Sprouting is reported to be associated with improvements of nutritional value of seeds, since it promotes accumulation of health-promoting phytochemicals (phenolic acids, flavonoids, and vitamins) (Oh and Rajashekar, 2009).
Cereal sprouts (i.e. caryopses within 3-4 days after germination) and microgreens (i.e. young seedlings of 1-2 weeks age, known as wheatgrass) may represent a valid alternative in human feeding (Benincasa et al., 2015). They allow avoiding processing treatments (i.e. milling and baking) thus maintaining the nutritional value of grains, mainly represented by substantial quantities of bioactive compounds – including phenolic acids – which are unevenly distributed in kernels, with aleurone layer having the highest antioxidant activity, followed by the bran fraction. The hulled wheat species, i.e. the diploid wheat einkorn (Triticum monococcum ssp. monococcum, genome AA) and the tetraploid emmer (Triticum turgidum ssp. dicoccum, genomes AABB), are less bred species and may be used for sprouting, to better satisfying target requirements of consumers compared to the most cultivated soft and durum wheat.
Recent works have focused on phenolic acids and flavonoids contents in ancient and modern wheat genotypes (Shewry and Hey, 2015), both on grains (Dinelli et al., 2011) and germinated sprouts (Benincasa et al., 2015; Hung et al., 2011). However, only few studies have investigated on the response in terms of phytochemicals accumulation during sprouting under sub-optimal conditions.
The content and composition of bioactive compounds in sprouts and wheatgrass depend on many factors such as genotype, growth stage and environmental conditions during sprouting (Cevallos-Casals and Cisneros-Zevallos, 2010). It is well acknowledged that plants under various abiotic stresses (i.e. drought, salinity, high light and chilling; Lim et al. 2012; Oh and Rajashekar, 2009) accumulate several phytochemicals, which play a major role in the adaptation of plants to unfavourable growing conditions (Guo et al., 2014; Yuan et al., 2010). Quality traits, as well as the level and activity of bioactive compounds occurring in foods of plant origin may be modified using a range of biotechnological and agronomic practices. However, to date little is known about the effect of abiotic stress on phytochemical accumulation in edible sprouts and the application of environmental stress to enhance their quality traits as well as the actual translatability of the enriched-bioactive compounds to biological benefits in lifestyle-related diseases. The effect of salt stress on bioactive compounds accumulation was studied in sprouts of radish (Yuan et al., 2010), broccoli (Guo et al., 2014), lentil (Świeca, 2015), peanut (Kavas et al., 2015), alfalfa (Wang et al., 2009) and buckwheat (Lim et al., 2012). In wheat (Triticum aestivum L.) the impact of salt stress was only evaluated in terms of germination rates and biometric determinations (Datta et al., 2009) while no quality trait was assessed.
Our hypothesis is that the content of bioactive compounds of ready-to-eat sprouts and wheatgrass of Triticum genotypes may be enhanced by both germination process and salt stress conditions applied during sprouting. The parameters assessed to characterise these effects were total phenolic content, radical scavenging activity and the content of some phenolic acids (both free- and boundfractions) in grains, sprouts and wheatgrass of three different Triticum genotypes (einkorn, emmer and durum wheat).
Materials and methods
Plant material and germination conditions
Grains of einkorn (Triticum monococcum L. ssp. Monococcum var Monlis, TMoM), emmer (Triticum turgidum L. spp. dicoccum (Schrank ex Schubler) Thell. var Zefiro, TDiZ) and durum wheat (Triticum turgidum L. ssp. durum (Desf.) Husnot var Creso, TDuC) were collected from different locations of Central Italy and were sown in autumn 2014 in small plots of 3 m2 without replicates at Mosciano S. Angelo (Central Italy, Province of Teramo, 42.75 °N, 200 m a.s.l., Mediterranean climate). The soil was uniform, silty-clay with 1.2% organic matter. Fertilisation inputs included 60 kg ha–1 of N; no pest and disease control was performed and weeds were controlled by hand. Grains were hand harvested in July 2015, at a grain water content lower than 15% and cleaned from dust and any other contaminants and stored in a dry and cool room until the beginning of the experiment, performed in laboratory.
