Salt-tolerance screening in Limonium sinuatum varieties with different flower colors

Limonium sinuatum, a member of Plumbaginaceae commonly known as sea lavender, is widely used as dried flower. Five L. sinuatum varieties with different flower colors (White, Blue, Pink, Yellow, and Purple) are found in saline regions and are widely cultivated in gardens. In the current study, we evaluated the salt tolerance of these varieties under 250 mmol/L NaCl (salt-tolerance threshold) treatment to identify the optimal variety suitable for planting in saline lands. After the measurement of the fresh weight (FW), dry weight (DW), contents of Na+, K+, Ca2+, Cl−, malondialdehyde (MDA), proline, soluble sugars, hydrogen peroxide (H2O2), relative water content, chlorophyll contents, net photosynthetic rate, and osmotic potential of whole plants, the salt-tolerance ability from strongest to weakest is identified as Pink, Yellow, Purple, White, and Blue. Photosynthetic rate was the most reliable and positive indicator of salt tolerance. The density of salt glands showed the greatest increase in Pink under NaCl treatment, indicating that Pink adapts to high-salt levels by enhancing salt gland formation. These results provide a theoretical basis for the large-scale planting of L. sinuatum in saline soils in the future.


Determination of physiological indicators. Determination of FW, DW and relative water content of
leaf.. After cleaning the leaves with 10 mM calcium chloride solution 25 followed by deionized water, FW of the leaves was measured immediately and DW was obtained following incubation at 105 °C for 15 min and drying to constant weight at 70 °C for 2 days 23 . Five replicates were performed per variety and treatment. The reduction rate was calculated as (FW under control condition-FW under saline condition)/FW under control condition × 100%. The same method was processed in calculating the reduction rate of DW. The relative water content is calculated as (FW-DW)/FW × 100% 26 .
Determination of sodium ion (Na + ), potassium ion (K + ), calcium ion (Ca 2+ ), and chloride ion (Cl − ) contents.. The ion contents in the samples were measured according to Higinbotham 27 . In brief, leaf tissue (0.5 g FW) was collected from plants under both 0 and 250 mmol/L NaCl treatment for all five varieties. The tissues were ashed, dissolved in HNO 3 , and the contents of Na + , K + , and Ca 2+ measured using a Flame photometer (FP6440, Yuanxi, Shanghai, China). Cl − content was measured by ion chromatography according to Wang 28 . Briefly, after boiling for 30 min and filtering through a 0.22 μm filter membrane, the ion solution was injected into an ion chromatograph (ICS-90A, ThermoFisher, Massachusetts, USA) to measure Cl − contents. The ion concentration is shown as mmol/g FW. Five replicates were performed for each variety and treatment. Given that Na + content increased under NaCl treatment, the increase rate was calculated as (Na + content under saline condition-Na + content of control)/Na + content of control × 100%. The same calculation method was applied in the increase rate of Cl − .
Determination of proline content, osmotic potential, malondialdehyde (MDA), hydrogen peroxide (H 2 O 2 ), soluble sugars, and chlorophyll content, and photosynthetic rate.. Proline content was determined in accordance with Demiral 29 . The plant tissue was ground, and ground tissue (0.5 g FW) was added to 10 mL of 5% acetic acid and 40 mL of distilled water. After filtering, the filtrate (8 mL) was mixed with 0.8 g zeolite with shaking for 5 min and centrifuged for 10 min (1500 g). The supernatant (3 mL) was combined with glacial acetic acid (3 mL) and ninhydrin reagent (3 mL) and boiled for 1 h. Benzene (3 mL) was used for static layering, and the upper colored liquid was collected and used to measure optical density at 515 nm. The proline content was calculated from a standard curve based on the optical density. Five replicates were performed for each variety and treatment.
The osmotic potential was measured as described by Tomlinson 30 . Fresh leaf tissue (0.5 g) was cut into small pieces, frozen in liquid nitrogen, and placed into a syringe to squeeze out and collect the cell sap. A freezing point osmometer (SMC 30C-1, Tianhe, Tianjin, China) was used to measure the osmotic potential of the plant cell sap. The formula used to calculate osmotic potential is − iCRT (R = 0.0083143 L Mpa Mol −1 K −1 , T = 298 K). Five replicates were performed for each variety and treatment.
