Effect of drying processes on prenylflavonoid content and antioxidant activity of Epimedium koreanum Nakai.

Epimedium koreanum Nakai is a famous Chinese herbal medicine for the treatment of impotence, osteoporosis, immune suppression and cardiovascular diseases. Drying is the most common and fundamental procedure in post-harvest processing of E. koreanum, which contributes to the variations of flavonoid content, especially prenylflavonoids, the bioactive components. In present study, effect of drying processes on flavonoid content and antioxidant activity were investigated. High performance liquid chromatography coupled with diode-array detection and electrospray ionization quadrupole time-of-flight tandem mass spectrometry methods were employed. Twenty seven compounds were identified and 11 of them, including eight prenylflavonoids and three other types of flavonoids, were further quantified. The antioxidant activity of E. koreanum was evaluated by the 2,2-diphenyl-1-picrylhydrazyl free radical scavenging method. The results showed that content of the eight prenylflavonoids exhibited significant variations after different drying processes, especially icariin and baohuoside I. The variation tendency of antioxidant activity was positively correlated with the content of total flavonoid, afzelin and icariin.

1. Introduction

Epimedium koreanum Nakai, one of the main origin of “Yinyanghuo”, a well-known Chinese herbal medicine, is mainly distributed in China, Korea and Japan [1]. It is widely used for the treatment of impotence [2], osteoporosis [3], immune suppression [4], cardiovascular diseases [5-7], and cancer [7]. Crude drugs and preparations of E. koreanum have also been used as tonic and health-care products. Phytochemical studies have shown that E. koreanum contains a variety of constituents, including flavonoids, lignans, and polysaccharides [8]. Among them, flavonoids, especially prenylflavonoids, are considered as the major bioactive profiles [9], which have been proved to possess various activities, such as sexual hormone like effects [10], antioxidant [11], preventing osteoporosis [12], antidepressant [13], and anti-tumor activity [14].

Quality of commercial Herb Epimedii is mainly controlled by content of icariin (0.5%, g/g DW) according to Chinese Pharmacopoeia (edition 2015) [1]. However, our previous research found that the content of icariin was strikingly different, and even could not be detected in some batches of commercial Herb Epimedii. The range of icariin content in 104 batches collected from all over the country was 0.01%–0.17% (g/g DW) and all of them were substandard [15]. Meanwhile, studies on quality evaluations also indicated that the content of flavonoids differs greatly in Herb Epimedii [16]. Reasons for this phenomenon were usually ascribed to culture environment, climate factors, and harvest time, or put the blame on germplasms [17]. Actually, in addition to the factors mentioned above, post-harvest processing is also a main reason that affects the quality of medicinal plants [18].

In our early pre-experiments, we found that post-harvest drying process could significantly affect the content of flavonoids, especially prenylflavonoids in E. koreanum. Whether the variation of the flavonoids could lead to the change of activity was still unknown. Therefore, it is necessary to study the relationship of the flavonoid content and activity after different drying processes.

So far, there has been a number of research focused on icariin [19,20], which is considered as the main active compound in Epimedium. However, some investigations indicated that other flavonoids, such as epimedin C, also exhibited superior bioactivities [21]. It is obvious that simply evaluating the content of icariin cannot reflect the true quality of E. koreanum. Moreover, there have been studies examining the changes of flavonoid content after processing in different methods recently, but they all mainly paid attention to the variation of icariin content [18]. Simultaneous determination of more kinds of prenylflavonoids and other types of flavonoids in E. koreanum after drying processes has not been reported. In this regard, a sensitive and rapid high performance liquid chromatography coupled with electrospray ionization quadrupole time-of-flight tandem mass spectrometry (HPLC-ESI-Q-TOF-MS) method was established and validated for the qualitative study of 27 flavonoids. A high performance liquid chromatography coupled with diode-array detection (HPLC-DAD) method was employed for the quantitative analyses of eight prenylflavonoids and three other types of flavonoids, namely epimedin A (1), epimedin B (2), epimedin C (3), icariin (4), baohuoside I (5), korepimedoside C (6), epimedokoreanoside I (7), korepimedoside B (8), hyperoside (9), quercitrin (10), and afzelin (11) (Fig. 1) in E. koreanum after different drying processes. Meanwhile, an 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity assay was also conducted to test the effect of drying processes on E. koreanum.

