practices, the fertilization regime has a significant impact on the yield, the chemical composition and bioactive properties, and the optimum fertilization has to be considered for the commercial production of final products with similar quality as the wild counterparts [
3,
52,
53,
54,
55]. Moreover, the existing genotypic diversity among the numerous ecotypes of these species suggests a wide variation in chemical composition, which along with the effect of growing condition, may result in significant differences in the chemical profile of wild edible greens [
56,
57].
Despite the prolific studies regarding the chemical properties and the cultivation practices of various wild edible species, there is scarce literature for
Scolymus hispanicus since most of the studies focus on the chemical characterization and bioactivities of plants collected from the wild [
18,
36,
58]. On the other hand, Papadimitriou et al. [
59] suggested that
S. hispanicus is moderately tolerant to salinity species that could be utilized in saline agriculture, while they also suggested its introduction in soilless cropping systems with reduced macronutrients requirements [
60]. Considering the limited information about
S. hispanicus cultivation, the aim of the present study was to evaluate the impact of different fertilization regimes that varied in the amounts of the main macronutrients (e.g., N, P, K) on the growth, nutritional and mineral profile and chemical composition of
S. hispanicus plants. The presented results will be helpful for the integration of the species as an innovative crop in the existing farming systems, especially in the small-scale farms of the Mediterranean, while they provide a best-practice guide for the fertilization of the species, focusing on high yields without compromising the quality of the final product.
3. Results and Discussion
The results of growth parameters are presented in
Table 1. Significant statistical differences regarding the morphological traits in relation to the different fertilization regimes were recorded. The highest number of leaves/plant was achieved for the SH4 (300:100:100; 19.46) fertilization regime, being significantly different from the rest of the treatments, whereas SH1 (100:100:100) and SH6 (300:300:300) treatments recorded the lowest mean values. The SH3 (200:200:200) and SH5 (300:200:200) treatments recorded the highest fresh weight of leaves/plant (116.41 g and 113.58 g, respectively), whereas plants treated with SH6 and the control treatment had the significantly lowest fresh weight. On the other hand, the highest dry matter content was observed for the control treatment, while the leaves obtained from plants treated with SH4 and SH5 had the lowest overall dry matter content. Our results indicate that apart from the increased amounts of nitrogen, the application of nutrient solution with balanced composition in P and K had a beneficial effect on fresh biomass yield, whereas excessive amount of macronutrients resulted in fresh yields similar to the untreated plants due to reduction in leaves number. These findings corroborate the aspect of minimum nutrients requirements for wild edible species, especially for the case of P and K, which were also confirmed by Polyzos et al. [
39] in
Cichorium spinosum cultivation. Moreover, Papadimitriou et al. [
60] suggested that a high ratio of N:K (e.g., 2.38 mol/mol) in nutrient solution resulted in increased yields of leaves and roots in hydroponically grown
Scolymus hispanicus compared to lower ratios (e.g., 1.59 mol/mol). This contradiction could be attributed to differences in the cropping systems (open hydroponic system vs. pot cultivation), since the availability and uptake of N are not comparable. Regarding the dry matter content, Polyzos et al. [
39] also reported increased values of dry matter in untreated plants
C. spinosum plants, suggesting stressful conditions due to nutrients deprivation, a finding that is in agreement with the results of our study.
Table 2 presents the results of chlorophyll content (SPAD index, leaf area and specific leaf area). SPAD index values were the highest when plants were treated with SH4 and SH1 treatments. Similarly, the plants that received SH4 treatment formed the highest leaf area (1721.63 cm
2), without being significantly different from SH1 treatment, while all the tested fertilization regimes had higher SAPD index and leaf area values than the untreated plants (SHC treatment). This finding is in line with the highest number of leaves per plant, which was recorded for the SH4 treatment, thus indicating that the leaf area increased due to the formation of more leaves instead of the development of bigger leaves. Similarly, control treatment recorded the lowest overall value of specific leaf area (12.65 m
2/kg), while the significantly highest values were measured for plants treated with SH5 and SH4 treatments (16.82 and 16.42 m
2/kg, respectively). The significant effect of fertilization regime on SPAD index value of wild edible leafy greens has been also reported by Polyzos et al. [
39] and Fidimundy et al. [
46], who studied the growth parameters of
Cichorium spinosum and
Sanguisorba minor plants, respectively. On the other hand, Tzortzakis and Klados [
64] and Papadimitriou et al. [
59] did not record any differences in
C. spinosum and
S. hispanicus plants treated with nutrient solutions of different salinity levels. According to Di Mola et al. [
65], SPAD index values of baby spinach and lamb’s lettuce leaves were positively correlated with nitrogen availability, while Karkanis et al. [
66] highlighted the importance of harvesting stage on this parameter. Moreover, Fontana et al. [
55] suggested that apart from total nitrogen availability, the nitrogen form may also affect chlorophyll content in purslane plants cultivated in a soilless hydroponic system. Based on our results, the increased amounts of nitrogen combined with low amounts of P and K (SH4) or a balanced solution of N:P:K (SH1) were beneficial to chlorophyll content and leaf area, whereas Polyzos et al. [
39] suggested that the untreated plants or those that received a nutrient solution that contained 200:200:200 ppm of N, P and K had the highest overall SPAD values. Therefore, it could be assumed that each species may respond differently to fertilization regime, while in our case, the excessive inputs of micronutrients in