Germination was carried out to obtain sprouts (growth stage: about 5 cm shoot length) and wheatgrass (growth stage: about 12 cm shoot length) of TMoM, TDiZ and TDuC. Caryopses (200 g for each genotype) were incubated in plastic trays containing filter paper laid over glass balls immersed in 0 (i.e. control), 25, 50 and 100 mM NaCl solutions (0_S, 25_S, 50_S and 100_S, respectively) throughout the whole experiment, i.e. until either sprout or wheatgrass stage. Additionally two recovery treatments were included consisting of sprouts obtained with 50 and 100 mM NaCl solutions and then transferred in trays containing distilled water until wheatgrass stage (50_R and 100_R, respectively). Each treatment consisted of 4 trays (replicates) arranged on a completely randomised design; replicates were then regrouped two by two for the chemical analysis.
The trays were placed in a growth chamber at 18°C and at light:dark regime of 10:14 hours. Light intensity was set at 200 μmol photons m–2 s–1, in order to approximate the average light intensity achievable indoor by natural light incoming through a window, with the aim of reproducing the likely environmental conditions for homemade production. Distilled water was periodically added to trays to keep the initial NaCl concentration of each treatment. For 0_S treatments, sprouts were harvested 5 days after the start of incubation (DAS), while wheatgrass was harvested 8 DAS; for all the other treatments, since salinity slowed germination and seedling growth, sprouts and wheatgrass were sampled when they reached the same growth stages as in the control (Table 1). Recovery-treatments (50_R and 100_R) were sampled only at wheatgrass stage, assuming same analysis values as in 50_S and 100_S for sprouts.
For sprouts, the whole plant material was collected, as it is usually in wheat sprouting. By analogy, the whole plant material was taken also for wheatgrass, although only the shoot is normally used to extract juice. Sampled material was stored at -20°C for analytical determinations.
2,2’-azinobis-(3-ethyl-benzothiazoline-6-sulfonic acid) (ABTS), acetic acid, acetonitrile (HPLC grade), ethyl acetate, were purchased from Carlo Erba (Milan, Italy). Sodium carbonate was provided by Panreac (Barcelona, Spain). Folin-Ciocalteau reagent, hydrochloric acid (37% w/v), methanol, sodium hydroxide, gallic acid (GA), caffeic acid (CA), syringic acid (SRA), P-coumaric acid (P-CA), trans-ferulic acid (trans-FA) and salicylic acid (SA) were purchased from Sigma Aldrich (St. Louis, MO, USA). All standards were prepared as stock solution at 1 mg mL–1 in methanol and stored at -20°C in darkness conditions. Ultra-pure water was obtained from a reverse osmosis system (Technolab, Milan, Italy).
Total polyphenols content
Total polyphenols content (TPC) determination was performed on grains, sprouts and wheatgrass of TMoM, TDiZ and TDuC, using the Folin-Ciocalteau reagent method (Singleton and Rossi, 1965). Grains (10 g) were milled using a laboratory miller and sieved through a 32-mesh screen. Five grams of milled (grains) or freeze-sampled (sprouts and wheatgrass) material for each replicate were treated with MeOH (10 mL) and homogenised with the T-25 Ultra-Turrax (IKA-LAB, Seneco S.r.l., Milano, MI, Italy), then sonicated with Sonis 4 for 1 hour in a cooled water bath. The extracts were centrifuged for 10 minutes at 15,000 g at 4°C and the supernatant was filtered through a Chromafill AO-22/25 polyamide filter. 0.8 mL of Folin-Ciocalteau reagent were added to aliquots of 0.2 mL of extracts; after 5 min, 0.8 mL of a 25% Na2CO3 solution was added and then deionised water up to 10 mL final volume. Solutions were maintained at room temperature under dark conditions for 60 min and the total polyphenols content was determined at 765 nm using a Perkin Elmer Lambda Bio20 spectrophotometer (Perkin-Elmer, Waltham, MA, USA). Gallic acid (GA) standard solutions were used to calibrate the method; the contents of total polyphenols in each extract was calculated and expressed as mg of GA equivalent (GAE) per 100 g (dry weight, DW) of sample (mg GAE 100 g–1 DW).