Fig. 1

Chemical structures of the 11 flavonoids. Compounds 1–8 are eight prenylflavonoids and compounds 9–11 are three other types of flavonoids.

2. Materials and methods

2.1. Reagents, chemicals and materials

Epimedin A, epimedin B, epimedin C, icariin, baohuoside I, korepimedoside C, epimedokoreanoside I, korepimedoside B, hyperoside, quercitrin, and afzelin were separated and purified in our lab. Peak area normalization method on HPLC-DAD indicated that their purities were all above 98.0%. The structures were confirmed by their UV, MS, 1H NMR, and 13C NMR data compared with the literatures [22-29]. DPPH and ascorbic acid (vitamin C) was purchased from Sigma–Aldrich (St. Louis, MO, USA). HPLC-grade acetonitrile were purchased from Fisher (Thermo Fisher Scientific, Fair Lawn, NJ, USA). The deionized water for HPLC analysis was obtained from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, China). Other solvents of analytical grade were purchased from Yuwang Group Co., Ltd. (Shandong, China). Then all HPLC-grade solvents were filtered through 0.22 μm micropore membranes (Tianjin Branch billion lung Experimental Equipment Co., Ltd, Tianjin, China).

E. koreanum leaf samples were collected in August 2014 from Caohekou, (Liaoning province, China). They were authenticated by Professor Jincai Lu, (Department of Medicinal Plant, School of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University). The voucher specimens were deposited in the Department of Medicinal Plant of Shenyang Pharmaceutical University.

Drying processes were conducted in the XMTH digital temperature controller (Yuandong NC instrument factory, Yuyao, China). Each sample was prepared in triplicate. The water content of all samples was determined by oven drying method according to Chinese Pharmacopoeia (edition 2015) and was less than 12% [1]. Drying processes are listed in Table 1.

Table 1

Different drying methods.

Drying method	Temperature (°C)	Time (min)	Drying method	Temperature (°C)	Time (min)
1	Drying in the shade		10	140	20
2	100	45	11	140	30
3	100	60	12	140	40
4	100	75	13	140	50
5	100	90	14	160	10
6	120	20	15	160	20
7	120	30	16	160	30
8	120	40	17	160	40
9	120	50

2.2. Preparation of standard solutions

Each standard stock solution was prepared by dissolving the accurately weighed standard in methanol solution to yield the initial concentration. They were stored in 10 mL volumetric flasks at 4 °C until use.

2.3. Preparation of sample solutions

2.3.1. Preparation of sample solutions for quantification

The dried leaf samples of E. koreanum were ground into fine powder. An amount of each sample (0.2 g) was placed in a 50 mL conical flask with stopper and 20 mL of 50% ethanol (v/v) was added. After accurate weighing, the sample was ultrasonically extracted for 1 h and cooled to room temperature. 50% ethanol solution was added to the extract to compensate for the solvent loss during extraction. Then the extract was fitered, sealed and stored at 4 °C followed by filtering through 0.22 μm micropore membranes prior to analysis.

2.3.2. Preparation of sample solutions for antioxidant activity assay

The fine powder of each sample (2.0 g) was ultrasonically extracted with 50% ethanol (200 mL) for 1 h. After cooled to room temperature, the extract was concentrated by a rotary evaporator under vacuum at 55 °C. The residue was scraped and stored at 4 °C until use.

2.4. HPLC-DAD analysis

2.4.1. HPLC conditions

HPLC was carried out on an Agilent (Santa Clara, CA, USA) Model 1260 liquid chromatography system with diode-array detection (DAD) at 270 nm. A Zorbax SB-C18 analytical column (250 mm × 4.6 mm, 5 μm, Agilent, Santa Clara, CA, USA) was used for separation. Mobile phase A contained 100% acetonitrile while mobile phase B contained 100% water. The gradient was set up as follows: 0–25 min, 24% (v/v) A; 25–30 min, 24%–30% A; 30–45 min, 30%–33% A; 45–50 min, 33% A; 50–59 min, 33%–50% A; 59–65 min, 50%–60% A; 65–70 min, 60%–72.5% A. The flow rate was 1.0 mL/min and the injection volume was 10 μL. The experiments were conducted at 30 °C.

