Improving water deficit tolerance of Salvia officinalis L. using putrescine

To study the effects of foliar application of putrescine (distilled water (0), 0.75, 1.5, and 2.25 mM) and water deficit stress (20%, 40%, 60%, and 80% available soil water depletion (ASWD)) on the physiological, biochemical, and molecular attributes of Salvia officinalis L., a factorial experiment was performed in a completely randomized design with three replications in the growth chamber. The results of Real-Time quantitative polymerase chain reaction (qRT-PCR) analysis showed that putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced cineole synthase (CS), sabinene synthase (SS), and bornyl diphosphate synthase (BPPS) relative expression. The highest concentration of 1,8-cineole, camphor, α-thujone, β-thujone, CS, SS, and BPPS were obtained in the irrigation regime of 80% ASWD with the application of 0.75 mM putrescine. There was high correlation between expression levels of the main monoterpenes synthase and the concentration of main monoterpenes. The observed correlation between the two enzyme activities of ascorbate peroxidase (APX) and catalase (CAT) strongly suggests they have coordinated action. On the other hand, the highest peroxidase (PO) and superoxide dismutase (SOD) concentrations were obtained with the application of 0.75 mM putrescine under the irrigation regime of 40% ASWD. Putrescine showed a significant increase in LAI and RWC under water deficit stress. There was an increasing trend in endogenous putrescine when putrescine concentration was increased in all irrigation regimes. Overall, the results suggest that putrescine may act directly as a stress-protecting compound and reduced H2O2 to moderate the capacity of the antioxidative system, maintain the membrane stability, and increase secondary metabolites under water deficit stress.

. Also, heating, cooling, relative humidity and vapor pressure deficit of growth chamber were set with 65 Ḟ, 75 Ḟ, %70 and 0.55 kPa, respectively. A factorial experiment was performed in a completely randomized design (CRD) with three replications. Treatments included irrigation regimes and putrescine, as follows: 1. Irrigation after depletion of 20% available soil water (control). 2. Irrigation after depletion of 40% available soil water. 3. Irrigation after depletion of 60% available soil water. 4. Irrigation after depletion of 80% available soil water.
Four concentrations of putrescine (distilled water (0), 0.75, 1.5, and 2.25 mM) were also applied. Foliar application of putrescine was performed one week before applying irrigation regimes. To investigate proper treatments, plants must have similar masses of foliage before treatments are applied. The plant height, number of nodes, number of leaves per plant and leaf area were 24 cm, 11, 52 and 8 cm 2 , respectively. Therefore, no water stress or putrescine were applied in the first 14 days of the growth cycle. During this period, all pots were irrigated when 20% of the available soil water was depleted (ASWD). The pots in unstressed control were irrigated daily and distance between upper and lower limits of water uptake was 20% available soil water which was determined by TDR. After leaf emergence, foliar application was conducted with 15 mL of putrescine for all aboveground parts of each plant. The depth of the irrigation water was assigned based on the available soil water and calculated using the following equations: where FC and PWP (%) are the volumetric soil water amounts, D (cm) is the soil layer depth, Id is the irrigation depth (cm), ρ is the fraction of ASW that can be depleted from the root zone (20%, 40%, 60%, and 80%), Ig is the coarse depth of irrigation (cm), and Ea is the irrigation efficiency (%) assumed at an average of 65% 31 . The applied irrigation water was measured (based on Eq. (3)) at each irrigation 32 . Irrigation treatments were implemented based on the maximum allowable depletion (MAD) from the percentage of ASWD. Each treatment was irrigated when the available soil water reached its threshold value 31 . The applied treatments were 20%, 40%, 60%, and 80% MAD of ASWD. A TDR probe (Time-Domain Reflectometry, Model TRIME-FM, Germany) was applied to measure soil water amount at a depth of 12.5 cm (root zone of sage). TDR probes were taken from the first experiment to the last experiment. Sampling was conducted from the sage leaves one week after applying the irrigation regimes. Also, samples from all treatments were collected on the same day in the morning. The expression rates were considerably higher in young leaves (first and second nodes) than in fully mature leaves (fifth and upper nodes) 30 . Accordingly, only first and second node leaves were used. These samples were cleaned with ethanol and paper tissues to remove any surface contamination, immediately frozen in liquid nitrogen, and stored at − 80 °C for use in measuring physiological attributes and RNA extraction. RWC, LAI (leaf area index) and putrescine endogenous. The length of time for the discs to reach saturation varies with species and conditions. Studies have shown that water uptake by leaf discs is initially rapid for several hours, followed by a slower rate uptake that can persist as long as the floated discs remain healthy. This usually occurs within 3 to 6 h. Our tests have shown that water uptake by leaf discs is initially rapid for four hours, followed by a slower rate uptake that could continue as 6 h as the floated discs remain healthy. On the other hand, errors arising from growth was minimized by growth inhibitors or by floating the discs at 4 °C. The water content of plants was easily determined by weighing the material immediately after sampling, drying at 75 °C and reweighing 24 h later.
