Introduction

In the face of ongoing climate change, agriculture is encountering new challenges that affect the growth and development of cultivated plants1. One significant issue is autumn droughts, which particularly impact winter cereals such as triticale. These early developmental impairments limit the plant’s ability to form sufficient tillers before winter dormancy, which is a key determinant of final yield in winter triticale2.

In response to environmental stresses, plants activate various defense mechanisms, including increased synthesis of phenolic compounds3. These compounds, produced via the phenylpropanoid pathway, play a key role in protecting plants against biotic and abiotic stresses. The level of phenolic compounds depends, among other factors, on the availability of carbohydrates. Therefore, under optimal growth conditions or during recovery after stress, a reverse process, an increase in carbohydrate content due to inhibited phenolic compound synthesis, would be more beneficial for plant productivity4. In this way, carbohydrates may become available for plant growth and development processes, although such reports are currently lacking in the literature. To date, only Rohde et al.5 showed in Arabidopsis thaliana mutants, pal1 and pal2, that disruption of phenylalanine ammonia-lyase (PAL) function leads to significant changes in carbohydrate metabolism, including increased activity of sugar metabolism-related enzymes. An increase was observed in the activity of trehalose-6-phosphate synthase, triosephosphate isomerase, and phosphoglycerate kinase.

Phenylalanine ammonia-lyase (PAL) is a key enzyme that initiates the biosynthetic pathway of phenolic compounds. PAL transforms L-phenylalanine to trans-cinnamic acid that initiates the phenylpropanoid pathway6. Phenolic compounds and carbohydrates are major carbon sinks in plants, playing essential roles in development and stress responses. It should be emphasized, that phenolic compounds interact with other signaling molecules, including polyamines, which form cross-links with phenolic compounds to reinforce cell walls7, auxins, whose transport and activity can be modulated by phenolic derivatives8, and proline, which is recognized as stress indicator and osmoprotectant9, highlighting the integrative role of phenolics in coordinating structural integrity, growth, and carbon-nitrogen metabolic homeostasis.

In this study, 4-hydroxybenzoic acid hydrazide (HBH) was used as a PAL inhibitor. The chemical structure of HBH is similar to that of L-phenylalanine, the natural substrate of PAL, and it acts as a competitive inhibitor by binding to the enzyme’s active site, thereby blocking its catalytic function4. This compound remains poorly characterized in plant physiology, with only a single study reporting its application in the context of biotic stress10. Its limited use to date makes HBH an interesting candidate for investigating PAL-related responses under optimal plant growth conditions.

Winter triticale is an important feed component in the nutrition of farm animals, especially pigs and poultry11. Meanwhile, in the context of climate change, the discussion mainly focuses on human food security, while an equally important yet often overlooked issue is feed security, which is crucial for the stability of livestock production12,13. A decline in winter triticale yields induced by autumn drought may lead to feed shortages, increased feeding costs, and reduced profitability of animal husbandry. Hence, there is an urgent need for research aimed at securing feed cereal production and ensuring the stability of livestock production in the context of projected climate change. To date, no studies have explored PAL inhibition in cereals under optimal growth conditions, particularly in the context of its broader metabolic implications. This research therefore provides novel insights into the metabolic functions of PAL beyond its well-established role in plant biology. The findings may support future strategies to enhance the resilience and nutritional quality of forage cereals. Moreover, the study aligns with national and EU-level priorities concerning sustainable feed production and overall food system stability14. This study addresses the modulation of phenylpropanoid metabolism under optimal growth conditions influences both primary and secondary metabolism, with particular emphasis on nitrogen metabolism, which is involved in stress adaptation and biomass quality15.