2.4.2. HPLC method validation

The method was validated in terms of linearity, precision, stability, repeatability, accuracy, limit of detection (LOD), and limit of quantification (LOQ).

2.4.2.1. Calibration curves, LOD and LOQ

Linearity was examined with mixed standard solutions (Fig. 2A). Appropriate volumes of the eight standard stock solutions which had been prepared were calculated separately and mixed together. The mixed standard solution of the three flavonoids was also prepared as that described above. Different volumes of the mixture solutions were dissolved with methanol to obtain a series of desired concentrations of standard working solutions. Every calibration curve contained six different concentrations and was performed in triplicate. An aliquot (10 μL) of each standard working solution was subjected to HPLC analysis. The linearities for the 11 compounds were all established by plotting the peak area (y) versus concentration (x) of each analyte.

Fig. 2

HPLC chromatograms of standard compounds (A) and sample extract (B). Fig. 2B was the HPLC chromatogram of the sample drying by method 12. The injection volumes of the standard mixture (A) and the sample extract (B) were both 10 μL. The concentrations of the standards (A) were as follows: epimedin A (1), 0.076; epimedin B (2), 0.107; epimedin C (3), 0.156; icariin (4), 0.236; baohuoside I (5), 0.057; korepimedoside C (6), 0.192; epimedokoreanoside I (7), 0.255; korepimedoside B (8), 0.177; hyperoside (9), 0.117; quercitrin (10), 0.149; afzelin (11), 0.0895 mg/mL. Concentration of the extract (B) was 2.0 mg/mL.

The LODs and LOQs under the present HPLC method were determined at signal-to-noise ratios (S/N) of 3 and 10, respectively. Each standard stock solution was diluted to a series of appropriate concentrations with methanol and an aliquot (10 μL) of the diluted solutions was injected into HPLC for analysis.

2.4.2.2. Precision, stability, repeatability, and accuracy

Intraday and Interday variations were used to evaluate the precision of the developed method. Mixed standard solutions of the eight prenylflavonoids and three flavonoids were injected in six replicates per day for three consecutive days, respectively. Variations were expressed as the relative standard deviation (RSD) of the replicates. The mixed standard solutions stored at room temperature were analyzed at 0 h, 2 h, 4 h, 8 h, 12 h and 24 h to assess the stability. For the repeatability, six replicates of each mixed standard solution were analyzed and it is measured by RSD values. The recovery test was carried out to examine the accuracy of the method. It was performed by adding a known amount of analytes to a certain amount of sample. The quantity of each analyte after injection was subsequently realized from the corresponding calibration curve. The recovery of each standard compound was calculated using the following formula: recovery (%) = (amount found - original amount)/amount added × 100%. Six replicates were performed for the test. The percent recovery rates for the analytes were presented as mean ± SD.

2.5. ESI-Q-TOF-MS analysis

The ESI-Q-TOF-MS spectra were acquired in negative ionization mode on a Micro TOF spectrometer (Bruker Co., Karlsruhe, Germany) and recorded on a mass range of m/z 50–1500. The capillary voltage and end plate voltage were set at 3800 V and −500 V, respectively. The dry gas was maintained at 180 °C and flowed at the rate of 8.0 L/min. The nebulizer pressure was set as 1.2 bar.

2.6. Determination of eight prenylflavonoids and three other types of flavonoids content

Content of each compound was calculated from its corresponding calibration curve. Data are presented as mean ± SD (mg/g) of three independent extracts. The HPLC chromatogram of the sample extract drying by method 12 was shown in Fig. 2B.

2.7. Determination of total flavonoid content

The total flavonoid content was determined according to Chinese Pharmacopoeia (edition 2015) [1]. After preparation of sample extract (see 2.3.1), each extract was taken precisely (0.5 mL), placed into a 50 ml conical flask and diluted with methanol to the scale line to obtain the test solution. The absorbance of the test solution was determined at 270 nm on a Shimadzu UV-2401 spectrophotometer (Shimadzu, Tokyo, Japan). Icariin of 10 μg/mL was prepared as the reference solution.

