Introduction

Critical low temperatures causing damage to flowers not only differ with the phonological stage of tree flowers, but also vary among species, cultivars, orchards, and even within the trees (Westwood 1993). Spring frost harm to blossoms has been a severe problem in pistachio-producing areas of Iran (Pakkish et al. 2011). The most recent frost events causing damage to pistachio flowers in these areas occurred in spring 2016 and 2018. Spring frost damage was more widespread in 2018 and had a severe adverse economic impact on pistachio producers. So, avoiding or reducing spring frost damage caused by different cultural practices is an important issue.

Many factors have been reported to play a role in the tolerance of tree flowers to spring frost. The most important factors are genotype (Westwood 1993), ice formation, flower sugar content (Ercisli 2003), and the nutritive status of the flowers (Rodrigo 2000). In addition, a number of strategies are being utilized to reduce the impact of low-temperature stress on crop plants. One of the best strategies for reducing temperature stress in crop plants is proper plant nutrition. The application of plant nutrients makes a substantial contribution to the improvement of freezing tolerance (Waraich et al. 2012). The current study is the first evidence that investigates the effects of potassium and zinc on the tolerance of pistachio flowers to spring frost damage.

Plants show different responses to decreased formation and extension of ice, such as increased soluble sugars and compatible solutes (Ershadi et al. 2016), modified cell wall and membrane lipid composition (Steponkus 1984), altered gene expression and protein synthesis (Salzman et al. 1996), and enhanced capacity of phenolic compounds and antioxidants (Valle 2002). Accumulated soluble carbohydrates in plants can be considered as cytoprotective compounds, avoiding or slowing the formation of ice crystals and can ultimately increase cold hardiness (Karimi and Ershadi 2015). By proline accumulation in trees when exposed to low temperatures, osmotic adjustment is induced, dehydrating cell turgor is sustained, and plant tolerance to dehydrative stresses is improved (Ghasemi Soloklui et al. 2012).

One of the injurious aspects of freezing stress is oxidative stress (Pakkish and Tabatabaienia 2016). The formation of reactive oxygen species (ROS) is facilitated under oxidative stress, which harms cell structures as a result of biomembrane lipid oxidation (Foyer and Noctor 2003). There are some effective intracellular mechanisms in plants to reduce the oxidative injuries from freezing stress containing both activations of enzymatic and non-enzymatic antioxidant systems (Mittler 2002). The main ROS scavenging enzymes that have been reported to be dramatically activated in cold-stressed plants are guaiacol peroxidase (GPX), catalase (CAT) and ascorbate peroxidase (APX) (Apel and Hirt 2004).

Mineral nutrition of the plants plays a crucial part in increasing tolerance to environmental stresses (Marschner 1995). In environmental stress conditions, potassium (K) among the mineral nutrients makes a significant contribution to crop survival. Under stress conditions, K is an essential element for many physiological processes, such as translocation of photosynthates into sink organs, maintenance of enzyme activation and turgidity, and photosynthesis (Marschner 1995; Mengel and Kirkby 2001). Under low potassium conditions, oxidative injuries induced by frost can be intensified, which diminish the plant growth and yield. High amounts of potassium supply can protect against frost oxidative damage (Waraich et al. 2012). In addition, zinc (Zn) plays a critical role in plant metabolism through affecting the practices of carbonic anhydrase and hydrogenase, ribosomal fraction stabilization, and cytochrome synthesis (Tisdale et al. 1984). Activated plant enzymes by Zn are concerned with cellular membrane integrity maintenance, carbohydrate metabolism, pollen formation, and protein as well as auxin regulation syntheses (Marschner 1995). The environmental stress tolerance in plants is Zn dependent, requiring maintenance and regulation of gene expression (Cakmak 2000). Under low Zn conditions, citrus trees were shown to be more vulnerable in cold temperatures (Cakmak et al. 1995).

To the best of our knowledge, there is no information regarding potassium and zinc fertilizers on freezing tolerance of pistachio flowers. The objective of the present research was to evaluate the effects of K2SO4, ZnSO4, and their combined use on antioxidant activities, soluble carbohydrates, proline, phenolic compounds, total soluble proteins, MDA, H2O2 content, and spring frost tolerance of ‘Kaleh Ghouchi’ pistachio flowers.

