Maize seeds of Meiyu No.3 with 10.1% moisture content were obtained from HaiNan Lv Chuan Seed Co., LTD. SA and H₂O₂ were purchased from Shanghai Dingguo Biotech Co., Ltd., Shanghai, China.

Seeds were surface sterilized with 0.5% NaClO solution for 15 min and then washed three times with sterilized distilled water to remove the residue of disinfectant. The sterilized seeds were primed with priming solutions (1:5, w/v) at 20°C in darkness for 24 h, without changing the priming solution during the process. Then all primed seeds were air-dried at 25°C for 48 h to their original moisture contents. In this study, four priming solutions including distilled water (hydropriming), 0.5 mM SA, 50 mM H₂O₂, and 0.5 mM SA+50 mM H₂O₂ were carried out.

All primed maize seeds were germinated in rolled towels moistened with water for 7 days in growth chambers at chilling stress (13°C), with a photosynthetic active photon flux density of 250 mmol m^-2⋅s^-1 and a photoperiod of 12 h light (L):12 h dark (D) (Wang et al., 2013). Those treatments were, respectively, indicated by HP+CS (hydropriming and chilling stress), SA+CS (SA priming and chilling stress), H₂O₂+CS (H₂O₂ priming and chilling stress), and SA+H₂O₂+CS (SA+H₂O₂ priming and chilling stress). Six replications of 50 seeds each for each treatment were used. The geminated seeds (5 mm radicle penetrated through seed coat) were counted daily for 7 days, and then germination energy (GE) and germination percentage (GP) was calculated on day 4 and day 7, respectively (ISTA, 2004). The germination index (GI) was measured as GI = Σ (Gt/Tt) (Hu et al., 2005). The mean germination time (MGT) was calculated as MGT = Σ (Gt × Tt)/ΣGt, where Gt is the number of new germinated seeds in time Tt. Shoot height and root length were measured manually with a ruler. The seedling dry weight (DW) was weighed directly after drying at 80°C for 24 h. Measurements were made on thirty randomly selected seedlings per replication. Vigor index (VI) was determined as VI = GI × DW (Hu et al., 2005).

During seed germination after sowing, 10 seed embryos per replication and 6 replications for each treatment at each sampling time (0, 6, 24, 48, and 72 h after sowing) were collected for physiological parameters determination. The H₂O₂ content was measured according to the method of Doulis et al. (1997). The activities of SOD, POD, APX, and CAT were determined through the methods described by Qiu et al. (2005). POD, APX, and CAT activity was, respectively, calculated as μmol ascorbate decomposition min^-1⋅g^-1⋅FW and μmol H₂O₂ decomposition min^-1⋅g^-1⋅FW; while one unit of SOD activity was defined as 50% inhibition of nitro-blue tetrazolium (NBT) photochemical reaction and the results were expressed in U g^-1⋅FW. GR was quantitated by measuring the decrease in absorbance at 340 nm due to the oxidation of nicotinamide adenine dinucleotide phosphate (NADPH), and GR activity was calculated as μmol NADPH oxidation min^-1⋅g^-1⋅FW (Smith and Johnson, 1988). The malondialdehyde (MDA) content was measured using the thiobarbituric acid reaction method (Gao et al., 2009). The activity of α-amylase was conducted using 3,5-dinitrosalicylic acid colorimetric method (Li et al., 2009). The soluble sugar content in seeds was measured by anthrone colorimetric method (Li and Li, 2013) with slight modification.

Endogenous SA content was analyzed using HPLC according to the method of Wang et al. (2013). Each sample was ground into powder with liquid nitrogen and then 0.5 g of powder was homogenized in 1 ml aliquot of 90% (v/v) methanol. After centrifugation at 10,000 × g for 15 min, the supernatant was dried by nitrogen purging at 35°C. The residue was dissolved in 0.25 ml of 5% (w/v) TCA and the upper phase containing SA was concentrated by nitrogen purging at 35°C. In addition, the released SA was extracted from a lower aqueous phase. The SA collected from the upper phase and the lower phase were mixed and dissolved in a 600 μl of mobile phase consisting of 0.2 M sodium acetate buffer (pH 5.5) (90%) and methanol (10%), which was then filtered through a 0.22 μm membrane filter and analyzed by HPLC (Waters 600, Waters 717 automatic sampler, Waters, Milford, United States) under fluorescence detection (Waters 474). The excitation and emission wavelengths were 305 and 407 nm, respectively. Six replications of embryos for each treatment were collected at each sampling time.