Radical scavenging activity
Radical scavenging activity was measured on TPC extracts, according to the method described by Re et al. (1999). ABTS was dissolved in water to a 7 mM concentration; the ABTS radical was obtained from reaction of ABTS stock solution with 2.45 mM potassium persulphate and allowing the mixture to stand in the dark at room temperature for 12-16 h before use. The ABTS radical solution was diluted with deionised water to reach an absorbance of 0.70}0.02 at 734 nm. Thirty μL of differently diluted extracts were added to 2.97 mL of diluted ABTS radical solution. The absorbance at 734 nm was evaluated by a Perkin Elmer Lambda Bio 20 spectrophotometer (Perkin-Elmer, Waltham, MA, USA). For each sample, the percentage of inhibition (I%) after 5 minutes of reaction was plotted as a function of concentration and the Trolox equivalent antioxidant capacity (TEAC) calculated as the ratio of the linear regression coefficient of the sample to that of the Trolox standard (Fluka, Buchs, Switzerland). Results were expressed as μmol of Trolox equivalents (TE) per g (DW) of sample (μmol TE g–1DW).
Extraction of the phenolic fractions (free and bound)
The extraction of free phenolic acids (free-PAs) was achieved on grains, sprouts and wheatgrass of TMoM, TDiZ and TDuC, following the procedure of Vaher et al. (2010) with some modifications. One gram of milled (grains) or freeze-sampled (sprouts and wheatgrass) material for each replicate was suspended in 5 mL of MeOH/water/acetic acid (70/29.5/0.5 v/v) and homogenised with the T-25 Ultra-Turrax (IKA-LAB, Seneco S.r.l., Milan, Italy) then sonicated with Sonis 4 for 40 minutes at room temperature. After centrifugation at 4,000 g for 10 min, the supernatant was removed and the extraction was repeated. The combined extracts were evaporated to dryness under a gentle flow of nitrogen and the residues were redissolved in 1 mL of MeOH. The final extract was filtered through 0.45 μm PTFE syringe filters (Phenomenex, Torrance, CA, USA) and stored at -20°C until HPLC analysis.
After extracting free-PAs, the solid residue was hydrolysed according to Kim et al. (2006) with minor modification. Briefly, the residue was digested with 10 mL 4M NaOH; the suspension was sonicated for 40 min and treated over night at room temperature. After alkaline hydrolysis the supernatant was adjusted to pH 2 with 6M HCl. The supernatant was extracted three times with ethyl acetate (20 mL each extraction); the mixture was vigorously vortexed for 1 min at room temperature, then centrifuged at 4,000 g for 10 min. The three fractions were pooled, evaporated to dryness under a rotary evaporator in vacuum, redissolved in 2 mL of MeOH and stored at -20°C until HPLC analysis.
All the extractions were performed under dim light to avoid sample degradation by photo-oxidation.
Identification and quantification of phenolic compounds
PAs were analysed using an HPLC system consisting of a Perkin Elmer series 200 LC equipped with a UV/VIS detector and controlled through the software TotalChrom Navigator (Version 6.3.1). The separation of the analytes was carried out with a Kinetex C18 column (250x4.6 mm ID, 5 μm; Phenomenex, Torrance, CA, USA). The mobile phases were water with 1% (v/v) acetic acid (A) and 100% acetonitrile (B) with a solvent flow rate of 1 mL min–1 for a total run time of 60 min. The gradient programme was as follows: from 95% to 84% solvent A in 20 min, from 84% to 30% solvent A in 20 min, from 30% to 95% solvent A in 10 min and then 10 min of post-run for reconditioning. Automatic injections of 20 μL of the standards and sample solutions were carried out; UV chromatograms were recorded at 280 nm (Zhou and Yu, 2004). Identification of PAs was based on their retention times, obtaining suitable chromatographic elution with no interferences or overlapping peaks. Quantitative analysis was based on peak area; the calibration curves, i.e. the peak area versus concentration, were linear in the range of the assessed concentrations (r2>0.992). The lines of regression, calculated in the range of 5-80 ppm, have been used for quantitative analysis of PAs in the extracts.
To test the effect of Triticum genotypes (TMoM, TDiZ and TDuC) on TPC, TEAC, free-PAs and bound-PAs in grains, oneway analysis of variance (ANOVA) was performed.