2.8. DPPH radical scavenging activity assay

The free radical scavenging activity of the extracts on DPPH was evaluated by the method described in the report with a slight modification [30]. The extract powder of each sample was dissolved in 50% ethanol and diluted to a series of concentrations. Then 1.0 mL of sample solution was added to 1.0 ml of DPPH solution (0.2 mM in methanol). The mixture was vigorously shaken and incubated in the dark for 60 min. In the control, 50% ethanol was substituted for the sample. Ascorbic acid (vitamin C), a stable antioxidant, was used as a positive control. The absorbance at 517 nm of the mixture was measured on a Varioskan Flash L-117 Microplate Reader (Thermo Scientific, MA, USA). Antioxidant activity is calculated according to the following formula: DPPH • radical scavenging capacity (%) = [(A control–A sample)/A control] × 100, where A control is the absorbance of control (DPPH solution without sample), and A sample is the test sample. The antioxidant capability was expressed by IC50 values which were defined as the concentration required scavenging 50% of the available free radicals. They were estimated by nonlinear regression using SPSS 19.0 software (SPSS Inc., Chicago, IL, USA). The results were reported as the mean ± SD (μg/mL) of three independent experiments.

2.9. Pearson correlation analysis

The correlations between the 11 constituents and total flavonoid content and IC50 values were tested by the Pearson Correlation Analysis using SPSS 19.0 software.

3. Results and discussion

3.1. HPLC method validation

The linearities were expressed by the regression equations. All calibration curves showed good linear regressions (r > 0.9996) within test ranges. LODs and LOQs were less than 0.045 μg/mL and 0.135 μg/mL, respectively. As shown in Table 2, RSDs of precision, stability, repeatability were all less than 2.0%. The average recoveries of the 11 analytes ranged from 95.46% to 99.62% and their RSDs were less than 2.0%. Therefore, the HPLC method was precise, accurate, and sensitive enough for simultaneously quantitative analysis of the 11 flavonoids in E. koreanum.

Table 2

Precision, stability, repeatability, and recovery of the 11 analytes (n = 6).

Compound	Precision (RSD, %)		Repeatability (RSD, %)	Stability (RSD, %)	Recovery (%)
	Intraday	Interday			Mean (%)	RSD (%)
1	0.36	0.59	0.35	0.15	98.25 ± 0.89	1.02
2	0.66	0.81	0.44	0.14	96.73 ± 1.03	0.67
3	1.08	0.89	1.01	0.26	97.19 ± 1.38	0.95
4	0.18	0.37	0.17	0.16	96.85 ± 1.26	0.59
5	0.45	0.62	0.45	0.48	95.46 ± 1.01	0.18
6	0.48	0.54	0.31	0.31	98.49 ± 1.64	1.11
7	0.31	0.45	0.29	0.26	98.16 ± 0.89	1.35
8	0.08	0.17	0.09	0.08	98.47 ± 0.74	0.31
9	0.57	0.77	0.51	0.47	96.79 ± 0.94	1.05
10	0.74	0.85	0.79	0.52	97.54 ± 1.08	0.97
11	0.62	0.69	0.58	0.39	99.62 ± 0.88	0.81

Data are presented as the mean ± standard deviation.

3.2. Quantification of the 11 flavonoids and total flavonoid

Content of epimedin A, epimedin B, epimedin C, icariin, baohuoside I, korepimedoside C, epimedokoreanoside I, korepimedoside B, hyperoside, quercitrin, afzelin, and total flavonoid under different drying processes are displayed in Figs. 4 and 5.

Fig. 4

Content of icariin and total flavonoid. The content of icariin was determined using an HPLC-DAD method. The total flavonoid content was measured on a Shimadzu UV-2401 spectrophotometer. The results were presented as the mean value ± standard deviation of three separate experiments.

Fig. 5

Base Peak Chromatogram (BPC) of the HPLC-MS analysis of the extract of E. koreanum. The BPC of the extract was the sample drying by method 10. Concentration of the extract was 1.0 mg/mL. The injection volume was 5 μL. Intens = Intensity.