Total RNA was isolated from seed embryo using Trizol reagent (Huayueyang, Beijing, China) and reverse transcribed using a Rever Tra Ace qPCR RT kit (Toyobo, Osaka, Japan) following the manufacturer’s instructions. Gene specific RT-PCR primers were designed based on their cDNA sequences (Supplementary Table S1). Six genes involved in stress responses, i.e., ZmSOD4, ZmAPX2, ZmCAT2, and ZmGR. ZmAMY (α-amylase gene) involved in seed germination. Six genes involved in GA signal, i.e., ZmGA20ox1, ZmGA3ox2, ZmGA2ox1, ZmGID1, ZmGID2, and ZmRGL2. Four genes involved in ABA signal, i.e., ZmNCED1, ZmCYP707A2, ZmCPK11, and ZmSnRK2.1. ZmPAL (encoding phenylalanine ammonia-lyase) involved in SA biosynthesis. The RT-qPCR was performed using Roche real-time PCR detection system (Roche life science, United States). Each reaction (20 μL) consisted of 10 μL of SYBR Green PCR Master Mix (Takara, Chiga, Japan), 1 μL of diluted cDNA and 0.1 μM forward and reserve primers. The PCR cycling conditions were as follows: 95°C for 3 min, followed by 40 cycles of 95°C for 10 s and 58°C for 45 s. The maize Actin gene was used as an internal control. Relative gene expression was calculated according to Livak and Schmittgen (2001).

Data were analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) (version 9.2) followed by calculation of the Least Significant Difference (LSD, p < 0.05). Percentage data were arc-sin-transformed prior to analysis.

As compared with NP+Cn, the seed germination was inhibited obviously under low temperature, and none of priming treatments reached the same germination speed and seedling growth compared to NP+Cn (Figure 1). The germination percentage of NP+Cn was 96% on the third day, when that of NP+CS was 0%. Seed priming treatments significantly promoted seed germination under chilling stress as compared with NP+CS. Seeds treated with SA+H₂O₂+CS started germinating at 1 day after sowing, whereas seeds treated with SA+CS or H₂O₂+CS began to germinate on the second day, while those seeds treated with HP+CS delayed germination until the third day after sowing (Figure 1 and Supplementary Figure S2). On the third day, the maximum germination percentage (80.0%) was recorded in SA+H₂O₂ under chilling stress, as well as GE, GI, DW, and VI, which were also significantly higher than other priming treatments (Table 1). All priming treatments owned longer RL than NP+CS did, in which SA+H₂O₂ had the relatively higher level (6.05 cm). Priming treatments showed less of an effect on SH, in which H₂O₂ and SA+H₂O₂ significantly heighten SH compared to NP+CS (Table 1).

Hydrogen peroxide concentration gradually increased under normal condition (NP+Cn) and were significantly higher than NP+CS from 24 to 72 h after sowing (Figure 2A). The H₂O₂ content in NP+CS kept a stably low level and showed no visibly difference during seed germination (Supplementary Figure S1A). As comparing with NP+CS, the H₂O₂ concentration of H₂O₂+CS was significantly higher at 0 and 48 h; while SA+CS obviously improved H₂O₂ concentration from 48 to 72 h. SA+H₂O₂+CS enhanced H₂O₂ level in general, but did not show distinct difference with NP+CS and HP+CS from 6 to 72 h.

In addition, NP+Cn and NP+CS showed relatively higher MDA levels during seed germination than all priming treatments did (Figure 2B and Supplementary Figure S1B). The MDA content of NP+Cn decreased significantly at 72 h compared to NP+CS. Except for 24 and 72 h, HP had obviously higher MDA content than other three priming treatments. SA primed seeds showed the lowest MDA levels from 0 to 72 h, followed by SA+H₂O₂ and H₂O₂ primed seeds under chilling stress.

Non-primed seeds had prominently higher SA concentrations under 25°C than 13°C during seed germination (Figure 3A), and the notable increase in SA contents happened after 24 h after sowing. Under chilling stress, SA concentrations treated by SA+CS, H₂O₂ +CS, and SA+H₂O₂+CS rapidly increased as compared with NP+CS which were all significantly higher than HP+CS. All priming treatments increased SA contents markedly from 0 to 72 h under chilling stress.

