In Jeju-native Citrus, flavonoids are the main contributors to the various types of biological activity, such as antioxidant, antitumor, and anti-inflammatory activity. Thus, we developed simultaneous quantification methods for the analysis of ten bioactive flavonoids in Jeju Citrus fruits (Dangyuja, Gamja, Jigak, Sadugam, and Soyuja) harvested at six different time points using a high-performance liquid chromatography–diode array detector (HPLC-DAD). Separation was performed using a flow rate of 0.8 mL/min, a column temperature of 40 °C, a mobile phase buffer of 0.5% acetic acid, and a detection wavelength of 278 nm. The established analytical method showed good linearity (R2 ≥ 0.9997), precision (inter-day < 0.599%, intra-day < 0.055%), and accuracy (recoveries 92.30–108.80%). The HPLC–DAD method was subsequently applied to analyze flavonoids in Citrus samples. Overall, the quantification results indicated that the compositions and content of flavonoids differed for each Citrus species. The harvesting period also influenced the changes in flavonoid content within each Citrus species. The analytical results with chemometrics revealed that higher flavonoid levels in early-harvested Citrus were derived from the improved fruit size and reduced flavonoid synthesis during maturation. This study provides a practical and reliable method for the analysis of ten flavonoids that can be further utilized in the quality assessment of Jeju Citrus.
Keywords: HPLC–DAD; method validation; flavonoids; Jeju native Citrus; harvest time; chemometrics
Citrus fruits are important in the global food, pharmaceutical, and cosmetic industries. On Jeju Island, South Korea, a total of 22 unique Citrus species have been grown on a large scale due to the island's subtropical climate since 476 A.D. [[
The beneficial effects of Jeju Citrus fruits are derived from their enriched phytochemicals, such as ascorbic acid, coumarins, carotenoids, limonoids, and dietary fiber [[
As most plant extracts consist of a complex mixture of phytochemicals, a simple, rapid, and reproducible analytical method is required to obtain better experimental results [[
The objective of this study was to develop a novel performance method for the efficient analysis of ten flavonoids (rutin, narirutin, naringin, hesperidin, neohesperidin, quercetin, naringenin, hesperetin, nobiletin, and tangeretin) in Jeju Citrus fruits (Dangyuja, Gamja, Jigak, Sadugam, and Soyuja) harvested at six different time points (from 3 September to 21 November). To thoroughly compare the flavonoid content of the five Jeju native Citrus fruits harvested at six different time points, we applied the established method and then combined the analysis results with chemometrics.
A total of five species of Jeju-native Citrus were used for this research, including 'Dangyuja' (Citrus grandis Osbeck), 'Gamja' (Citrus benikoji Hort. ex Tanaka), 'Jigak' (Citrus aurantium L.), 'Sadugam' (Citrus pseudogulgul Hort. ex Shirai), and 'Soyuja' (Citrus junos Sieb. ex Tanaka). All Citrus species were cultivated under the same agronomic and environmental conditions at the experimental farm of the Jeju Special Self-Governing Province Agricultural Research & Extension Services (Jeju, Republic of Korea). Samples were collected on six dates: 3 September, 18 September, 4 October, 21 October, 6 November, and 21 November in 2019. The six times of harvesting were decided after referring to previous studies [[
The powdered sample (10 g) was mixed with 200 mL of 70% ethanol (v/v) for 24 h, and the mixture was filtered through filter paper (No. 2; Advantec, Tokyo, Japan). After evaporation and lyophilization, the powdered sample extracts (10 mg) were redissolved in 4 mL of 70% methanol (v/v) and sonicated for 10 min. The resulting extracts were filtered through a 0.50 µm PTFE filter (Advantec) and then used for HPLC analysis.
