Introduction

Drought stress is one of the major constrains in plant productivity that influences the growth and physiological characteristics of plants (Santisree et al. 2015). Exposure of plants to water stress could produce reactive oxygen species such as superoxide anion radicals (O2−), hydroxyl radicals (·OH), hydrogen peroxide (H2O2), and singlet oxygen (1O2) (Csiszar et al. 2007; Liao et al. 2012). The accumulation of ROS in plant tissues makes serious physiological and biochemical dysfunctions including damages of photosynthetic apparatus, disorders in electron transfer of chloroplasts and mitochondria, and damages the cellular membranes (Gill and Tuteja 2010). The responses of plants to drought depend on duration and severity of water shortage, genotype and stage of development (Bray 1997), but the plant cells are generally protected from damages of different stresses by an antioxidant defense system which detoxify reactive oxygen species (Hayat et al. 2011; Nalousi et al. 2013). The plant antioxidant defense machinery could be either enzymatic e.g. superoxide dismutase, ascorbate peroxidase, catalase and peroxidase or non-enzymatic (e.g. glutathione, proline, a-tocopherols, glycin betaein and phenols) (Gill and Tuteja 2010; Freschi 2013). The imbalance between ROS production and the capacity of plant antioxidant system causing drastic oxidative damages (Farooq et al. 2009). In addition, genetic factors playing important roles in diverse signaling pathways related to plant defense responses (Li et al. 2015).

An increasing rarity of water in the future will cause improving adaptation to drought stress as a major objective of plant breeding efforts. Many methods have been applied to alleviate drought stress effects on plants such as identification and cultivation of drought tolerant genotypes or transformation of drought resistant genes to a host plant that need high technology, investment, and long time (Liao et al. 2012; Nazar et al. 2015). However, application of plant growth regulators in a right time and concentration is inexpensive, accessible and simple alternative way to mitigate harmful effects of different abiotic stresses. Application of a suitable PGR on plants could improve water use efficiency and acclimation to drought stress (Hao et al. 2007; Qiao and Fan 2008). Nitric oxide (NO), is considered as a signaling molecule that involved in various physiological processes from promotion of seed germination (Kovacs and Lindermayr 2013) to regulation of plant maturation (Liu and Guo 2013). Some studies have reported an increased production of NO under environmental stresses such as water deficit, salinity, high irradiance, heat, cold and pathogen attack (Qiao and Fan 2008; Siddiqui et al. 2011). It has been shown that application of NO at low concentrations commonly has an acclimation-like effect and resulting in improved plant growth and tolerance towards drought stress, which is primarily due to increased plant antioxidant capacity (Siddiqui et al. 2011; Nalousi et al. 2013; Rahimian Boogar et al. 2014). The most common NO donors are sodium nitroprusside (SNP), S-nitroso-N-acetylpenicillamine (SNAP), S-nitrosoglutathione (GSNO), and diethylamine NONOate sodium (DEA-NONOate) (Freschi 2013; Santisree et al. 2015).

Plants may be simultaneously exposed to multiple stresses in the field or greenhouse that may result in oxidative stress which cannot be attributed to a specific stress (Farooq et al. 2009). In vitro plant tissue culture allows stringent control of the physical environment, nutrition, stress level, and monitoring plant responses to a stress (Pérez-Clemente and Gómez-Cadenas 2012; Piwowarczyk et al. 2014). Polyethylene glycol is a neutral, non-ionic, inert, non-penetrating, and osmotically active polymer (Piwowarczyk et al. 2014) that can be used to mimic osmotic stress in plant tissue cultures to study mechanisms of tolerance. PEG induces drought stress by reducing water potential of the media (Elmaghrabi et al. 2017).