Materials and methods

Orchard site and treatments

Field studies were conducted in a commercial orchard located in Jafarieh, Ghom province (latitude. 34° 40′; longitude. 51° 0′ E; altitude. 936 m above sea level), Iran. Forty-five trees were selected (15-year-old trees) for this experiment. This region has long and hot summers, but the spring temperatures may plunge to − 4 °C or even lower (Norozi et al. 2019), causing severe damage to pistachio flowers. The pistachio orchard, cv. ‘Kaleh Ghouchi’, was planted at 7 × 4 m in clay loam soil with similar cultural practices such as fertilization or irrigation received by all trees. The experiment was conducted as a completely randomized design (CRD) with nine treatments and five replications (trees) per treatment. Treatments included applying three levels of K2SO4 (0, 1, and 2%) and three levels of ZnSO4 (0, 0.5, and 1%). The nutrient treatments were applied with 0.2% Tween 20 on each tree in two consecutive seasons 2017 and 2018. The nutrient solutions were sprayed to a runoff on each tree in two times (bud swell stage and green tip stage) in March 2017 and 2018. Distilled water with 0.2% Tween 20 was sprayed as the control. In a preliminary study, the pistachio trees were sprayed with different concentrations of K2SO4 and ZnSO4. It was found that treatment with K2SO4 at 1 and 2% and ZnSO4 at 0.5 and 1% significantly alleviated freezing injury in flowers, and these concentrations, therefore, were used in the following experiments.

Controlled freezing procedure

The excised 1-year-old twigs were collected during the anthesis phase in March. The shoot segments with three or four flowers were placed in polyethylene bags kept moist with ice. Freezing was accomplished in a programmable freezing chamber (Kimia Rahavard, Tehran, Iran) based on a stepwise temperature reduction program. The shoot segments were tested by exposing to three different low test temperatures (0, − 2 and, − 4 °C). The starting temperature was 5 °C with the freezing rate of 2 °C h−1, and before removing from the freezing chamber, samples were held for 60 min at each test temperature. For slow thawing, the samples were placed on ice for 60 min and then kept at 6 °C in a refrigerator. The freezing damage was visually rated by discoloring and browning of flowers and pistils. Flowers were individually cross-sectioned by slicing horizontally through the middle part of the flower with a razor blade. These slices were examined under a stereomicroscope (Leica MS5; Wetzlar, Germany) (Karimi and Ershadi 2015). The samples were immediately frozen in liquid nitrogen for biochemical and antioxidant analyses.

Freezing tolerance was measured based on LT50 (the lethal temperature at which 50% of the flowers are dead) by fitting response curves with the following logistic sigmoid function (Fiorino and Mancuso 2000):

$$R = \frac{a}{{1 + e^{{b \left( {x - c} \right)}} }} + d,$$

where R is the flower browning percentage based on the LT50 estimation method used, x is the treatment temperature, b is the slope of the function at inflection point, and a, c and d determine the asymptotes of the function.

Proline

Proline concentration was assessed according to Bates et al.’s (1973) method with slight modification. About 1 g of fresh weight of flowers was ground in liquid nitrogen, and 0.5 g of ground tissue was homogenized in 10 mL of 3% (w/v) aqueous sulfosalicylic acid. Then the homogenate was filtered via a Whatman No.1 filter paper. Two mL of filtered extract plus 2 mL ninhydrin and 2 mL glacial acetic acid were taken for the analysis. The reaction mixture was incubated in a boiling water bath for 1 h, and the reaction was finished in an ice bath. Four mL of toluene was added to the mixture, and the organic phase was extracted. While toluene was used as a blank, absorbance was measured at 520 nm spectrophotometrically (Cary Win UV 100, Varian, Australia). The concentration of proline was calculated by the use of a calibration curve and presented as μmol g−1 FW.

