Improving seed germination and seedling growth of guava under heat and osmotic stresses by chemical and hormonal seed treatments

: This study, consisting of two independent experiments, was conducted to optimize presowing guava seed treatments and evaluate the optimized treatments in improving germination and seedling growth of guava under heat and osmotic stresses. In the first experiment, seeds of guava cultivar White Flesh Local I were soaked in water, gibberellic acid (GA 3 ) (0.05 and 0.1%), hydrochloric acid (HCl) (5 and 10%) and sulfuric acid (H 2 SO 4 ) (5 and 10%) for 24 and 48 h (for water and GA 3 ), and 2 and 5 min (for HCl and H 2 SO 4 ). Seed soaking (priming) with GA 3 (0.1%, at 48 h) and HCl (10%, 2 min) were the most effective treatments to improve seed germination and early seed growth of guava and were used in the second experiment. In the second experiment, treated and untreated seeds were sown in plastic boxes between two layers of filter papers maintained at osmotic potentials of 0, −1.5 and −3 MPa. The germination boxes were incubated at optimal (25 °C) and higher (32 °C) temperatures. Germination was significantly suppressed at a higher temperature and with an increase in the osmotic potential. However, seed treatments with GA 3 and HCl were effective to improve the germination and seedling growth of guava under both temperature and osmotic stresses. In conclusion, chemical and hormonal seed treatments may help improve the seed germination and seedling growth of guava under heat and osmotic stresses by modulation of antioxidant enzymes and leaf proline. Seed treatment with GA 3 (0.1%, 48 h) was the most effective in this regard. evaluated under less than optimal conditions. This study was, therefore, conducted to evaluate the potential of chemical and hormonal seed treatments in improving seed germination and seedling growth of guava under heat and osmotic stresses.


INTRODUCTION
obtained from local fresh-cut fruit. Seed cores were placed in a blender, excess water was added; and then seeds were separated from the fruit, cleaned, and dried in shadow (one week).
All seeds were pretreated with a 0.1% sodium hypochlorite (NaOCl) solution for 5 min and then subsequently rinsed with distilled water and air dried to avoid fungal attack. The study consisted of two independent experiments.

Experiment I
Guava seeds were soaked in distilled water, GA 3 (0.05 and 0.1%), HCl (5 and 10%), and H 2 SO 4 (5 and 10%). Seeds were soaked in water and GA 3 for 24 and 48 h, and were soaked in HCl and H 2 SO 4 for 2 and 5 min maintaining a seed to solution ratio of 1:5 (w/v). After treatment, seeds were rinsed with distilled water and air dried at 20 °C under shade until reaching original weight. Seeds (30) were placed evenly on sterilized filter paper moistened with 15 mL of distilled water in Petri dishes (12 cm). The filter papers were replaced every three days to ensure even moisture. Dishes were sealed with Parafilm to minimize water loss from evaporation. Petri dishes were closed to prevent moisture loss and avoid contamination and were placed in a germinator at 25 °C. The experiment was conducted as a factorial completely randomized design with three replications.

Experiment II
Based on the results of experiment I, guava seeds were soaked in distilled water, 10% HCl for 2 min, and GA 3 (0.1%, 48 h) maintaining a seed to solution ratio of 1:5. The treated seeds were rinsed with distilled water and air dried at 20 °C under shade until the original weight (~ 14%). Seeds (30 in each box) were germinated in plastic boxes between two layers of filter papers moistened with 15 mL of distilled water (0 MPa) and with solutions of polyethylene glycol (PEG 6000 ) having an osmotic potential −1.5 and −3.0 MPa. The target osmotic potential solutions were prepared following Michel and Kaufmann (1973). Distilled water, containing 0.02% fungicide (Captan; C 9 H 8 C l3 NO 2 S) was used to prepare the solutions. The boxes were placed in a germinator at 25 (optimal) and 32 °C (heat stress) with 10/14 h day/night duration. The filter papers, in each box, were replaced every two days to maintain a uniform water potential. The experiment was conducted as a factorial completely randomized design with three replications.

