1. Introduction
Soil salinity is an increasing problem in many areas worldwide, particularly in the semi-arid and arid Mediterranean [
1,
2]. Over 7% of the world’s total land and approximately 20% of irrigated land is affected by high salinity. As the extent of global soil salinization and drought events are expected to increase as a result of the global climate change [
3], systematic research on salinity and drought stress tolerance mechanisms in plants and breeding tolerant crops is paramount for future food security. Salinity induces alterations in the growth and development of plants due to its cumulative effect on several physiological as well as biochemical processes such as water balance, mineral ion homeostasis, osmolyte accumulation, antioxidant metabolism, photosynthetic capacity of plants, etc. [
4]. All these changes ultimately lead to huge economic losses in crop production.
Brassica vegetables (Brassicaceae) include many economically important species grown worldwide. These vegetables have received public and scientific attention for their health potential due to their wealth of “healthy phytochemicals” (carotenoids, phenolics, glucosinolates) [
5,
6,
7,
8,
9]. They are recognized as a functional food since different epidemiological and meta-analyses have shown that metabolites found in Brassicas have anti-inflammatory, anti-oxidant, anti-mutagenic, and anti-carcinogenic activities [
10,
11,
12].
As
Brassica crops are commonly grown in the Mediterranean area, their production is greatly affected by unfavorable environmental conditions (abiotic stresses including increased salinity). Therefore, current pressing questions are, for example, how increased salinity affects the growth of commercially important
Brassica crops and which of the metabolites in their tissues correlate with the tolerance to salinity of these species. These natural substances are the products of plant interaction with the environment, have limited occurrence, and some are reported to mediate abiotic stress tolerance. Their role is usually associated with protection against oxidative damage induced by abiotic stresses. However, their ecological value in plant adaption to the environment has been relatively poorly investigated. Glucosinolates, a class of specialized metabolites, almost exclusively found in Brassicaceae, have been shown to increase in plants when salinity is higher than the tolerance levels, while their production is inhibited under severe stress conditions that depend on plant species/variety [
13]. Carotenoids and polyphenolic compounds are widely abundant in many plant species including Brassicaceae. Bose et al. summarized the current knowledge of the role of carotenoids and polyphenolic compounds in salinity tolerance using halophytes, a salinity-tolerant species, as a plant model for understanding complex salinity tolerance mechanisms [
14]. The authors reported that halophytes have a higher concentration of carotenoids than glycophytes (a salinity-sensitive species) under control conditions and showed a much lower reduction in carotenoid concentration after salt treatment. In addition, halophytes, such as
Cakile maritima, (Brassicaceae) accumulate polyphenolic compounds that participate in salinity tolerance due to reactive oxygen species (ROS) scavenging ability [
15].
A group of phenolic compounds that may participate in abiotic stress responses are phenolic acids. Phenolic acids, including hydroxybenzoic and hydroxycinnamic acids, and their derivatives may be present in soluble forms in which they are conjugated with sugars or organic acids, as well as bound to more complex structures such as hydrolysable tannins or lignins [
16]. There are reports of beneficial effects of the exogenous application of some phenolic acids to salinity-stressed plants. Miura and Tada reported that salicylic acid (SA) has great agronomic potential to improve the stress tolerance of various agriculturally important crops [
17]. However, the applicability of SA is dependent on the concentration used, the mode of application, the plant species, and stage of growth. SA is a phenolic acid that acts as a stress hormone, mediating plant responses to biotic and abiotic stresses. In addition to SA, increased salinity tolerance of wheat seedlings was also obtained after treatment with sinapic, caffeic, ferulic, and p-coumaric acids [
18]. Further, endogenous ferulic and p-coumaric acid are purported to be involved in the tolerant mechanism against salinity stress in rice [
19]. Soil salinity also increased the concentrations of leaf phenolics, including chlorogenic acid, in honeysuckle as a mechanism for acclimation to saline stress [
20]. Martinez et al. reported that cinnamic and p-coumaric acid and p-coumaryl-CoA, in addition to flavonols, were several times higher in tomato due to salinity, heat, and combined stress (heat + salinity) compared to control plants [
21]. The role of phenolic acids in salinity tolerance is therefore still unclear, especially in
Brassica crops, and needs further investigation.