The MDA content was determined as reported in Hong 31 . Leaf tissue (0.5 g) was collected and homogenized in 5 mL 0.1% TCA. The homogenate was transferred to the test tube, combined with 5 mL 0.5% thiobarbituric acid solution, and boiled for 10 min. The sample was centrifuged at 1500 g for 15 min, and the optical density of the supernatant was measured at 532 nm and 600 nm. MDA content (mmol/g FW) = ΔAN/155 W, ΔA is the difference between A 532 and A 600 ; N is the total volume of the supernatant; 155 is the absorption coefficient of www.nature.com/scientificreports/ 1 mmol reaction product at 532 nm; W is the fresh weight of the plant material (g). Five replicates were performed for each variety and treatment. The content of H 2 O 2 was determined as described by Vergara 32 . In brief, fresh leaves (0.3 g) was grinded in 5 mL precooled acetone before centrifuged at 500 g for 8 min. Afterward the supernatant (1 mL) was mixed with  in the sample checked on the standard curve (μmol), V T was the total volume of sample extract (mL), V 1 was the volume of sample extract (mL), and FW was the fresh weight of plant tissue (g). Five replicates were performed for each variety and treatment. Soluble sugars were measured following the protocol of Prado 33 . Fresh leaf tissue (0.3 g) was dissolved in 10 mL of double distilled H 2 O (ddH 2 O) in a boiling water bath for 50 min, filtered, and brought to a volume of 25 mL. Afterward 0.5 mL of extract solution was combined with 1.5 mL distilled water, 0.5 mL ethyl anthrone acetate, and 5 mL concentrated sulfuric acid, shaken thoroughly, boiled in water bath for 1 min, and cooled. The optical density of the solution was measured at 630 nm. The soluble sugars content was calculated from a standard curve. Five replicates were performed for each variety and treatment.
Chlorophyll levels were determined referring to Maxwell 34 . Leaf tissue (0.3 g) was combined with 5 mL dimethyl sulfoxide in 5 mL 80% acetone and incubated in a 65°C water bath at 24 h (protected from the light) to fully decolorize. Afterward bring to 25 mL after filtration and the solution was used to measure the optical density at 663 nm, 645 nm, and 470 nm. Chlorophyll content (mg g −1 or mg dm −2 ) = CV/1000A, C is chlorophyll concentration (mg L −1 or mg dm −2 ); V is the total volume of extract solution (mL); A is fresh weight of the sample (g) or sampling area (dm −2 ). The pigment concentration (mg/L) was calculated as C a = 12.7A 663 − 2.69A 645 ; C b = 22.9A 645 − 4.68A 663 ; C total = 20.0A 645 + 8.02A 663 ; C XC = (1000A 470 − 3.27C a − 104C b )/229; C a , C b are the concentrations of chlorophyll a and b, C total is the concentration of total chlorophyll; C XC is the total concentration of carotenoids. Five replicates were performed for each variety and treatment.
The photosynthetic rate was measured on the basis of Wang 35 . In this experiment, a photosynthetic instrument (LI-6400XT, LI-COR, Nebraska, USA) was used to measure the photosynthetic parameters of leaves. The photosynthetic effective quantum density, U PAR (μmol m −1 s −1 ), μ is 4.55 36 , was measured at a temperature of 23°C, and the leaf area of each cultivar was 1 cm 2 . Five replicates were performed for each variety and treatment.
Determination of salt gland density in different varieties. The density of salt glands was measured according to Yuan 37 . The leaves were fixed in a mixture of ethanol and acetic acid (3:1; v/v), rinsed with 70% ethanol to decolorize, and cleared in Hoyer's solution. Afterward cleaned leaves were fixed on a glass slide for DIC microscopy (ECLIPSE 80i, Nikon, Tokyo, Japan). The salt gland density was calculated according to Ding 38 and expressed as number per mm 2 . Five replicates were performed for each variety and treatment.