The chromatographic analysis was performed using a Dionex UltiMate 3000 HPLC-DAD system (Thermo Fisher Scientific, Waltham, MA, USA). Flavonoids were separated on a Cadenza CD-C18 column (4.6 × 150 nm, 3 µm; Imtakt Corp., Kyoto, Japan) with gradient elution for 39 min. Mobile phases A and B were composed of water containing 0.5% buffer solution and acetonitrile, respectively. The gradient program used in this study was as follows: 0 min, 20% B; 4 min, 20% B; 16 min, 35% B; 28 min, 75% B; 32 min, 75% B; 34 min, 20% B; and 39 min, 20% B. The injection volume was 10 µL. The optimal HPLC analysis condition was selected after performing the elution programs with different flow rates, column temperatures, mobile phase buffers, and detection wavelengths as follows: flow rates of 0.8, 1.0, and 1.2 mL/min; column temperatures of 30, 35, 40, and 45 °C; mobile phase buffers of acetic acid and formic acid; and detection wavelengths of 266 and 278 nm. The validated chromatographic conditions used for sample analysis were a flow rate of 0.8 mL/min, a column temperature of 40 °C, a mobile phase buffer of 0.5% acetic acid, and a detection wavelength of 278 nm.
For method validation, rutin, narirutin, naringin, hesperidin, neohesperidin, quercetin, naringenin, and hesperidin were purchased from Sigme-Aldrich (www.sigmaaldrich.com (accessed on 4 November 2023)). Hesperetin, nobiletin, and tangeretin were obtained from ChemFaces (www.chemfaces.com (accessed on 4 November 2023)). Each standard stock solution was prepared at a concentration of 1000 µg/mL. The stock solutions were diluted with 70% methanol (v/v), and those for hesperidin, neohesperidin, and tangeretin were diluted with dimethyl sulfoxide. The HPLC method was validated by determining a series of parameters, including linearity, precision, accuracy, and chromatographic factors (resolution and asymmetry), in compliance with the International Conference of Harmonization (ICH) Q2 (R1) guidelines [[
Calibration curves were constructed using standard mixtures dissolved in 70% methanol (v/v) at a concentration of 2.5–100 µg/mL. Linear regression equations were used to calculate the peak area versus concentration of each flavonoid. Linearity was demonstrated using correlation coefficients. The limit of detection (LOD) and limit of quantification (LOQ) were measured based on the linear regression approach with the formulas [LOD = 3.3 × σ/S] and [LOQ = 10 × σ/S], where σ represents the standard deviation of the y-intercept and S represents the slope of the calibration curve.
Inter- and intra-day precision tests were performed three times for each flavonoid at high, medium, and low concentrations (
Accuracy was estimated by performing a recovery study. Experimental data were obtained by analyzing the Dangyuja sample harvested on 3 September, after adding standard solutions at three different concentrations (
For the visualization of the overall pattern of Citrus species with various harvest periods, principal component analysis (PCA) was performed after normalization with unit-variance scaling using the Soft Independent Modeling of Class Analogy (SIMCA) software (version 17.0; Umetrics, Umeå, Sweden). To compare the content of each flavonoid in Citrus species according to the harvest period, one-way analysis of variance (ANOVA) was conducted using GraphPad Prism 8 (San Diego, CA, USA).
Ten flavonoids (rutin, narirutin, naringin, hesperidin, neohesperidin, quercetin, naringenin, hesperidin, nobiletin, and tangeretin) were chosen as analytes to develop the HPLC-DAD method for 70% ethanol extracts derived from five Citrus species at six different harvest time points. To improve the resolution and selectivity, we first modified the flow rates with three different conditions (0.8, 1.0, and 1.2 mL/min). The overlaid chromatogram shows that the retention time (RT) was shortened by increasing the flow rate (Figure 1A). In addition, the asymmetry factors and resolution were considered to evaluate the perfect peak shape. Generally, asymmetry factors closer to 1.0 represent good Gaussian peaks, and resolution values above 2.0 indicate good separation quality. Considering the asymmetry factors and resolutions, our results indicate that a higher flow rate worsens the peak asymmetry and resolution (Table S1). Consequently, 0.8 mL/min was selected as the flow rate.
Controlling the column temperature is a potential means of improving the method's reproducibility and RT adjustment. Thus, the effect of the column temperature on the flavonoid analysis was also tested under column temperatures of 30, 35, 40, and 45 °C. With an increase in the column temperature, the analytical results were obtained faster, but the peak asymmetry became slightly worse (Figure 1B, Table S2). The resolution was constant at all column temperatures. However, a higher column temperature has the advantage of reducing the column back pressure. Considering all the results and factors, 40 °C was chosen for our method.