Our recent paper, based on the comparative analysis of three
Brassica crops with global economic importance, i.e., Chinese cabbage (
Brassica rapa ssp.
pekinensis), white cabbage (
Brassica oleracea var.
capitata), and kale (
Brassica oleracea var.
acephala) in relation to sensitivity/tolerance to salinity, identified Chinese cabbage as sensitive, white cabbage as mildly tolerant, and kale as the most tolerant species [
22]. We have also shown that plant hormones play an important role in mediating salinity tolerance in the above
Brassica crops [
22,
23]. In this article, we have extended our research and evaluated the effect of salinity on the levels of specialized metabolites in the same
Brassica species (
B. rapa spp.
pekinensis,
B. oleracea var.
capitata, and
B. oleracea var.
acephala). We analyzed three groups of metabolites (carotenoids, glucosinolates, and phenolics) in three
Brassica seedlings at increased salt concentrations (0–200 mM NaCl). More detailed analysis of phenolic acids was then carried out by UPLC–MS/MS. Our hypothesis was that differently tolerant
Brassica species would respond diversely to salinity stress in terms of profile and phenolic acid levels. Correlation studies between
Brassica crops with different tolerance to salinity stress and levels of specialized metabolites are also discussed, with a particular focus on phenolic acids.
4. Materials and Methods
4.1. Plant Material and Experimental Conditions
Chinese cabbage (B. rapa var. pekinensis) was obtained from International Seeds Processing GmbH, Germany, while white cabbage (B. oleracea var. capitata cv. Varaždinski) was obtained from the Agricultural Advisory Service of Varaždin Region, Croatia, and kale seeds (B. oleracea var. acephala) from a family farm from Vrgorac, Croatia. Before germination, the seeds were surface-sterilized in 3% Izosan G (Pliva, Croatia) for 10 min, washed with sterile water (5 times), transferred to 1% agar plates, and left at +4 °C for three days in the dark. The seeded plates were placed in a growing chamber, in a vertical position, under control conditions of 16/8 h light/dark photoperiod, light intensity of 115 µmol m−2 s−1, and temperature 22 °C. After the seedlings reached about 1 cm in length, they were placed on 1% agar plates containing NaCl (in concentration range 50–200 mM). Corresponding controls were placed on 1% agar without salt. Both control and experimental plates were incubated for 24 h. Biomass, root growth, and ROS production were determined in in vivo seedlings. For biochemical analysis, five biological replicates of seedlings were collected, immediately frozen using liquid nitrogen, and stored at −80 °C. The plant material was then freeze-dried and stored until analysis.
4.2. Determination of Sodium and Potassium Content
Levels of Na
+ and K
+ in
Brassica seedlings were determined by high-resolution inductively coupled plasma mass spectrometry (HR-ICP–MS, Element 2, Thermo, Bremen, Germany) in connection with an autosampler ESI-a SC-2 DX FAST (Elemental Scientific, Omaha, NE, USA). The measurement parameters and instrument conditions were set as described earlier [
49]. Indium was used as an internal standard. Lyophilized seedling tissues (about 100 mg) were subjected to microwave- (Anton Paar Multiwave 3000, Graz, Austria) assisted acidic digestion in HNO
3/HF (60:1,
v/
v) at 1400 W. The measurements were performed in four replicates.
4.3. Proline Quantification
Proline concentrations were assayed according to [
50] with some modifications. In brief, extraction was performed using 30 mg of the freeze-dried tissue in 70% ethanol. A volume of 100 µL of the extract was mixed with 1000 µL of the reaction mixture (1% ninhydrin [
w/
v], 60% acetic acid [
v/
v], and 20% ethanol [
v/
v]) and then heated to 95 °C for 20 min. Proline levels were measured at 520 nm using a UV–VIS spectrophotometer (BioSpec-1601 E, Shimadzu) and calculated using a standard curve (y = 0.0015x, R
2 = 0.9991; serial concentrations of proline standard (Sigma): 0.04, 0.1, 0.2, 0.4, 1.0, 1.5 mM). The results are expressed in µM L-proline mg
−1 dw (dry weight).