Allium hirtifolium is a wild medicinal plant distributed from north–west to central and south–west of Iran, which is also known as Persian shallot. For many years, fresh and dry bulbs of A. hirtifolium are used to treat for inflammatory, arthritis, diarrhea, rheumatic, and stomach pains (Asili et al. 2010). Allicin is identified as the chief biologically active constituent of sulphur containing compounds of Persian shallot (Asili et al. 2010) that has antifungal (Mahboubi and Kazempour 2015), antimicrobial (Ismail et al. 2012), antiparasitic and antioxidant activities (Taran and Izaddoost 2010). To meet A. hirtifolium demand in herbal medicine and improve its culture, there is an urgent need for improving the A. hirtifolium responses to different environmental stresses such as water deficit.

Drought stress mitigation by NO treatment has not been studied in Allium species so far. Thus, this research was aimed to investigate the growth, physiological and biochemical responses of in vitro grown A. hirtifolium to PEG-induced drought and exogenous SNP as a NO donor.

Materials and methods

Plant materials and experimental design

The bulbs of A. hirtifolium were collected from natural habitats of Lorestan province in Iran. After washing with running tap water for 2 h, and removing outer dry scales, they were soaked in a detergent for 15 min, in 70% ethanol for 5 min and in 5% sodium hypochlorite for 15 min. Then, these bulbs were rinsed three times with sterile distilled water. For shoot induction, 3–5 mm of the sterilized bulbs were excised and cultured on MS media with 0.5 mg l−1 1-naphthalene acetic acid, 1.5 mg l−1 6-benzylaminopurine (Farhadi et al. 2017), and different concentrations of PEG (0, 2, 4, 8 and 16 mM), with or without SNP (0, 10, 40 and 70 µM). All the cultures were sub-cultured every 4 weeks, using the same media. Treatments with 2, 4, 8 and 16 mM PEG are considered as slight, mild, moderate and severe drought stress, respectively. The shoot induction was estimated by percentage of explants forming shoots. The number of formed shoots were recorded after 8 weeks of culture, and thereafter these shoots were excised and transferred to the same media with the same concentrations of PEG and SNP. Tissue samples for all analyses were collected from in vitro plantlets under different treatments after 8 weeks of culture. For all biochemical analyses, only leaves were used, while allicin content was determined form bulbs. This experiment was laid out as factorial on the bases of completely randomized design with four replicates.

Basal medium and culture conditions

Cultures were kept at a temperature of 25 ± 1 °C under a 16/8 h light/dark conditions. Light was provided by white fluorescent tubes at 40 μmol m−2 s−1. All media consisted of MS mineral salts, vitamins (Murashige and Skoog 1962) and 3% (w v−1) sucrose, which were solidified with 0.8% agar. The pH of the media was adjusted to 5.7 and autoclaved for 20 min at 121 °C. Sub-culturing was carried out on the same media every 4 weeks. Naphthalene acetic acid and benzylaminopurine were added to the media before autoclaving, while SNP was added after autoclaving through filter sterilization.

Relative water content

Relative water content of regenerated plantlets was determined at the end of experiment according to Barrs and Weatherley (1962) as: (FW−DW/TW−DW) × 100; where FW is fresh weight of shoots, DW is dry weight recorded after drying the samples at 80 °C for 24 h; and TW is turgor weight, determined by subjecting shoots to rehydration in distillated water for 2 h at room temperature.

Pigments and proline

Total chlorophyll and carotenoids were extracted from leaves using 80% acetone. The absorbance of the pigments was measured by UV visible spectrophotometer (Analytikjena Spekol 1500) at 470, 648, and 664 nm. Total chlorophyll and carotenoid contents were estimated using the equations proposed by Lichtenthaler (1987) and expressed as mg of each pigment per g leaf fresh weight.

Proline content was determined by a spectrophotometer according to Bates et al. (1973). Each sample (0.3 g) was homogenized in 3 ml of 3% sulfosalicylic acid and centrifuged at 2000 g for 5 min. The filtered homogenate was reacted with equal volume each of acid ninhydrin and glacial acetic acid for 1 h in a test tube placed in a water bath at 100 °C and the reaction was terminated on an ice bath. The mixture was extracted by toluene and the absorbance was recorded at 520 nm. The proline content was determined from the standard curves of l-proline and reported as µg per g leaf fresh weight.