To calculate RWC, 20 leaf discs (around 1 cm in diameter) were punched, and their fresh weight (FW) was recorded. The same leaf discs were kept in petri dishes containing distilled water for six hours to record saturation weight (SW), and after that discs were oven dried at 75 °C for 24 h to record the dry weight (DW). Calculations were performed using the formula: RWC = (FW -DW)/ (SW − DW) 33 .
Leaf tissue was ground under liquid nitrogen using a mortar and pestle. Putrescine extraction followed by HPLC quantification were carried out in accordance with Lütz et al. 34 .
RNA preparation and cDNA synthesis. Total RNAs were extracted from the sage leaves using a RNeasy Plant Mini Kit (QIAGEN GmbH, Hilden, Germany) according to the manufacturer's instructions. RNA quality and concentrations were determined using 1.0% agarose gel electrophoresis and analysis by a NanoDrop 2000/2000c Spectrophotometer (Wilmington, DE 19,810 U.S.A.). Moreover, cDNA was synthesized with a SuperScript III Reverse Transcriptase reagent kit (Thermo Fisher Scientific, USA). Essential oil extraction. The shade-dried foliage of collected sage was extracted by hydro-distillation in a Clevenger device with double-distilled water. The obtained essential oil was separated from the aqueous layer using a 100 ml capacity separatory hopper (a piece of laboratory glassware used in liquid-liquid extractions). The collected surplus aqueous essential oil was dried over anhydrous sodium sulfate. After extraction, the essential oil was conserved in vials sealed at 4 °C until GC/MS evaluation.

Identification of essential oil composition (GC-MS analysis).
Gas chromatography/mass spectrometry (GC/MS) analysis was accomplished using a Thermoquest-Finnigan TRACE GC-MS instrument (Manchester, UK) provided with the same gas chromatography situation as mentioned for the GC analyses. The percentages of compounds were calculated by the area normalization method on GC-MS chromatogram, without considering response factors. Helium was utilized as the vector gas at a flow rate of 1.1 ml min −1 with a split ratio of 1:50, and ionization voltage was 70 eV. The constituents of the essential oil were analyzed based on the retention index (RI) of the series of n-terpenes (C5h8)n and the oil on a Ph-5 column under the same chromatographic conditions. Single constituents were identified based on comparisons of their mass spectra fragmentation designs with standard ones found in the literature in wiley libraries or with those of valid compounds 37 .
Glycine betaine (GB), proline and total reducing sugars (TRS). GB was determined using the method of Grieve and Grattan 38 , in which 5 ml of toluene-water mixture (0.5% toluene) was added to 0.1 g dried ground material. Afterwards, tubes were shaken for 24 h at 25 °C, the extract was filtered, and the volume was made to equal 100 ml. Then, 1 ml of 2 N HCl solution was added to 1 ml of filtrate, and an aliquot of 0.5 ml from this solution was mixed with 0.1 ml of potassium tri-iodide solution. After placing the mixture in an ice bath for 90 min, 4 ml of 1,2 dichloroethane was added to it. The optical density was determined spectrophotometrically at 365 nm. Proline was estimated using the method described by Bates et al. 39 .
To measure total reducing sugars, the method introduced by Dubois et al. 40 was employed.