Adjusting the pH of the mobile phase by adding a buffer is essential in achieving desirable separation. Among the developed HPLC methods, acidified water solvents such as acetic acid, formic acid, and trifluoroacetic acid are common choices [[
Additionally, the effect of the detection wavelength was inspected at 266 and 278 nm, where the 10 flavonoids had common absorption characteristics. When comparing each peak height, rutin and quercetin demonstrated stronger absorbability at 266 nm but narirutin, naringin, hesperidin, neohesperidin, naringenin, hesperetin, nobiletin, and tangeretin showed stronger absorbability at 278 nm (Figure 1D). Based on these results, 278 nm was selected for the HPLC method in this study.
Chromatograms of ten flavonoid standards were obtained by using the optimized method (flow rate of 0.8 mL/min, column temperature of 40 °C, mobile phase buffer of 0.5% acetic acid, and detection wavelength of 278 nm) (Figure S1). The HPLC method successfully separated the 10 compounds within the range of 4.88–26.70 min. Compared with previous studies, the total separation time was shortened, and more flavonoids could be analyzed [[
To validate this method, the linearity, LOD, and LOQ were assessed by drawing calibration curves for each flavonoid. Generally, a correlation coefficient (R
To determine the instrumental precision, inter- and intra-day precision tests were conducted. The stability of analytical methods is expressed by low % RSD values. In this study, the inter-day precision for the RTs and peak areas was less than 0.116 and 0.318%, respectively, and the intra-day precision for the RTs and peak areas was less than 0.012 and 0.034%, respectively (Table 2). These results indicated that the proposed method was reliable and reproducible.
The optimized method was subsequently applied to a recovery test to analyze its accuracy. Accuracy is defined as the closeness of agreement between the value accepted as a conventional true value and the value identified [[
The flavonoid content of five types of Citrus species (Dangyuja, Gamja, Jigak, Sadugam, and Soyuja) harvested on six different dates (3 September, 18 September, 4 October, 21 October, 6 November, and 21 November) was analyzed using the optimized HPLC method (Figure S2). The content of each flavonoid in Citrus fruit extracts was calculated according to the calibration curves shown in Table 1. After comparing the RT and UV spectra of the peaks with those of the standards, a maximum of eight flavonoids were identified in the samples, excluding naringenin (aglycone form of naringin) and hesperetin (aglycone form of hesperidin) (Table 3). Independently of the harvest date, the amounts of total flavonoids were compared among the five Citrus species. The highest concentration was found in Jigak and ranged from 95.59 to 209.72 mg/g. Dangyuja and Sadugam also contained high levels of total flavonoids that ranged from 59.03 to 147.55 mg/g and from 52.86 to 159.82 mg/g, respectively. Soyuja and Gamja showed comparatively low levels of total flavonoids in the range of 24.03 to 67.29 mg/g and 22.36 to 34.93 mg/g, respectively. Previous work found that the total flavonoid content was highest in Jagak peels (ranging from 135.30 to 145.33 mg/g), whereas it was lowest in Sadugam peels (ranging from 2.79 to 8.35 mg/g) [[
For the individual compounds, hesperidin and narirutin were the two main compounds in Gamja, Sadugam, and Soyuja extracts (Table 3). Our findings are in agreement with previous reports that have revealed that hesperidin is one of the most abundant flavonoids in most Citrus fruits, and narirutin and hesperidin are the two major compounds in Gamja, Sadugam, and Soyuja juice [[
PCA was subsequently applied to outline all the data and find novel patterns in the complex datasets formed from 30 Citrus samples (five Citrus species harvested at six different periods) with eight variables (flavonoids detected in the samples). The two highest-ranking principal components (PCs) of the score plot accounted for 77.0% of the total variance, indicating that the two PCs were sufficient to explain the variability (Figure 2). The percentage of variables for the PCs was 50.6% for the first and 26.4% for the second. In the score plot, the samples were grouped according to similarities in their metabolic properties, and both PC1 and PC2 largely contributed to the separation of Citrus species. In particular, the highest-ranking PC separated Dangyuja and Jigak from the other three species. This result supports previous reports regarding genetic variations occurring within Citrus species based on sequence analysis [[