H2O2 and malondialdehyde

The hydrogen peroxide (H2O2) levels were determined as described by Velikova et al. (2000). The leaf samples were homogenized in 5 ml of trichloroacetic acid (TCA, 0.1% w v−1) at 4 °C and centrifuged at 12,000 g for 10 min. From the resulting supernatant, 0.5 ml was added to 0.5 ml of phosphate buffer (pH 7.0) and 1 ml of 1 M KI. The absorbance of the reaction mixture was measured at 390 nm and the H2O2 level was reported as µmol H2O2 per g leaf fresh weight.

Lipid peroxidation was determined by the level of decomposition product of the peroxidized membrane polyunsaturated fatty acids as malondialdehyde (MDA), described by Cakmak and Horst (1991). Fresh leaf samples (0.3 g) were ground in 3 ml of trichloroacetic acid (TCA, 0.1% w v−1) and centrifuged at 10,000 g for 20 min at 4 °C. Then 0.5 ml from the supernatant was added to 1.5 ml of thiobarbituric acid (TBA, 0.5% w v−1) with 20% TCA (w v−1). The samples were heated at 90 °C for 30 min. Thereafter, the reaction was stopped in ice bath and the absorbance of the supernatant was recorded at 532 and 600 nm. MDA concentration was determined using the specific extinction coefficient of 155 mM−1 cm−1. The results were expressed as nmol MDA per g leaf fresh weight.

Antioxidant enzymes assays

For enzymes assays, 1 g of explants was homogenized in 10 ml of 50 mM potassium phosphate buffer (pH 6.8). The homogenates were centrifuged at 12,000 g for 20 min at 4 °C and the supernatant fraction used as the source of protein and enzymes. The amount of soluble protein in the extract was determined using the Bradford (1976) protein assay, with bovine serum albumin (BSA) as the standard. The activity of superoxide dismutase was estimated by the method of Giannopolitis and Ries (1977). The SOD activity was assayed by its ability to inhibit the photochemical reduction of nitro blue tetrazolium (NBT), and it was reported as units per mg soluble protein. The ascorbate peroxidase activity was determined according to Nakano and Asada (1981) and expressed as mmol ascorbate oxidized per min per mg soluble protein (U mg−1) with 2.8 mM−1 cm−1 as the extinction coefficient. The catalase activity was estimated by the method of Aebi (1983). The activity of catalase was reported as mmol decomposed H2O2 per min per mg soluble protein (U mg−1), using the H2O2 extinction coefficient of 39.4 mM−1 cm−1. The activity of peroxidase was determined by guaiacol and H2O2 substrates as described by Chance and Maehly (1955). With the extinction coefficient of tetraguaiacol product (26.6 mM−1 cm−1), the activity of POX was expressed as mmol produced tetraguaiacol per minute per mg soluble protein (U mg−1).

Total phenol and allicin

Total phenol content was measured by the Folin–Ciocalteu reagent (Slinkard and Singleton 1977). Briefly, 0.5 g of a explant was extracted with 5 ml 85% methanol at room temperature of 22–25 °C and centrifuged at 12,000 g for 10 min. 200 μl of the supernatant was diluted with 300 μl of distillated water, and subsequently 2.5 ml of freshly prepared 50% Folin–Ciocalteu reagent was added. The solution was incubated in darkness at room temperature for 3 min, and then 2 ml of 7.5% sodium carbonate solution was added. After 30 min, the absorbance was measured at 760 nm. Different concentrations of gallic acid were used as a standard, and the total phenol content was expressed as mg of gallic acid equivalent per g leaf fresh weight.

