In this study, the chemical diversity of polyphenols in Iris lactea var. chinensis seeds was identified by combined MS/MS-NMR analysis. Based on the annotated chemical profile, the isolation of stilbene oligomers was conducted, and consequently, stilbene oligomers (
Keywords: Iris lactea var. chinensis; polyphenols; stilbene oligomers; chemical profiling; neuraminidase; MS/MS molecular networking; SMART
Stilbene is a class of natural polyphenols that can construct complex oligomeric structures [[
Iridaceae comprises 1800 species in 60 genera, representing one of the largest families of the superorder Lilianae [[
Mass spectrometry and NMR spectroscopy are the most common techniques in metabolomics studies, and each brings its strength and weakness. In mass spectrometry, the structure information is described by the size of the molecular ion and the fragmentation patterns of the molecular ions with high sensitivity. The diversity of fragmentation ion patterns belongs to molecular structures, so analysis of these patterns. However, results from mass spectrometry depend on the ionizabilities of targeted molecules. Unlike mass spectrometry, NMR spectroscopy is quantitative and produces highly reproductive results. Even the sensitivity of NMR is lower than that of mass spectrometry, NMR data is not influenced by the type of molecules. Moreover, the sensitivity and acquisition time have been improved by the development of cryo- and microprobes [[
The MS/MS molecular networking organizes MS/MS spectra into a spectral similarity-based network, which groups specialized metabolites into several molecular families based on possible structural similarity [[
SMART 2.0 is a convolutional neural network-based approach for the rapid annotation of molecularly diverse natural products [[
Here, we described the combined MS/MS-NMR chemical discovery on the ethanolic extract of I. lactea var. chinensis seeds, and then based on the information generated by molecular networking and SMART analyses, further isolation and identification of two new stilbene oligomers (1 and 2) and eight known stilbene oligomers (3–10), followed by the quantitative analysis using three major compounds (3, 6, and 9) obtained in this study in order to evaluate the potential of I. lactea var. chinensis as the source of stilbene oligomers. Furthermore, the isolated compounds (1–10) were tested for their neuraminidase inhibitory activities [[
To analyze chemical diversity of I. lactea var. chinensis-derived secondary metabolites, the MS/MS molecular network analysis was carried with I. lactea var. chinensis extract together with trans-ε-viniferin, which is a typical stilbenoid from natural products. The result showed three major molecular families (MFs) (Figure 1). MF A was composed of nodes with m/z 500–900 molecular weight ranges. Three red nodes in A were identified as procyanidin B type dimers of catechin by matching MS/MS fragmentation patterns with the GNPS (global natural products social molecular networking) spectral library. The difference of 288 Da between spectral nodes of 577 and 865 was deduced from the catechin block. MF B represented stilbene dimers, and one node was identified as trans-ε-viniferin (3) by matching MS/MS fragmentation and retention time of the standard compound. MF C was a large molecular family in which precursor ion of spectral nodes was in the range of m/z 600–1200. Even though none of the nodes in C were annotated by spectral library matching, precursor mass differences shown at 162 Da (hexose) or 452 Da (two stilbene moieties) inferred that spectra in MF C came from stilbene oligomers and their glycosides.
To confirm the scaffold of MF C with the range of m/z 600–1200, the SMART analysis was carried out with ethyl acetate extract of I. lactea var. chinensis in which polar primary metabolites within sugars or organic acids were removed during the solvent extraction. From SMART analysis with the HSQC spectra of the fraction, 100 nearest structural neighborhoods were suggested. This result was filtered by molecular weight range of 600–1200 Da, which was the molecular weight range of MF C. The result annotated that the secondary metabolites in the MF C were stilbene oligomers, supporting the prediction from molecular network analysis (Figure 2).