To determine whether the characteristic flavonoids in each Citrus fruit varied depending on the harvest time, ANOVA was conducted for flavonoids that were commonly detected in samples over the whole period. Significantly different samples among harvest dates were identified with p-values of less than 0.05 and then visualized in box plots (Figure 3). In the Dangyuja samples, all detected flavonoids, except rutin, showed decreasing patterns over time. In the Soyuja samples, the narirutin, naringin, hesperidin, and neohesperidin levels gradually decreased from the 3 September to 21 November time points. These patterns for narirutin and hesperidin were similar in the Sadugam samples. For the Jigak samples, all analyzed compounds in the mid to late time points showed fluctuating patterns, but the levels of these compounds were significantly higher in the earliest harvest time point. Regarding the Gamja species, we observed a significantly decreasing pattern for the narirutin content based on the harvesting date. These results indicate that the decreasing patterns of flavonoid components were consistent among the five Citrus groups and were significantly correlated with fruit maturation. Unlike other plants that tend to accumulate flavonoids in mature tissues and organs, Citrus species produce large amounts of flavonoids during the immature season [[
This study investigated an efficient method for the simultaneous determination of ten flavonoids (rutin, narirutin, naringin, hesperidin, neohesperidin, quercetin, naringenin, hesperidin, nobiletin, and tangeretin) in five types of Citrus species fruits (Dangyuja, Gamja, Jigak, Sadugam, and Soyuja) harvested on six different dates (3 September, 18 September, 4 October, 21 October, 6 November, and 21 November) using HPLC–DAD. The parameters affecting the chromatographic conditions, such as the flow rate, column temperature, mobile phase buffer, and wavelength, were optimized in the robustness study. The established analytical method was validated with good linearity, precision, and accuracy, indicating its reliability and reproducibility. In addition, this method was applied to quantify flavonoids from five Citrus species harvested at six different time points. Quantification results showed that hesperidin and narirutin were dominant in Gamja, Sadugam, and Soyuja, whereas naringin and neohesperidin were dominant in Dangyuja and Jigak. The PCA and ANOVA results presented significant variances among the five Citrus groups and revealed that higher amounts of flavonoids were obtained from unripe fruits in each Citrus group. These results suggest that immature Jeju Citrus fruits containing large amounts of flavonoids could be utilized as health-promoting materials. In conclusion, we have demonstrated that this analytical method is suitable for the quantification of Citrus flavonoids and that this method can be further applied to the quality assessment of various Citrus fruits.
Graph: Figure 1 Overlaid chromatogram of 10 flavonoids with different (A) flow rates (0.8, 1.0, and 1.2 mL/min), (B) column temperatures (
Graph: Figure 2 Principal component analysis (PCA) score plot for five Citrus species harvested at six different time points. Each Citrus species is represented on the plot by a unique color, and each harvest time is represented on the plot by a unique symbol: Dangyuja, orange; Gamja, green; Jigak, yellow; Sadugam, blue; Soyuja, violet; 3 September, circle; 18 September, box; 4 October, triangle; 21 October, inverted triangle; 6 November, diamond; 21 November, pentagon.
Graph: Figure 3 One-way ANOVA results as boxplots showing flavonoid content based on different harvest times and the growth for each Citrus fruit species.
Table 1 Calibration curves for limit of detection (LOD) and limit of quantification (LOQ) of 10 flavonoids.
No. Compound a RT b Linear Regression Equation c R2 LOD LOQ 1 Rutin 4.88 y = 0.3374x + 0.1633 0.99979 0.0845 0.2560 2 Narirutin 7.86 y = 0.1563x + 0.0451 0.99993 0.0527 0.1598 3 Naringin 8.97 y = 0.1567x + 0.0278 0.99996 0.0713 0.2161 4 Hesperidin 9.77 y = 0.1458x + 0.0182 0.99998 0.0610 0.1849 5 Neohesperidin 10.69 y = 0.1582x + 0.0105 0.99998 0.0649 0.1965 6 Quercetin 16.04 y = 0.5322x − 0.0979 0.99998 0.4395 1.3317 7 Naringenin 19.44 y = 0.2367x + 0.0222 0.99999 0.1281 0.3882 8 Hesperetin 20.61 y = 0.2147x + 0.0267 0.99999 0.1363 0.4130 9 Nobiletin 25.00 y = 0.5082x + 0.0120 1.00000 0.0992 0.3005 10 Tangeretin 26.70 y = 0.5802x + 0.0468 0.99999 0.0762 0.2310
Table 2 Results of inter-day and intra-day precision and recovery test for 10 flavonoids.