Data on TPC, TEAC, free-PAs and bound-PAs in sprouts and wheatgrass were subjected to two-way ANOVA, with genotypes (TMoM, TDiZ and TDuC) representing the main factor and sprouting conditions (0_S, 25_S, 50_S, 100_S, 50_R and 100_R) the secondary factor.
All the statistical analyses were performed with R software (R Development Core Team, 2013). If the ANOVA detected significant differences, means separation was obtained through the Tukey’s honestly significant difference (HSD) test. Prior to ANOVA, data were analysed to test the normality and homoschedasticity assumptions.
Results and discussion
The results about total polyphenols content in grains of TMoM, TDiZ and TDuC are reported in Figure 1A. TPC values ranged from 30.8 (TDuC) to 40.3 (TMoM) mg GAE 100 g–1 DW (Figure 1A); significant differences among genotypes were not detected. These data match with the range reported in literature (Shewry and Hey, 2015; Vaher et al., 2010), while differ from Benincasa et al. (2015), due to higher TPC in TDuC grains. Probably the particular environmental conditions of 2015 (i.e. temperature and rainfall regimes during grain filling) and some abiotic stress factors occurring during kernel development, significantly affected the biosynthesis and accumulation of phenolic compounds (Herbinger et al., 2002).
The effect of genotype was significant in terms of radical scavenging activity, with TMoM and TDuC showing the highest and lowest TEAC values (1.49 and 0.57 μmol TE g–1 DW, respectively; Figure 1B), confirming data by Yilmaz et al. (2015) although with higher and more variable values probably due to the different extraction’s assays.
In order to have a better characterisation of wheat genotypes, providing more detailed information when compared with spectrophotometric methods (such as total polyphenols by Folin- Ciocalteau reagent method), grains extracts were analysed to assess their free and bound phenolic acids fractions. Five phenolic compounds, CA, SRA, P-CA, trans-FA and SA were investigated.
In no-sprouted grains, the bound forms accounted for most of the total phenolic acids content (Table 2) as bound-PAs are crosslinked with the mainly structural components such as hydrolysable tannins, lignins, cellulose and proteins (Gawlik-Dzikiet al., 2012). Averaged over genotypes, trans-FA showed the highest concentration among total bound-PAs in kernels (Table 2), as previously observed (Benincasa et al., 2015).
Regarding the effect of genotype, the content of total free-PAs in grains of TMoM and TDiZ was more than double compared to that of durum wheat, principally due to higher contents of both PCA and trans-FA (Table 2). An analogous trend was observed for the bound-fraction and the content of total bound-PAs of TMoM and TDiZ which were, on average, over three- fold higher than TDuC (Table 2). Bound trans-FA content lowered by about 52 and 56% in TDuC with respect to TMoM and TDiZ, while P-CA was on average 97% lower than the other genotypes, in agreement with Benincasa et al. (2015).
Effect of sprouting and salt stress on polyphenols content and radical scavenging activity
Figures 2 and 3 report trends of TPC and TEAC as observed respectively in sprouts and wheatgrass of TMoM, TDiZ and TDuC germinated under normal and salt stress conditions.
Averaged over both genotypes and salt treatments, sprouting enhanced TPC from 6-fold (in sprouts) up to over 20-fold (in wheatgrass) and similar increases were also observed for TEAC values (Figure 4A and B), confirming their significant linear correlation (r2=0.98), as previously observed in some fruits and vegetables (Gorinstein et al., 2007; Kugler et al., 2007). The highest increase was recorded for TDuC, the genotype with the lowest TPC and TEAC values in grains (TPC: up to 7-fold in sprouts, up to 27-fold in wheatgrass; TEAC: up to 13-fold in sprouts, up to 38- fold in wheatgrass) (Figure 4A).
The observed increase in total phenolic content and antiradical activity confirmed evidence obtained in cereals (Benincasa et al., 2015; Donkor et al., 2012) as well as in several other species belonging to Leguminosae and Brassicaceae families (Cevallos- Casals and Cisneros-Zevallos, 2010; Duenas et al., 2009). Germination leads to significant changes in phenolic composition, mainly due to the activation of endogenous hydrolytic enzymes, which in turn affect the antiradical activity of sprouts (Duenas et al., 2009).