Therefore, the aim of the study was to evaluate the effect of a phenylalanine ammonia-lyase (PAL) inhibitor (HBH) on the metabolism of winter triticale seedlings. Although the potential off-target effects of HBH on other metabolic pathways cannot be completely excluded, this compound has been only marginally studied in the context of plant metabolism. Therefore, our research is intended to provide new insights into its biological activity and to evaluate both the beneficial and adverse consequences of PAL inhibition. We hypothesize that limiting the synthesis of phenolic compounds through PAL inhibition may lead to increased sugar accumulation, as well as changes in the metabolism of nitrogen-containing bioorganic compounds, polyamines and proline. In addition, to gain a deeper understanding of the regulatory mechanisms and metabolic interactions induced by the PAL inhibitor, the content of indole-3-acetic acid (IAA), a phytohormone that plays a key role in regulating plant growth and development, was determined.

Results and discussion

HBH uptake and water relations

To date, PAL inhibitors (e.g. 2-aminoindan-2-phosphonic acid – AIP, α-aminooxyacetic acid – AOA, α-aminooxy-β-phenylpropionic acid – AOPP, O-benzylhydroxylamine - OBHA) were applied mainly in the context of biotic stress16,17,18,19,20,21, with relatively little attention given to their potential in enhancing crop productivity or mitigating yield loss. Our study addresses this gap by evaluating the metabolic consequences of PAL inhibition under optimal growth conditions.

The application of the PAL inhibitor, 4-hydroxybenzoic acid hydrazide (HBH, 10–3 M)10, had a differential effect on water status and HBH accumulation in winter triticale seedlings (Table 1). There was no statistically significant difference in leaf water potential (ΨW) between the control and HBH-treated seedlings. The average ΨW values were − 0.75 MPa in the control and − 0.76 MPa in the HBH treatment. Conversely, a significant increase in leaf osmotic potential was observed in seedlings exposed to HBH, as indicated by a decrease in ΨO values from − 1.28 MPa in the control to -2.41 MPa in HBH-treated plants. Additionally, the content of HBH was confirmed in the leaves of treated seedlings [4.82 µg/g (d.w.)], whereas no HBH was detected in the control group (Table 1). The detection of HBH confirms its uptake through the roots and subsequent translocation within the plant. The pronounced decrease in osmotic potential (ΨO) suggests that HBH may have influenced the accumulation of osmotically active compounds, such as soluble sugars22.

Table 1 Leaf water potential (ΨW), osmotic potential (ΨO), and content of the PAL inhibitor 4-hydroxybenzoic acid hydrazide (HBH) in winter triticale seedlings under control and HBH treatment.
Full size table

Phenylpropanoid pathway metabolites

Inhibition of phenylalanine ammonia-lyase led to changes in the levels of phenolic compounds (Table 2). PAL inhibition was primarily reflected by a significant decrease in cinnamic acid content, the direct product of PAL enzymatic activity23. A significant decrease was observed in the levels of cinnamic acid [from 80.15 to 39.47 ng/mg (d.w.)], benzoic acid [from 62.22 to 34.18 ng/mg (d.w.)], p-coumaric acid [from 0.66 to 0.40 ng/mg (d.w.)], chlorogenic acid [from 78.52 to 54.39 ng/mg (d.w.)], rosmarinic acid [from 145.3 to 115.0 ng/mg (d.w.)], and ferulic acid [from 32.41 to 18.84 ng/mg (d.w.)]. A statistically significant increase was observed for salicylic acid [from 0.93 to 1.16 ng/mg (d.w.)], p-hydroxybenzoic acid [from 2.31 to 29.6 ng/mg (d.w.)], and syringic acid [from 4.99 to 7.88 ng/mg (d.w.)]. These increases may reflect the activation of other conversions of phenolic compounds within the phenylpropanoid pathway triggered by PAL inhibition. Although PAL inhibition often results in a decreased content of certain phenolic compounds, other studies have demonstrated that the levels of others may increase24,25. Liu et al.26 showed that the increase in the content of some phenolic compounds after AIP inhibition of PAL could result from upregulation of phenylpropanoid pathway genes (such as 4CL5, C4H1) to counteract the decrease in some phenylpropanoid metabolites.