The ZmPAL expression of NP+Cn reached its maximum level at 24 h after sowing (160-fold higher than NP+CS) (Figure 3B); while ZmPAL expression of NP+CS stayed at relatively low levels during 72 h after sowing. Under chilling stress, priming seeds significantly enhanced ZmPAL expression compared to NP+CS and NP+Cn at 0 h, and also up-regulated ZmPAL expression compared to NP+CS at 24 h. ZmPAL expression was obviously higher in case of SA+H₂O₂ combination than other three priming treatments from 6 to 48 h after sowing.

All of the antioxidant enzymes activities including SOD, APX, CAT, and GR of six treatments showed a stepped upward tendency during seed germination and reached highest levels at 72 h after sowing (Figure 4). These four enzymes activities of NP+CS were generally lower and had significant differences at 72 h as compared with those of NP+Cn. On the whole, all priming treatments improved four tested antioxidant enzymes activities compared to NP+CS, with a few exceptions such as HP+CS at 24 h in SOD and H₂O₂+CS at 72 h in GR. The relative higher levels of four antioxidant enzymes were all recorded in SA+H₂O₂ among four priming treatments under chilling stress.

The ZmSOD4 expression of NP+Cn increased and got the highest level at 48 h, which was about six-fold higher than NP+CS (Figure 5A). At the end of seed priming, SA+H₂O₂+CS, H₂O₂+CS, and HP+CS significantly enhanced ZmSOD4 expression compared to NP+Cn and NP+CS, and the higher ZmSOD4 level was recorded in SA+H₂O₂. At 48 h, ZmSOD4 expression in primed seeds increased fleetly, and SA+H₂O₂ had significantly higher ZmSOD4 level than other treatments during seed germination except 6 and 24 h.

The expressions of ZmCAT2 and ZmAPX2 in NP+Cn and NP+CS remained at low levels during seed germination after sowing; however, sudden increases happened in NP+Cn at 24 h, which were notably higher than those in NP+CS and priming treatments (Figures 5B,C). All four priming treatments improved ZmCAT2 and ZmAPX2 expression compared with NP+Cn and NP+CS at 0 h, of which SA+H₂O₂+CS kept the higher levels during seed imbibition than other priming treatments.

The ZmGR expression of NP+Cn increased from 0 to 24 h after sowing, and the highest expression were recorded at 24 h in NP+Cn which was about 14-fold higher than NP+CS (Figure 5D). At the end of seed priming, the expression levels of ZmGR were significantly up-regulated, of which SA+H₂O₂ had the highest ZmGR level (32-fold higher than NP+CS) among the four priming treatments (Figure 5D, 0 h). In addition, ZmGR expression in SA+H₂O₂+CS was visibly higher than other three priming treatments from 0 to 24 h after sowing.

The activity of α-amylase in NP+Cn increased gradually during seed germination, and reached higher level at 48 and 72 h than those in NP+CS and other priming treatments (Figure 6A). At the end of priming (0 h), α-amylase activities were improved obviously by primed treatments compared to those non-primed seeds, and seeds treated by SA, H₂O₂, and SA+H₂O₂ owned higher α-amylase levels than those treated by HP.

The expression of ZmAMY in NP+Cn was increased gradually from 0 to 48 h, and the highest value was recorded at 48 h which was about 90-fold higher than that in NP+CS (Figure 6B). ZmAMY expression were up-regulated significantly at the end of priming treatments (Figure 6B, 0 h), then sharply decreased from 0 to 6 h. SA+H₂O₂ remained ZmAMY expression above 100-fold more than NP+CS from 0 to 48 h. After sowing for 72 h, ZmAMY expression of all treatments fell back to a low level.

In addition, the soluble sugar content of NP+Cn increased rapidly to a maximum value at 48 h after sowing, and then accompanied by a slightly decline (Figure 6C). Except 24 h after sowing, priming treatments obviously improved soluble sugar content in maize seeds compared with NP+CS. During seed germination, SA+H₂O₂+CS kept a relatively higher level than other priming treatments except 48 h (Figure 6C).