Observations and measurements
Germination and seedling growth The number of germinated seeds was counted daily until reaching a constant count. Both experiments were terminated 30 days after sowing. Seeds with a radicle length of 2 mm were considered as germinated. Germination percentage was calculated as the ratio of the number of germinated seeds to the number of seeds planted and was expressed as a percentage. Germination rate was estimated using the following equation by Ranal et al. (2009) where R s is the germination rate, S i is the number of germinated seeds per day, D i is the number of days passed since the beginning of the experiment. At the final harvest, five seedlings were randomly chosen from each replication to record morphological parameters. Root and shoot length were determined. Root and shoots were separated, and fresh weight was recorded immediately and was then oven-dried at 70 °C for 24 h to record dry weight. A digital weighing balance, with a precision of 0.001 g, was used to record fresh and dry weights. Leaf area was recorded using the image software Image J (National Institutes of Health, Maryland, USA).

Proline extraction and measurement
Total free proline contents were measured following a modified method of Khare et al. (2012). Samples (0.1 g of root and shoot) were homogenized in 0.5 mL of 3% (w/v) sulphosalicylic acid. Each homogenate (0.2 mL) was combined with 0.2 mL of glacial acetic acid to which, 0.2 mL of ninhydrin was added. The mixture was boiled in a water bath at 100 °C for 30 min and cooled in an ice bath immediately. The chromophore containing toluene was separated and absorbance of the red color was read at 520 nm on the UV-visible spectrophotometer (Cary 100 UV-Visible Spectrophotometer, Agilent, United States).

Total soluble carbohydrate
The soluble carbohydrate contents were estimated using the Anthrone method (Yemm and Willis 1954). The sample (0.5 g each of root and shoot) was homogenized with hot aqueous ethanol (80%). After centrifugation, 0.2 mL of supernatant was transmitted into another test tube and distilled water was added to reach a volume of 1 mL. Anthrone reagent (0.2%) was then added into the tubes. The samples were heated in a boiling water bath for 8 min and were then cooled rapidly; the intensity of the green to dark green color was estimated at 625 nm by a digital spectrophotometer (Cary 100 UV-Visible Spectrophotometer, Agilent, United States).

Enzymatic antioxidants and oxidative stress markers
Guava seedling samples were ground in liquid nitrogen and the powder was extracted with 10 mL of 50 mmol·L -1 phosphate buffer (pH 7.0), then centrifuged at 14,000 × g for 15 min at 4 °C. Peroxidase (POD) activity was assayed spectrophotometrically at 470 nm using guaiacol as a substrate according to Hemeda and Klein (1990). Superoxide dismutase (SOD) activity was measured by estimating its ability to prevent the photochemical reduction of nitro blue tetrazolium (NBT) according to Beauchamp and Fridovich (1971). The SOD activity was measured spectrophotometrically at 560 nm. The catalase (CAT) activity was measured following the method of Aebi (1974). Tissues (0.1 g) were disrupted in 0.4 mL of extracting buffer (50 mmol·L -1 phosphate buffer, pH 7, and 1% Triton X-100). The homogenates were centrifuged for 20 min at 12,000 × g. Then, 0.1 mL of the supernatant was mixed with 0.9 mL of the buffered substrate (20 mmol·L -1 H 2 O 2 in 50 mmol·L -1 phosphate buffer, pH 7), and the OD of the mixture was measured spectrophotometrically at 240 nm.
Hydrogen peroxide (H 2 O 2 ) was determined spectrophotometrically at 390 nm according to Alexieva et al. (2001) to quantify H 2 O 2 content. The accumulation of malondialdehyde (MDA) in leaf samples due to lipid peroxidation was measured spectrophotometrically at 532 nm following Dhindsa et al. (1981).

Statistical analyses
The data were statistically analyzed by analysis of variance using the Statistical Analysis Software (SAS, version 9.1). Duncan's new multiple range test was used for mean separation at a significance level of p < 0.05. Principal component analysis, Pearson correlation coefficient, and dendrogram clustering were carried out using statistical software R v3.4.3.