Combining the results of MS/MS molecular network and SMART analysis, major chromatographic peaks were tentatively annotated in a scaffolding level (Figure 3A and Table 1). Especially fourteen peaks were annotated as stilbenoids targeted to isolate: two dimers (peaks H and I), four dimer glycosides (peaks C–F), five tetramers (peaks K, M–P), and three tetramer glycosides (peaks G, J, and L). Stilbene monomers were not observed in the result. To confirm the annotation on stilbenoids, targeted isolation and structural elucidation were conducted thereafter.
Based on the presence of stilbenoids predicted from the MS/MS-NMR combined analyses, the EtOAc- and BuOH-soluble extracts of I. lactea var. chinensis seeds (Figure 4) were selected for the isolation of stilbenoids and their glycosides. As a result, two new compounds 1 and 2 were isolated along with eight known structures, which were assigned by comparison of MS and NMR data with literature, as trans-ε-viniferin (3) [[
Compound 1 was obtained as a brown amorphous powder. Its molecular formula is C
Compound 2 was isolated as a brown amorphous powder, and HRESIMS data in the negative mode determined its molecular formula as C
The contents of stilbene oligomers in I. lactea var. chinensis seeds were quantified by using HPLC-UV. From HPLC-UV chromatogram of I. lactea var. chinensis extract, three major stilbene peaks were chosen for the quantification of their contents in the extract (Figure S8 and S9, Supplementary Materials). The representative HPLC-UV chromatogram is depicted in Figure 3B. The analysis method was validated by recoveries and reproducibility against the three constituents. The quantitative HPLC-UV analysis of trans-ε-viniferin (
The results showed good linearity between the peak area and concentration. All calibration curves were linear with correlation coefficients (R2) of over 0.9999. Recoveries of the three compounds were tested by the analysis of the I. lactea var. chinensis seed extract spiked with the major compounds. As shown in Table 3, the average recoveries of trans-ε-viniferin (3), vitisin A (6), and vitisin B (9) were 96.7%, 92.7%, and 100.8%, respectively. Precision was evaluated by inter- and intra-day analytical precision for an ethanolic crude extract of I. lactea var. chinensis seed at the concentration of 3.33 mg/mL (Table 4). Both intra-day and inter-day precisions were found to be within accepted criteria with the relative standard deviation (RSD) (<3%). From HPLC-UV analysis with the validated method, the contents of trans-ε-viniferin (3), vitisin A (6), and vitisin B (9) in I. lactea var. chinensis seed extract were quantified as 2.32 ± 0.06 (3), 4.95 ± 0.14 (6), and 1.64 ± 0.01 (9) mg/g DW (dry weight), respectively (Table 4). The stilbene contents of I. lactea var. chinensis seeds suggested that this natural product could be considered as a great source of stilbene oligomers with over 8 mg of stilbene oligomers per 1 g of the seeds. This stilbene contents were similar to those of Vitis vinifera (7.9 mg/g), which is known to be a stilbene-rich plant [[
The viral neuraminidase inhibition assay was carried out to evaluate neuraminidase inhibitory activities of isolated compounds (1–10) against neuraminidase from the H1N1 influenza virus (Figure S10, Supplementary Materials). Tamiflu (Oseltamivir), a well-known flu therapeutic molecule and a neuraminidase inhibitor [[
Optical rotation was recorded on a JASCO P-2000 polarimeter (JASCO, Easton, MD, USA). UV and ECD spectra were obtained using a Chirascan and ECD spectrometer (Applied photophysics, Surrey, UK). NMR spectra were recorded on AVANCE-600 and AVANCE III HD (Bruker, Billerica, MA, USA) at 25 °C with a cryogenic probe. All HRESIMS data were measured on a Waters Xevo G2 QTOF mass spectrometer (Waters Co., Manchester, UK). Column chromatography was performed with Sephadex LH-20 (25–100 μm, Pharmacia, Piscataway, NJ, USA). TLC was carried out with Kieselgel 50 F254 coated normal silica gel TLC plate (Merck, Darmstadt, Germany). MPLC was carried out by Grace Reveleris MPLC system (Grace, IL, USA) with the C18 column (120 g, Grace, IL, USA). The preparative HPLC system was equipped with a G-321 pump (Gilson, Middleton, WI, USA), a G-151 UV detector (Gilson, WI, USA), and the Kintex C18 column (250 mm × 10 mm i.d.; 5 μm, Phenomenex, CA, USA). The analytical HPLC system was UltiMate™ 3000 equipped with UltiMate™ 3000 variable wavelength UV detector (Thermo Scientific, CA, USA) and YMC Triart C18 column (250 mm × 4.6 mm i.d.; 5 μm, YMC Co. Ltd., Japan). All solvents were purchased from Daejung Chemicals & Metals Co. Ltd. (Si-Heung, Korea). The reagents for aldose discrimination (L-cysteine methyl ester hydrochloride, ο-tolyl isothiocyanate) were purchased from Tokyo Chemical Industry CO. Ltd. (Tokyo, Japan).
