Variations in morphology, physiology, and multiple bioactive constituents of Lonicerae Japonicae Flos under salt stress

Lonicerae Japonicae Flos (LJF) is an important traditional Chinese medicine for the treatment of various ailments and plays a vital role in improving global human health. However, as unable to escape from adversity, the quality of sessile organisms is dramatically affected by salt stress. To systematically explore the quality formation of LJF in morphology, physiology, and bioactive constituents' response to multiple levels of salt stress, UFLC-QTRAP-MS/MS and multivariate statistical analysis were performed. Lonicera japonica Thunb. was planted in pots and placed in the field, then harvested after 35 days under salt stress. Indexes of growth, photosynthetic pigments, osmolytes, lipid peroxidation, and antioxidant enzymes were identified to evaluate the salt tolerance in LJF under different salt stresses (0, 100, 200, and 300 mM NaCl). Then, the total accumulation and dynamic variation of 47 bioactive constituents were quantitated. Finally, Partial least squares discrimination analysis and gray relational analysis were performed to systematically cluster, distinguish, and evaluate the samples, respectively. The results showed that 100 mM NaCl induced growth, photosynthetic, antioxidant activities, osmolytes, lipid peroxidation, and multiple bioactive constituents in LJF, which possessed the best quality. Additionally, a positive correlation was found between the accumulation of phenolic acids with antioxidant enzyme activity under salt stress, further confirming that phenolic acids could reduce oxidative damage. This study provides insight into the quality formation and valuable information to improve the LJF medicinal value under salt stress.

www.nature.com/scientificreports/ VIX DS-11, USA). The contents of photosynthetic pigments (chlorophyll a, b, total chlorophyll, and carotenoids) were calculated according to the following formula: Proline, soluble proteins, total AA, and MDA assay. Fresh leaf samples (0.1 g) were powdered with a 1 mL sulfosalicylic acid dihydrate (SA) and transferred to a tube for 10 min heating in the boiling water bath. Afterward, centrifuged at 10,000 g for 10 min, then the supernatant was collected to detect proline concentrations at 520 nm using UV-Vis spectrophotometer (DENOVIX DS-11, USA) 25,26 . Dilute 1 mg standard substance to 15, 10, 8, 6, 4, 2, 1, 0 μg/ml, respectively. Then establishing a standard curve according to the absorption value and concentration of the standard substance. Fresh leaf samples (0.5 g) were homogenized and extracted by 4.5 mL PBS (0.1 mol/L, PH: 7.4) in an ice-water bath. After centrifugation, the supernatant, 563 μg/mL protein standard solution, and blank solution (double distilled water) were detected using a UV-Vis spectrophotometer (DENOVIX DS-11, USA) at 562 nm, which could display the concentration of protein 27 .
Fresh leaf tissues (0.5 g) were homogenized and 4.5 mL normal saline (NS) was added, then centrifuged at 685 g for 10 min, and then took the supernatant to be measured. 1 mL supernatant mixed with 2 mL amino acid reaction solution and 1 mL amino acid chromogenic agent, and the mixture was centrifuged at 685 g for 10 min, the supernatant was then taken to measure the absorption value at 650 nm using UV-Vis spectrophotometer (DENOVIX DS-11, USA) 25 .
The content of MDA was measured by Thibabituric Acid (TBA) method 28 , which represents lipid peroxidation. 0.5 g fresh leaf tissues were grounded with 4.5 mL TBA, and centrifuged at 685 g for 10 min; 2 mL aliquot of the supernatant was mixed with 0.5% TBA, and then heated in boiling water for 40 min, quickly cooled with ice-bath. After centrifugation at 685 g for 10 min, the supernatant was prepared for the assay of MDA and the absorbance at 532 nm was recorded.
Enzyme extraction and activities assay. The fresh leaf of L. japonica under different salt stresses were collected for determining the antioxidant activity. The leaves (0.5 g) were homogenized with 4.5 mL PBS in an ice water bath and centrifuged at 350 g (catalase, CAT), 685 g (superoxide dismutase, SOD and peroxidase, POD), and 5595 g (ascorbate peroxidase, APX) for 10 min, respectively. The supernatant was used to examine the activities of the enzyme.
The activity of SOD was assayed at 550 nm using the hydroxylamine method; the activity of POD was determined at 420 nm according to the colorimetric method 25,26 ; the activity of CAT was measured at 405 nm based on the ammonium molybdate method, and the activity of APX was determined at 290 nm. All of the experiments were carried out by assay kits. The amount of 200 μL reaction solution was measured by multi-mode microplate reader using UV-Vis spectrophotometer (DENOVIX DS-11, USA).