Among cell wall-bound phenolics, the most noticeable decrease was observed in ferulic acid, which dropped from 2661.7 in the control to 1388.5 ng/mg (d.w.) under HBH treatment. It should be emphasized that ferulic acid is one of the major phenolic components of the plant cell wall, contributing to its rigidity and structural integrity27. Similarly, cinnamic acid content significantly declined from 969.1 to 605.9 ng/mg (d.w.), and p-coumaric acid from 1121.9 to 1016.7 ng/mg (d.w.). Other decreases were also observed for benzoic acid [from 15.91 to 5.15 ng/mg (d.w.)], vanillic acid [from 13.56 to 11.11 ng/mg (d.w.)], and chlorogenic acid [from 7.18 to 4.44 ng/mg (d.w.)]. These changes reflect a general suppression of phenolic acid biosynthesis under PAL inhibition. However, statistically significant accumulation of salicylic acid, p-hydroxybenzoic acid, syringic acid and caffeic acid in the cell wall was also observed. Several other phenolic acids, e.g. gentisic acid, 3,4-dihydroxybenzoic acid, gallic acid, rosmarinic acid, sinapic acid, and homovanillic acid, were not detected (Table 2). The observed reduction in cell wall-bound phenolic compounds may reflect the inhibition of their incorporation into cell wall structures16,28. Similar effects were linked to reduced lignin biosynthesis in PAL-deficient or PAL-inhibited plants, as phenolic intermediates are essential for lignin polymerization and overall cell wall architecture5,29.

Table 2 Content of free and cell wall-bound phenolics [ng/mg (d.w.)] in winter triticale seedlings under control and PAL inhibition.
Full size table

Carbon-nitrogen balance

The HBH treatment significantly changed the carbohydrate content (Table 3). An increase in glucose [from 30.2 to 42.0 µg/mg (d.w.)] and trehalose [from 0.013 to 0.024 µg/mg (d.w.)] contents was observed under HBH treatment, while a significant decrease for fructose [from 19.6 to 15.0 µg/mg (d.w.)], sucrose [from 48.9 to 36.5 µg/mg (d.w.)], raffinose [from 0.155 to 0.087 µg/mg (d.w.)], 1-kestose [from 4.99 to 2.66 µg/mg (d.w.)], and nystose [from 2.23 to 0.44 µg/mg (d.w.)]. These findings clearly indicate that the HBH induces an indirect effect on carbohydrate metabolism. Rohde et al.5 showed that the disruption of PAL led to transcriptomic adaptation of components of the carbohydrate metabolism. In turn, Cass et al.30 demonstrated that attenuation of the phenylpropanoid pathway increases carbohydrate availability.

Table 3 Content of soluble carbohydrates [µg/mg (d.w.)] in winter triticale seedlings under control and PAL inhibition.
Full size table

In addition, it was also important to determine whether HBH influences nitrogen metabolism. Hence, the contents of proline, a key osmoprotectant and signaling molecules31, and polyamines, which play regulatory roles in plant growth32 and stress responses33, were also analyzed. The HBH-treated seedlings exhibited significantly higher level of proline (Fig. 1). Proline accumulation suggests that PAL inhibition may affect certain stress-related metabolic pathways, possibly due to altered carbon allocation or changes in osmotic adjustment mechanisms.

Fig. 1
figure 1

Proline content in winter triticale seedlings under control and PAL inhibition. Mean values ± SE (n = 3). Asterisks mark differences significant at p < 0.05 vs. control, Student’s t-test.

Full size image

The HBH significantly affected the profile of free polyamines in triticale seedlings (Table 4). A statistically significant decrease was observed in agmatine content, which dropped from 0.019 in the control to 0.004 ng/mg (d.w.) in HBH-treated plants. In contrast, cadaverine and putrescine contents increased significantly, rising from 0.37 to 0.63 ng/mg (d.w.) and from 19.8 to 22.6 ng/mg (d.w.), respectively. For spermidine and spermine, the changes were not statistically significant. Polyamines occur in free or conjugated forms34. They are usually combined with cinnamic acid, forming amide linkages resulting in hydroxycinnamic acid amides such as caffeic acid, p-coumaric acid, and ferulic acid, which are all the major components of phenolic groups present in different part of plants35. Thus, changes in polyamine metabolism under PAL inhibition may result from altered availability of phenolic compounds, which serve as components of conjugated forms.