Dried seeds of Iris lactea var. chinensis were purchased from the oriental herb market, Health Wisdom (https://www.healthwisdom.shop). The dried seeds (Figure S12, Supplementary Materials), of which the moisture content was 9.1%, used in the present study (SNU-1609), were identified by H.-S. Chae, along with DNA-based authentication (Tables S3–S5, Supplementary Materials) performed by Life Sciences Research Institute, Biomedic Co., Ltd. in South Korea, and deposited at the Medicinal Plant Garden, College of Pharmacy, Seoul National University.
The air-dried seeds of I. lactea var. chinensis (2 kg) were grounded and macerated with 70% (v/v) ethanol in water (3.0 L) at room temperature for 2 days. This was repeated three times. The filtrate was concentrated in vacuo and lyophilized to yield a brown solid (50.0 g, yield: 2.5%). The extract was ground to powder and stored at 4 °C in refrigerator.
This crude extract powder (50.0 g) was suspended in water (1.5 L) and successively extracted with n-hexane (3 × 1.5 L, 2.9 g), ethyl acetate (3 × 1.5 L, 20.0 g), and n-butanol (3 × 1.5 L, 7.1 g), sequentially. Each extract was concentrated in vacuo and stored at 4 °C in refrigerator.
Lyophilized crude extract of I. lactea var. chinensis was dissolved in LC-MS grade MeOH with sonication and filtered with a 0.2 μm PVDF membrane syringe filter (Pall Gelman Sciences, Ann Arbor, MI) with a concentration of 2 mg/mL.
UPLC-QTOF/MS data were acquired from the Waters Acquity UPLC system (Waters Co., Milford, MA, USA), which consists of a binary solvent delivery system, autosampler, and photodiode array (PDA) detector. UPLC column was the Waters Acquity UPLC BEH C18 (150 mm × 2.1 mm, 1.7 μm). The mobile phase was 20 mM formic acid in water (A) and acetonitrile (B), with the following gradient: 10–90% B (0–14 min, v/v). The flow rate was set at 300 μL/min, and the injection volume was 2.0 μL. The temperatures of the autosampler and column oven were maintained at 15 °C and 40 °C, respectively. The MS experiments were performed on a Waters Xevo G2 QTOF mass spectrometer (Waters MS Technologies Manchester, UK) connected to the UPLC system through an electrospray ionization (ESI) interface. The ESI condition was set as follows: negative ion mode, capillary voltage 2.5 kV, cone voltage 40 V, source temperature 120 °C, desolvation gas temperature 350 °C, cone gas flow 50 L/h, and desolvation gas flow 800 L/h. The ion acquisition rate was 0.2 s. Data were centroided during acquisition using independent reference lock-mass ion via the LockSpray
LC-MS/MS data were analyzed using a feature-based molecular networking workflow, which is available on the GNPS web platform (https://gnps.ucsd.edu) with a spectral preprocessing by MZmine2 software [[
The HSQC spectrum of the ethyl acetate extract of I. lactea var. chinensis seeds (20 mg/mL) was preprocessed using MestReNova, phase-corrected, and referenced to the solvent peak (CDCl
The ethyl acetate extract (20.0 g) was fractionated with a reversed-phase MPLC with a MeCN-H
The n-butanol extract (7.1 g) was fractionated into two fractions (BU1, BU2) by a reversed-phase MPLC with the MeCN-H
cis-ε-viniferin-11a,13b-di-O-β-d-glucopyranoside (
cis-vitisin B-13b-O-β-d-glucopyranoside (
Three isolated and purified compounds—trans-ε-viniferin (3), vitisin A (6), and vitisin B (9)—were used as reference compounds. Stock solutions (10 mg/mL) were prepared by precisely weighing and dissolving each in 50% (v/v) aqueous methanol before every experiment day. A mixed standard stock solution was prepared by adding a precise volume of each stock solution to make a solution containing 1 mg/mL of each analyte. The mixed solution was diluted via serial dilution. The crude extract power (100 mg) was precisely weighed and dissolved in 50% (v/v) aqueous methanol with sonication before every experiment day, and the solution was diluted to sample solution concentration (3.33 mg/mL). All solutions were filtered through a 0.2 μm PVDF membrane filter before injection.
