Bermudagrass ‘WBD128’ collected from Xiaojiang City, Zhejiang Province, China, was used in this study. Bermudagrass seeds were surface-sterilized with 75% ethanol for 1 min followed by 10% NaClO for 3 min then washed with sterile water for four times. The sterilized bermudagrass seeds were sown on MS medium containing 1.5% sucrose, pH 5.8. After stratification at 4°C for 3 days in darkness, the MS plates were transferred to the growth chamber (16 h light/8 h dark, 26°C) for seed germination.

Uniform stolons of the genotype ‘WBD128’ were planted in plastic pots, which were kept in the greenhouse for 1 month to establish under growth conditions of 12 h light/12 h dark, 28°C.

For LK stress treatment, 6-day-old bermudagrass plants on MS plate were transferred to LK or MS medium with different concentrations of melatonin and treated for the indicated times described in the figure legends section, and then the fresh weight and primary root length were recorded. The LK medium was modified from MS medium according to the method described previously (Xu et al., 2006). Briefly, 2.99 mM CaCl₂ and 1.25 mM KH₂PO₄ were replaced by 2.99 mM Ca(NO₃)₂ and 1.25 mM NH₄H₂PO₄, 18.79 mM KNO₃ and 20.6 mM NH₄NO₃ were removed, and 1.5 mM MgSO₄ was unchanged. The final K⁺ concentration in LK agar medium was adjusted by adding KCl. In the present study, the K⁺ concentration was 100 μM in the LK medium.

For analysis of low K⁺ stress on the PSII activities of bermudagrass, the established ‘WBD128’ plants were transferred to LK or MS liquid medium with or without melatonin for the indicated times described in figure legends.

The pulse-amplitude modulation (PAM) fluorometer (PAM 2500, Heinz Walz GmbH) was employed to measure chlorophyll (Chl) a fluorescence transient. The measurement was performed as described previously (Chen et al., 2014; Hu et al., 2016). After the collected bermudagrass leaves were pre-adapted in the dark for 30 min, the OJIP transients were triggered by red light (3,000 μmol photons m^-2 s^-1) to guarantee closure of all reaction centers of PS II and to evaluate a real fluorescence intensity. The Chl a fluorescence emission induced by the strong light pulses was subsequently tested and digitized between 10 μs and 320 ms. The OJIP transient data was processed using the JIP-test as reported by Chen et al. (2014).

The measurement of endogenous melatonin contents was performed by enzyme-linked immunosorbent assay (ELISA) as previously described with some modifications (Fan et al., 2015; Hu et al., 2016). Briefly, a 0.3 g of leaf tissues was ground into fine powder in liquid nitrogen, transferred into tubes containing 5 ml of extraction solution (Acetone:Methanol:Water = 89:10:1), and then homogenized on ice for 1 h. After centrifugation at 4500 × g for 5 min at 4°C, the supernatant was transferred to a new tube containing 0.5 ml of 1% trichloric acid. The mixture was centrifuged at 4500 × g for 10 min at 4°C, and the extract was used for quantification of endogenous melatonin using the Melatonin ELISA Kit (EK-DSM; Buhlmann Laboratories AG, Schonenbuch, Switzerland) according to the manufacturer’s instruction.

After growth on MS or LK medium for 2 weeks, shoots and roots of the bermudagrass seedlings were harvested for K⁺ content measurements. The plant samples were washed thoroughly with distilled water, placed in oven at 105°C for 30 min to deactivate the enzymes, and then dried at 80°C for 48 h. After weighing, the samples were ground into a fine power, and then digested in a solution (HClO₄:HNO₃ = 1:4, v/v). Subsequently, the K⁺ content assays were performed using ICP-MS (Thermo Fisher Scientific, Waltham, MA, United States) as previously described (Xie et al., 2017).

The total RNA was isolated from the bermudagrass roots using TRIzol reagent (Invitrogen, United States) according to the manufacturer’s instruction. DNaseI was used to remove any contaminated genomic DNA, and then a 2 μg of total RNA was reversely transcribed into cDNAs using M-MLV reverse transcriptase (Promega, United States). Real-time quantitative RT-PCR was performed by the method described previously by Chen et al. (2012), and the CdACT2 gene was used as a quantitative control. The specific primer sequences were listed in Table 1.

All experiments were performed with at least three repetitions. The significance of differences was determined by ANOVA or Student’s t-test (P < 0.05), as indicated in the figure legends.

To investigate the role of melatonin in plants response to LK stress, the endogenous melatonin production in bermudagrass leaves was measured after being treated with LK stress for 12, 24, and 72 h. The results revealed that the endogenous melatonin contents were remarkably increased for 2- to 4.7-fold during the 12–72 h of LK treatment (Figure 1). The significant increase of endogenous melatonin content by LK stress suggests that melatonin may be involved in plant response to LK stress in bermudagrass.