The PAL inhibitor also influenced the levels of cell wall-bound polyamines (Table 4). A significant decrease in agmatine was observed, from 0.0073 in the control to 0.0013 ng/mg (d.w.) in HBH-treated seedlings. Putrescine content increased significantly, rising from 0.24 to 0.54 ng/mg (d.w.), indicating enhanced accumulation of this diamine in the cell wall matrix. Cadaverine was not detected in either treatment. Although spermidine showed a decreasing trend [from 1.12 to 0.64 ng/mg (d.w.)], the change was not statistically significant. By contrast, spermine content was significantly reduced, dropping from 0.81 to 0.26 ng/mg (d.w.) in HBH-treated plants. These results indicate that PAL inhibition indirectly modulates nitrogen metabolism, promoting the accumulation in the cell wall of certain diamines like cadaverine and putrescine while decreasing agmatine. Cell wall-localized polyamines are essential for maintaining wall thickness and stability, as they strengthen the bonds between structural elements of the wall (polycarbohydrates, lignin, phenolic compounds)36.

Table 4 Content of free and cell wall-bound polyamines [ng/mg (d.w.)] in winter triticale seedlings under control and PAL inhibition.
Full size table

Auxin homeostasis and seedling growth parameters

A significant decrease in indole-3-acetic acid (IAA) content observed in HBH-treated seedlings (Fig. 2) suggests that PAL inhibition may interfere with auxin homeostasis. This reduction could result from altered tryptophan-dependent IAA biosynthesis, or from disrupted signaling due to changes in phenolic compound levels, which modulate auxin transport and stability37. Therefore, to clearly determine which mechanism underlies the observed decrease in IAA content, further studies are required, including tryptophan quantification and gene expression analyses of auxin biosynthesis and transport pathways.

Fig. 2
figure 2

Indole-3-acetic acid (IAA) content in winter triticale seedlings under control and PAL inhibition. Mean values ± SE (n = 3). Asterisks mark differences significant at p < 0.05 vs. control, Student’s t-test.

Full size image

The application of 4-hydroxybenzoic acid hydrazide significantly affected the morphological parameters of winter triticale seedlings after 7 days of treatment (Table 5). HBH caused a decrease in both shoot and root length compared to the control, with reductions of approximately 10.6% and 16.9%, respectively. In contrast, the dry mass of both shoots and roots increased significantly in response to HBH application, by about 8.9% and 13.6%, respectively. The increase in dry mass observed in HBH-treated seedlings is likely linked to enhanced carbohydrate utilization for biomass production rather than for phenolic synthesis. Part of the carbohydrates was also allocated to root development, indicating carbon partitioning between above- and below-ground organs. Li et al.38 indicated that environmental factors, such as nutrient availability and drought, can significantly alter carbon allocation patterns between above- and below-ground organs. While biomass measurements provide useful information on overall growth, they do not fully reflect the dynamics of assimilated carbon, which can be allocated to storage, tissue growth, or metabolic activity. Our observations of increased root and shoot biomass in HBH-treated seedlings suggest that carbon partitioning may be influenced by changes in the secondary metabolism of phenolic compounds.

Table 5 Seedling traits (length/dry mass of roots and shoots) under control and PAL inhibition.
Full size table

Metabolite profiling – heat maps

To summarize the above results, the application of HBH significantly altered phenolic content, accompanied by changes in carbohydrate levels, suggesting a rearrangement of carbon metabolism. Additionally, alterations in proline and polyamine levels indicate modifications in nitrogen metabolism in response to HBH treatment. The observed decline in IAA further supports interference with hormonal regulation. Taken together, these results highlight the multifaceted role of PAL activity in carbon–nitrogen dynamics through direct effects on phenolic metabolism and indirect impacts on carbohydrate content, as well as polyamine and proline metabolism. The distribution of free and cell wall-bound phenolic compounds, free and cell wall-bound polyamines, soluble carbohydrates, proline, and indole-3-acetic acid (IAA) across treatments was visualized in the form of heat maps (Fig. 3), with values normalized within each compound group.