Experiment I
All seed priming and soaking treatments significantly improved the germination and germination rate of guava seeds as compared to the control (Table 1). In this regard, the highest germination percentage (93.97%) and germination rate (21.1) were observed with seed priming by GA 3 (0.1% for 48 h), followed by seed soaking in HCl (10% for 2 min) ( Table 1).
All seed priming and soaking treatments significantly improved radicle and plumule length and dry weight, and total plantlet length and weight of guava than water soaking alone ( Table 1). The highest radicle length (2.49 cm) and plumule length (4.79 cm) was observed in GA 3 seed priming by 0.1% for 48 h. Likewise, the highest plantlet length (7.29 cm) and the total weight of plantlet (31 g) were recorded with GA 3 seed priming (0.1% for 48 h) (Table 1). Likewise, the highest root and shoot dry weights were also recorded with GA 3 seed priming (0.1% for 48 h).

Germination percentage and rate
Osmotic stress significantly reduced the germination percentage and germination rate of guava seeds with an increase in the level of stress (Table 2). Heat stress (32 °C) was effective to promote germination percentage. However, seed priming with GA 3 and soaking in HCl significantly improved the germination percentage and rate at all levels of osmotic stress and at both temperature regimes. In this regard, GA 3 seed priming with 0.1% for 48 h improved the germination percentage and rate at all stress levels and at both temperature regimes (Table 2).

Seedling growth parameters
Osmotic and temperature stresses significantly reduced seedling growth parameters, namely radicle, and plumule length, root and shoot dry weight, root and shoot fresh weight and leaf area as compared to the respective controls. Seed treatments significantly improved seedling growth parameters under both stress treatments, however, GA 3 seed priming was the most effective (Table 2).

Enzymatic antioxidants and oxidative markers
Activities of enzymatic antioxidants (peroxidase, superoxide dismutase and catalase) and oxidative markers (hydrogen peroxide, and malondialdehyde) significantly increased under both osmotic and temperature stresses on guava seedlings. However, seed treatments further increased the activity of enzymatic antioxidants under both stresses. Nonetheless, seed treatments caused a significant reduction in hydrogen peroxide and malondialdehyde. Seed priming with GA 3 was the most effective treatment in increasing the activity of enzymatic antioxidants while decreasing for hydrogen peroxide and malondialdehyde (Table 3).

Proline and carbohydrate contents
Proline and carbohydrate contents in guava seedlings (shoot and root) significantly increased under both osmotic and temperature stresses. Seed treatments further improved the proline and carbohydrate contents under both stresses, however, seed priming with GA 3 was the most effective (Fig. 1).

Principal component analysis, correlation and dendrogram clustering of guava seeds under osmotic conditions
All morphological and biochemical traits were loaded into two principal components (PC1 and PC2), explaining 94.50 of the total variances (Fig. 2a). Most of the examined traits were discriminated by PC1, and thus explained by the larger proportions of variances (78.30%); while the lower proportions of variances (16.2%) were indicated by PC2 in experiment II (Fig. 2a). Principal component analysis (PCA) plots separated the treatment groups into three main groups: (i) guava seeds under -1.5 and -3 MPa (osmotic stress) at 25 and 32 °C, (ii) guava seeds treated with HCl soaking (10%, 2 min) and GA 3 priming (0.1%, 48 h) under -1.5 and -3 MPa at 25 and 32 °C, (iii) guava seeds soaked in water, HCl (10%, 2 min) and GA 3 (0.1%, 48 h) without osmotic stress at 25 and 32 °C (Fig. 2b). All the enzyme activities were positively associated with GA 3 priming (0.1%, 48 h at 25 and 32 °C) under osmotic stress, while H 2 O 2 were positively linked with guava seeds under -1.5 and -3 MPa (osmotic stress) at 25 and 32 °C (Fig. 2c). In addition, germination and seedling growth parameters were positively associated with GA 3 priming (0.1%, 48 h) at 25 and 32 °C (Fig. 2c). Means sharing the same letters, for a parameter, don't differ significantly different at p < 0.05 as per Duncan's new multiple range test. GM: germination; GMR: germination rate; RL: radicle length, PL: plumule length; RDW: root dry weight; SDW: shoot dry weight; RFW: root fresh weight; SFW: shoot fresh weight; LA: leaf area; HCl: seed soaking in 10% HCl solution for 2 min; GA 3 : seed priming in 0.1% gibberellic acid solution for 48 h.