Chemicals and reagents. The purity of 47 standard compounds was more than 98%; the structures and source of 47 standard compounds were shown in Fig. S2 and Table S1; All of the chemicals and reagents were reported in our previous study 29,30 . Preparation of sample solutions. Fresh flower buds were harvested on the 20th, 25th, and 35th day, respectively, and then placed under natural dry conditions. All dried flower buds were powdered and sieved through 50 mesh. After accurately weighed powder samples (1.0 g) were extracted using ultra-sonication with 40 mL 70% methanol for 45 min, the loss of solvents was compensated with 70% methanol and mixed well. Then centrifuged at 8050 g for 10 min and filtered 0.22 nm membrane filters. The supernatants were stored in a sample bottle at 4 ℃ before UFLC-QTRAP-MS/MS analysis.
Method validation and sample determination. The standard solution containing 47 compounds was prepared and the calibration curves were constructed using serially diluted standards. The LODs and LOQs of compounds were defined as signal-to-noise (S/N) ratios of 3:1 and 10:1, respectively. Method precision was calculated as the RSD of compound concentrations determined by inter-day precision. Moreover, repeatability was achieved by evaluating the closeness of a set of analytical results, which were obtained for six different rep- The mean values and standard deviation were calculated from the measurements of five replicates. The PLS-DA was made by SIMCA-P 13.0 (https ://www.sarto rius.com/en/produ cts/proce ss-analy tical -techn ology /data-analy tics-softw are/mvdasoftw are/simca ). The heat map was made by the software of Heml 1.0 Heatmap Illustrator (http://www1.heatm apper .ca/expre ssion /). One-way ANOVA followed by the Student-Newman-Keuls test was used to compare the means with the significance level by IBM SPSS Statistics 22 (https ://www.ibm.com/analy tics/spss-stati stics -softw are). Pearson Correlation Coefficient was also conducted using IBM SPSS Statistics 22 (https ://www.ibm.com/ analy tics/spss-stati stics -softw are), and then the difference was considered significant at p < 0.05. The other data were calculated by Origin Pro 9 (https ://www.origi nlab.com/index .aspx?go=Suppo rt&pid=2051).

Results
Changes in morphology under salt stress. The growth parameters were shown in Table 1 and phenotype profiles of representative L. japonica presented in Fig. S3. Both leaf diameter and plant height decreased with the increasing salinity. Although average flower buds length and fresh/dry weight per 100 flower buds under high salt stress were the lowest compared with the control group, they all increased at low salt stress, indicating that salt stress had a remarkable effect on the growth parameters of L. japonica. According to the description about the grade of salt stress injury to plants, low salt stress led to the minimum damage. Fig. 1, the content of photosynthetic pigments increased firstly and then decreased except carotenoid; but the increments were insignificant. Among them, low salt stress occupied the highest values, indicating that this treatment significantly increased chlorophyll a, chlorophyll b, total chlorophyll, and carotenoids concentration.

Changes in osmolytes, and lipid peroxidation under salt stress.
In the experiment, we identified osmolytes including soluble sugars, soluble proteins, and proline; both soluble proteins and proline had higher values under low salt stress ( Fig. 2A,B). The contents of proline also increased compared with the control group. This increase occurred in the early stage of salt stress with low salinity, while with the lasting salinity increase, the proline concentration showed the opposite trend. It is well acknowledged that the content of MDA was quantified as the estimate of membrane lipid peroxidation, which represented the degree of cell membrane damage. As shown in Fig. 2C, MDA continued to accumulate when salinity increases from 100 to 300 mM. This phenomenon indicated that L. japonica tolerated salinity, and the oxidative damage peaked under 300 mM salt stress. Amino acids can provide the material basis for normal growth, metabolism, and life support. It can be seen from Fig. 2D that the content of total amino acids reached the maximum 43.31 μmol/mgprot, and then almost straight down. Fig. 3 showed that the activity of POD, SOD, CAT, and APX varied with different concentrations of NaCl which peaked at 100 mM NaCl treatment. POD activity peaked at 61.68 U/mg, but suddenly declined by 2.98-fold from 100 to 300 mM. CAT activity induced by 3.2-fold at 100 mM compared with controls. And APX activity peaked at 1.26 U/mg firstly, then declined by 7.38-fold compared with controls which accompanied with increasing salinity from 100 to 300 mM.