Fig. 3
figure 3

Heat map illustrating the relative levels of free and cell wall-bound phenolic compounds, free and cell wall-bound polyamines, soluble carbohydrates, proline and indole-3-acetic acid (IAA) in the analyzed samples (C - Control, I - Inhibitor). Color intensity represents compound abundance (from the lowest - yellow to the highest - green values), with values normalized independently within each compound group to highlight intra-group variation. Proline and IAA, as single representatives of amino acids and auxins, were normalized separately.

Full size image

Conclusions

The present study demonstrated that 4-hydroxybenzoic acid hydrazide (HBH), a PAL inhibitor, when applied via the root system, can modify the biochemical profile of triticale seedlings, indicating its potential systemic activity. These findings demonstrate that PAL inhibition has broader implications indirectly affecting carbohydrate and nitrogen-related pathways. At the level of primary metabolism, altered sugar and polyamine profiles, along with increased proline content, suggest a disruption of the carbon–nitrogen balance, indicating that PAL activity contributes to maintaining metabolic homeostasis under non-stress conditions. Regarding secondary metabolism, the reduction in phenolic compounds confirms PAL’s crucial role in phenylpropanoid biosynthesis and highlights its regulatory function in coordinating secondary metabolite production.

However, given the complexity of soil-HBH interactions, including possible microbial degradation or chemical modification of HBH in the soil, additional studies are required to verify its stability, uptake efficiency, and mode of transport in planta. It is also important to assess whether HBH treatment affects morphological traits such as leaf area, and ultimately yield components. Hence, it is essential to assess both physiological and agronomic outcomes under controlled and field conditions to evaluate the practical applicability of HBH in modifying phenylpropanoid metabolism in cereals.

Further studies should also aim to clarify the signaling mechanisms that connect PAL activity with carbon and nitrogen metabolic pathways, particularly under varying environmental conditions, and should include detailed analyses of PAL, C4H, and 4CL gene expression. Transcriptomic and metabolomic analyses could help identify key regulatory factors affected by PAL inhibition. Moreover, investigating the long-term physiological and agronomic consequences of PAL attenuation under field conditions may provide valuable insights into its potential application in crop improvement strategies focused on optimizing biomass composition, enhancing stress resilience, and improving yield stability.

Materials and methods

Plant material and plant growth conditions

The winter triticale cultivar used in this study was ‘Moderato’, distinguished by a pronounced utilization of carbohydrates for the synthesis of phenolic compounds, which are subsequently incorporated into the cell wall39. This metabolic trait makes ‘Moderato’ an appropriate model for studying the effects of PAL inhibition on seedling metabolism.

Plants vegetation was conducted in air-conditioned greenhouse in plastic containers (0.7 L). The containers were filled with a 3 cm layer of sterile vermiculite (fine-grade vermiculite, 1–3 mm), in which triticale seeds were sown (10 seeds per container). The 3 cm layer of vermiculite was intentionally used to limit the substrate volume and ensure effective exposure of the root system to the PAL inhibitor. Plants growth conditions: air temperature 23/18 °C (± 2 °C) day/night; relative air humidity about 40%. The plants were illuminated by high pressure sodium lamps (400 W; Philips SON-T AGRO, Brussels, Belgium), PPFD (photosynthetic photon flux density) at the level of leaves reached about 200 µmol m- 2 s- 1 (QSPAR Quantum Sensor, Hansatech Instruments LTD, Kings Lynn, England) and 10 h/14 h (day/night) photoperiod was set.