To further investigate the possible role of melatonin involvement in plant growth and development and LK stress response, we examined the phenotypic changes of bermudagrass under normal (MS) and LK stress conditions with or without melatonin treatment. The 6-day-old seedlings were transferred and vertically cultured on MS or LK medium containing 20 or 50 μm melatonin for 1 week. It was found that under both MS and LK conditions, melatonin-treated plants grew better as characterized by greater fresh weight and higher root length compared with those of non-melatonin-treated plants (Figures 2A–D). The primary root length of 20 or 50 μm melatonin-treated plants (MS+20 or 50 μm melatonin) was 130 and 125% of those of non-treated plants, while under LK condition, the root length of melatonin-treated plants were increased to 156 and 158% of those of non-treated plants (Figures 2A–D), suggesting that melatonin was involved in regulating both plant growth and development and LK response. In addition, the K⁺ contents of bermudagrass plants were measured after they were grown on MS or LK medium with or without 50 μm melatonin application. Under LK condition, melatonin-treated plants accumulated more K⁺ in both shoot and root tissues compared with those of melatonin non-treated plants. By contrast, when plants were grown on MS medium, the higher K⁺ accumulation in the shoots of melatonin-treated plants were observed compared with those of melatonin non-treated plants, whereas there was no significant difference in K⁺ content in roots between melatonin and non-melatonin treated plants (Figure 3). The results suggest that the greater K⁺ content accumulation in melatonin-treated plants may contribute to the LK-tolerant phenotype and improved growth and development of bermudagrass.

Next, the expression of genes involved in potassium transport was investigated by RT-PCR. The results revealed that expression levels of the three tested gene were significantly induced by LK stress or melatonin treatment, and further increased to higher levels after LK stress and melatonin combined treatment (Figure 4). These results suggested that the higher K⁺ content accumulation in melatonin-treated plants may result from the increased expression of K⁺ transport genes.

To investigate whether melatonin is involved in regulation of PSII activities under LK stress, OJIP fluorescence transient curves were tested. As shown in Figure 5, the OJIP fluorescence transients were significantly affected by LK stress. The OJIP curves dropped remarkably after LK treatment compared with those of plants that were grown in control condition (MS liquid medium). There were no significant differences in the OJIP transient curves between plants grown in MS medium with or without melatonin application for 1 week or 2 weeks. However, the OJIP curves were obviously higher in plants grown in LK with melatonin (LK+Mel) than those of plants grown in LK medium without melatonin application for 1 or 2 weeks (Figure 5), suggesting that melatonin played important roles in PSII against LK stress. Besides, the OJIP curves of plants grown in LK medium for 2 weeks were significantly lower than those of plants grown in LK medium for 1 week, revealing that PSII activity was further decreased during plants exposure to LK stress for prolonged days.

The JIP-test was further applied to study the photosynthetic behavior of PSII. The results showed that the photosynthetic behavior of PSII was significantly different between plants treated with or without melatonin under low K⁺ stress, but not under control condition (MS) with or without exogenous melatonin (Figures 6–8). In brief, the structural and functional parameters of photosynthetic behavior including performance indexes, flux ratios and specific energy fluxes were improved in melatonin-treated plants under LK condition.

The performance indexes (PI) such as PI_ABS, PI_Total, and PI_CS are widely used to describe the overall activity of PSII. As shown in Figure 6, the values of PI_ABS, PI_Total, and PI_CS were all significantly lower in plants treated by LK stress than those of plants grown under MS condition. For instance, the value of PI_ABS reduced remarkably after LK treatment for 1 week, and it was further decreased after LK for 2 weeks. However, it was reversed by melatonin application during LK treatment (Figure 6). Similar results were also observed in the parameters, such as PI_Total and PI_CS (Figure 6). These results suggested that exogenous melatonin played a protective role in PSII activity under LK stress.

For the flux ratios, φP₀ (maximum quantum yield of primary photochemistry) decreased significantly under LK stress, and it further declined as the LK treatment time prolonged. After melatonin application during LK stress, the parameter of φP₀ was higher than that of LK treatment (Figure 7). Similarly, the parameters of other flux ratios such as ΨE₀ (Efficiency/probability with which a PSII trapped electron is transferred from Q_A to Q_B) and φE₀ (quantum yield of the electron transport flux from Q_A to Q_B) displayed the alteration pattern similar to φP₀ (Figure 7). These results imply that melatonin plays positive roles in improving the behavior in both electron donor side and acceptor side of PSII when plants were treated by LK stress. The parameters of specific energy fluxes, including ET_O/RC [Electron transport flux (further than QA^-) per RC] and DI_O/RC (Energy fluxes for dissipation) were further calculated. The ET_O/RC value was lower in the plants of LK treatment than that of MS treatment, and a slight but not significant increase was observed in the plants treated with melatonin when the plants were treated for 1 week (Figure 8). After the plants were treated for 2 weeks, a remarkable decrease of ET_O/RC was observed in the plants by LK treatment, and melatonin mitigated the damage effectively (Figure 8). On the contrary, DI_O/RC increased slightly in the plants after they were treated by LK stress for 1 week. When the plants were treated for 2 weeks, DI_O/RC showed the highest value under LK condition among four different treatments, while DI_O/RC decreased in LK+Mel regime (Figure 8).