Optimization of UFLC-QTRAP-MS/MS conditions. The optimum UFLC-QTRAP-MS/MS conditions
were modified based on our previous work 29 . The most abundant and specific fragment ions were selected as MRM transition from MS/MS and the detailed information of 47 compounds was summarized in (Table S2). The representative extract ion chromatograms of 47 active constituents in the MRM mode were shown in Fig. 4A; the total ion chromatograms (TIC) were presented in Fig. 4B. www.nature.com/scientificreports/ Method validation. The detailed information of the method validation for UFLC-QTRAP-MS/MS was listed in Table 2. All of the compounds showed good linearity with r > 0.9990. The LODs and LOQs of 47 constituents ranged from 0.002-167.42, 0.008-552.48 ng/mL, respectively. The RSD values of precision, repeatability, stability of 47 constituents were less than 5%. The recoveries were measured in the range of 94.2-105.2%.

The accumulation of bioactive constituents of all groups in LJF.
The detailed changes of bioactive constituents during 35 days in LJF were obtained. According to Table S3 and Fig. 5, the variation trends of amino acids, nucleosides, organic acids, and flavonoids increased and then decreased. Conversely, iridoids concentration peaked under 300 mM, such as loganin secoxyloganin and morroniside increased 3.98, 3.10, and 9-folds compared to controls, respectively. The contents of chlorogenic acid and its derivatives were also enhanced under salt stress, including isochlorogenic A and B were elevated 2.32 and 2.31-folds, respectively. Among them, the trends of chlorogenic acid, neochlorogenic acid, isochlorogenic A, and isochlorogenic B were accordant, which indicated that the accumulation and changes of pathway-related constituents interacted with each other. From the chemical fingerprint and the content data, the chemical fingerprint was similar, but the content of individual constituents or total contents was different under different salt stress. The contents of organic acids were the most abundant under different salt stress. And the total LJF biomass reached a peak by 1.1-fold after 35 days under low salt stress compared with controls. Altogether, we found that the environment had an obvious effect on the accumulation of bioactive constituents, the accumulation of active constituents was the highest under low salt stress.
Dynamic accumulation of multiple bioactive constituents. As shown in Fig. 6, although the accumulation of bioactive constituents under four salinity levels increased with time (20, 25, and 35 days, respectively), the dynamic accumulation of these constituents still exhibited some differences. As the accumulation of bioactive constituents peaked at 100 mM, we supposed that low salt stress might be the optimum condition  www.nature.com/scientificreports/ for accumulating bioactive constituents of LJF, which represented the highest medicinal. As one of the indicator constituents to evaluate the quality of LJF, the dynamic accumulation and changing trend of chlorogenic acid was firstly increased and then decreased in the whole process. And the incremental accumulation became greater with time prolonging under 100 mM treatment. This also suggested that the environment has a great impact on the accumulation of active constituents and the different accumulation patterns were presented in the four different salt stress levels. In the long run, low salt stress (100 mM) instead of high salt stress (300 mM) was beneficial for biomass accumulation and growth of LJF. According to the traditional viewpoint, phenolic constituents were regarded as secondary antioxidants and played a vigorous role in detoxifying ROS. In this study, we found that soil salinity considerably affected phenolic acid contents and antioxidant activity. In Fig. 7, the contents of phenolic acids were positively correlated with antioxidant enzymes activity (SOD, POD, CAT, and APX activity) under salt stress, which were decreased with the reducing phenolic acid contents from 100 to 300 mM, indicating that the phenolic acids positively correlated with oxidative damage reduction. Pearson Correlation Coefficient (r) was performed to further analyzed the correlations between phenolic acid contents and antioxidant enzymes under salt stress. the results showed that the r-value of SOD, POD, CAT, and APX activity were 0.994, 0.989, 0.532, and 0.416, respectively. Among them, the correlation between phenolic acid contents and SOD, POD activity was significant correlation at 0.01, 0.05 level, respectively.
Multivariate statistical analysis. The hierarchical clustering heat map displayed the changes in the accumulation of 47 constituents under different salt stresses. As shown in Fig. 8A, the changes of 47 constituents in LJF indicated that the environment influenced the accumulation of bioactive constituents. And the specific cases in the control group and salt stress groups were separated. In the different salt stress groups, 100 mM and 200 mM group were firstly clustered together and then separated. Finally, the four different salt stress groups were completely separated. www.nature.com/scientificreports/ The PLS-DA was performed to cluster and classify LJF under different salt stresses. Samples were divided into four groups with R 2 X = 0.959, R 2 Y = 0.974, and Q 2 = 0.871, respectively (Fig. 8B); of which the values represented excellent fitness (R 2 X/ R 2 Y) and high predictive ability (Q 2 ). And in the PLS-DA loading plot (Fig. 8C), L-glutamate, L-lysine, neochlorogenic acid, sweroside, loganin acid, secologanic acid, and secoxyloganin possessed great loading values of ions, enhancing to separate samples.