The content of allicin as an organosulfur compound was measured by the high performance liquid chromatography (Shimadzu, 150 × 4.6 C18 column with a Bishoff pump) with a spectrophotometer (KNAUER), using an internal standard of butyl peroxy benzoate (Iberl et al. 1990). The results were reported as mg allicin per g bulb fresh weight.

Statistical analysis

Results were subjected to analysis of variance using the statistical analysis program (SPSS ver.16.0). The mean values were compared by Duncan’s multiple range test at p ≤ 0.05.

Results

Explant regeneration

Significant interaction of PEG × SNP (Table 1) showed that the days to shoot induction up to 8 mM PEG was statistically similar at different levels of SNP application. The highest days to shoot initiation was recorded for 16 mM PEG, with or without SNP treatments. Application of 10 and 40 µM SNP decreased the average days to regeneration in explants under severe drought stress in comparison with untreated explants, but 70 µM SNP had no beneficial effect on this trait.

Table 1 Effects of sodium nitroprusside (SNP) on regeneration and physiological characters of A. hirtifolium under drought stress imposed by polyethylene glycol (PEG)
Full size table

Increasing water stress resulted in a continuous reduction of the explant regeneration in all SNP treatments. This reduction was more pronounced under 8 and 16 mM PEG, particularly in 0 and 70 µM SNP treated explants. SNP treatment significantly improved regeneration of explants under moderate and severe water deficit, but this was not consistent over all SNP concentrations and 70 µM SNP showed negative effects on shoot induction (Table 1).

The most prominent growth effect on shoot development was observed at 40 µM SNP. The unstressed explants had up to 7.59% more shoot regeneration when treated with SNP, whereas in stressed explants this improvement was up to 11.97%. Application of PEG and SNP significantly affected the number of regenerated shoots per explant (p ≤ 0.01), but their interaction was not significant (p > 0.05). The number of regenerated shoots per explant decreased from 11.56 to 6.94 under severe drought stress (Fig. 1a). SNP application of 40 μM resulted in an increase of 33.33% in number of shoots per explant (Fig. 1b). The effects of PEG and SNP levels on shoot regeneration in in vitro cultured basal plates of A. hirtifolium are shown in Fig. 2.

Fig. 1
figure 1

Effects of polyethylene glycol (a) and sodium nitroprusside (b) on number of regenerated shoots in A. hirtifolium explants

Full size image
Fig. 2
figure 2

Effects of polyethylene glycol (PEG) and sodium nitroprusside (SNP) on shoot regeneration in in vitro cultured basal plates of A. hirtifolium. a Cultured basal plate explants. b Shoot regeneration in control without PEG and SNP. c Shoot regeneration under severe drought stress (16 mM PEG). d Shoot regeneration under severe drought stress (16 mM PEG) with 40 µM SNP. e Shoot regeneration in non-stressed explants with 40 µM SNP. f Regenerated shoots on a fresh medium

Full size image

Relative water content

Drought stress and SNP application significantly affected the relative water content of regenerated plantlets (p ≤ 0.01). PEG imposed drought stress caused significant losses in RWC up to 22.01% in comparison with control. Exogenous SNP application at 10 and especially at 40 µM significantly increased RWC in non-stressed and stressed plantlets. Treatment of plantlets with 40 µM SNP under slight, mild, moderate and severe drought stress increased RWC by 4.22, 4.41, 4.95 and 16.29%, respectively (Table 1).

Photosynthetic pigments and proline

Mean squares of PEG and SNP were significant for photosynthetic pigments and proline (p ≤ 0.01), but their interaction was only significant for chlorophyll and proline contents (p ≤ 0.01). Drought stress reduced the content of total chlorophyll down to 16.03% in comparison with control. The SNP treatments of 10 and 40 µM slightly improved total chlorophyll content in control and PEG-treated shoots, however high concentration of SNP (70 µM) exaggerated the negative effects of drought stress. Application of 40 µM SNP in medium was not only able to prevent chlorophyll decline in water-stressed plants, but also increased total chlorophyll content at all levels of water availability, compared with untreated explants (Table 1). Drought stress reduced the carotenoids contents down to 24.2% in comparison with control plantlets (Fig. 3a). The highest and lowest carotenoids contents were obtained by 10–40 and 70 µM SNP, respectively (Fig. 3b).