3.2.1. Quantification of the 11 flavonoids

The most noticeable result of this research was the increase of content of the diglycoside and monoglycoside, including icariin and baohuoside I (Fig. 3A). Low amounts (0.37%, g/g DW) of icariin were found in the materials drying in the shade; while a surge of icariin content was observed during the drying processes at all temperatures, especially above 120 °C. Its maximal value of around 1.82% (g/g DW) was occurred after drying at 140 °C for 20 min, which was much higher than the Chinese Pharmacopoeia standard (0.5%, g/g DW). The increasing rates of icariin content accelerated distinctly along with the elevation of drying temperatures in the range of 120–140 °C and showed a significant positive correlation. The variation of icariin as a whole exhibited a V shape. Variation of baohuoside I content showed a similar trend with that of icariin (Figs. 4 and 5). However, unlike icariin, content of baohuoside I reached its maximum value after drying at 140 °C for 50 min. It is obvious that drying temperature also has a significant influence on the baohuoside I content.

Fig. 3

Content of eight prenylflavonoids (A) and three other types of flavonoids (B). The results were presented as the mean value ± standard deviation of three separate experiments.

Variations of triglycosides content showed completely different trends compared with the diglycoside and mono-glycoside (Fig. 3A). Epimedin A, epimedin B, and epimedin C exhibited the same changing trend after drying processes. The results revealed that content of the three prenylflavonoids kept their maximal values in the shade sample and they were all subjected to a minimum after drying at 160 °C for 40 min. Content of korepimedoside C showed no significant change with different drying temperatures. The peak value and least value occured after drying at 100 °C and 160 °C, respectively. Content of korepimedoside B, which was the second highest among the eight prenylflavonoids, showed the similar changing trend with epimedokoreanoside I. The maximal value of the two compounds occured after drying at 100 °C for 75 min. Then content of korepimedoside B hit a slump when drying in method 14.

The maximum content of the other three flavonoids all occured in the sample drying by method 12. Then their content reduced to the minimum when drying at 160 °C. The amounts of hyperoside and quercitrin presented the similar variation tendency. However, because of their low content, the results (Fig. 3B) exhibited no significant change under different drying processes.

3.2.2. Quantification of total flavonoid

The total flavonoid content ranged from 11.3% to 14.9% (g/g DW) (Fig. 4), which were all above the standard of Chinese Pharmacopoeia (edition 2015) (5%, g/g DW). The changing trend of total flavonoid content was similar to that of icariin. However, the variation of total flavonoid content was not as significant as that of icariin, indicating that content of icariin actually had little influence on it.

In the research above, the effect of drying processes on prenylflavonoid content of E. koreanum was investigated. The results showed that drying processes could have a significant influence on the content of prenylflavonoids in E. koreanum. Prenylflavonoid diglycoside and monoglycoside content in E. koreanum increased significantly while triglycosides decreased. However, the total flavonoid content changed little under different drying processes. These results indicated that raising temperatures and extending drying durations may lead to the transformation from triglycosides to diglycosides and monoglycosides [31], thus making the total flavonoid content almost unchanged. Nevertheless, the increment of icariin and baohuoside I was not equal to the decrement of the other six prenylflavonoids. This phenomenon could be explained by the hypothesis that the new generation of icariin and baohuoside I was not entirely transformed from the six prenylflavonoids, but from other multiple components in a complex conversion process.

3.3. DPPH radical scavenging activity assay

DPPH assay was used to measure the antioxidant activities of E. koreanum after different drying processes. This method is based on chemical reactions and the lower IC50 values, the better activities. The IC50 values were ranged from 87.44 to 118.12 μg/mL. The highest scavenging potency was observed after 140 °C drying process with an IC50 value of 87.44 μg/mL, and under this treatment condition the sample contained the highest content of icariin and total flavonoid. Meanwhile, the sample drying in the shade with the lowest content of icariin and total flavonoid showed the poorest antioxidant activity (118.12 μg/mL). In general, the variation tendency of DPPH free radical scavenging capability of the extracts was the same with that of icariin and total flavonoid content.