Total phenol

The total phenolic content was determined colorimetrically using Folin–Ciocalteu reagent, as described by Karimi and Ershadi (2015). In sum, 0.3 mL of each diluted methanolic extract (10%) and 1 mL of Folin–Ciocalteau reagent (10%) were mixed and vortexed. A volume of 1 mL from 7% sodium carbonate solution was added to the mixture after 5 min. Shaking the final solution for 90 min at ambient temperature, the absorbance was spectrophotometrically measured at 765 nm. The total phenolic values were measured by applying a calibration curve drawn for the gallic acid standard solution and expressed as mg gallic acid g −1 FW.

Soluble carbohydrates

Soluble carbohydrates were estimated according to the approach presented by Ershadi et al. (2016) with some modification. Until analysis, flowers were oven-dried for 3 days at 80 °C, placed in a coffee grinder to pass a 40 mesh, and stored in the dark inside airtight containers at ambient temperature. Soluble carbohydrates were extracted three times from 1 g of ground tissue with 5 mL of 80% ethanol and centrifuged for 15 min at 9000 rpm. One mL of 0.2% anthrone reagent (2 g anthrone in 1 L of 72% sulfuric acid) was added to 100 μL of the extract of ethanolic. The reaction mixture was heated in a boiling water bath for 10 min and then rapidly cooled on ice. Using a Cary WinUV 100 spectrophotometer at 620 nm, the extract absorbance was measured. Glucose was used as a blank sample. Using a calibration curve, the concentration of soluble carbohydrates was finally measured and exhibited as mg g −1 FW.

Total soluble proteins

Total soluble proteins were extracted from flowers, and their content was determined by measuring the absorbance at 595 nm using the colorimetric approach of Bradford (1976), considering bovine serum albumin as the standard. Total soluble protein values were represented as mg g−1 FW.

Hydrogen peroxide (H2O2)

After reaction with potassium iodide, H2O2 concentration was spectrophotometrically measured based on Velikova and Loreto (2005) method. 1 g of fresh flower tissue was ground and homogenized in a mortar, including 10 ml of 0.1% (w/v) trichloroacetic acid (TCA). The homogenate was centrifuged at 6000×g for 15 min, where 0.5 ml of the supernatant, 0.5 ml of 10 mM potassium phosphate buffer, pH 7.0, and 0.1 ml of reagent were mixed together (0.1 M KI in double-distilled fresh water). The supernatant absorbance was read at 390 nm. In the absence of flower extract, a blank sample was provided using 0.1% (w/v) TCA. From a standard curve of known H2O2 concentrations, the concentration of H2O2 was obtained and expressed as μmol g−1 FW.

Lipid peroxidation

Lipid peroxidation of membrane was observed for 100 mg of each flower sample, which was homogenized in 5 ml of 0.1% (w/v) trichloroacetic acid (TCA) and centrifuged at 6000 rpm for 5 min. The supernatant was collected, and lipid peroxidation was measured in terms of malondialdehyde (MDA) concentration based on Heath and Packer (1968).

Antioxidant activities

About 0.5 g of flower tissue was weighed and then macerated in a mortar with liquid nitrogen until a fine powder was obtained. From each sample, 100 mg of the frozen flower powder was homogenized in 1.0 ml of sodium phosphate buffer (0.5 M, pH 7.8), including 1 mM EDTA and PVP-40 (2% w/v). Samples were homogenized and centrifuged at 10,000×g for 20 min at 4 °C, and all enzyme activity was assessed using the supernatant.

According to Herzog and Fahimi (1973), the guaiacol peroxidase (GPX) activity measurement was done following guaiacol oxidation by H2O2 at 470 nm. The 1 ml volume of each crude flower enzyme extract was added to 3 ml volume of the reaction mixture, including 0.8 μl guaiacol (25 Mm) and 1.3 μl H2O2 (30% V/V) in a volume of 3 ml sodium phosphate buffer (50 mM, pH 7.0). The GPX activity unit was defined as the amount of enzyme that oxidizes 1.0 μmol guaiacol ml−1 min−1. The specific activity of the enzyme is defined as unit mg−1 protein.