Data analysis
Statistical analysis and correlation were performed using SPSS 13.0 software (SPSS Software Inc., USA). The results were subjected to a one-way analysis of variance (ANOVA), and Duncan's test was used to determine significant differences between the means (P = 0.05). In the figures, the error bars represent the means ± standard deviations (n = 5) and different letters indicate significant differences at P = 0.05. Correlation is processed at P = 0.05 and 0.01 using Pearson correlation analysis. The figures were generated using SigmaPlot 12.5 (Systat Software, Chicago, IL, USA).

Results
Identification of the salt-tolerance threshold. Plant biomass is an important measure of salt tolerance. Figure 2 shows the biomass of the aerial parts of plants under a gradient of different NaCl concentrations (0, 100, 200, 300, and 400 mmol/L) after 2 weeks of treatment. FW and DW were measured in the five varieties of L. sinuatum seedlings and constructed a regression curve based on the means of five data under different treatments as independent variables. Most studies use the salt concentration at which plant growth or biomass decreases by 50% of the control value as the salt-tolerance threshold 39 . Here, when the FW and DW of the www.nature.com/scientificreports/ upper parts of the seedlings were reduced by 50% of the non-NaCl treatment value, different varieties showed different salt-tolerance thresholds. The highest threshold was obtained for Pink (250 mmol/L) (Supplementary Fig. 1), suggesting that Pink is the most salt-tolerant variety. To identify the optimal salt concentration for further experiments, we calculated the average salt-tolerance threshold, i.e., 228 mmol/L for FW and 233 mmol/L for DW (Fig. 2). Therefore, a salt-tolerance threshold of 250 mmol/L was utilized in subsequent experiments.
Pink shows the best growth under 250 mmol/L NaCl treatment. All varieties showed inhibited growth under 250 mmol/L NaCl treatment (Fig. 3), but the changes in FW and DW showed no significant trends among varieties ( Supplementary Fig. 2). In order to make effective comparison among different varieties, FW and DW reduction rate are calculated to compare the changes between control and saline condition (Fig. 4). Pink showed the least FW reduction, followed by Yellow, Purple, White and Blue, while White has the least DW reduction, afterward Pink, Yellow, Purple and Blue. Based on the reduction rate of FW and DW, Pink is considered the most salt tolerance variety, followed by Yellow, Purple, White, and Blue. Biomass can be used as a measure of plant growth, and various physiological processes could be responsible for the ability of Pink to maintain growth in the presence of salt. Therefore, we measured the physiological indicators of the different varieties under NaCl treatment in order to reveal the underlying salt-tolerance mechanisms.

Effect of NaCl treatment on different physiological indicators in five varieties.
Comparisons of the Na + , K + , Ca 2+ , and Cl − contents; MDA, soluble sugars, proline contents, H 2 O 2 content and relative water content of leaf; chlorophyll contents; and osmotic potential and photosynthetic rate are shown in Figs. 5, 6, 7 and 8, respectively. Each variety showed significant changes under NaCl treatment. Figure 5 shows a comparison of the relative Na + , K + , Ca 2+ and Cl − contents among varieties under 250 mmol/L NaCl treatment. Na + and Cl − content under NaCl treatment increased compared with the control in all varieties (Fig. 5a,c), while K + and Ca 2+ showed various trends in different varieties (Fig. 5e,f). Given that Na + and Cl − are considered the stress ion to protoplast and different varieties have various basal level under control, the increase rate of Na + (Fig. 5b) and Cl − (Fig. 5d) under NaCl treatment are calculated in each variety in order to make intuitive comparison among different varieties. Pink shows the least Na + increase (1.87%) under saline condition, followed by Yellow, Blue, Purple and White. In the aspect of Cl − increase, Purple (96.04%) has the least, afterward Yellow, Pink, White and Blue. Together with the increase rate of Na + and Cl − , these results indicate that Pink accumulates less Na + and Cl − than the other varieties under high-salt conditions, which should lead to less injury than the other varieties. Figure 6 shows a comparison of the relative MDA, soluble sugars, proline contents, H 2 O 2 and relative water content of leaf among varieties. Though the MDA increase rate of Pink under saline condition (Fig. 6b) shows the most, the absolute value of MDA (Fig. 6a) is the least accumulation in Pink, indicating that Pink suffered the  (Fig. 6a), which may explain the serious damage level.