The scavenging potency of E. koreanum was weaker than ascorbic acid (13.21 μg/mL) and other flavonoids, such as luteolin and hyperoside [32]. Flavonoids are powerful antioxidants against free radicals because of their phenolic groups [33,34]. The main structural features of flavonoids required for efficient radical scavenging can be summarized as follows: an ortho-dihydroxy (catechol) structure in the B ring, for electron delocalization; 2, 3-double bond in conjugation with a 4-oxo function in the C ring provides electron delocalization from the B ring, and hydroxyl groups at positions 3 and 5 provide hydrogen bonding to the oxo group [35]. Nevertheless, C-3′ and C-4′ in the B ring of prenylflavonoids in E. koreanum, are often replaced by methoxyl groups, thus resulting in their weak antioxidant capability. The glycosylation of the 3-OH group also has a strong suppressive effect on the antioxidant activity.

3.4. Pearson correlation analysis

The correlation coefficients between the 11 compounds and total flavonoid content and IC50 values were reported in Table 3. The one which showed negative correlation with IC50 values was positively correlated with antioxidant activity. Analysis of the results exhibited a significant positive correlation between the antioxidant potency and total flavonoid content. The content of icariin and afzelin also presented strong positive correlation with antioxidant activity. Baohuoside I and hyperoside showed weak correlation with the antioxidant ability. Conversely, almost no correlation was recorded between anti-oxidant capability and the other compounds. Therefore, content of the total flavonoid, afzelin, and icariin should be responsible for the antioxidant activity of E. koreanum.

Table 3

Correlation coefficients between the 11 compounds and total flavonoid content and IC50 values.

	1	2	3	4	5	6	7	8	9	10	11	Total flavonoid
IC50	0.058	0.056	0.183	−0.497*	−0.362	−0.068	0.257	0.115	−0.326	−0.126	−0.574*	−0.759**

p < 0.01,

p < 0.05.

In the present research, we only explored the free radical scavenging potential of E. koreanum in vitro. However, the metabolic mechanism of drugs in vivo is quite different from the reaction mechanism of in vitro experiments. Furthermore, E. koreanum mainly possessed the activities of estrogenic effect and antiosteoporosis [3,10], which was contributed by prenylflavonoids presented in the herb [10,36]. These two pharmacological activities will be the focus in our future research.

Pharmaceutical companies usually adopt the way of drying in the shade or drying in the sun for the treatment of fresh Herb Epimedii [37]. Drying in the shade takes about a week, while drying in the sun takes three to five days. The disadvantages of these drying methods are time-consuming compared with baking, and the icariin content in the herbs treated with these methods is sometimes unqualified. One of the reasons for the unstable quality of the herbs is probably attributed to the processing methods. According to the results above, baking E. koreanum at a temperature greater than 100 °C could reach the quality regulated by Chinese Pharmacopoeia (edition 2015). However, the quality of Chinese herbal medicines should be evaluated by a variety of indicators and the standard of Chinese pharmacopoeia (edition 2015) is only part of it. In our investigation, merely the flavonoids content and DPPH free radical scavenging capability of E. koreanum was concerned. Changes in other properties of herbs, such as some important bioactivities, are still unknown. Therefore, in addition to the influence on the flavonoids content, it is uncertain whether baking at a temperature above 100 °C could make a positive effect on other properties of E. koreanum. It is necessary to perform further investigation.

Some existing literatures suggested that higher temperatures and extending durations of drying could lead to the decomposition of bioactive profiles in herbs [38,39]. Our results demonstrated that a suitable drying process could have a positive impact on the content of bioactive components in E. koreanum. Our research may provide a new approach to improve the quality of E. koreanum, especially the prenylflavonoid content.

3.5. Identification of 27 compositions from E. koreanum by HPLC-ESI-Q-TOF-MS

The base peak chromatogram of the sample extract drying by method 10 was displayed in Fig. 5. The chemical compositions were identified by retention time, accurate molecular mass, and fragmentation using the target compound screening method in the mass spectrometry software. Under the optimized condition, 27 compounds were identified based on the database of Scifinder, standard compounds and comparison with literatures. The identifications of 27 compounds were listed in Table 4.