The activity of catalase (CAT) was calculated by measuring the drop of H2O2 absorbance at 240 nm wavelength (Bergmeyer 1970). The 3 ml volume of each reaction mixture included sodium phosphate buffer (0.05 M, pH 7.0) with H2O2 (3% v/v) and EDTA (1.0 mM). The drop in absorption at 240 nm occurred for 3 min. One unit of CAT activity was determined by the enzyme amount, which resulted in 1.0 μmol of degraded H2O2 ml−1 min−1. The specific activity of the enzyme is presented as unit mg−1 protein.

Statistical analysis

The data were analyzed using the GLM procedure of SAS software (Version 9.1), and significant differences were tested at P ≤ 0.05 using Duncan’s multiple ranges. Before statistical analysis, the expressed data as percentages were subjected to arcsine transformation, and the original values of all transformed data were presented. Correlation analysis between physicochemical variables and LT50 values of flowers was performed using the CORR procedure of SAS.

Results

The experiment was done two 2 years (2017–2018), and the results are means of 2 years.

Freezing tolerance

Based on flowers’ LT50 values, K2SO4 and ZnSO4 treatments significantly affected the freezing tolerance of pistachio flowers. Flowers of trees treated with 1% K2SO4 + 1% ZnSO4 showed the highest freezing tolerance, reached at 50% mortality at − 4.5 °C, whereas flowers of the control treatment (0% K/0% Zn) indicated 50% mortality at − 1.9 °C. Flowers of other nutrient treatments showed 50% mortality at − 2.9 to − 4 °C (Fig. 1).

Fig. 1
figure 1

Effect of K and Zn application on LT50 values of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Proline

The effect of K2SO4 and ZnSO4 treatments on the proline content of the flowers is presented in Fig. 2. There was a significant effect of K2SO4 and ZnSO4 treatments on the proline content of flowers exposed to − 2 and − 4 °C (P ≤ 0.0001), whereas there was no significant effect in proline content of those exposed to 0 °C (P = 0.3220; Fig. 2). The highest proline content (6.44 μmol g−1 FW) was obtained for 1% K2SO4 + 1% ZnSO4 treatment at − 2 °C, and this effect was superior to that of the most other nutrient treatments, whereas the lowest values (4.18 μmol g−1 FW) were observed in the control. However, no significant difference was found between the control and some other nutrient treatments regarding this trait. Control treatment showed the lowest proline content (5.36 μmol g−1 FW) at − 4 °C, although no significant difference was found between the control and 0% K2SO4 + 1% ZnSO4 treatment regarding this trait, while, the highest proline content of the flowers (7.19 μmol g−1 FW) was found in trees treated with 1% K2SO4 + 1% ZnSO4. Nevertheless, no significant difference was found between this treatment and the most other nutrient treatments regarding this physiological index (Fig. 2).

Fig. 2
figure 2

Effect of K and Zn application on proline content of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Total phenolic content

Results show that K2SO4 and ZnSO4 treatments caused significant changes in the total phenolic content of pistachio flowers exposed to 0, − 2, and − 4 °C (Fig. 3). The lowest total phenolic content (31.60 mg g−1 FW) occurred in flowers of the control treatment at − 4 °C, whereas the highest one (56.70 mg g−1 FW) was found with 1% K2SO4 + 1% ZnSO4 treatment. Similarly, the highest total phenolic content (46.12 mg g−1 FW) was obtained for 1% K2SO4 + 1% ZnSO4 treatment at − 2 °C, although no significant difference was found between this treatment and some other treatments regarding this trait, whereas the lowest value (26.29 mg g−1 FW) was observed in the control treatment of flowers. In addition, flowers of trees treated with 2% K2SO4 + 0% ZnSO4 and 0% K2SO4 + 1% ZnSO4 showed the lowest total phenolic content at 0 °C, whereas the highest one was observed in flowers of % K2SO4 + 1% ZnSO4 and 2% K2SO4 + 1% ZnSO4 treatments, and this effect was superior to that of the other treatments (Fig. 3).