To cope with the damage caused by NaCl treatment, cells usually accumulate organic osmotic regulating substance such as soluble sugars (Fig. 6c) and proline (Fig. 6e). High accumulation of soluble sugars is shown in Pink and Purple, and the increase rate (Fig. 6d) under saline treatment indicates the comparison between varieties. Highest increase rate is detected in Purple, followed by Pink and Yellow. Besides, high proline accumulation is shown in Blue, Yellow and Pink in descending order of actual value (Fig. 6e), while the increase rate has the opposite trend with the most in White, Blue and Pink (Fig. 6f). Proline reduces the osmotic potential in the cell 41 , allowing it to resist external osmotic stress, thereby improving plant survival in adverse environments 29 . Proline content depends on the catabolism of sugar 42 .Combined the absolute value and the increase rate, Pink is considered to accumulate a large amounts of soluble sugars and proline to improve the osmotic adjustment ability under salt treatment. The large accumulation of osmoregulation substances can effectively reduce the osmotic potential under NaCl treatment 43 . Figure 8 (a) shows that the osmotic potential decline markedly in all varieties, and Yellow and Pink have the most reduction rate (Fig. 8b).
Moreover, H 2 O 2 , as a kind of superoxide, can cause oxidative stress to plants under various stresses 44 . In Fig. 6g, under salt stress, the lowest H 2 O 2 generation is detected in Pink, while higher in White and Purple, which indicates that Pink suffers the least oxidative stress under salt treatment and is more suitable for saline environment. In addition, relative water content is also measured (Fig. 6h) and no significant difference is detected among different varieties, indicating that all varieties of L. sinuatum can keep normal moisture condition to cope with physiological drought of NaCl 45 .
Plants always produce large amounts of pigments under saline environment to maintain normal photosynthetic efficiency. In addition, a positive correlation was detected between chlorophyll content and net photosynthetic rate 46 . Figure 7 shows a comparison of the relative chlorophyll content among varieties. In order to show the changes in pigment content in more detail, the changes in total chlorophyll, chlorophyll a, chlorophyll b, and carotenoid contents are shown. The pigment contents of the Pink and Purple varieties were high, which help improve the photosynthetic rate. Under NaCl treatment, net photosynthetic rate is obviously inhibited in all varieties (Fig. 8c). Pink shows the highest under saline condition, and the reduction rate under salt treatment is also the lowest in Pink (Fig. 8d). These results further explain why Pink has the highest biomass under NaCl treatment, which may be due to high accumulation in osmoregulation substance and high photosynthetic efficiency.

Effects of NaCl on salt gland density of five varieties of Limonium sinuatum. Salt glands are
structures for salt secretion that are specifically produced by recretohalophytes 47 . We therefore performed statistical analysis of the salt gland densities of expanded leaves of the five L. sinuatum varieties under NaCl treatment. Figure 9a shows the changes in salt gland density in the leaves of the five varieties of L. sinuatum under salt stress (images shown in Supplementary Fig. 3). Salt gland density increased in all varieties under NaCl treatment compared to the control. The density of salt glands in Pink increased by 225.86% (Fig. 9b), while in Purple only 7%. The increased salt gland can help the plants to excrete more Na + outsides in order to further decrease the Na + accumulation in vivo.
Finally, given that Pink showed the greatest tolerance to NaCl treatment, in order to verify the optimal variety suitable for growing in field, we examined flowering in the five varieties grown in Yellow River Delta (salt content: 0.2%). After six months of growth, only Pink and Yellow plants flowered consistently, whereas the three other varieties rarely flowered and only showed vegetative growth (Fig. 10). These results suggest that Pink and Yellow www.nature.com/scientificreports/ are the optimal varieties for the development of saline horticulture and further planting in saline soil, which is consistence with the formal results in laboratory conditions.