No. Compound Inter-Day Precision Intra-Day Precision Accuracy b RT Area RT Area Recovery RSD 1 Rutin 0.116 0.318 0.012 0.026 105.82 2.709 2 Narirutin 0.093 0.263 0.010 0.020 102.19 1.521 3 Naringin 0.072 0.231 0.008 0.023 100.92 1.371 4 Hesperidin 0.055 0.236 0.007 0.019 107.97 1.480 5 Neohesperidin 0.045 0.221 0.006 0.020 100.02 1.391 6 Quercetin 0.036 0.321 0.006 0.034 102.45 2.820 7 Naringenin 0.024 0.241 0.005 0.023 95.62 1.221 8 Hesperetin 0.016 0.194 0.003 0.025 95.06 0.993 9 Nobiletin 0.010 0.221 0.002 0.025 96.61 1.737 10 Tangeretin 0.009 0.220 0.002 0.030 106.30 4.834
Table 3 Quantification results for five Citrus species harvested at six different time points.
Species Harvest Date Compound (mg/g) Rutin Narirutin Naringin Hesperidin Neohesperidin Quercetin Nobiletin Tangeretin Total Dangyuja 3 September 2.756 ± 0.021 5.902 ± 0.015 63.063 ± 0.027 7.029 ± 0.147 62.301 ± 0.049 4.053 ± 0.040 1.253 ± 0.006 1.197 ± 0.010 147.553 ± 0.073 18 September 2.423 ± 0.050 4.426 ± 0.077 60.059 ± 0.967 6.410 ± 0.113 62.342 ± 0.985 3.717 ± 0.042 1.216 ± 0.021 1.138 ± 0.020 141.731 ± 2.273 4 October 2.850 ± 0.042 4.645 ± 0.072 47.479 ± 0.759 5.214 ± 0.090 41.826 ± 0.662 2.661 ± 0.061 0.888 ± 0.014 0.924 ± 0.013 106.488 ± 1.710 21 October 2.778 ± 0.001 3.607 ± 0.004 42.156 ± 0.015 3.569 ± 0.010 39.650 ± 0.034 2.116 ± 0.029 0.904 ± 0.001 0.979 ± 0.003 95.760 ± 0.028 6 November 2.639 ± 0.012 2.566 ± 0.016 26.616 ± 0.190 2.791 ± 0.013 28.159 ± 0.201 1.527 ± 0.011 0.618 ± 0.004 0.536 ± 0.004 65.452 ± 0.446 21 November 2.609 ± 0.054 2.496 ± 0.040 27.054 ± 0.358 1.697 ± 0.026 22.972 ± 0.294 1.226 ± 0.036 0.478 ± 0.005 0.501 ± 0.004 59.033 ± 0.805 Gamja 3 September 3.572 ± 0.057 12.135 ± 0.221 * N.D. 17.210 ± 0.814 0.159 ± 0.022 N.D. 0.935 ± 0.022 0.922 ± 0.033 34.933 ± 1.091 18 September 2.543 ± 0.114 8.787 ± 0.329 N.D. 15.101 ± 0.685 0.100 ± 0.006 N.D. 0.615 ± 0.021 0.662 ± 0.021 27.808 ± 1.138 4 October 2.520 ± 0.017 8.451 ± 0.042 N.D. 14.324 ± 0.627 N.D. N.D. 0.612 ± 0.004 0.716 ± 0.006 26.623 ± 0.661 21 October 3.139 ± 0.045 8.215 ± 0.093 N.D. 16.903 ± 0.513 N.D. N.D. 0.828 ± 0.017 0.973 ± 0.020 30.058 ± 0.598 6 November 2.866 ± 0.054 6.226 ± 0.093 N.D. 11.855 ± 0.362 N.D. N.D. 0.537 ± 0.014 0.654 ± 0.020 22.138 ± 0.457 21 November 3.331 ± 0.102 5.482 ± 0.141 N.D. 11.999 ± 0.822 N.D. N.D. 0.700 ± 0.027 0.847 ± 0.041 22.359 ± 1.116 Jigak 3 September 1.846 ± 0.091 1.211 ± 0.047 126.091 ± 4.586 2.330 ± 0.045 76.289 ± 2.738 0.743 ± 0.017 0.786 ± 0.034 0.424 ± 0.018 209.719 ± 7.564 18 September 1.250 ± 0.066 0.888 ± 0.010 76.538 ± 0.801 0.577 ± 0.018 38.794 ± 0.381 N.D. 0.409 ± 0.005 0.233 ± 0.006 