Fig. 3
figure 3

Effect of K and Zn application on total phenolic content of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Soluble carbohydrates

The effect of K2SO4 and ZnSO4 treatments on soluble carbohydrates of the flowers is presented in Fig. 4. Nutrient treatments significantly affected the soluble carbohydrates of flowers exposed to 0, − 2, and − 4 °C. The highest soluble carbohydrates (5.36 mg g−1 FW) were obtained with the 1% K2SO4 + 1% ZnSO4 treatment at 0 °C, and this effect was superior to that of the most other nutrient treatments, whereas the lowest one (2.89 mg g−1 FW) was observed in flowers of 2% K2SO4 + 0% ZnSO4 treatment, although no significant difference was found between this treatment and some other treatments regarding this trait. Soluble carbohydrates in the flowers increased from 4.16 mg g−1 FW in 2% K2SO4 + 0% ZnSO4 treatment to 7.31 mg g−1 FW under 1% K2SO4 + 1% ZnSO4 treatment at − 2 °C. Soluble carbohydrates of the control flowers were 6.23 mg g−1 FW at − 4 °C, whereas the highest ones (11.03 mg g−1 FW) were obtained for 1% K2SO4 + 1% ZnSO4 treatment, and this effect was superior to that of the other treatments except for 1% K2SO4 + 0.5% ZnSO4, 0% K2SO4 + 1% ZnSO4, 2% K2SO4 + 0.5% ZnSO4 (Fig. 4).

Fig. 4
figure 4

Effect of K and Zn application on soluble carbohydrates content of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Soluble proteins

K2SO4 and ZnSO4 treatments significantly affected soluble proteins of flowers exposed to 0, − 2, and − 4 °C (Fig. 5). Soluble proteins in the flowers markedly increased from 5.37 in control to 7.46 mg g−1 FW in the flowers under the 1% K2SO4 + 1% ZnSO4 treatment at − 4 °C, and the effect of 1% K2SO4 + 1% ZnSO4 treatment was significantly higher than that of some other nutrient treatments. Similarly, as shown in Fig. 5, nutrient treatments significantly increased the levels of soluble proteins from 5.09 in the control to 6.94 mg g−1 FW in the flowers under 1% K2SO4 + 1% ZnSO4 treatment at − 2 °C. However, no significant difference was found among most nutrient treatments. In addition, the lowest soluble proteins (3.35 mg g−1 FW) were obtained with the 2% K2SO4 + 0% ZnSO4 treatment at 0 °C, whereas the highest one (5.61 mg g−1 FW) was observed in flowers of 1% K2SO4 + 1% ZnSO4 treatment, and this effect was superior to that of the other treatments (Fig. 5).

Fig. 5
figure 5

Effect of K and Zn application on soluble proteins content of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Hydrogen peroxide (H2O2)

The results showed hydrogen peroxide of control treatments were higher than those of nutrient-treated flowers. Furthermore, the lowest hydrogen peroxide was found with 1% K2SO4 + 1% ZnSO4 treatment under all three freezing temperature stress conditions. However, no significant difference was found between 1% K2SO4 + 1% ZnSO4 treatment and the most other nutrient treatments regarding concentration of hydrogen peroxide at − 2 °C (Fig. 6).

Fig. 6
figure 6

Effect of K and Zn application on hydrogen peroxide (H2O2) concentrations of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Lipid peroxidation (MDA)

K2SO4 and ZnSO4 treatments significantly affected the MDA concentration of flowers exposed to − 2 and − 4 °C, whereas there was no significant effect in MDA concentration at 0 °C (Fig. 7). The highest MDA concentrations was observed in the control treatment of flowers at − 2 and − 4 °C, although no significant difference was found between control and some nutrient treatments regarding this index, whereas the lowest ones were observed in flowers of 1% K2SO4 + 1% ZnSO4 treatment. However, no significant difference was found between this treatment and most other nutrient treatments regarding concentration of MDA (Fig. 7).