Discussion
L. sinuatum is a pioneer plant that could be used for the improvement of saline lands due to the high salt resistance and various colors 48 . Therefore, it is important to identify the best salt-tolerant varieties for cultivation in these areas. In the current study, Pink showed the highest biomass and the strongest salt resistance among the five varieties examined. Our analysis of physiological indicators including Na + , K + , Ca 2+ , and Cl − contents; MDA, soluble sugars, proline contents, H 2 O 2 content, relative water content and chlorophyll contents, osmotic www.nature.com/scientificreports/ potential, photosynthetic rate, and salt gland density under 250 mmol/L NaCl (salt-tolerance threshold) treatment explained why Pink has the greatest salt tolerance. Moreover, analysis of plants grown in the field (Fig. 10) confirmed the superior salt resistance of Pink. Pink is recommended as the optimal varieties for extensive planting and greening in saline soil, followed by Yellow. Plant biomass is an important indicator of salt tolerance 43 : the greater the increase in biomass under salt stress, the higher the salt tolerance. As shown in Fig. 4, Pink showed the minimal reduction under 250 mmol/L NaCl treatment, and photosynthetic efficiency of Pink was also the highest among five varieties (Fig. 8), indicating that Pink suffered the least damage under saline condition. The three basic components of salt stress are usually considered as ionic toxicity, osmotic stress and oxidative stress 49 .
How Pink cope with high ionic toxicity? On the one hand, under salt stress, Na + content increases, which affects the absorption of K + and Ca 2+ . K + plays an important role in the osmotic regulation of cells 50 . Ca 2+ regulates the ionic balance and reduces the absorption of Na +51 . In salt-tolerant plants, K + efflux is significantly inhibited under salt stress to maintain high intracellular K + /Na + levels, thereby reducing the damage from salt stress 52 . After salt treatment, the Pink variety had the relatively low Na + and Cl − contents among the five varieties, whereas K + and Ca 2+ showed the opposite trend. Therefore, Pink regulates ionic balance under salt stress, maintaining high K + /Na + levels, thus showing strong salt tolerance. On the other hand, salt gland is the typical and specific epidermal structure of recetohalophytes 53 , which can excrete the excessive Na + out of the plants to avoid damage 54 . The most salt gland was induced in Pink under salt treatment (Fig. 9), so it is speculated that Na + can be effectively transferred out of the cell to further avoid ionic toxicity.
NaCl can induce the physiological drought due to the osmotic stress. Compatible media was always generated to deal with the osmotic stress, such as proline and soluble sugars. Proline is an important compatible solute in plant cells that protects enzymes from inactivation by NaCl and reduces the osmotic potential in the cells 55 , thereby helping plants resist external osmotic stress and tolerate adverse environments 29 . Sugar content under stress condition is intricately associated with carbohydrate content of plant 25 . Though not always the highest   www.nature.com/scientificreports/ accumulation in proline and soluble sugars, Pink keep the relatively high level of osmotic adjustment substance to reduce the osmotic potential, which allowed the plant to re-absorb water from the NaCl solution to maintain normal growth. Oxidative stress is inevitable induced by salt stress due to the generation of superoxide, with the typical representative H 2 O 2 56 . Pink suffered the smallest oxidative stress under salt treatment and is more suitable for saline environment. MDA content can reflect the degree of membrane damage and the effects of stress on plants 57 . The least accumulation of MDA also insist the opinion that Pink suffered less oxidative stress.
Given that 16 physiological indicators were measured under salt treatment in addition to DW and FW, which one is most closely related to biomass? Correlation analysis was performed between FW and the other 14 indicators. As shown in Table 1, photosynthetic rate showed the strongest positive correlation with salt resistance in L. sinuatum.
Together, the results of our experiments show that the Pink variety has the strongest salt tolerance. Based on these results, the level of salt tolerance of the five varieties of L. sinuatum from high to low is Pink, Yellow, Purple, White, and Blue. Further analysis of the performance of these varieties in saline soil also validate the results of these preliminary experiments. Nonetheless, our findings provide a theoretical basis for the cultivation of L. sinuatum in saline-alkali areas, which could be widely planted to facilitate the greening and transformation of saline soils.