Antioxidant enzymes (SOD, PO, APX and CAT) and 2,2-diphenyl-1-picrylhydrazyl (DPPH). Leaf samples (500 mg) were pulverized in Na-P buffer (pH 7.0) containing 1 mM EDTA and 1% soluble polyvinyl pyrrolidone (PVP) using a mortar and pestle. The leaf extract was then centrifuged at 12,000 g for 20 min at 4 °C. The enzyme extracts obtained were used to determine the activity levels of SOD, CAT, PO, and APX enzymes. Enzyme activity was expressed as enzyme unit (EU) mg −1 protein. Furthermore, the extraction buffer for ascorbate peroxidase enzyme contained 1 mM ascorbic acid. The protein concentration was analyzed using the method reported by Bradford 41 .
To estimate SOD activity (EC 1.15.1.1), the Bayer and Fridovich 42 method was used. Briefly, 3 ml reaction mixture (containing 100 mM phosphate buffer with pH 7.8), 13 mM methionine, 75 µM NBT, 0.1 mM EDTA, 2 µM riboflavin, and 100 µL enzyme extract was incubated under a light intensity of 3600 lx for 15 min, and the reaction was stopped by switching off the light. The absorbance was read at 560 nm.
The free radical scavenging activity of DPPH was determined spectrophotometrically as described by Hung et al. 46 . In brief, 2 mL of different extract solutions (16-500 g/ml) were mixed with 2 mL of DPPH solution. The mixture was allowed to stand for 30 min to achieve complete reaction at room temperature. Finally, the absorbance of samples was determined at 515 nm.

H 2 O 2 and lipid peroxidation (MDA).
To calculate leaf H 2 O 2 47 , 100 mg fresh tissue was extracted with 5 mL trichloro acetic acid (0.1%) and then centrifuged at 10,000 rpm for 10 min. Finally, supernatant was mixed with potassium phosphate buffer (pH 7.0) and potassium iodide (KI), and absorbance was read at 390 nm. Fresh plant was ground into a fine powder in liquid nitrogen and homogenized with 0.1% (w/v) trichloroacetic acid (TCA). The homogenate was then centrifuged at 12,000 rpm. Aliquots of the supernatant were mixed with 0.5% 2-thiobarbituric acid (TBA) (prepared in 20% TCA). First, the mixture was heated to 95 °C and then cooled on ice. The mixture was further centrifuged at 12,000 rpm. The absorbance of the supernatant was read at 532 and 600 nm.
The MDA content was calculated using its absorption coefficient at 155 mM −1 cm −1 , after the non-specific absorbance was reduced to 600 nm. Finally, MDA content was expressed as nmol g -1 of fresh weight 48 .
Statistical analysis. Amplification data was analyzed using Rotor-Gene 6000 Series software (version 1.7).
The threshold cycle (Ct) values of the triplicate PCRs were averaged and the relative quantification of the transcript levels was determined using the comparative Ct method 49 . The Ct value of the calibrator (the sample with the highest Ct value) was subtracted from every other sample to produce the ∆∆Ct value, and 2 -∆∆Ct was taken as the relative expression level for each sample. The data was subjected to analysis of variance (ANOVA) using SAS 9.3 software (SAS Institute, Cary, NC, USA). Mean comparisons were evaluated using the LSD Test at 5% probability levels. Pearson's correlation coefficients were determined using the CORR procedure.

RWC, LAI and endogenous putrescine.
Based on the results of analysis of variance, putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced RWC, LAI and endogenous putrescine ( Table 2).
The current results showed a decreasing trend in RWC with increasing water deficit stress. The highest RWC was obtained with the application of 0.75 and 1.5 mM putrescine under the irrigation regimes of 80% and 20% ASWD, respectively (Fig. 1A). Generally, the highest RWC (92.93%) was observed with the application of 1.5 mM putrescine under the irrigation regime of 20% ASWD (Fig. 1A). LAI was decreased under water deficit stress. The highest LAI was observed with the application of 0.75 mM putrescine under the irrigation regimes of 40%, 60%, and 80% ASWD (Fig. 1B). Generally, the highest LAI (0.54) was shown with the application of 0.75 mM putrescine under the irrigation regime of 40% ASWD (Fig. 1B). The current results revealed an increasing trend in endogenous putrescine when putrescine concentration was increased under irrigation regimes of 20%, 40%, 60%, and 80% ASWD. Moreover, the highest concentration of endogenous putrescine was obtained with the application of 2.25 mM putrescine under irrigation regimes of 20%, 40%, 60%, and 80% ASWD (Fig. 1C). Generally, the highest concentration of endogenous putrescine (480.00 nmol g −1 FW) resulted from the application of 2.25 mM putrescine under the irrigation regime of 80% ASWD (Fig. 1C).