Table 4

Identification of 27 compositions in E. koreanum by HPLC-ESI-Q-TOF-MS.

No.	Identification	TR (min)	Formula for M	[M–H]−			Fragmentation
				Measured mass	Calcd mass	Accuracy (ppm)
1	Hyperoside	6.3	C21H20O12	463.0955	463.0882	−15.8	301.0455 [M-H-gal]
2	Diphylloside A/Ikarisoside C	7.9	C38H48O20	823.2702	823.2666	−4.4	677.2140 [M-H-rha]
3	Trifolin	8.6	C21H20O11	447.0949	447.0933	−3.7	–
4	Epimedoside E	8.7	C37H46O19	793.2579	793.2561	−2.3	631.2026 [M-H-glu]
5	Quercitrin	8.9	C21H20O11	447.0968	447.0933	−7.9	301.0390 [M-H-rha]
6	Icariside C 1	10.2	C21H38O8	417.2492	417.2494	0.4	–
7	Epimedoside A	11.0	C32H38O15	661.2139	661.2138	−0.1	499.1631 [M-H-glu]
8	Anhydroicaritin-3,7-di-O-glucoside	12.7	C33H40O16	691.2238	691.2244	0.7	529.1709 [M-H-glu]
9	Afzelin	13.7	C21H20O10	431.1002	431.0984	−4.3	285.0436 [M-H-rha]
10	Epimedin A	27.9	C39H50O20	837.2800	837.2823	2.7	675.2276 [M-H-glu]
11	Epimedin B	31.9	C38H48O19	807.2582	807.2717	16.7	645.2153 [M-H-glu]
12	Icariin	39.8	C33H40O15	675.2269	675.2283	3.8	513.1741 [M-H-glu]
13	Tricin	41.0	C17H14O7	329.0682	329.0667	−4.6	–
14	Sutchuenmedin A	43.4	C33H38O14	657.2184	657.2189	0.7	–
15	Anhydroicaritin-3-O-rhamnoside (1–2)-furan acid-7-O-glucoside	44.2	C39H48O19	819.2716	819.2717	0.1	657.2197 [M-H-glu]
16	Korepimedoside C	44.8	C41H52O21	879.2923	879.2928	0.6	717.2413 [M-H-glu]
17	Epimedokoreanoside I	50.5	C43H54O22	921.3009	921.3034	2.7	759.2504 [M-H-glu]
18	Ikarisoside B	53.2	C32H38O15	661.2138	661.2138	0.0	–
19	Epimedoside C	53.9	C26H28O11	515.1579	515.1559	−3.9	–
20	Icariside I	55.8	C27H30O11	529.1736	529.1715	−3.8	–
21	Ikarisoside F	58.2	C31H36O14	631.2031	631.2032	0.3	–
22	Caohuoside C	58.5	C27H30O11	529.1729	529.1715	−2.6	–
23	Epimedokoreanin A/Epimedokoreanin C	61.2	C25H26O8	453.1575	453.1555	−4.3	–
24	Ikarisoside A/Baohuoside II	64.8	C26H28O10	499.1640	499.1610	−6.1	–
25	Korepimedoside B	65.1	C45H56O23	963.3114	963.3140	2.7	801.2649 [M-H-glu]
26	Korepimesoside A/Korepimesoside B	66.0	C39H46O18	801.2623	801.2611	−1.4	–
27	Baohuoside I	78.4	C27H30O10	513.1796	513.1766	5.8	367.1249 [M-H-rha]

4. Conclusion

In this paper, the proposed HPLC-DAD and HPLC-ESI-Q-TOF-MS method was used for the quantitative and qualitative analysis of E. koreanum under different drying processes. We revealed a significant variation of prenylflavonoids content in E. koreanum after different drying processes. The content of prenylflavonoid diglycoside and monoglycoside, icariin and baohuoside I, showed significant positive correlations with drying temperatures while prenylflavonoid triglycosides profiles exhibited negative correlations. The reason for these results was ascribed to the loss of glycosyls in prenylflavonoids. Meanwhile, the DPPH assay results showed that the changing trend of antioxidant activity was positively correlated with the content of total flavonoid, afzelin, and icariin.