I. lactea var. chinensis seeds extract was quantitatively analyzed by HPLC-UV using the UltiMate™ 3000 system (Thermo Scientific, USA) equipped with a binary pump, an auto-sampler, and a photodiode array detector. The analytical column was a YMC Triart C18 column (250 mm × 4.6 mm i.d.; 5 μm, YMC Co. Ltd., Japan), and the mobile phase consisted of 0.1% formic acid in water (A) and acetonitrile (B), with the following gradient: 20–80% B (0–30 min). The flow rate was set at 1 mL/min, and the injection volume was 5 μL. The temperatures of the autosampler and column oven were maintained at 15 °C and 30 °C, respectively. The detection wavelength was set at 280 nm. The results were expressed as milligrams per gram of the crude extract (mg/g).
The HPLC method was validated by the linearity, limit of detection and quantification, accuracy, precision, and repeatability. The linearity of the method was assessed from the correlation coefficients (R2) of the regression curves obtained for each standard. The limits of detection (LOD) and limits of quantification (LOQ) were calculated from the standard deviation of y-intercepts (σ) and the slopes (S), according to the following formulas: LOD = 3.3 × σ/S and LOQ = 10 × σ/S. Accuracy was evaluated by spiking I. lactea var. chinensis extract (3.33 mg/mL), with three concentrations of the quantified standard compounds. Precision (intra- and inter-day) of the analysis was accomplished by analyzing I. lactea var. chinensis extract (3.33 mg/mL) 3-fold on three consecutive days.
The viral neuraminidase activity was screened using 2´-(4-methylumbelliferyl)-α-D-N- acetylneuraminic acid (4-MU-NANA) as the fluorescent substrate. DMSO solvent was used to dilute the concentrations of all compounds and corresponding concentrations in the enzyme buffer MES (32.5 mM 2-(N-morpholino) ethanesulfonic acid, 4 mM CaCl
Previously, combined analyses between MS/MS and NMR were difficult because each method was used in different stages in natural product research; LC-MS was used in the early stage to dereplicate secondary metabolites, and NMR was used in the last stage to elucidate molecular structures. However, the SMART analysis allowed us to use NMR analysis in early-stage within extract or fraction levels to annotate what kinds of metabolites inside. Using both analyses together, fourteen stilbenoids were rapidly annotated, and ten stilbenoids (1–10) were identified by further isolation and purification studies.
To evaluate the value of I. lactea var. chinensis seeds as a source of stilbene oligomers, contents of three major stilbene oligomers—trans-ε-viniferin (3), vitisin A (6), and vitisin B (9)—were quantified by HPLC-UV analysis, which was validated by method validation. The stilbene contents of I. lactea var. chinensis seeds suggested that this natural product could be considered as a great source of stilbene oligomers with over 8 mg of stilbene oligomers per 1 g of the seeds. All the isolates were evaluated for their viral neuraminidase inhibitory activities, and compounds 7–10 showed moderate inhibitory activities. From these results, I. lactea var. chinensis seed extract might be proposed as a sustainable and rich source of stilbene oligomers with viral neuraminidase inhibitory effect.