Gene Expression Assay. Expression analysis revealed that putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced the relative expression levels of CS, SS, and BPPS ( Table 2).
The CS, SS, and BPPS relative expression levels were increased under water deficit stress ( Fig. 2A,B,C). The highest CS and SS relative expression levels were obtained with the application of 1.5 mM putrescine under the irrigation regime of 20% ASWD ( Fig. 2A,B). The highest CS and BPPS relative expression level was achieved with the application of 0.75 mM putrescine under the irrigation regime of 40% ASWD ( Fig. 2A,C). The highest SS and BPPS relative expression levels were obtained with the application of 1.5 and 0.75 mM putrescine under the irrigation regime of 60% ASWD, respectively (Fig. 2B,C). Generally, the highest relative expression levels of CS (18.06), SS (18.48), and BPPS (11.46) resulted from the application of 0.75 mM putrescine under the irrigation regime of 80% ASWD ( Fig. 2A,B,C).

Glycine betaine (GB), proline and total reducing sugars (TRS). The analysis of variance results
showed that putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced significantly GB, proline, and TRS ( Table 2). The highest GB and proline contents were obtained under irrigation regime of 80% compared to irrigation regime of 20% (Fig. 4A,B). The lowest GB, proline and TRS contents were obtained with the application of 1.5 and 0.75 mM putrescine under irrigation regimes of 20% and 40% ASWD, respectively (Fig. 4A,B,C). Moreover, the lowest GB and proline content were observed with the application of 0.75 mM putrescine under irrigation regime of 60% ASWD (Fig. 4A,B). Generally, the lowest GB (0.06 mmol g −1 DW), proline (0.26 mg g −1 FW), and TRS (0.18 mg g −1 FW) contents were obtained with the application of 1.5 mM putrescine under the irrigation regime of 20% ASWD (Fig. 4A,B,C).
Antioxidant enzymes (SOD, PO, APX, and CAT) and DPPH. Based on the results of the analysis of variance, putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced antioxidant enzymes (SOD, PO, APX, and CAT) and DPPH levels (Table 2). SOD, PO, APX, CAT, and DPPH values were increased under water deficit stress (Fig. 5A,B,C,D,E). The highest SOD, PO, and APX values were obtained with the application of 1.5, 0.75, and 0.75 mM putrescine under the irrigation regimes of 20%, 40%, and 60% ASWD, respectively (Fig. 5A,B,C). Moreover, the highest SOD and PO values were observed with the application of 0.75 mM putrescine under the irrigation regime of 80% ASWD (Fig. 5A,B). Furthermore, the highest CAT resulted from the application of 1.5 and 0.75 mM putrescine under the irrigation regimes of 20% and 60% ASWD, respectively (Fig. 5D). Generally, the highest SOD (0.33 μmol mg −1 protein) and PO (3.32 μmol mg −1 protein) values were obtained with the application of 0.75 mM putrescine  www.nature.com/scientificreports/ under the irrigation regime of 40% ASWD (Fig. 5A,B), and the highest APX (0.11 μmol mg −1 protein) and CAT (1.20 μmol mg −1 protein) levels were obtained with the application of 1.5 mM putrescine under the irrigation regime of 20% ASWD (Fig. 5C,D). The highest DPPH was also obtained by applying 0.75 mM putrescine under the irrigation regimes of 40%, 60%, and 80% (Fig. 5E). Generally, the highest DPPH (91.89%) was observed with the application of 0.75 mM putrescine under the irrigation regime of 80% (Fig. 5E).

H 2 O 2 and MDA.
