Effect of Different Extraction Methods on Major Bioactive Constituents at Different Flowering Stages of Japanese Honeysuckle (Lonicera

Japanese honeysuckle (Lonicera japonica Thunb.) is a traditional Chinese medicinal herb usually used to treat inflammatory disorders. The contents of two major active constituents of Japanese honeysuckle, chlorogenic acid and luteolin, vary with the plant’s development stage. This study aims to analyze the antioxidants and major active constituents of Japanese honeysuckle using different extraction methods at various flowering stages. To achieve the highest efficacy of major active constituents, methanol extraction was better than water extraction. After treatment with microwaves for 3 Min (M3), the mean contents of chlorogenic acid and luteolin obtained using methanol were 42.96 and 1.22 mg/g DW, respectively, which were highest of all treated groups. Of the different flower bud stages, total phenolic content was highest during the Silver Flower (SF) stage, with mean values of 26.91 and 25.82 mg/g DW for water and methanol extraction, respectively. The highest trolox content was also observed at this stage, with mean values of 27.02 and 28.62 mg/g DW for methanol and water extraction, respectively. The level of chlorogenic acid increased with the development of flower buds from the Juvenile Bud (JB) stage to the Gold Flower (GF) stage. The highest value obtained using methanol extraction was 24.45 mg/g DW at the SF stage and the highest values obtained using water extraction were 11.28 and 11.25 mg/g DW at the White Bud (WB) and SF stages, respectively. On the other hand, the content of luteolin obtained using methanol extraction decreased with flower bud development. The highest luteolin levels (0.60 mg/g DW) were found at the JB and Green Bud (GB) stages. For water extraction, however, luteolin content increased with flower bud development, reaching its highest level (0.142 mg/g DW) at the GF stage. The present results can be applied to determine the optimal harvest stage and postharvest treatment method for Japanese honeysuckle.


Introduction
In Asia, Chinese Traditional Medicine (CTM) has been used to treat various diseases for thousands of years. Functional phytochemicals are acquired from many natural herbal sources with potent bioactivity. Lonicera (Caprifoliaceae) is a traditional medicinal herb genus comprised of more than 150 species that is widely distributed in Eastern Asia, including China, Japan, Korea and Taiwan [1]. They are perennial climbing plants with white-to-golden flowers and a rich, sweet smell. The Japanese honeysuckle (L. japonica Thunb., also known as Jin Yin Hua in Chinese) is a main species of Lonicera widely used in CTM [2].
The dried flower buds of L. japonica Thunb. have been used not only for tea but also for treating wind-heat, fever and respiratory tract infection [3]. Furthermore, L. japonica Thunb. is often combined with other medicinal herbs to treat many diseases [4]. For example, it is often combined with Gingyo-San and used as a traditional antipyretic treatment for the common cold and influenza [5]. When combined with Shuang-Huang-Lian, it is used to treat respiratory diseases [6,7] and when combined with Jiangtang tablets, it is an anti-diabetic treatment [8]. Several pharmacological studies have demonstrated that the constituents of L. japonica Thunb. Possess much bioactivity and many functions, such as antibacterial [9], anti-inflammatory [10], antioxidant [11] antiviral [12] and hepatoprotective functions [13,14]. Recently, due to the outbreak of COVID-19 worldwide, L. japonica Thunb. has received increasing attention because it is reported to be capable for the prevention of COVID-19.
Phytochemicals are natural chemicals originally extracted from alga, vegetables, fruits and herbs. Dietary phytochemicals may have The quality and quantity of phytochemicals in plant tissue or organs may be affected by many factors, such as fertilization, growing season and harvest time. Figueiredo et al. indicated that several factors affect the volatile components and essential oil of plant extracts [28]. Ramakrishna and Ravishankar demonstrated that the secondary metabolites of plants are influenced by abiotic stress [29]. Jaime et al., found that temperature changes the volatile accumulation of Salvia lavandulifolia [30]. Novak et al. indicated that thymol increases with decreasing temperatures, while carvacrol increases with increasing temperatures in Origanum spp [31]. The accumulation of different compounds in Eleutherococcus senticosu is affected by light quality [32], whereas light intensity affects anthocyanin accumulation in Melastoma malabathricum [33]. In addition, the phenolic content of blueberries is influenced by maturation status; all phenolic compounds are degraded in fully ripe berries [34]. For strawberries, two active compounds comprised of anthocyanin and cinnamic acid derivatives are increased when the fruits develop from white to red [35].
Previous research has demonstrated that the content of phytochemical compounds and constituents of floral organs can be influenced by maturation status or development stage. For example, Acacia cyclops is most volatile at the yellow bud stage, which increases aliphatic compounds and irregular terpene content and decreases monoterpenoid levels [36]. After stage 3 (when the petals start to split), when Camellia sinensis has more phytochemical compounds than at other stages, the plant's phytochemical compounds decreased until the full bloom stage [37]. The maximum contents of chlorogenic acid, cichoric acid, echinacoside and isobutylamide in Echinacea purpurea occur at stages 1, 1, 2 and 3, respectively, but total content is highest at stage 3 [38]. The maximum contents of gentiopicroside, sweroside, swertiamarin and longanic acid in Gentiana macrophylla occur at stages 2, 3, 3 and 4, respectively [39]. The most flavonols and anthocyanin in Helleborus niger are present at stages B and POF, respectively [40]. Ascorbic acid, β-Carotene, flavone and rutin in Hemerocallis fulva increase with the stage of development, whereas flavan-3-ol, quercetin and wogonin decrease with the stage of development [41,42]. The contents of anthocyanin and flavonol in Lilium increase during the late stages of flower bud development [43]. The anthocyanin content of Paeonia suffruticosa increases with development stage after the half-opening or full-opening stage [44]. Sandersonia aurantiaca has the highest flavonol content at stage 1, but the content does not have a monotonic relation with development stage [45].
This study aims to determine the antioxidant and major phytochemical compounds (chlorogenic acid and luteolin) of L. japonica Thunb. at different flowering stages. Furthermore, the effects of different extraction methods on major active constituents' content at different flowering stages are also investigated. The results of the present study can provide strong evidence regarding the optimal harvest period and postharvest treatment strategy to enhance the value of CTM. Flower buds at different stages were harvested in October 2014. Five hundred flowers were collected for each group. The fresh weight was recorded and the dry weight was investigated after the flowers were oven-dried at 40°C for 96 hours until fully dehydrated. The water content was calculated using the following equation:

Antioxidant activity analysis
Determination of total phenolic content: The total phenolic content of the extract was determined according to the method proposed by Sato et al. [46]. Fresh gallic acid solution (0-500 mg L -1 ) was prepared with 80 % methanol and 20 % sodium carbonate. Twenty-µL extracted samples of L. japonica were mixed with 1 mL Folin-Ciocalteu reagent and left to stand for 5 min. The mixture was added to 0.9 mL of 20 % sodium carbonate and left to stand for 10 min. Then, it was centrifuged at 3,000 rpm for 10 min. The absorbance was measured to be 760 nm. The total phenolic content was calculated from the calibration curve and expressed as mg of gallic acid equivalent to g in dry weight.
Determination of DPPH radical scavenging activity: Analysis of DPPH (1,1-diphenyl-2-picryl-hydrazyl) radical scavenging activity was conducted according to Shimada et al.'s method [47]. Fresh DPPH-methanol solution (26.68 µM) was prepared and stored at 4°C until use. Four mL of the solution was mixed with 20 µL of extracted solutions of L. japonica. The mixture was incubated in the dark at 4°C. Methanol solution was used as a blank. The absorbance change at 517 nm was measured 10 min later. All measurements were performed in triplicate. DPPH radical scavenging activity and 50 % scavenging concentration (EC50) were calculated using Equations (2) and (3), respectively. The relative trolox content of samples was calculated from the calibration curve and expressed as trolox equivalents.

Determination of chlorogenic acid and luteolin contents
In the present study, chlorogenic acid and luteolin contents were determined by HPLC according to the method used by Lu et al. with some modification [48]. Briefly, quantification of chlorogenic acid and luteolin in the flower buds of L. japonica was performed using the Waters 1525 binary HPLC Pump (Waters Corp.) equipped with a Waters 2487 UV-Visible Detector (Waters Corp.). The standard solutions and extracts were separated on a SunFire C18 (5 µm 4.6x250 mm, Waters Corp.). The mobile phase was a methano l-0.4 % phosphoric acid solution (50:50, v/v), the flow rate was 1 mL min -1 , the ultraviolet spectrum change was 350 nm and the injection volume was 10 mL. Next, 100 mM of prepared chlorogenic acid and luteolin was mixed with Milli-Q ® Water (Millipore, Merck, Germany) through a syringe filter with PES of 0.22 µm (Sterlitech Corp.). Then, chlorogenic acid was diluted to 0, 0.25, 0.5, 1, 1.5 and 2 mM and luteolin was diluted to 0, 0.5, 1, 2, 5 and 7.5 mM. All samples were through syringe filters with PES of 0.22 µm until use. The chlorogenic acid and luteolin contents of samples were calculated from the calibration curve.