Treatments

The inhibitor was applied daily in the morning for 7 days, starting on the sixth day of seedling growth (autotrophic phase of growth). The treatment involved direct application into the vermiculite, with a daily volume of 50 ml. The concentration of the inhibitor, 4-hydroxybenzoic hydrazide (HBH), was 10− 3 M. The HBH concentration (10− 3 M) was selected based on a study conducted on wheat under biotic stress conditions10. To ensure its suitability for triticale seedlings and to exclude any phytotoxic effects, preliminary tests were carried out prior to the main experiment. These tests confirmed the effectiveness and safety of the selected dose under the applied conditions. Control plants were treated with 50 ml of distilled water.

Sample preparation

Each biological replicate was based on a bulk sample of seedling leaves randomly collected from multiple containers within each treatment. The samples were immediately frozen in liquid nitrogen, lyophilized (Freeze Dry System/Freezone® 4.5, LABCONCO Kansas City, Missouri, USA), and ground to a fine powder (MM400, Retsch, Haan, Germany). In total, three bulk samples were prepared. One subsample (5–50 mg) was taken from each bulk sample for individual biochemical analyses. Thus, the number of biological replicates for each biochemical measurement was three (n = 3). Technical replicates were randomly performed during the analytical procedures to ensure data accuracy and reproducibility because of the limited amount of plant material available.

Measurements

Water potential (ΨW)

The measurements of water potential were performed with a psychrometer HR 33T (WESCOR, Inc., Logan, Utah, USA) equipped with leaf chambers of C-52 (WESCOR) type and digital meter Metex M − 3640D. The measurements were carried out on fresh leaf discs (diameter 5 mm) cut out from the middle part of the leaf. Samples were placed in chambers for 45 min in order to saturate the leaf chamber with water vapor. The measurement was done with the dew point method.

Osmotic potential (ΨO)

The measurements of osmotic potential were carried out with a psychrometer HR 33T (WESCOR, Inc., Logan, Utah, USA) equipped with leaf chambers of C-52 (WESCOR) type and digital meter Metex M − 3640D. The measurements were performed on blotting paper discs (diameter 5 mm) saturated with cell sap squeezed from leaves with a medical syringe. Samples were kept in chambers for 30 min in order to saturate the leaf chamber with water vapor. The measurement was done with the dew point method.

The level of 4-hydroxybenzoic acid hydrazide (HBH)

Approximately 50 mg of the plant material underwent double extraction with 1 ml of extraction buffer (methanol/water/concentrated ammonia; MeOH/H2O/NH4; 20/70/10 v/v/v) while shaking at 30 Hz for 10 min. The samples were then centrifuged for 5 min at 22,000 × g and 10 °C using a Universal 32R centrifuge (Hettich, Tuttlingen, Germany). The supernatant was collected, combined, and evaporated under a nitrogen (N2) stream using a TurboVap LV (Capillary, MA, USA). The residue was re-suspended in 1.2 ml of 0.1 M NH4, centrifuged, and the supernatant was evaporated to dryness in a clean test tube under N2. The final residue was dissolved in 150 µl of sample buffer, which consisted of 65% acetonitrile (ACN) in 100 mM ammonium acetate (NH4CH3COO), v/v. The samples were centrifuged and analyzed using an Ultra-High-Performance Liquid Chromatography (UHPLC) system (Agilent Infinity 1260, Agilent, Germany) coupled with a triple quadrupole mass spectrometer (Agilent 6410, Agilent, USA) utilizing electrospray ionization (ESI). Chromatographic separation was performed on an ACE HILIC-N analytical column (5 μm, 2.1 mm × 100 mm; ACE, Aberdeen, Scotland) using a linear gradient of H2O (A) versus ACN (B), both containing 10% and 100 mM NH4CH3COOH at pH 9, along with 5 µM medronic acid. The gradient was set to transition from 100% to 60% B in 1.5 min, followed by a 0.5-minute hold at 60% B, and then returned to 100% B in 0.2 min, with a flow rate of 0.5 ml/min at 40 °C. The optimal conditions for analysis included a capillary voltage of 4 kV, gas temperature of 350 °C, drying gas flow of 12 L/min, and nebulizer pressure of 35 psi. Samples were analyzed using dynamic multiple reaction monitoring (MRM), targeting the precursor ion ([M+H]+) at 153.1 m/z, with a fragmentor voltage of 84 V, and product ions at 95.1 m/z (collision energy, CE, 11 V) and 81.1 m/z (CE, 35 V). External calibration was performed using a pure HBH standard. All reagents were purchased from Sigma-Aldrich (Poznan, Poland).