The analysis of variance results showed that putrescine concentration, irrigation regime, and the two-way interaction between irrigation regime and putrescine concentration significantly influenced significantly H2O2 and MDA ( Table 2). The values of H 2 O 2 and MDA increased under water deficit stress (Fig. 6A,B). The lowest H2O2 and MDA were obtained with the application of 1.5 and 0.75 mM putrescine under irrigation regimes of 20% and 60% ASWD, respectively (Fig. 6A,B). Moreover, the lowest H2O2 was obtained with the application of 0.75 mM putrescine under irrigation regime of 80% (Fig. 6A). Generally, the lowest H 2 O 2 (0.45 μmol g −1 FW) and MDA (0.16 μmol g -1 FW) content were obtained with the application of 1.5 mM putrescine under the irrigation regime of 20% ASWD (Fig. 6A,B).

Discussion
The current results showed a significant increase in compatible osmolytes (proline and GB) under the irrigation regime of 80% ASWD. Compatible osmolytes aid in the maintenance of turgor and stabilize macromolecular structures in response to stress 50 . The increase in sage proline in irrigation regimes of 40%, 60% and 80% indicates the important role of this osmolyte under water deficit stress. Indeed, the accumulation of proline and GB in stressed sage plants maintain membrane integrity, reduce oxidation of lipid membranes, stabilize ROS scavenging enzymes, and scavenge free radicals 51 . Generally, increases in compatible osmolytes stabilize redox potential and NAD(P) + /NAD(P)H ratio to prevent oxidative damage under stress conditions 52 . www.nature.com/scientificreports/ In present study, LAI reduced under water deficit stress. On the other hand, this decrease was greater in irrigation regimes of 60% and 80% than 40% ASWD, which could be due to leaves with a smaller size and a higher falling rate. Exogenous application of 0.75 mM putrescine under irrigation regimes of 40%, 60% and 80% ASWD improved leaf area index. The highest LAI was obtained in irrigation regime of 40% ASWD under exogenous application of 0.75 mM putrescine. This improvement may be attributed to maintenance of LAI (assimilatory surface) due to high turgor and sustained photosynthetic ability of the crop by protecting the photosynthetic machinery from reactive oxygen species produced during drought stress and by increasing Rubp content under drought condition. Similarly, the positive effects of exogenously putrescine on LAI and tolerance to abiotic stresses in different plant species have been reported by others 53 .
In response to water deficit stress, RWC decreased in sage plants. Decreases in RWC subject the cell membranes to changes such as penetrability and decreased sustainability under water deficit stress 54 , that probably aims to create osmotic adjustment 55 . There was a negative correlation between proline and RWC (− 0.56; P < 0.05) ( Table 3). Indeed, the decrease in RWC in irrigation regime of 80% AWSD were possibly due to the accumulation of proline and GB. Foliar application of putrescine alleviated the detrimental effects of water deficit stress and increased RWC considerably. These results are concordant with those of Farooq et al. 53 . The possible effect of foliar spray is related to the fact that while in direct contact with leaf surface, PAs improved the water status of epidermal cells and underlying cells 53 . It seems that the contribution of putrescine to osmotic adjustment can be considered as a mechanism to retain RWC for better growth and productivity. Two reasons for the responses of PAs under different adverse environmental conditions might be the ability to scavenge ROS and adjust osmosis 53 .