Due to GRA prefers to deal with complex relationship problems with multiple factors and variables like traditional Chinese medicine, it was carried out to evaluate the quality of LJF based on the contents of 47 bioactive constituents. Grey comprehensive evaluation values (ri) and quality ranking were listed in Table 3, it showed that the quality ranking of LJF was 100 mM at the top, followed by 0, 200, and 300 mM; and the ri values were 0.75, 0.52, 0.29, and 0.25, respectively. It manifested LJF could be grown in soil with 100 mM NaCl and obtained the optimum quality.

Discussion
The changes in morphology, physiology, and metabolites of medicinal plants are adaptors to adversity and provide a self-protection mechanism. Salinity has become a worldwide problem that serves as ubiquitous abiotic stress affecting the growth, quality, and yield of medicinal plants. When plants suffer from salt stress, the absorption, accumulation, and distribution of water and nutrients are affected, triggering morphological changes of plants, this is a visual change 31 . Growth parameters and biomass are the most direct indicators of salt tolerance, and then plant height and fresh weight can reflect the biomass of plants 32 . According to these issues, it is inferred that flower buds treated with 100 mM NaCl might be the maximum accumulation of biomass. The decrease at 200 and 300 mM NaCl might be due to the injury caused by osmotic stress, and ion toxicity caused by the accumulation of Na + ions, which hinder absorption of K + ions 33 .Therefore, we think that different salt stresses had different influences on the process of growth and development.   www.nature.com/scientificreports/ The content of MDA is regarded as a reliable indicator of membranes' oxidative damages and has antioxidant activity through the protection of cellular membranes and enzyme integrity 34 . In this study, the contents of MDA increase with salinity, which indicate salt aggravates oxidative damage of LJF. Proline is an important osmolyte in plants, its accumulation is a defense response to salt stress 35 . Proline not only promotes osmotic homeostasis, but also associated with the salt tolerance of LJF. This is consistent with previous results on wheat 36 , wild chicory 37 . Besides, as a protein precursor, it is the energy source of the stress recovery process 38 . Amino acids and biological enzymes are also primary metabolites of plants and play a role in maintaining osmotic balance 39 . According to the results, total amino acids content and photosynthesis/antioxidant enzyme activity peak at 100 mM salt condition, which is optimal salt concentration.    www.nature.com/scientificreports/ Photosynthesis is a prerequisite for keeping plant normal growth. Therefore, to study the effects of salt stress on photosynthetic parameters is of great significance for improving salt tolerance and the quality of medicinal plants. The strength of photosynthesis can be used as an indicator for judging plant growth and stress resistance 40 , and the content of Chlorophyll b directly affects the photosynthetic capacity of plants. Here, we find that chlorophyll b concentration peak at 100 mM NaCl. Compared with the control group, the contents of photosynthetic pigments are elevated under low salt stress, but reduced under high salt stress, indicating that low salt stress could boost the photosynthesis of LJF. Conversely, high salt stress inhibits 41 .
It is known that high salt concentration causes osmotic pressure, resulting in an oxidation imbalance in the plant cells; sequentially, causing damage to proteins and cell membrane integrity 42 . In order to mitigate oxidative damage, plants have developed a series of antioxidant responses, including osmotic balance, antioxidant enzymes, and metabolites. The changes in antioxidant enzymes (POD, SOD, APX, and CAT) imply that salt stress stimulates the antioxidant system. The increase of APX activity is accompanied by the increase of SOD, CAT activity, which is referred to as a common regulation for active oxygen scavenging. Also, APX is a key enzyme to protect chloroplasts and other cellular components from H 2 O 2 and hydroxyl radicals. Our findings revealed that POD, SOD, APX, and CAT activity are induced under low salt stress, suggesting that antioxidant enzymes perform a protective effect on preventing adverse reactions. Also, the correlation between phenolic acid contents and antioxidant enzymes under salt stress is consistent with the previous research 18 ; our results add further support for this conclusion.