Phenolics

About 16 phenolics compounds (free and released from the cell wall by hydrolysis) were determined using the UHPLC method in its modified form in a triple quadrupole mass spectrometer (TQMS) (triple quadrupole, QQQ)39.

About 25 mg of lyophilized and fine ground sample was weighted and extracted in 1 ml of 10% formic acid in methanol for 1 h at 250 rpm on rotary shaker. Then samples were centrifuged (15 min, 2200 × g, 10 °C) and supernatant collected. Extraction was repeated, methanolic fractions pooled and aspirated to dryness under nitrogen stream (TurboVap LV, Capiler Ltd., MA, USA) – the fraction of soluble phenolics. The pellet was hydrolyzed with 3 M NaOH then formed suspension was neutralized with concentrated hydrochloric acid and released phenolics extracted and evaporated as above - the fraction of cell wall bounded phenolics. Both phenolic fractions were reconstituted in 2 ml of sample buffer, i.e. acidified methanolic water (4.5% methanol and 4.5% formic acid) and cleaned-up on Discovery DPA-6 S SPE cartridges (3 ml, 250 mg, Supelco). SPE cartridges were activated with 3 ml of methanol followed by 3 ml of sample buffer. After that samples were applied, and slowly aspirated. Cartridge were flush with 2 ml of sample buffer and dried under vacuum for 3 min. Phenolics were washed out with 2 ml of 0.35 M ammonia in 60% methanol in water. Then samples were evaporated to dryness under nitrogen, reconstituted in 250 µl of methanol and after filtration (0.22 μm nylon membrane) injected on to HPLC column. Agilent Infinity 1260 system equipped with binary pump, autosampler and fluorescence detector (FLD) was used. Phenolics were separated on Zorbax Eclipse Plus Phenyl-Hexyl 3.5 μm 3.0 mm×100 mm column under linear gradient of 2% formic acid aqueous solution versus methanol. Beginning from 100% and ending at 35% formic acid solution in 14 min, at flow rate of 0.8 ml/min. Two sets of excitation (Ex) and emission (Em) wave length were used. Excitation and emission maxima were chosen based on absorption and emission spectra of pure phenolic acids standards. Sixteen compounds were monitored, arranged according to order of elution are: gallic acid, 3,4-dihydroxobenzoic acid, p-hydroxobenzoic acid, gentisic acid, caffeic acid, vanillic acid, homovanillic acid, chlorogenic acid, syringic acid, p-coumaric acid, ferulic acid, benzoic acid, sinapic acid, salicylic acid, rosmarinic acid, trans-cinnamic acid.

Carbohydrates

Sugars were analyzed according method reported by Hura et al.39. About 10 mg of lyophilized and homogenized samples were extracted in 1 ml of ultra-pure water (Option R, Elga, UK) by shaking for 15 min at 30 Hz ( MM 400, Retsch, Germany). Then samples were centrifuged for 5 min at 21,000 × g (Universal 32R, Hettich, Germany). Supernatant was collected, diluted with acetonitrile 1:1 (v/v) filtered (0.22 μm nylon membrane, Costar Spin-X, Corning, USA) and analyzed on HPLC for soluble sugar content.