In the present study, water deficit stress increased the production of H2O2 and MDA in plants, which is an indication of tissue damage, thus enhancing SOD, CAT, PO, APX and DPPH activities under water deficit conditions. The effectiveness of the antioxidant defense system function depends on the intensity of the water deficit stress 56 . The highest APX and DPPH concentrations for scavenging H2O2 and MDA protecting biomolecules were obtained in the 80% ASWD irrigation regime. This higher enzyme activity did not provide enough protection against ROS (H2O2) and MDA. Indeed, to improve the damage oxidative stress imposes on sage, foliar applications of putrescine (0.75 and 1.5 mM) are required. Similar results were reported by Tajti et al. 57 (Table 3). There were also negative correlations between SOD and APX and proline (− 0.56 and − 0.52; P < 0.05, respectively) ( Table 3). Also, there was high correlations between H2O2 and MDA (0.74; P < 0.01) ( Table 3). The lowest compatible osmolytes and H2O2 and MDA and highest antioxidant enzymes including CAT and APX were obtained with the application of 1.5 mM putrescine; this showed the role of putrescine in decreasing of lipid peroxidation, stabilizing the cell membranes, preventing degrading of cell membranes by free radicals like H2O2. This response indicates a good H2O2 scavenging ability in the application of putrescine. Indeed, putrescine inhibits NADPH oxidase enzymes in cell walls, which ultimately leads to less H2O2 production in putrescine-treated plants 58 . The observed correlation between the two enzyme activities strongly suggests coordinated action between APX and CAT. Yiu et al. 59 reported similar results. In the present study, with the application of 0.75 mM putrescine, SOD and PO played an important role in protecting H2O2 under the 40% ASWD irrigation regime. On the other hand, SOD (as the first line of defense against ROS) and APX have significant negative correlations with H 2 O 2 (− 0.66, − 0.74; P < 0.01, respectively) ( Table 3). Indeed, probably putrescine substantially improved the impacts of water deficit stress on the membrane stability index in sage plants by binding to the negativelycharged phospholipid head group 60 . It is well documented that PAs (e.g., putrescine) are able to induce adaptive changes to maintain plasma membrane integrity under water deficit stress. Moreover, the enzymatic antioxidant activity enhanced by putrescine seems to be the result of de novo synthesis and/or the activation of the enzymes, mediated by the transcription and/or translation of specific genes 61 , that potentially aids stressed plants to resist against oxidative stress induced by water deficit stress. There were also high correlations between LAI, SOD, PO and CAT and RWC (0.92, 0.54, 0.70 and 0.65; P < 0.01, respectively) ( Table 3). Moreover, there were positive correlations between PO and CAT and LAI (0.65 and 0.64; P < 0.01) ( Table 3). So, the highest RWC, LAI, PO, CAT and SOD were obtained with the application of 0.75 mM putrescine and the 40% ASWD irrigation regime. Also, the highest RWC and CAT were obtained with the application of 1.5 mM putrescine and the 20% ASWD irrigation regime.
Elevation of ambient CO 2 concentration results in a massive enhancement of photosynthesis, quite often the rate is doubled or even tripled. The absolute demands for ATP and NAD(P)H by the plant cell as well as www.nature.com/scientificreports/ the demand for ATP relative to NAD(P)H vary in response to growth, development and stress conditions 62 . A dominant concept is that water deficit stress, similar some other stresses, favors electron flux to O 2 based on the idea that the regeneration of NADP + cannot keep pace with NADPH generation. However, any significantly increased in the NADPH: NADP + ratio will have deep effects on the regulation of photosynthesis due to the vital coupling of electron transport to ATP synthesis 63 . An important point (In terms of the sustainability of alternative electron flow) is the mean turnover times of ATP and NADPH in the chloroplast stroma, which do not trespass a few seconds at moderate to high rates of steady-state photosynthesis 64 . In addition to any raise of the Mehler reaction, over reduce of the stroma will favor engagement of other ATP generating processes such as cyclic electron transport and chlororespiratory pathways. Classical energy dissipating mechanisms for eliminating the overflow of electrons comprise non-photochemical quenching, photorespiration, and the xanthophyll cycle 65 . As a result, photosynthetic control over plastoquinol oxidation at the cytochrome b6f. complex will limit overreduced of PSI and join to build electron pressure in PSII 66 . Increased photosynthetic control will not only promote energy dissipation in PSII but will also tend to increase the likelihood of singlet oxygen production. Furthermore, the synthesis of secondary metabolites and biosynthesis of highly reduced compounds like isoprene are significantly involved in the dissipation of excess photosynthetic energy 67 . The amount of energy decomposed by isoprene emission could account for up to 25% of the net photosynthesis energy store under stress conditions 68 . Indeed, synthesis and increased secondary metabolites could be considered as the machinery to minimize the reduction equivalents.