118.689 ± 1.208 4 October 1.132 ± 0.043 0.645 ± 0.019 67.263 ± 1.527 0.611 ± 0.032 40.043 ± 0.919 N.D. 0.434 ± 0.011 0.237 ± 0.012 110.366 ± 2.555 21 October 1.322 ± 0.015 0.758 ± 0.004 76.695 ± 0.251 0.618 ± 0.015 41.622 ± 0.184 N.D. 0.527 ± 0.003 0.295 ± 0.004 121.837 ± 0.427 6 November 0.952 ± 0.021 0.560 ± 0.016 58.641 ± 0.463 0.591 ± 0.065 27.640 ± 0.180 N.D. 0.321 ± 0.006 0.210 ± 0.004 88.915 ± 0.654 21 November 1.103 ± 0.018 0.539 ± 0.016 56.356 ± 0.428 0.430 ± 0.029 36.403 ± 0.322 N.D. 0.491 ± 0.004 0.270 ± 0.006 95.591 ± 0.813 Sadugam 3 September 2.740 ± 0.054 126.41 ± 2.671 N.D. 30.666 ± 0.595 N.D. N.D. N.D. N.D. 159.815 ± 3.186 18 September 2.417 ± 0.029 87.971 ± 0.952 N.D. 20.372 ± 0.777 0.158 ± 0.002 N.D. N.D. N.D. 110.760 ± 1.755 4 October 2.891 ± 0.046 73.851 ± 0.702 N.D. 17.574 ± 0.175 0.107 ± 0.018 N.D. N.D. N.D. 94.316 ± 0.921 21 October 2.765 ± 0.063 54.398 ± 0.949 N.D. 13.850 ± 0.257 N.D. N.D. N.D. N.D. 71.013 ± 1.266 6 November 2.608 ± 0.017 48.014 ± 0.252 N.D. 12.663 ± 0.094 N.D. N.D. N.D. N.D. 63.285 ± 0.361 21 November 2.414 ± 0.024 40.890 ± 0.623 N.D. 9.553 ± 0.133 N.D. N.D. N.D. N.D. 52.858 ± 0.779 Soyuja 3 September 0.395 ± 0.032 21.344 ± 0.121 10.822 ± 0.057 24.063 ± 0.123 10.671 ± 0.052 N.D. N.D. N.D. 67.294 ± 0.384 18 September 0.245 ± 0.013 17.161 ± 0.283 8.177 ± 0.129 18.802 ± 0.544 8.211 ± 0.080 N.D. N.D. N.D. 52.596 ± 1.024 4 October 0.289 ± 0.021 15.529 ± 0.063 7.104 ± 0.046 16.940 ± 0.074 6.917 ± 0.039 N.D. N.D. N.D. 46.778 ± 0.241 21 October N.D. 12.485 ± 0.026 6.317 ± 0.033 15.082 ± 0.064 6.457 ± 0.065 N.D. N.D. N.D. 40.341 ± 0.164 6 November N.D. 11.755 ± 0.127 5.814 ± 0.093 13.753 ± 0.165 5.755 ± 0.082 N.D. N.D. N.D. 37.078 ± 0.464 21 November N.D. 6.675 ± 0.020 3.816 ± 0.005 9.301 ± 0.024 4.242 ± 0.014 N.D. N.D. N.D. 24.033 ± 0.062
Conceptualization, H.B.H., B.G. and S.-A.Y.; methodology, H.H. and H.B.H.; validation, H.H.; formal analysis, H.H.; investigation, H.H. and S.C.K.; resources, H.B.H., S.C.K. and B.G.; data curation, H.H.; writing—original draft preparation, H.H.; writing—review and editing, Y.-M.H.; visualization, H.H.; supervision, Y.-H.J. and Y.-M.H.; project administration, Y.-H.J. and Y.-M.H. All authors have read and agreed to the published version of the manuscript.
The data presented in this study are available upon request from the corresponding author.
The authors declare no conflict of interest.
The following supporting information can be downloaded at: https://
By Hyejin Hyeon; Ho Bong Hyun; Sung Chun Kim; Boram Go; Seon-A Yoon; Yong-Hwan Jung and Young-Min Ham
Reported by Author; Author; Author; Author; Author; Author; Author