Fig. 7
figure 7

Effect of K and Zn application on malondialdehyde (MDA) concentrations of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Antioxidant activities

The effect of nutrient treatment on antioxidant activities of the pistachio flowers is presented in Figs. 8 and 9. Nutrient treatments significantly increased the antioxidant activities of the flowers in all three freezing temperature stress conditions. The lowest antioxidant activities were observed in the flowers of control treatment. The highest guaiacol peroxidase activity (5.48 units mg−1 protein) was obtained for 1% K2SO4 + 1% ZnSO4 treatment at 0 °C, and this effect was superior to that of the other treatments. In addition, the highest levels of guaiacol peroxidase activity of flowers were observed in 1% K2SO4 + 1% ZnSO4 and 2% K2SO4 + 1% ZnSO4 treatments at − 2 °C, and the effect of the two treatments was more considerable than that of the other nutrient treatments. Similarly, flowers of trees treated with 1% K2SO4 + 1% ZnSO4 and 2% K2SO4 + 1% ZnSO4 showed the highest levels of guaiacol peroxidase activity at − 4 °C, and this effect was superior to that of the other treatments except for 2% K2SO4 + 0.5% ZnSO4 (Fig. 8). Furthermore, the maximum catalase activity was found in 1% K2SO4 + 1% ZnSO4 and 2% K2SO4 + 1% ZnSO4 treatments at 0, − 2 and − 4 °C, and the effect of the two treatments was superior to that of the other treatments (Fig. 9).

Fig. 8
figure 8

Effect of K and Zn application on guaiacol peroxidase (GPX) activities of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image
Fig. 9
figure 9

Effect of K and Zn application on catalase (CAT) activities of pistachio flowers cv. ‘Kaleh Ghouchi’. Different letters at the top of columns indicate significant differences (P ≤ 0.05) among treatments. Error bars indicate standard deviation. Values are means of three replicates

Full size image

Correlations

There was a significant negative correlation between the LT50 value and proline content of flowers (r = 0.54, P ≤ 0.01; Table 1). Phenolic compounds, soluble carbohydrates, and soluble protein contents of flowers presented significant negative correlations with LT50 values. In contrast, MDA and H2O2 contents of flowers had significant positive correlations with LT50 values (P ≤ 0.01; Table 1). In addition, correlation coefficients between LT50 values and antioxidant activities were highly significant (P ≤ 0.01; Table 1).

Table 1 Correlation of different variables with LT50 values in flowers of ‘Kaleh Ghouchi’ pistachio under different nutrient treatments
Full size table

Discussion

Spring late frost damage has been one of the main problems in pistachio-producing areas of Iran. Pakkish et al. (2011) indicated that more than 88% of pistachio flowers at − 4 °C and more than 43% at − 2 °C were destroyed during growth season. According to this study, K2SO4 and ZnSO4 treatments significantly increased the freezing tolerance of flowers (expressed as LT50) from − 1.9 °C in the control to − 4.5 °C in 1% K2SO4 + 1% ZnSO4 treatment flowers (Fig. 1). Potassium has been applied to avoid chilling harm development in a number of horticultural crops, including grapevine (Wang et al. 2013; Karimi 2017), tomato, pepper, and eggplant (Hakerlerler et al. 1997). This finding could be attributed to the cold stress, which causes regulation of osmotic and water potential and the decline of electrolyte leakage (Webster and Ebdon 2005). The high K concentrations protected the plant against freezing by lowering the cell solution freezing point down. Furthermore, for enzyme activities involved in regulating freezing tolerance, an adapted cytosol K+ concentration is necessary (Kant and Kafkafi 2002). Additionally, plants with zinc deficiency show increased susceptibility to cold damage, including both winter and frost harm to flowers (Schupp et al. 2001). Less tolerance of citrus trees to low temperatures under Zn-deficient conditions was reported by Cakmak et al. (1995). Therefore, potassium and zinc can significantly reduce frost damage in plants. The results of this study also showed that K2SO4 and ZnSO4 treatments were effective in reducing freezing injury in pistachio flowers (Fig. 1).