The ZmNCED1 expression of NP+Cn had its relatively higher value at 48 h as compared with other times after sowing (Figure 7A). At the end of priming (0 h), four priming treatments increased obviously the ZmNCED1 expressions than non-primed seeds, among which the highest level was recorded in HP. During seed germination from 0 to 6 h under chilling stress, ZmNCED1 expression in priming treatments rapidly decreased; however, from 24 to 72 h, SA, H₂O₂, and SA+H₂O₂ presented a gradual upward trend.

ZmCYP707A2 expressions of priming treatments were up-regulated significantly at 0 h as compared with NP+Cn and NP+CS (Figure 7B), in which SA+H₂O₂ up-regulated the expression of ZmCYP707A2 by almost 16-fold than NP+CS. However, ZmCYP707A2 expression decreased rapidly and stayed at low levels during time after sowing in different treatments.

The changing trends of ZmSnRK2.1 and ZmCPK11 expressions in four priming treatments were similar during seed germination (Figures 7C,D). At 6 h after sowing, they reached their highest levels, andZmSnRK2.1 expression in SA+CS, H₂O₂+CS, SA+H₂O₂+CS, and HP+CS were, respectively, 11-fold, 10-fold, 10-fold, and 6-fold higher than NP+CS; while ZmCPK11 expression was, respectively, 33-fold, 50-fold, 65-fold, and 22.5-fold higher than NP+CS. In addition, ZmSnRK2.1 and ZmCPK11 expressions in NP+Cn reached maximum at 24 h, which was, respectively, 35-fold and 58-fold higher than NP+CS.

During seed germination, ZmGA20ox1 and ZmGA3ox2 genes were involved in GA biosynthesis. ZmGA20ox1 expression of NP+Cn reached the highest level at 24 h (about 11-fold higher than NP+CS), and was significantly higher than NP+CS from 24 to 48 h (Figure 8A). At the end of priming treatments (0 h), the expression levels of ZmGA20ox1 were up-regulated visibly. SA+H₂O₂+CS reached the highest ZmGA20ox1 expression level (about 25-fold higher than NP+CS) which were distinctly higher than H₂O₂+CS, SA+CS, and HP+CS treatments. At 6 h after sowing, ZmGA20ox1 expression remained increasing and reached obviously highest levels which were, respectively, 110-fold in SA+H₂O₂+CS, 100-fold in H₂O₂+CS, 90-fold in SA+CS, and 70-fold in HP+CS higher than NP+CS, and then they decreased rapidly.

The highest ZmGA3ox2 expression of NP+Cn was recorded at 48 h, which was significantly higher than NP+CS and other treatments (Figure 8B). Four priming treatments up-regulated ZmGA3ox2 expression compared with NP+CS. Seeds treated with SA, H₂O₂, and SA+H₂O₂ owned higher levels than those treated by HP from 6 to 48 h under chilling stress. And the significantly highest value was recorded in SA+H₂O₂+CS.

The ZmGA2ox1 expression in NP+Cn was always significantly lower than that in NP+CS from 6 to 72 h (Figure 8C). Four priming treatments increased ZmGA2ox1 expression after priming, which, however, decreased rapidly at the beginning of seed germination and significantly down-regulated as compared with NP+CS from 24 to 72 h. SA+H₂O₂+CS owned the lowest value among the four priming treatments.

The ZmRGL2 expressions in NP+Cn and NP+CS stayed at low levels after priming (Figure 8D), the markedly highest level was showed at 24 h in NP+Cn which was about 22-fold higher than NP+CS and other treatments. The expression levels of ZmRGL2 were significantly up-regulated at the end of priming treatments as compared with NP+CS, but then decreased rapidly.

The ZmGID1 and ZmGID2 genes are two positive regulators to mediate GA signaling pathways. The maximum levels of ZmGID1 and ZmGID2 expressions in NP+Cn were observed at 48 h, which were, respectively, about 13-fold and 3-fold higher than NP+CS (Figures 8E,F). After four priming treatments, ZmGID1 and ZmGID2 expressions were significantly up-regulated as compared with NP+Cn and NP+CS at 0 and 6 h, respectively. Seeds treated with SA+H₂O₂, H₂O₂, and SA had higher ZmGID1 and ZmGID2 expression values than those treated by HP (at 48 and 6 h, respectively) and the highest value was recorded in SA+H₂O₂ under chilling stress.