Fig. 3
figure 3

Effects of polyethylene glycol (a) and sodium nitroprusside (b) on carotenoids content of A. hirtifolium leaves

Full size image

Proline accumulation significantly increased with increasing water limitation and also with SNP application up to 40 µM. Severe drought stress increased proline content by 84.56% in comparison with control. Under non-stress condition, SNP did not affect proline accumulation, but application of 40 µM SNP under mild, moderate and severe drought stress increased proline content by 44.76, 24.67 and 18.97%, respectively, compared with untreated plantlets (Table 1).

Accumulation of H2O2 and malondialdehyde

The effect of PEG-imposed drought stress on accumulation of reactive oxygen species was evaluated by H2O2 and lipid peroxidation levels in leaves of regenerated plantlets. Water stress caused a significant increase in H2O2 content up to 47.75%, compared with non-stressed plantlets (p ≤ 0.01). Exogenous application of up to 40 µM SNP significantly reduced H2O2 accumulation at all PEG levels, but 70 µM SNP significantly increased H2O2 content compared with other treatments. Results showed that SNP can reverse the effects of drought stress on H2O2 accumulation with dose depended manner. The lowest H2O2 accumulation under all watering levels was recorded for plantlets treated with 40 µM SNP (Table 1).

Water stress caused a significant increase in MDA content in A. hirtifolium leaves up to 2.04 times, compared with the plantlets grown on media without PEG. Exogenously applied SNP had no significant effect on MDA accumulation in unstressed plantlets. However, application of 10 and 40 µM SNP reduced MDA by 22.02 and 34.69%, respectively (Table 1).

Antioxidant enzymes activities

Since water shortage elevates ROS levels, the activities of antioxidant enzymes were determined as indicators of oxidative stress. The interaction of PEG × SNP for antioxidant enzymes activities was significant (p ≤ 0.05). All measured antioxidant enzymes (SOD, APX, CAT and POX) showed significant increase in response to drought stress at all levels of SNP treatment. The highest increase in enzymes activities was observed under moderate and severe drought stress, while non - low stressed explants showed little differences in all enzymes activities (Table 2).

Table 2 Effects of sodium nitroprusside (SNP) on antioxidant enzymes activities and allicin content of A. hirtifolium under drought stress imposed by polyethylene glycol (PEG)
Full size table

The enzymes responded differently to exogenous application of SNP. The activities of APX and SOD increased, but the activities of CAT and POX decreased with increasing SNP under all PEG levels. Treatment of plantlets with 70 µM SNP reduced CAT and POX activities by 7.96 and 9.66% in non-stressed plantlets, and by 18.42 and 25.51% in explants grown under severe drought stress, respectively (Table 2).

Phenol and allicin contents

The results indicated that phenol content of plantlets linearly increased with decreasing water availability (Fig. 4a). However, application of SNP had no significant effect on total phenol content in non-stressed and stressed plantlets (Fig. 4b).

Fig. 4
figure 4

Effects of polyethylene glycol (a) and sodium nitroprusside (b) on total phenol content of A. hirtifolium leaves

Full size image

The interaction of drought stress × SNP for allicin content of A. hirtifolium bulbs was also significant (p ≤ 0.01). The results showed that the allicin content of bulbs at almost all levels of SNP treatment enhanced with declining water supply up to moderate stress (8 mM PEG), and then slightly diminished under severe water stress. The application of SNP up to 40 µM increased the amount of allicin under all PEG levels, but there were little differences between untreated and treated bulbs with 70 µM SNP. The highest allicin content was obtained in the plantlets treated with 40 µM SNP under moderate water stress (40.41% increase in comparison with control plantlets) (Table 2).