Statistical analysis
All experiments were conducted in triplicate. All results were expressed as means ± Standard Deviation (SD). The significance of differences was determined via LSD (Least Significant Difference) using SAS version 9.4 with a significance level of p<0.05. The bud length of L. japonica Thunb. in this study was a little longer than in Wang et al.'s study [18]

Effects of different extraction methods
The present study showed that methanol's extraction efficiency was superior to that of water for both antioxidants (i.e. scavenging  Means in the same column followed by the same letter are not significantly different at the 5 % level according to the Least Significant Difference (LSD).
DPPH radical activity, equivalent trolox content and total phenolic content) and major compounds (i.e. chlorogenic acid and luteolin). For US treatment, methanol (3.81 %) showed significant DPPH radical scavenging activity that was 20.05 times higher than that of water (0.19 %) ( Table 2). For H55 treatment, no significant difference was observed between water and methanol extraction of equivalent trolox content (3.48 and 3.48 mg/g DW, respectively). For H95 treatment, water and methanol extraction of equivalent trolox content increased to 19.75 and 28.10 mg/g DW, respectively, indicating that heat can significantly improve extraction efficiency. For microwave treatment, DPPH radical scavenging activity and equivalent trolox content was significantly influenced by both extraction solvent and treatment time ( Table 2). Compared with water extraction, methanol extraction with M1 and M3 treatment increased DPPH radical scavenging activity by 1.49 and 1.40 times, respectively. In addition, compared to water extraction, methanol extraction with M3 and M1 treatment increased the equivalent trolox content by 8.42 and 11.11 mg/g DW, respectively (Table 2). Regarding total phenolic content, H95 and M1 and M3 treatment resulted in the most efficient extraction of gallic acid content for both methanol and water ( Figure 2). However, the extraction efficiency of methanol extraction with M3 was not as high. Lower extraction efficiency was also observed for H55 and US treatment. For these two treatments, methanol had better extraction efficiency than water.
The effects of different treatment methods and extraction solvents on the major biochemical compounds, chlorogenic acid and luteolin, are displayed in figures 3 and 4, respectively. For all treatments, methanol had higher extraction efficiency than water. Similar to the results for antioxidants, the extraction efficiency of major compounds increased with temperature and time of extraction. As a result, H95 and M3 treatment with methanol extraction led to the highest content of major compounds. Compared to H55 treatment, H95 significantly increased chlorogenic acid and luteolin extraction using water (12.51 and 0.06 mg/g DW, respectively) and methanol (13.93 and 0.38 mg/g DW, respectively). In addition, compared to M1 treatment, M3 increased chlorogenic acid and luteolin extraction by 4.89 and 0.03 mg/g DW, respectively, using water and 5.58 and 0.27 mg/g DW, respectively, using methanol.
As mentioned above, antioxidants and chlorogenic acid and luteolin content were substantially increased with temperature and time of extraction (   Means for the same solvent followed by the same letter are not significantly different at the 5 % level according to the LSD.  Means for the same solvent followed by the same letter are not significantly different at the 5 % level according to the LSD.

DPPH radical scavenging activity at different flower development stages
The results of the present study showed that DPPH scavenging radical activity and trolox content increased from the JB stage to the SF stage and then slightly decreased at the GF stage (Table 3). In other words, the highest antioxidant content was found at the SF stage (85.01 % DPPH scavenging effect and 27.02 mg/mL DW trolox content for water extraction; 90.04 % DPPH scavenging effect and 28.62 mg/mL DW trolox content for methanol extraction) and the lowest antioxidant content was observed at the JB stage (7.07 % DPPH scavenging effect and 2.29 mg/g DW trolox content for water extraction; 19.30 % DPPH scavenging effect and 6.14 mg/g DW trolox content for methanol extraction). Table 3 again shows that methanol had better overall extraction efficiency than water.