High performance liquid chromatography analyses of free sugars were performed using an Agilent 1200 chromatograph equipped with degasser, binary pump, automated liquid sampler and thermostated column compartment (Agilent, Germany) and ESA Coulochem II electrochemical detector with 5040 Analytical Cell (ESA, USA) with analogue-to-digital converter. Separation of soluble sugars (glucose, fructose, sucrose, trehalose, raffinose, 1-kestose, nystose) was achieved on RCX-10; 7 μm; 250 × 4.1 mm column (Hamilton, USA) at flow rate of 1.5 ml min- 1 in gradient mode. Mobile phase A: 75 mM aqueous NaOH solution, B: 500 mM sodium acetate in 75 mM aqueous NaOH solution, 0% A (0–3.5 min), 0–35% B (3.5–19 min), then 0% B (19–20 min). Injection volume was 0.01 ml, column temperature 40 °C; integration time 20 min. Pulse amperometric detection was employed (analytical potential 200 mV; oxidizing potential 700 mV, reducing potential – 900 mV with reference to a palladium electrode) on a gold electrode.

Proline

Proline content was measured spectrophotometrically according Gadzinowska et al.40. About 5 mg of lyophilized and homogenized samples were extracted in 0.5 ml of 3% 5-sulfosalicylic acid (Sigma) for 5 min. After that samples were centrifuged at 21,000 × g for 15 min. The aliquot of 200 µl of clear supernatant was transferred to polypropylene screw cap vials and 200 µl concentrated formic acid (Sigma) and 400 µl 3% ninhydrin reagent in 2-methoxyethanol (Sigma) were added. Samples were heated for 0.5 h at 95 °C in water-bath, then samples were transferred to 96-well plates. The absorbance was measured at 514 nm on micro-plate reader (BioTek).

Polyamines

Free and cell wall-bound polyamines (agmatine, cadaverine, putrescine, spermidine, spermine) extraction and dansylation were performed according to a method described by Hura et al.33. The HPLC system used was Agilent Infinity 1260 system equipped with a binary pump, an autosampler and a fluorescence detector (FLD). Separation was achieved on Poroshell 120 EC-C18 3.0 × 50 mm 2.7 μm analytical column (Agilent Technologies), under linear gradient of water with 1% formic acid (A) and methanol (B), from 51% to 80% B in 8 min. Integration time was 10 min. Fluorescence detection was conducted at 350 nm for excitation and 510 nm for emission.

Indole-3-acetic acid (IAA)

Indole-3-acetic acid was assessed according to the procedure described by Hura et al.41. Freeze-dried and pulverized samples were extracted in 1 ml methanol/water/formic acid, (CH3OH/H2O/HCOOH, 15/4/1 v/v/v) after the addition of a stable isotope-labeled internal standard solution (ISTD). Evaporated pooled supernatant was resuspended and cleaned up on mixed-mode SPE cartridges (BondElutPlexa PCX, Agilent, Santa Clara, CA, USA). Ultrahigh performance liquid chromatography (UHPLC) using an Agilent Infinity 1260 system coupled with 6410 Triple Quad LC/MS with an electrospray interface (ESI) (Agilent Technologies, USA) was used. The separation was achieved on an AscentisExpres RP-Amide analytical column (2.7 μm, 2.1 mm × 150 mm, Supelco, Bellefonte, PA, USA) at a linear gradient of H2O vs. acetonitrile with 0.01% of HCOOH. Further technical details are given by Dziurka et al.42. Quantification was carried out using the linear range of a standard curve constructed with known amounts of IAA taking account ISTD recovery. The standard was supplied by OlChemIm, (Olomouc, Czech Republic) wears other chemicals were from Sigma-Aldrich (Poznań, Poland).

Seedling growth parameters

Seedlings were dissected into roots and shoots (stems + leaves). Seedling length was measured using a ruler. Dry mass was determined by drying roots and shoots in an oven at 70 °C until constant weight, followed by weighing with an analytical balance.

Statistical analyses

The statistical software used was STATISTICA 13.0 (Stat-Soft, Inc., Tulsa, OK, USA). The data were represented as means ± standard error (SE). Statistical differences between the experimental series were evaluated using Student’s t-test. Statistical significance denoted by asterisks was attributed to p-values ≤ 0.05.