Modifications of sprouting conditions (i.e. salt-stress during germination) aimed at overproducing antioxidants inter alia phenolic compounds, usually involve induction of the natural mechanisms of plant resistance. In our study, the interaction Genotype x NaCl was significant in terms of total polyphenols contents in sprouts and wheatgrass of TMoM, TDiZ and TDuC (P<0.05; Figures 2A and 3A, respectively). A similar genotype-specific response to different magnitudes of salt stress was also observed by Danai-Tambhale et al. (2011) in 21-day old seedling of two indica rice cultivars, indicating the importance of genetic variability on salt tolerance (Yuan et al., 2010). In general, sprouts of TMoM did not shown differences on TPC values when exposed to different salinity conditions; moreover, at wheatgrass growth stage TPC values were significantly higher only at 100 mM NaCl solution (1082 mg GAE 100 g–1 DW; Figure 3A). This trend is in agreement with the studies of Guo et al. (2014) and Yuan et al. (2010) who found a significant increase in TPC in radish and broccoli sprouts, but only at the highest stressful condition. Conversely, the effect of salt stress significantly enhanced TPC of both TDiZ and TDuC genotypes already during the very early stages (Figure 2A) and the highest values were reached by 25_S treatments (295 and 274 mg GAE 100 g–1 DW, in TDiZ and TDuC, respectively). For these genotypes, at the highest salt concentration (100_S treatment) the TPC levels were similar to the unstressed control (Figure 2A). The prolonged exposure to salinity, as observed at wheatgrass stage, enhanced TPC values in TDuC up to 100_S treatment (Figure 3A); no differences between salt-treatments were observed in TDiZ (Figure 3A). Regardless of genotypes, recovery-treatments lead to a reduction on TPC, especially for those previously subjected to the highest stressful condition; however, it has to be pointed out that TPC values in 50_R and 100_R treatments were generally higher than in the unstressed control (Figure 3A).
The trend of the radical scavenging activity, as a function of salt stress levels, resulted often unclear and not always in agreement with TPC data, strictly depending on growth stage and genotypes; for this reason it does not seem meaningful and worth speculating on. Anyway, averaged over genotypes, we found a significant linear correlation between TPC and TEAC values at sprout stage (r2=0.81), which was reduced at wheatgrass stage (r2=0.72).
Effect of sprouting and salt stress on free- and boundphenolic acids
Both free and bound fractions content of the detected PAs in sprouts and wheatgrass of TMoM, TDiZ and TDuC are presented in Tables 3 and 4, respectively.
The total free-PAs content ranged from 10.5 to 27.2 μg g–1 DW in sprouts (Table 3) and from 18.4 to 85.8 μg g–1 DW in wheatgrass (Table 4). During germination the total free-PAs content increased significantly by over 6-fold at sprout stage (averaged over genotypes and salt treatments) (Figure 5A). Then, a further increase from sprout to wheatgrass stage was observed and the free-PAs content averagely increased by about 17-fold with respect to grains (Figure 5A), confirming results on a FW basis reported by Benincasa et al. (2015). Genotypes gave different response to germination process with the highest free-PAs increase observed in TDuC.
Total bound PAs ranged from 334.2 to 1898.8 μg g–1 DW in sprouts (Table 3) and from 208.0 to 7769.8 μg g–1 DW in wheatgrass (Table 4), contributing heavily to the total detected PAs (free- PAs plus bound-PAs). Also in this case, bound-PAs content tended to increase over germination although not at the same extent as the free- ones, from 0.5-fold to about 2.5-fold in sprouts and wheatgrass, respectively (averaged over genotypes and salt treatments) (Figure 5B).
The effect of genotype was particularly clear: in TMoM and TDiZ the content of total-bound PAs were lower at sprout stage with respect to grains (Figure 5B). An early reduction in the TPC of bound phenolic extracts was also observed by Hung et al. (2012) in waxy wheat germinated grains, and it was attributed to a first lost hydrolysis of polyphenolic compounds bound to cell walls (Yang et al., 2001). Moreover, during the first stages of germination the decline of bound phenolic concentration coincided with a rise of the free-fraction, indicating a shift between two forms (Xu et al., 2009). However, it is improbable to compare our results with the literature because of the different germination time (5-days vs 2-days old sprouts).