In the present research, there was a steady rise in the proline content of flowers in all low-temperature treatments, but the treatment with a higher concentration in both elements, in particular 1% K2SO4 + 1% ZnSO4, significantly increased the rise of this characteristic (Fig. 2). In addition, there were correlations between LT50 values and proline contents of pistachio flowers (Table 1). The relationship between proline increase and low-temperature tolerance has been documented in other previous studies (Beheshti Rooy et al. 2017; Karimi 2017). One of the most critical roles of proline in the field of low-temperature tolerance is the protection of the plant against osmotic changes, protection of cell membranes and cellular enzymes, as well as energy storage for post-freezing restorations (Ashraf and Foolad 2007). Although high proline content in stressed tissues did not essentially indicate high freezing tolerance, its accumulation in tissues is necessary for triggering an array of different metabolic changes in plants to cope with abnormal growth conditions (Ghasemi Soloklui et al. 2012).

Total phenolic content of pistachio flowers increased positively from 0 to − 4 °C, and the treatment with a higher concentration in both elements, in particular 1% K2SO4 + 1% ZnSO4, significantly increased the levels of the total phenolic content. Phenolic compounds are a category of the non-enzymatic antioxidant systems (Apel and Hirt 2004) activated in plants during exposure to low-temperature stress (Karimi 2017). Phenolic compounds’ biosynthesis is partly controlled by genetical factors and mainly by the ecological stimulus, including low temperature (Dixon and Paiva 1995). Phenolic compounds increase membrane integrity by cellular osmotic regulation (Pennycooke et al. 2005). The relationship between phenolic compounds and low-temperature tolerance was confirmed in several fruit tree species, including petunia (Pennycooke et al. 2005), olive (Cansev et al. 2012) and grapevine (Karimi and Ershadi 2015), which is in line with the results observed in the present study (Table 1).

A continuous increase in soluble carbohydrates of the flowers was observed in all low-temperature treatments, but the treatment with a higher concentration in both elements, in particular 1% K2SO4 + 1% ZnSO4, significantly improved the rise of soluble carbohydrates content (Fig. 4). In addition, negative correlations were found between LT50 values and concentration of soluble carbohydrates (Table 1). This correlation was confirmed in other plants such as walnut (Aslamarz et al. 2010), oak (Thomas et al. 2004; Morin et al. 2007), and pomegranate (Ghasemi Soloklui et al. 2012). Soluble carbohydrates increase the concentration of the cytoplasmic solution, and decrease the cell freezing point and stability of cell membranes (Morin et al. 2007). These osmoregulants also can protect plasmalemma and cell wall integrity during exposure to cold stress (Santarius 1992).

The concentrations of flower’s soluble protein increased noticeably under low-temperature stress, but the treatment with a higher concentration in both elements, in particular 1% K2SO4 + 1% ZnSO4, significantly increased the rise of this characteristic (Fig. 5). Soluble protein accumulation under low temperatures was confirmed by Zhang et al. (2012) in wild grape, by Eris et al. (2007) in olive and by Tamura et al. (1998) in pear. It has been argued that soluble proteins accumulate in plant tissues and serve as cryoprotective and antifreeze compounds, which can inhibit or slow down ice crystal formation in cell membranes and several cell structures (Atici and Nalbantoğlu 2003). Meanwhile, the soluble protein can regulate the expression of freezing tolerance genes. The relationship between soluble protein content and low-temperature tolerance was confirmed in other plants such as grapevine (Beheshti Rooy et al. 2017), which is in line with the results observed in this study (Table 1).

The first place of low-temperature injury is mostly cell membranes of the plant, whose extent is usually indicated by lipid peroxidation (MDA) (Guo et al. 2012). In this study, a continuous increase in MDA concentration of the flowers was observed in all low-temperature treatments, but K2SO4 and ZnSO4 application lessened MDA concentration increase (Fig. 7). Furthermore, the variation in membrane permeability (depending on H2O2 concentration) showed similar trends to MDA concentration, i.e., H2O2 concentration of flowers increased with the decrease in temperature, but K2SO4 and ZnSO4 significantly reduced this increase (Fig. 6), which led to a rise in freezing tolerance (Table 1). Mineral nutrition of plants plays a crucial role in increasing the endurance of the plant to environmental stresses (Marschner 1995). Among the mineral nutrients, potassium has a central role in the survival of plants under abiotic stresses. Potassium is necessary for various physiological processes, such as photosynthesis, photosynthate into sink organ translocation, turgidity maintenance, and enzyme activation under stress conditions (Waraich et al. 2012). Zinc is also critical to the growth of hormones, which have huge impacts on the regulation of plant growth in fruit trees. It seems that the required zinc level to assure optimum cold tolerance may be somewhat higher than the sufficient level assumed for normal growth of the crop. Zinc-deficient plants demonstrate increased susceptibility to cold damage, including both frost and winter flower damage (Schupp et al. 2001).