Discussion

Drought is an important environmental factor influencing the physiological and phytochemical characteristics of plants (Santisree et al. 2015). Stress conditions cause a shift in the antioxidant balance of plant cells as a result of increasing ROS generation, which induces lipid peroxidation in the membrane of cells. Compounds that are able to reduce the damaging effects of various stresses are prominent in both theoretical and practical points of views (Qiao and Fan 2008). In our research, A. hirtifolium resistance to drought stress in an in-vitro culture was increased by adding SNP as a NO donor in media. These findings revealed a significant link between SNP treatment under drought stress and the enhancement of almost all studied characteristics.

PEG in the medium retarded the regeneration of cultivated A. hirtifolium explants and decreased the regeneration percent and number of formed shoots (Table 1; Fig. 1). Addition of PEG to the medium decreases the water potential and makes osmotic stress (Aazami et al. 2010) which negatively affects plant growth. Induced drought stress causes turgor loss that obstructs cell expansion and disorders mitosis division, so retards growth (Farooq et al. 2009). Decreased growth in the PEG-containing media is one of the common consequences of water stress that was also reported in tomato (Aazami et al. 2010), onion (Kielkowska et al. 2012), marigold (Liao et al. 2012), sugarcane (Rao and FTZ 2013) and strawberries (Mozafari et al. 2017). In our study, SNP at low and medium concentrations accelerated the regeneration of A. hirtifolium explants, while high concentration of SNP (70 µM) had adverse effects on this generation under moderate and severe water deficit (Table 1). Similarly, Xu et al. (2009) found that high concentrations of SNP (100 and 150 µM) decreased the explant regeneration. It was reported (Sarropoulou and Maloupa 2017) that SNP intensifies the stimulatory effect of the cytokinins on shoot proliferation in terms of shoot number and shoot multiplication percentage. Tubic et al. (2016) also stated that cytokinin has an essential role in plant regeneration. Plant growth responses to exogenous SNP vary depending on species, developmental phase, and the applied SNP concentration (Santisree et al. 2015). Liao et al. (2012) reported that NO promoted multiplication and regeneration of in-vitro marigold plantlets under imposed drought stress. In unstressed Dioscorea opposite exogenously applied NO-donor SNP improved the cell survival of explants, thereby promoted frequency of in-vitro proliferation (Xu et al. 2009). This effect attributed to induced nitrosative stress under high concentration of SNP. NO mediates plant hormones signaling transduction in plant growth and development, but the mechanisms by which SNP improves the micro-propagation efficiency are not fully understood. It seems that NO involves in cytokinin signal transduction and regulation of plant cell division, therefore supplementation with the NO improves the micro-propagation efficiency (Otvos et al. 2005).

Relative water content reflects plant water status. Decreasing RWC of plantlets with increasing drought stress (Table 1) is one of the common consequences of water stress (Farooq et al. 2009), which was also observed in several species (Shallan et al. 2012; Liao et al. 2012; Ahmad et al. 2014). In the present study, SNP application at 40 µM was quite effective in improving RWC of stressed plantlets. The potential of SNP to sustain RWC of plantlets was more pronounced under severe drought stress. It has been proved that NO has important role in plant water relations (Cechin et al. 2015). NO is an effective factor in abscisic acid mediated signaling pathways during stomatal closure to prevent water losses from plant tissues under drought stress (Santisree et al. 2015).