HPLC analyses showed that the qualitative compositions of the free- and bound- phenolic fractions of TMoM, TDiZ and TDuC during germination did not substantially change (Tables 2, 3 and 4). In sprouts, trans-FA contributed to 63% and 57% of the total free- PAs and bound-PAs, respectively (Table 3). The trans-FA was also the most represented PA in wheatgrass (65% of the total free-PAs and 64% of the total bound-PAs; Table 4) (Hung et al., 2011).
Both free- and bound-PAs fractions in sprouts extracts were significantly affected by Genotype x NaCl interaction (P<0.05; Table 3). In TMoM sprouts, salt-stress induced significant differences in terms of total free-PAs only at 25_S, in part confirming TPC data. However, salinity significantly enhanced total bound- PAs (highest value at 50_S; Table3). Interestingly, the increase in the bound-fraction was related to a significant induction of both PCA and trans-FA. Cell-wall-bound FA has been demonstrated to cope with drought tolerance in triticale seedlings, acting as regulator of cell growth by ester and decreasing cell-wall extensibility and sprout elongation at the cost of osmotic adjustment (Hura et al., 2009). Conversely, salt-stress raised up total free-PAs till 50_S in sprouts of TDuC, whilst significantly lowered total bound-PAs (Table 3). The negative variation of total bound-PAs mainly depended on the variation of both P-CA and trans-FA. In general, salt-stress treatment induced an increase in the concentration of both free- and bound-phenolic fractions in sprouts extracts of TDiZ, although the differences were not always significant (Table 3). The contribution to total bound-PAs increase is also attributable to SA, which showed the highest values in 50_S treatment (Table 3) and which could act as a signaling molecule inducing salt tolerance in plants (Jayakannan et al., 2015).
With respect to wheatgrass extracts, the content of total free- PAs in response to NaCl treatments showed similar trends to sprouts ones, although differences on specific PAs were observed (Table 4). Regarding the bound-fraction, total detected bound-PAs in TMoM showed an upward trend until 50_S treatment; further NaCl level significantly constrained total bound-PAs values (Table 4). The same trend was observed for P-CA and trans-FA. However, the recovery-treatment 100_R allowed increasing total bound-PAs with respect to 100_S, thanks to the contribution of trans-FA and SA in their bound forms (Table 4). Interesting, SA became detectable in its bound form also in 50_R and 100_R treatments (Table 4). We hypothesised that TMoM, given optimal growing conditions, is promptly adaptable to stressful levels during germination with SA being involved in this process. On the other hand, in TDuC P-CA, trans-FA and, consequently, total-PAs in their bound forms significantly decreased in response to prolonged saltstress treatments (Table 4). For TDiZ, despite an increase in total bound-PAs until 50_S, any significant difference was detected (Table 4). This could be attributable also to the reduction of bound SRA that became undetectable under severe stressful conditions, even if the optimal growing conditions were restored (see 50_R and 100_R treatments; Table 4). Interestingly, TDiZ represented the genotype characterised by higher SA content with respect to TMoM and TDuC (Table 4).
Sprouting significantly enhanced TPC and detected boundand free-PAs, regardless of genotype. Response to salt stress depended on genotype sensitivity, growth stage and salt concentration. In general, moderate salinity levels (not exceeding 50 mM NaCl solution) induced accumulation of phenolic acids in both sprouts and wheatgrass of Triticum genotypes with a pronounced variability. In particular, in TMoM the bound-PAs fraction of both sprout and wheatgrass extracts increased until the dose of 50 mM NaCl, probably thanks to the induction of P-CA and trans-FA. Conversely, in TDuC, NaCl determined an increase of PAs in their free form (as confirmed by TPC data) and a reduction of the bound ones. In TDiZ salinity generally enhanced the TPC, TEAC, free- PAs and bound-PAs content, although the responses were not always significant.
However, to better understand the species-specific response to salt, further researches on a wider number of both genotypes and growing environments, are needed.
These results indicate that salt stress could be efficiently modulated in order to improve the nutritional value of cereal sprouts and wheatgrass, through the accumulation of their bioactive compounds. This could provide the basis for further in vivo investigations to understand the impact of the enriched-quality sproutsbased products on human health.