4.4. ROS and GSH Fluorescent Measurements
The amount of ROS (SO, H
2O
2) as well the GSH content of the seedlings was determined in vivo using specific dyes (dihydroethidium, DHE, dichlorodihydrofluorescein diacetate, DCFH-DA, and monochlorobimane, MCB) according to reported methods [
51]. All measurements were performed with the roots of stressed seedlings, compared to their appropriate controls. Briefly, the roots of stress and control seedlings were incubated in 10 μM DHE (30 min), 50 μM DCFH-DA (30 min), and 50 μM MCB (40 min) for the determination of SO, H
2O
2, and GSH, respectively. After incubation, the samples were washed with water to remove the dye surplus. Fluorescent signals that appeared as a result of the reaction between fluorescent dyes and substrates were determined using a fluorescent microscope (Olympus BX51, Olympus Optical Co. (Europa) GmbH) connected to the camera (Olympus DP70, Tokyo, Japan). The accumulated fluorescent products were quantified using Lucida 6.0 software (Kinetic Imaging Ltd., Wirral, UK). Twenty-five fields on each image were analyzed, and the results are presented as the mean of the fluorescence intensity of five images per treatment.
4.5. Pigment Content Determination
Plant pigments, chlorophylls
a and
b, and carotenoids were measured in fresh cotyledons of seedlings upon treatments, and their contents were calculated according to Lichtenthaler and Buschmann [
52]. Pigments levels were measured at three different wavelengths, 663.2 nm for chlorophyll
a, 646.8 nm for chlorophyll
b, and 470 nm for carotenoids. The results are presented in mg g
−1 fw (fresh weight).
4.6. Glucosinolate Measurements
Total glucosinolate content was measured according to Aghajanzadeh et al. [
53] with certain adjustments. Lyophilized tissue (30 mg) was extracted in 80% methanol. In order to inactivate the myrosinase enzyme, the extracts were subsequently heated in a thermobloc at 95 °C for 2 min and then were cooled and centrifuged (5 min at 13 000 rpm). Glucosinolate levels were determined in a reaction mixture (930 µL) containing 30 µL methanolic plant extract and 900 µL 2 mM disodium tetrachloropalladate (Na
2PdCl
4) using a UV–VIS spectrophotometer (BioSpec-1601 E, Shimadzu) at 425 nm. The samples were incubated for 30 min at room temperature before the measurements. The results were calculated using a standard curve (y = 0.0003x, R
2 = 0.998; serial concentrations of sinigrin standard (Carl Roth GmbH, Karlsruhe, Germany): 0.1, 0.25, 0.5, 1.0, 1.5, 3.0 mg mL
−1) and are presented as sinigrin equivalents per dry weight (μg sinigrin mg
−1 dw).
4.7. Determination of Polyphenolic Compounds
For the measurement of polyphenolic compounds, extractions were carried out in 2 mL 80% methanol using 60 mg of freeze-dried tissue. For tissue homogenization, a Mixer Mill MM 400 (Retsch, Haan, Germany) was used for 5 min at 30 Hz, after which the extracts were placed in a sonicator (10 min) and further mixed in a tube rotator (1 h, 15 rpm). The extracts were then centrifuged (Eppendorf centrifuge, 10 min, 13,000 rpm), and the supernatants were used for all analyses described below. All extractions were carried out in five biological replicates for all three species. The measurements were adapted to small volumes.