Drought imposed by PEG caused a decrease in total chlorophyll (Table 1) and carotenoids (Fig. 3a) contents of A. hirtifolium plantlets. A reduction in photosynthetic pigments is a typical symptom of oxidative stress that causes photo-oxidation of pigments and chlorophyll degradation. Significant reduction of chlorophyll content is also recorded in drought-stressed plants of cotton (Shallan et al. 2012), marigold (Liao et al. 2012), onion (Ahmad et al. 2014) and sunflower (Cechin et al. 2015). Chlorophyll (Table 1) and carotenoids (Fig. 3b) contents of explants were significantly improved by application of 40 µM SNP under different levels of water availabilities. It was reported that exogenous SNP invariably increases photosynthetic pigments in water-stressed plants (Hayat et al. 2011). In contrast, Cechin et al. (2015) found that SNP did not affect the chlorophyll and carotenoids contents of sunflower under drought stress. Releasing iron from dissociation of SNP can most likely enhance chlorophyll content of plants (Santisree et al. 2015). However, the effect of SNP on photosynthetic pigments depends not only on the stress conditions and species, but also on the applied SNP concentration, and sampling time. At high concentrations of SNP, NO stimulates chloroplast deterioration and the degradation of plastid pigments (Hao et al. 2007; Hayat et al. 2011), as also observed in our research (Table 1; Fig. 3b).

Proline, as the most common compatible solute, has a main role in plant tolerance of different stresses via contribution in osmotic adjustment, antioxidation, and ROS scavenging (Farooq et al. 2009; Shallan et al. 2012). ROS are part of abscisic acid signaling that induces proline accumulation (Jariteh et al. 2015). PEG-induced water stress increased the accumulation of proline in A. hirtifolium plantlets (Table 1), which is a common response of plants to drought stress (Hao et al. 2007; Hayat et al. 2011). Increasing proline content due to water deficit may be related with increasing activity of pyrroline-5-carboxylate reductase (P5CR), which converted pyrroline-5-carboxylate to proline (Nazar et al. 2015). Exogenous SNP up to 40 µM stimulated accumulation of proline under all watering levels (Table 1). Proline accumulation could alter different transcripts associated with gene expression and cell division (Maggio et al. 2002), therefore, the enhanced effect of SNP on proline accumulations could promote the cell proliferation and regeneration rate in in vitro culture (Xu et al. 2009).

Water limitation caused oxidative stress in A. hirtifolium plantlets that was confirmed by high H2O2 accumulation and MDA content (Table 1). Disruption in electron transport chains in chloroplasts, mitochondria, and plasma membrane under drought stress may result in increased formation of ROS (Farooq et al. 2009). The accumulated ROS via oxidative stress may damage proteins, nucleic acids, membrane lipids, and other cellular components (Farooq et al. 2009). One of the prominent effects of SNP in A. hirtifolium was reduction of H2O2 and MDA levels under all stress levels. The reduction of MDA was a consequence of enhancement in ROS scavenging and proline accumulation in SNP treated plantlets. The level of H2O2 and MDA in water-stressed plantlets treated with 40 µM SNP was statistically similar to non-treated unstressed explants (Table 1). This is a clear indication that SNP reverses the promoter effects of water limitation on these parameters. NO can act as a chain breaker during lipid peroxidation by interacting with lipid alkoxyl and peroxyl radicals (Santisree et al. 2015). Enhancing H2O2 and MDA level of plantlets by 70 µM of SNP in comparison with untreated explants under moderate and severe water deficit (Table 1) suggests that excess NO induces nitro-oxidative stress, so can act synergistically with ROS and increase the undesirable effects of drought stress in plant cells (Santisree et al. 2015).

Tolerant cells also activate their antioxidant enzymes such as SOD, APX, CAT and POX with increasing ROS accumulation. This response was evident in A. hirtifolium plantlets under PEG induced drought stress, where the activities of antioxidant enzymes increased with decreasing water availability (Table 2). Drought stress also induced theses antioxidant enzymes activities in different Allium species (Csiszar et al. 2007; Kielkowska et al. 2012).