The current results showed that monoterpene concentration, monoterpene synthesis, and H2O2 were increased under water deficit stress conditions with the highest contents obtained with the 80% ASWD irrigation regime. Indeed, it is well known that ROS as signal components are involved in the activation of monoterpene  www.nature.com/scientificreports/ water under the irrigation regime of 80% ASWD. So, in explaining this result, it should be said that putrescine enters the leaves by penetrating the cuticle or through the stomata before entering the plant cell, where they can be practical in metabolism and are further transported to other parts through plasmodesmata. Therefore, putrescine and monoterpenes were probably produced in the MEP pathway (plastid) and essential oil content affected by putrescine. Schmiderer et al. 13 achieved similar results; they reported that the expression of monoterpene synthase and contents of 1,8-cineole and camphor were increased with the application of gibberellins. Methyl jasmonate and SA have also been applied as abiotic elicitors to induce secondary metabolite biosynthesis as terpene metabolism in Tanacetum parthenium 73 . The highest amount of DPPH was observed with the application of 0.75 mM putrescine under the irrigation regime of 80% ASWD. High correlations were observed between DPPH and CS, BPPS, SS, 1,8-cineol, camphor, α-thujone, and β-thujone (0.64, 0.71, 0.77, 0.54, 0.55, 0.75 and 0.74; P < 0.05, 0.01, respectively) ( Table 3). There were also high correlations between SOD and CS, BPPS, and 1,8-cineol (0.58, 0.64, and 0.55; P < 0.05, 0.01, respectively), between PO and BPPS (0.74; P < 0.01), and between APX and BPPS, SS, α-thujone, and β-thujone (0.63, 0,56, 0.52, and 0.52; P < 0.05, 0.01, respectively) ( Table 3). High correlations were also observed between DPPH and SOD, PO, CAT, and APX (0.74, 0.64, 0.50, and 0.76, respectively) ( Table 3).
According to the current results, the content of compatible osmolytes and concentration of antioxidant enzymes were increased and decreased, respectively, with the application of 2.25 mM putrescine. There is evidence that SA causes a rise in the quantity of ROS in the cell, suggesting the existence of a self-induced SA H2O2 cycle 74 . This is not surprising, as a close correlation was recently reported between the endogenous PAs and SA contents 75 , which may be responsible for the negative effects of greater concentrations of putrescine. Some reports on SA have been in agreement with the current results 76 . The treatment of 2.25 mM putrescine resulted in an inhibitory effect compared with the 0.75 mM putrescine treatment. This is why stress-induced H2O2 accumulation was lower in the 0.75 mM than the 2.25 mM putrescine-treated plants. Putrescine reduced the accumulation of total monoterpenes in concentrations of 2.25 mM which, to some extent, can be attributed to excessive oxidative burst-induced putrescine in high concentrations. Indeed, there was a decreasing trend in the concentration of monoterpenes and the expression of monoterpene synthases genes with increasing putrescine concentrations under the irrigation regime of 80% ASWD; however, further studies are needed to determine whether the application of putrescine in high concentrations at the cellular level decrease monoterpenes in sage under water deficit stress conditions.

Conclusion
The current results showed a foliar application of putrescine alleviated the detrimental effects of water deficit stress and considerably increased LAI and RWC. The lowest compatible osmolyte and H 2 O 2 contents and the highest antioxidant enzymes including CAT and APX were obtained with the application of 1.5 mM putrescine, which showed the role of putrescine in decreasing of lipid peroxidation, stabilizing cell membranes and preventing the degradation of cell membranes by free radicals like H 2 O 2 . This response indicates a good H 2 O 2 scavenging ability with the application of putrescine. In the present study, SOD and PO in the application of 0.75 mM putrescine played important roles in decreasing H 2 O 2 under the irrigation regime of 40% ASWD. The highest levels of expression of the main monoterpenes synthase, concentrations of main monoterpenes, and DPPH were obtained with the application of 0.75 mM putrescine and the irrigation regime of 80% ASWD. PAs and monoterpenes were probably produced in the (MEP) pathway, and the essential oil content affected by putrescine. In general, the adequate putrescine concentration to trigger a best response is 0.75 mM. Indeed, putrescine could be a useful strategy to increase the main monoterpenes in sage plants. However, further researches are needed to determine the appropriate concentrations of exogenous putrescine under water deficit stress conditions in sage.