S. hispanicus (e.g., SH6) were not positively correlated with visual quality and fresh yield of leaves.
The nutritional profile of
S. hispanicus leaves in relation to fertigation regime is presented in
Table 3, where a variable response was detected. Total fat, crude protein and ash content were negatively affected by fertigation, since the highest overall values were recorded for the untreated plants. In contrast, dietary fiber and carbohydrates content were significantly higher for plants treated with SH3 and SH5 treatments, resulting in a higher energy content for the latter treatment. These results are within the range of the values reported by García-Herrera et al. [
56] who evaluated the proximate composition of wild golden thistle plants collected from different sites and in different years. However, the authors of that study detected a great variation among the tested samples, which indicates a significant effect of the genotype and the growing conditions on the nutritional value of the species. Wild
S. hispanicus plants are considered a rich source of dietary fiber [
3,
56] and total carbohydrates [
23], which in the case of our study, its content was significantly increased by fertigation regimes, whereas fat and ash contents were low compared to other wild edible species [
23]. According to the literature, commercial cultivation practices may affect the nutritional value of wild species depending on the species [
61,
67], allowing us to regulate the quality of the final product and improve the content of beneficial compounds. However, this is not always the case, as for example, Disciglio et al. [
18] did not observe a significant difference in protein content between wild and cultivated plants of
C. intybus,
Borago officinalis and
Diplotaxis tenuifolia, whereas Polyzos et al. [
39] recorded a decrease in protein and ash content when
C. spinosum plants were treated with nutrient solutions similar to our study.
Organic acid composition is presented in
Table 4. The main detected compounds were oxalic, quinic and malic acid followed by shikimic and citric acid, while traces of fumaric acids were also present in the studied samples. A varied response to fertigation regime was observed, with higher amounts of oxalic, quinic and total organic acids being detected in the untreated plants. Malic acid was the highest in SH treatment, while the SH3 treatment resulted in the highest amounts of shikimic and citric acid. On the other hand, the SH4 treatment resulted in the lowest values for all the detected compounds (except for the case of quinic acid where the lowest content was recorded for SH2 treatment) and consequently in the lowest content of total organic acids. According to the literature, the increased inputs of nitrogen are associated with high amounts of oxalic acid [
61,
67], while the nitrogen form may affect the accumulation of this particular organic acid or total organic acids [
68]. Moreover, Dias et al. [
69] reported a decrease in oxalic acid content in cultivated
Achillea millefolium plants compared to unattended ones. This contradiction could be due to the fact that in our study, the control plants were subjected to stress conditions due to nutrient deprivation and P in particular. According to Le Roux et al. [
70], P deficiency is associated with the synthesis and accumulation of organic acids which tend to decrease nitrogen assimilation. Another possible explanation could be related with the ratios and the total amounts of macronutrients applied in the tested fertigation regimes, which may result in synergistic or antagonistic effects that consequently affect nutrient assimilation and impair plant physiology and metabolism [
71]. According to Aboyeji et al. [
72], high amounts of P may have detrimental effects on the yield of groundnut plants due to antagonism between P and Zn that may affect plant growth and development. Moreover, excessive amounts of nitrogen are associated with reduced uptake of other nutrients, which result in stressful conditions and consequently in the accumulation of organic acids [
73,
74]. Therefore, based on our results and the literature reports, further studies are needed to reveal uptake and translocation of nutrients from roots to upper parts in order to reveal the mechanisms involved in organic acids biosynthesis as part of the antioxidant and osmoregulatory mechanisms of plants. In any case, the reduction of organic acids and oxalic acid in particular after the application of the tested fertigation regimes can be considered a positive impact on the quality of the final product, since oxalic acid is an antinutritional factor, and high intake (>5 g per day) may have severe health effects [
51].
Tocopherol and free sugar compositions are presented in
Table 5. Alpha-tocopherol was the only detected isoform of vitamin in all the studied samples, while except for the case of SH5 treatment, the rest of the fertigation regimes resulted in a significant decrease compared to untreated plants. The detected amounts were different from those reported by Petropoulos et al. [
21] (0.68 μg/100 g fw) and Marmouzi et al. [
75] (0.54 mg/100 g) and in the same range with the study of Morales et al. [
24] (0.02 mg/100 g fw) and Vardavas et al. [
22] (0.038 mg/100 g fw). Moreover, in contrast to our study, β- and γ-tocopherols were also detected in
S. hispanicus leaves, which could be due to different growing conditions and genotypic variability [
76]. The variable response to fertilization regime was also reported in the study of Polyzos et al. [
39] who evaluated the response of
C. spinosum plants as similar to our study fertigation regimes and recorded the highest α- and total tocopherols content for the treatment of 300:200:200 ppm of N:P:K. It seems that high amounts of nitrogen combined with a balanced content of P and K in the nutrient solution may improve tocopherol composition, a finding which has been confirmed in other species [
67,
77].