Increasing SOD and APX activities and decreasing CAT and POX activities with enhancing SNP application on A. hirtifolium explants under different watering levels (Table 2) indicate that the response to SNP may differ depending on species and enzyme. Depending on the physiological condition of the cell, NO stimulates or inhibits the antioxidant enzymes activities (Kovacs and Lindermayr 2013). Cechin et al. (2015) reported that SNP did not affect SOD activity in sunflower under drought stress, but Hayat et al. (2011) and Rahimian Boogar et al. (2014) stated that exogenous SNP increased both SOD and APX activities in drought stressed turfgrasses and tomato, respectively. Similar to our results, SNP reduced CAT and POX activities in unstressed Dioscorea opposite plantlets (Xu et al. 2009), while in other studies SNP generally had a stimulating effect on these enzymes as in cotton (Shallan et al. 2012), turfgrasses (Rahimian Boogar et al. 2014) and sunflower (Cechin et al. 2015).

Total phenolic compounds are well known as antioxidant constituents of plants inducing chemical and biological activities in different crops (Manuel Beato et al. 2011). Total phenol content was increased in A. hirtifolium leaves as drought stress increased, but it was not affected by SNP application (Fig. 4). Bettaieb et al. (2011) also showed that polyphenol content of in Salvia officinalis was significantly enhanced due to mild and severe water limitations, compared with control. Increasing phenolic compound (flavonoid) biosynthesis improved drought tolerance in transgenic tobacco (Yuan et al. 2015). In contrast, Shallan et al. (2012) reported that drought stress decreased the total phenol content of cotton, but it was significantly increased by SNP treatment. Manuel Beato et al. (2011) showed that different types of phenolic compounds varied with water deficit, which affected the total phenol content.

Allicin as an important secondary metabolite and antioxidant compound of A. hirtifolium increased under mild and moderate drought stress (Table 2). This is supported by previous study by Hassan et al. (2015) on garlic, who reported that high allicin content might be closely associated with the water deficit. Secondary metabolites as partially plant defense system increase in response to different environmental stresses.

SNP application enhanced allicin content of A. hirtifolium bulbs under different levels of water availability (Table 2). Improving allicin content of plantlets by application of SNP might be due to the increase in cell cycle growth, nutrients uptake or changes in biosynthesis pathways of sulfur compounds via the affecting of sulfur assimilation enzymes. The synthesis of ginkgolides, the main active compound of Ginkgo biloba was also significantly increased by drought stress and SNP application (Hao et al. 2007). Hao et al. (2007) reported that SNP as NO donor remarkably raised the proline accumulation and flavonoids and ginkgolides synthesis in Ginkgo biloba plants under drought stress. They proposed that the roles of plant secondary metabolites under drought stress are the same as proline and other osmo-regulators.

It seems that exogenous SNP by improving the potential of plant cells for proline and allicin accumulation, and increasing antioxidant enzymes activities inhibited lipid peroxidation in cell membranes and reversed the effects of PEG imposed drought stress on growth and physiological performance of in vitro cultured A. hirtifolium. According to Hayat et al. (2011) treatment with NO donor generally increases plant stress tolerance either by increasing antioxidant capeability or by inducing a set of stress defense mechanisms by NO, even after removal or degradation of the NO donor. However, it is important to understand that SNP exhibits an ambivalent or biphasic action in several stress models, including drought, so the exogenous application of SNP may lead to different results in different species and environmental conditions (Santisree et al. 2015).

Conclusion

These results indicate that SNP has a powerful drought-ameliorating potential on in vitro cultured A. hirtifolium, with no growth-retarding or other negative effects on this species. Exogenous application of SNP can help to maintain membrane stability under drought stress by preventing lipid peroxidation via increasing antioxidant enzymes activities and also enhancing proline and allicin contents. In addition, SNP improves proliferation frequency and allicin content of A. hirtifolium, and would be useful for its propagation and phytochemical production under in vitro condition. In future, combination of drought stress (8 mM PEG) and SNP (40 µM) could be used to increase commercial benefits of this plant.