The Folin–Ciocalteu method for the assessment of TP was used according to Singleton and Rossi [
54]. The results were calculated using a standard curve (y = 0.0011x, R
2 = 0.998; serial concentrations of gallic acid (Alfa Aesar, Haverhill, MA, USA): 50, 100, 150, 250, 500 mg L
−1) and are presented as equivalents of gallic acid per dry weight (mg GAE mg
−1 dw). TPA were determined using Arnow’s reagent according to the European Pharmacopoeia [
55]; the results were calculated by using a standard curve (y = 0.0042x, R
2 = 0.9936; serial concentrations of caffeic acid (Sigma-Aldrich, St. Louis, MO, USA): 10, 50, 100, 250, 500 mg L
−1) and are expressed as equivalents of caffeic acid per dry weight (mg CAE mg
−1 dw). TF were measured using the AlCl
3 method [
56]. The results were calculated by using a standard curve (y = 0.0031x, R
2 = 0.9898; serial concentrations of catechin standard (Kemika, Zagreb, Croatia): 50, 100, 150, 200, 250 mg L
−1) and are presented as equivalents of catechin per dry weight (mg CE mg
−1 dw). TFL were analyzed by the p-dimethylaminocinnamaldehyde (DMACA) method [
57]. The results were calculated by using a standard curve (y = 0.1414x, R
2 = 0.9996; serial concentrations of catechin standard (Kemika, Zagreb, Croatia): 0.5, 1, 2, 4, 6, 8, 10 mg L
−1) and are presented as equivalents of catechin per dry weight (mg CE mg
−1 dw).
4.8. Principle Component Analysis (PCA)
Relations between the measured values of specific metabolites (total phenolics, flavonoids, phenolic acids, flavanols, glucosinolates, and carotenoids) and the experimental variants of Brassica crops were examined by PCA. PCA was performed using a correlation matrix of the average values of traits after standardization (autoscaling). Linear correlations among variables were determined by Pearson coefficients (p < 0.05). The XLSTAT software (ver. 2017.01.40777) implemented in Microsoft Office Excel 2010 was used for all statistical procedures.
4.9. Phenolic Acid Analyses
The extraction of phenolic acids was performed using 30 mg of freeze-dried plant material in 80% methanol. Internal standards of deuterium-labeled 4-hydroxybenzoic and salicylic acids were added to all samples at a final concentration of 10
−6 mol L
−1. The fractions of soluble free acids, soluble ester-bound phenolic acids, and cell wall-bound phenolic acids were prepared by a previously published method [
58]. Quantification and identification of phenolic acids were performed using UPLC–MS/MS as described earlier [
59].
4.10. Statistical Analysis
The data were analyzed with the STATISTICA program (Version Stat Soft. Statistica.v 10.0. Enterprise). ANOVA was used to analyze the relevant factors, and values were considered to be significant at p < 0.05. Post-hoc multiple mean comparison (Tukey’s HSD test) was used for multiple comparisons.
5. Conclusions
In this work, we evaluated the effect of short-term (24 h) salt stress (influence of NaCl at concentrations of 50, 100, or 200 mM) on selected metabolites, with a special focus on phenolic acids, in Chinese cabbage, white cabbage, and kale seedlings. The negative effects of increased salinity were confirmed in all experimental variants. Seedlings showed root growth inhibition, reduced biomass production, and increased Na+/K+ ratio, elevated ROS and GSH, and proline content. Based on these parameters, a high level of stress was demonstrated due to increased salinity, especially in Chinese cabbage, suggesting that this Brassica is highly sensitive to salinity stress. PCA analysis showed a grouping of specific metabolites (carotenoids, polyphenols, and glucosinolates) closer to more tolerant species, suggesting a positive role for these natural substances in stress management. The phenolic acid analysis further confirmed that the more tolerant Brassicas, kale and white cabbage, contained a significantly higher level of total phenolic acids, especially of total hydroxycinnamic acids with respect to hydroxybenzoic acids, compared to Chinese cabbage. Total hydroxycinnamic acids tended to increase, while hydroxybenzoic acids tended to decrease under the applied salinity conditions. Furthermore, the more tolerant white cabbage and kale contained higher levels of cell wall-bound phenolic acids, particularly SiA, compared to the salt-sensitive Chinese cabbage. The most marked decrease in phenolic acids (especially PA, pCoA, SA, and CaA) was observed in salt-treated Chinese cabbage. White cabbage showed no significant changes in phenolic acid levels. In addition to the decrease in CaA, SA, and pCoA, there was also a significant increase in FA in kale under stress conditions. Our results suggest that phenolic acids are species-specific in Brassicaceae and can contribute to their stress tolerance. Salt-tolerant species exhibit a higher level of phenolic acids and suffer less from metabolic disorders under salinity stress.