Primary metabolites not only constitute essential nutrients for plant growth, but also protect cellular membranes integrality under salt stress 43 . In this study, the contents of serine, glutamate, and lysine are elevated, which could play a positive role in osmotic adjustments. Parts of nucleosides and their derivatives have significant physiological functions. For instance, inosine is beneficial in the treatment of rheumatic heart disease, acute, and chronic hepatitis. Nucleoside is the main component of nucleic acid. And nucleic acid is the precursor to the synthesis of RNA and DNA. In the study, the contents of nucleoside are relatively lower under salt stress, as salt could cause protein structure changes, increase cytoplasmic RNA enzyme activity, and then lead to a decrease in DNA synthesis 44 . Additionally, Li et al. 45 discussed the important secondary metabolites and their variation in plants under salt stress in detail, which reveal salt stress improves the accumulation of various secondary metabolites possessing antioxidant activity; such as phenolics and flavonoid metabolites in this study. Phenolic cooperates with the POD to scavenge H 2 O 2 . Flavonoid metabolites clear free radicals with the help of phenolic hydroxyl groups. The phenylpropanoid pathway is the main metabolic pathway for the synthesis of phenolics and flavonoids. Chlorogenic acid and its derivatives are the main phenolic acid constituents and have important quantitative significance both in LJF and other medicinal plants. Previous studies 18,46,47 have demonstrated that the accumulation and synthesis of chlorogenic acid and luteoloside are directly affected by the variation in hydroxycinnamoyl-Co A quinate hydroxycinamoyl transferase (HQT) and chalcone flavanone isomerase (CHI) activity. Therefore, the correlation between variations of main effective constituents (phenolic acids, flavonoids, and iridoids) contents and the activities of their associated regulatory enzymes under salt stress, which is a promising point and worthy of in-depth study. It is worth mentioning that the content of morroniside increase ninefold under 100 Mm NaCl treatment and plays a protective role in oxidative stress and myocardial injury. This inspires that even less bioactive constituent content, reprogramming related metabolic pathways through salinity induction also induces the production of these active substances, which provides new clues for improving the quality of medicinal plants. According to the traditional view, the production of secondary metabolites consume a large chunk of energy, which is evident from the plants' limited growth and lipid peroxidation under high salt 48 . Phenolic constituents are regarded as secondary antioxidants and play a vigorous role in detoxifying ROS. In our results, the accumulation of phenolic constituents is induced in low salt stress, while the accumulation is reduced under high salt stress. And Pearson Correlation Coefficient is performed to further verified the correlations. The result show that between phenolic acid contents and SOD, POD activity is significant correlation at 0.01, 0.05 level, respectively. Our findings are in accordance with that the low salinity is beneficial to the accumulation of secondary metabolites and causes little oxidative injury, while high salinity impedes the secondary metabolites production and reduce antioxidative activities in LJF.
Both hierarchical clustering analysis and PLS-DA distinguish LJF samples from different salt stresses, of which compounds marked on the PLS-DA loading plot make it distinguish in different samples. Through GRA analysis, the best quality of LJF is obtained at 100 mM NaCl, followed by 0, 200, and the 300 mM. The results illustrate that the quality of LJF treated with 100 mM NaCl is optimal. The superiority of the control group to other salt stress groups also indicates that excessive salt is not conducive to the quality formation of LJF. Briefly, multivariate statistical analysis provides a basis for the complete differentiation of samples under different salt stresses and LJF quality evaluation.
Generally, salt stress tempestuously affects plant growth, physiological parameter, and metabolic process. However, plants can maintain osmotic balance, avoid oxidative damage, and keep normal growth through a series of adaptive responses. This experiment cultivated LJF under different salt stresses (0, 100, 200, and 300 mM NaCl) in the field to investigate the effects on the quality of LJF from the variations of morphology, photosynthesis, osmotic balance, antioxidant activity, bioactive constituents' content. We find that the growth parameters, physiological indicators, and bioactive constituents content get the optimal values at 100 mM NaCl, indicating that salt stress could induce LJF remodeling transcriptional networks and metabolic pathways to regulate its growth and the accumulation of secondary metabolites. This will be verified with the multi-omics approach in the subsequent research. It will deepen acknowledge the mechanism of quality formation in LJF and provide a foundation for cultivation to improve desired characteristics. www.nature.com/scientificreports/

Conclusions
In the study, a reliable method combining morphology, physiology, and multiple bioactive constituents was proposed to evaluate the quality of LJF under different salt stresses (0, 100, 200, and 300 mM NaCl) based on UFLC-QTRAP-MS/MS and multivariate statistical analysis. The results demonstrated that LJF treated with 100 mM NaCl had the best quality; the increased accumulation of phenolic acids reduced oxidative damage; and low salt stress (100 mM) promote quality improvement, conversely, high salt (300 mM) inhibited that. Therefore, LJF cultivated in soil with 100 mM NaCl might be the optimal choice in the long run. Overall, this study sheds light on the effects of different salt stresses on the quality of LJF from the changes in morphology, physiology, and multiple bioactive constituents through field trials. The results might provide an integrated reference for the quality evaluation of other Chinese herbal medicines under salt stress and novel clues for improving the quality of medicinal plants in the pharmaceutical industry. www.nature.com/scientificreports/