DISCUSSION
Chemical (acid) and hormonal (GA 3 ) treatments were effective to improve the germination and early seedling growth of guava seeds. Acid scarification is an effective method for breaking seed dormancy due to hard seed coats. However, soaking seeds in concentrated acid for longer durations may damage the seed embryo. Similarly, soaking in diluted acids for short durations may not make the hard seeds permeable to oxygen and water (Tanaka-Oda et al. 2009). The same pattern was noted in this study; seed soaking in HCl (10%, 2 min) was the most effective treatment to break the seed dormancy in guava. While GA 3 is involved in the activation of germination process, seed soaking in GA 3 was effective in improving seed germination as well.
The current research thus indicates that GA 3 promotes germination in guava. In maize, a GA 3ox deletion mutant could not germinate even under normal conditions while the transcript level of GA 3ox2 improved germination 40-fold in dormancy-broken seeds as compared to dormant ones. Also, GA 2ox1 expression levels were higher in dormant seeds than in nondormant seeds (Li et al. 2017). Germination and seedling growth are promoted by the homeostasis of plant growth hormones (Dissanayake et al. 2010). Gibberellic acid increases the activity of the protease enzyme and changes proteins to amino acids such as tryptophan (auxin precursor). Gibberellic acid also modulates the dry weight of plants because it increases the leaf area and photosynthesis (Lester et al. 2002). Gibberellic acid functions as a regulator for plant growth through increasing cell division and cell elongation and replication (Salehi Sardoei et al. 2014). Gibberellic acid also modulates the hydrolysis of starch into monosaccharides, which promotes germination and early seedling growth (Arteca 1996;Farooq et al. 2018). It also stimulates the activation and/or de novo synthesis of hydrolytic enzymes involved in the germination metabolism (Paleg et al. 1964;Farooq et al. 2006).
Osmotic stress suppresses germination and morphological traits such as fresh and dry weight of shoots and roots and the leaf area in guava (Shaukat et al. 2015). The present study showed that severe drought stress and an increase in temperature led to the Table 3. Influence of chemical and hormonal seeds treatments on activities of antioxidant enzymes, hydrogen peroxide and malondialdehyde contents in guava under different osmotic and temperature regimes. accumulation of some osmolytes and increased activity of some enzymes. The accumulation of organic osmolytes in the cytoplasm to maintain the water potential of the plant during periods of stress is one of the osmotic defense mechanisms ). The exposure to stress may increase the accumulation of proline, soluble sugars and other organic osmolytes in the plant shoots and roots (Farooq et al. 2018;Zahedi et al. 2019). Accumulation of these solutes led to a decrease in the water potential of olive organs and water uptake in the plant became possible (Gholami and Zahedi 2019). As an osmotic adjustment option, proline accumulates more in plants at a higher temperature and under drought stress. However, GA 3 treatment may reduce the proline accumulation in plant tissues. This could stimulate vegetative growth of plant grown under drought stress (Kaya et al. 2006). Shoots (especially leaves) are a major sink for the accumulation of proline under stressful conditions (Chun et al. 2018). Carbohydrate accumulation is induced by respiratory changes under stress conditions (Gholami and Zahedi 2019). Zawaski and Busov (2014) reported that DELLA proteins and GA 2oxs not only repress growth but enhance plant resistance to stress via activation of ROS-detoxification enzymes, increased carotenoid production as a nonenzymatic antioxidant and ABA biosynthesis. Plants alter the activity of antioxidant enzymes such as SOD, CAT, APX and POX; antioxidant enzymes and endogenous GAs to avoid drought-induced oxidative damages (Liu et al. 2013). Carbohydrate accumulation in plant tissues under stress conditions may help in effective osmoregulation. In this regard, soluble organic compounds may act as osmoprotectants in addition to their role in osmoregulation (Gill et al. 2003;Farooq et al. 2018).

CONCLUSION
Chemical and hormonal seed treatments improved seed germination and seedling growth of guava under heat and osmotic stresses by modulation of antioxidant enzymes and leaf proline. Seed treatments with GA 3 (0.1%, 48 h) were the most effective in this regard. Germination and early growth are the key to successful plant establishment under optimal and less than optimal conditions. In this study, seed treatment with GA 3 was effective in improving guava seed germination and seedling growth under heat and osmotic stresses. However, further studies are needed to understand the role of hormonal homeostasis in germination under less than optimum conditions.