Total phenolic content at different flower development stages
Total phenolic content at different flower development stages followed the same trend as DPPH radical scavenging activity; however, no significant difference was found between water and methanol extraction ( Figure 5). Total phenolic content was highest at the SF stage (26.91 and 25.82 mg/g DW for water and methanol extraction, respectively) and then decreased dramatically at the GF stage (13.55 and 14.46 mg/g DW for water and methanol extraction, respectively).

Chlorogenic acid content at different flower development stages
Chlorogenic acid is a major biochemical compound that accounts for approximately 1-2 % of the composition of dried L. japonica buds. In the present study, chlorogenic acid content substantially increased with flower development stage until the SF stage, similar to antioxidant content ( Figure 6). The highest chlorogenic acid content obtained using methanol extraction (24.45 mg/g DW) was found at the SF stage and the highest obtained using water extraction (11.28 mg/g DW) was found at the WB stage. The maximum extraction efficiency of methanol was 2.17 times higher than that of water. At the GF stage, the chlorogenic acid content was decreased to 15.94 and 8.20 mg/g DW for methanol and water extraction, respectively.

Luteolin content at different flower development stages
The efficiency of methanol extraction of luteolin was 2.08 to 4.23 times higher than that of water ( Figure 7). Luteolin content slightly increased with flower development stage and reached its highest value (0.192 mg/g DW) at the GF stage after water extraction. However, it decreased from 0.60 mg/g DW at the JB stage to 0.4 mg/g DW at the GF stage.

Stages
Scavenging Effects (%) EC 50     Means for the same solvent followed by the same letter are not significantly different at the 5 % level according to the LSD. Means for the same solvent followed by the same letter are not significantly different at the 5 % level according to the LSD.
Microwave extraction is an approach that heats the solvent by microwave energy, disruption of hydrogen bonds, and enforcement of ion migration [49]. This approach is widely applied to rapidly extract biochemical compounds from many plants, such as the green tea [50], pistachio [51] and tomato plants [52]. In the present study, M3 treatment with methanol extraction led to the highest chlorogenic acid content (40 mg/g DW) in L. japonica buds (Figure 3). Zhang et al. used microwaves to extract chlorogenic acid from L. japonica and proposed that the maximum extraction content could reach 61.4 mg/g [53]. With regards to the extracting solvent, the results of the present study revealed that methanol was better than water for increasing antioxidant activity and major active constituents' content. Similar results were also obtained for the antioxidant activity of olive leaves [54] and the total phenolic content of cactus flowers [55].
It is generally recognized that the biochemical compounds of plants change depending on the flower development stage [36,39,40]. In this study, antioxidant activity and major biochemical compounds (except luteolin) varied with the flower development stages of L. japonica and reached their highest value at the SF stage. According to Joshi et al., antioxidant activity in Camellia sinensis flowers increased with development stage and reached its maximum value when petals started to split, whereas chlorogenic acid content was highest at the stage of flower opening [37]. This result was in line with the present study. In a study investigating Echinacea purpurea, the maximum chlorogenic acid content was observed at the initial stage of flower bud development [38]. However, Zhu and Xiong found the highest chlorogenic acid content during the GWB stage of L. japonica [56]. Wang et al. also investigated the chlorogenic acid content at different development stages of L. japonica flowers and found that the highest content was present at the GB stage and decreased thereafter [18]. The present study found the highest chlorogenic acid content at the WB and SF stages. This controversy may be due to several factors, such as plant species, growing environment, growing season and extraction methods. Therefore, further studies are necessary to study the effects of these factors on the extraction of bioactive components when determining optimal harvest time for commercial demand.

Conclusion
Analysis of various treatment methods showed that the M3 treatment was the most efficient for extraction of antioxidant and bioactive components of Japanese honeysuckle (Lonicera japonica Thunb.). Except for luteolin content, methanol had better extraction efficiency than water. We believe that the SF stage is the optimal harvest stage because DPPH scavenging radical activity, equivalent trolox content and chlorogenic acid content were at their highest. The present results can be applied to determine the best harvest time and treatment method for enhancing the value of Chinese Traditional Medicine (CTM). However, further investigations are still needed to study the influence of environmental and biogeochemical factors on the bioactive components of L. japonica Thunb.