Low temperature alters the metabolism balance of reactive oxygen species (ROS), resulting in their accumulation and scavenging enzyme destruction such as SOD, GPX, CAT, and APX damage (Kang and Saltveit 2002; Beheshti Rooy et al. 2017). Moreover, the ROS accumulation would induce and accumulate lipid peroxidation, harm membrane structure, and bring solute leaking (Pakkish and Tabatabaienia 2016). It has been suggested that the development of tolerance to chilling and freezing in plants are concerned with the enhancement of activates of antioxidant enzyme (Pakkish and Tabatabaienia 2016). Beheshti Rooy et al. (2017) detected that the cold-tolerant grapevines have a higher antioxidant enzyme activity than the cold-sensitive plants. Various treatments that induce frost and chilling tolerance and alleviate frost and chilling damage also enhance antioxidant enzyme activity (Zheng et al. 2008). The results of the present study showed that under low-temperatures stress, K2SO4 and ZnSO4 treatments significantly increased CAT and GPX (Figs. 8 and 9). The increased antioxidant enzyme activity could improve the tissue's ability to remove H2O2, which accounts for the lower level of H2O2 in the flowers detected under nutrient treatments (Fig. 6). On the other hand, the results showed that increasing the freezing tolerance influenced by K2SO4 and ZnSO4 was correlated with increasing antioxidant enzyme activity (Table 1). Our findings were in agreement with previous reports that exogenous application of potassium and zinc is effective in protecting fruit trees against cold stress (Schupp et al. 2001; Karimi 2017). However, based on our knowledge, this report is the first to show the preventive the effects of K and Zn against freezing damage of pistachio flowers. In this paper, the mechanism by which potassium and zinc induced chilling tolerance in pistachio flowers was investigated. When plants are subject to low-temperatures stress, a great content of intracellular ROS is produced (Pakkish and Tabatabaienia 2016). The ROS detoxification is subordinate to the antioxidant enzymes such as SOD, GPX, CAT, and APX (Beheshti Rooy et al. 2017). The increase in the activity of these enzymes ameliorates oxidative injury and contributes to the plant adaptation to low-temperature stress (Aghdam et al. 2012). In this research, we found that the activity of GPX and CAT enzymes in pistachio flowers was induced by K2SO4 and ZnSO4 treatments during cold stress (Figs. 8 and 9). Previous researches have reported similar findings. Devi et al. (2012) revealed the effectiveness of potassium application in reducing cold injury in ginseng by GPX, APX, and CAT elicitation. Zago and Oteiza (2001) figured out that zinc may contribute to the modulation of free radicals and their corresponding damaging effects through enhancement of antioxidant systems of the plants. We propose that the antioxidant enzyme activity induction by K2SO4 and ZnSO4 in pistachio flowers may determine the reduction of oxidative damage resulting from cold stress and thus improvement of freezing tolerance as well as alleviation of flower cold injury exposed to 0, − 2 and − 4 °C.

Conclusions

Nutrient treatments, notably 1% K2SO4 + 1% ZnSO4, improved freezing tolerance of pistachio flowers that may be due to increased antioxidant enzymes, soluble proteins, soluble carbohydrates, total phenolic, and proline contents as well as decreased lipid peroxidation and H2O2 content in flowers. A tight correlation was confirmed between freezing tolerance of pistachio flowers and measured accumulation of metabolite. The improvement of freezing tolerance of pistachio flowers is likely due to the accumulation of osmolytes. The results of the present study could have implications for improving the freezing tolerance of temperate fruit trees cultivated in cold regions by the use of potassium and zinc.

Author contribution statement

BV and RK conceived and designed the research. MMN and RK conducted the experiments. BV analysed the data. BV and MS wrote the manuscript. All authors read and approved the manuscript.