Regarding the free sugar composition, glucose was the main detected free sugar, followed by fructose, sucrose and trehalose. A variable response was recorded in relation to the tested fertigation regimes. In particular, the highest glucose content was detected in control plants, while fructose content was the highest in SH3 and SH4 treatments. Similarly, sucrose and trehalose were significantly higher for the SH6 and SH1 and SH6 treatments, respectively. Finally, the highest and lowest total free sugar content was recorded for SH4 and SH% treatments, respectively. To the best of our knowledge, this is the first report regarding the free sugar composition and no references are available for comparison purposes. However, Polyzos et al. [
39] who studied the same fertigation regimes in
C. spinosum plants also reported a varied response, while similar fluctuations have been suggested in other crops due to growing conditions and nutrients availability [
78,
79]. Regulation of free sugar composition through the application of tailor-made fertigation regimes could be a cost-effective means to improve the quality of the final products of wild edible species, since the increased sugar content could be associated with improved taste and organoleptic properties.
Mineral composition of
S. hispanicus leaves is presented in
Table 6. A varied response was recorded without specific trends in the effects of the tested fertigation regimes being observed. In particular, the untreated plants had the highest content of K and Zn, while plants treated with SH4 treatment recorded the highest content of Na, Ca and Mg. Moreover, SH1 and SH2 treatments had significantly higher Fe content compared to the rest of the treatments, while Mn and Cu content was significantly higher for SH1 treatment. The range of minerals detected in our study was in the same range as the values reported by García-Herrera et al. [
56] with slight variations in the case of Ca and Cu, while Papadimitriou et al. [
59] reported higher values for Na, K, Ca and Mg. However, it has to be noted that García-Herrera et al. [
56] recorded a high variability in minerals profile depending on the collection site and year, and they suggested a significant impact on growing conditions, while they determined the mineral content of the midribs
S. hispanicus instead of whole leaves, which were evaluated in our study. According to Rietra et al. [
72], significant interaction may occur among plant macro- and micronutrients, which may negatively or positively affect plant growth and crop yield. This is evident in our study where the varied amounts of N:P:K applied through fertigation resulted in a varied response regarding the mineral composition of
S. hispanicus leaves. According to Fageria and Oliveira [
80], P is the most influential nutrient since its imbalance may severely affect crop yield, a finding that is in agreement with our study where the highest overall fresh yield was recorded for the treatments where 200 ppm of P was applied (SH3 and SH5). Moreover, other studies report significant antagonistic effects between P and Mg or Ca [
81,
82], which coincide with the findings of our study where the highest Mg and Ca contents were recorded in SH4 treatment where 300:100:100 ppm of N:P:K was applied. Therefore, it could be suggested that the regulation of fertilization regime may favor crop yield and improve the mineral profile of
S. hispanicus leaves at the same time. However, the impact of growing conditions and genotype should be further investigated.
The fatty acid composition is presented in
Table 7. The most abundant compounds were α-linolenic acid (47.8–59.4%), palmitic acid (19.4–24.5%) and linoleic acid (10.44–12.72%), while the polyunsaturated fatty acids (PUFA) were the major class of fatty acids (60.3–70.8%). A varied response was recorded in relation to the fertigation regime without specific trends being observed among the tested treatments. The highest content of the major fatty acids was recorded either for untreated plants (e.g., α-linolenic acid) or for the treatments with low nutrient amounts (e.g., SH1 and SH2 in the case of linoleic and palmitic acid, respectively). The recorded composition was in the same range as the fatty acids profile reported by Morales et al. [
83] who detected high amounts of PUFA (57.66%), followed by monounsaturated and saturated fatty acids (MUFA: 34.16% and SFA: 8.19%) in peeled basal leaves. However, the content of individual compounds varied compared to our study, with α-linolenic acid being the most abundant one (30.55%), followed by linoleic and palmitic acids (26.44 and 20.65%, respectively). In contrast to our study, Vardavas et al. [
25] reported a different composition of the main fatty acids groups, suggesting MUFA (54.8%) as the most abundant class, followed by SFA (33.7%) and PUFA (11.4%). Moreover, they recorded a balanced ratio of n6/n3 (1.06) and different amounts of individual fatty acids (linoleic acid 33.8%; α-linolenic acid 32% and palmitic acid 30.3%). The variable-reported results could be associated with the tested raw material (whole leaves in our study were compared to peeled basal leaves) as well as to growing conditions and agronomic practices. Considering that the plants in our study were grown in a growth substrate under different fertigation regimes, this could also be a possible explanation for the observed differences as already confirmed by the literature reports [
39,
45].