Graph: Figure 1 The MS/MS molecular network built from the LC-MS/MS data of the ethanolic extract of I. lactea var. chinensis seeds. Molecular families (MF) A (procyanidins), B (stilbene dimers), and C (stilbene oligomers) were annotated based on the spectral library matching and manual inspection on MS/MS spectra. Properties of m/z and mass difference were tagged on each of the nodes and edges.
Graph: Figure 2 Digitized HSQC (heteronuclear single quantum coherence) spectra of ethyl acetate extracts of I. lactea (A) and their top 5 annotated structures from SMART analysis (B). Annotated structures were filtered by the molecular weight range of 600–1200 Da.
Graph: Figure 3 Base peak chromatogram (A) and HPLC-UV chromatogram (B) of ethanolic extract of I. lactea var. chinensis seeds.
Graph: Figure 4 Structures of isolated stilbene oligomers from I. lactea var. chinensis.
Table 1 Major chromatographic peaks in the LC-MS/MS profile of the I. lactea var. chinensis seeds extract.
Peak Retention Time (min) Precursor Ion Molecular Formula Error (ppm) MS/MS Fragments Compounds A 1.74 577.1344 C30H26O12 −0.3 425, 407, 289, 125 procyanidin B3 B 1.86 289.0701 C15H14O6 −3.8 245, 203, 125 catechin C 3.06 777.2402 C40H42O16 0.9 615, 453, 359, 347 vatalbinoside C ( D 3.64 777.2381 C40H42O16 −1.8 633, 575, 453 E 3.80 615.1884 C34H32O11 2.9 475, 453, 359, 347 F 4.48 615.1891 C34H32O11 4.1 537, 475, 453, 435, 359, 347 Stilbene dimer glycoside G 5.24 1067.3105 C62H52O17 1.3 905, 799, 663, 573, 453, 359, 347, 253 H 5.27 453.1336 C28H22O6 −0.4 435, 411, 359, 347, 253, 225 Stilbene dimer I 5.40 453.1335 C28H22O6 −0.7 435, 411, 385, 369, 359, 347, 253, 225 J 5.50 1067.3103 C62H52O17 1.3 905, 799, 663, 573, 453, 359, 347, 253 Stilbene tetramer glycoside K 5.51 905.2580 C56H42O12 −2.0 811, 799, 453, 359, 347 L 5.63 1067.3110 C62H52O17 1.5 905, 799, 663, 573, 453, 359, 347, 253 Stilbene tetramer glycoside M 5.66 905.2591 C56H42O12 −0.8 811, 799, 675, 545, 451, 439, 359, 347, 333 vitisin A ( N 6.06 905.2562 C56H42O12 −4.0 811, 799, 679, 573, 545, 477, 451, 359, 347 O 6.36 905.2609 C56H42O12 1.2 811, 799, 693, 545, 359, 347 vitisin B ( P 6.54 905.2591 C56H42O12 −0.8 811, 799, 693, 545, 451, 439, 359, 347, 333 vitisin C (
Table 2
1 2 Position δC δH ( δC δH ( 1a 133.5 133.5 2a 128.5 6.96 (d, 8.6) 128.5 7.00 (d, 8.6) 3a 116.3 6.73 (d, 8.6) 116.3 6.74 (d, 8.6) 4a 158.5 158.6 5a 116.3 6.73 (d, 8.6) 116.3 6.74 (d, 8.6) 6a 128.5 6.96 (d, 8.6) 128.5 7.00 (d, 8.6) 7a 94.9 5.24 (d, 6.1) 95.2 5.26 (d, 6.5) 8a 57.6 3.33 (d, 6.1) 57.9 3.91 (d, 6.5) 9a 147.0 146.6 10a 108.4 6.22 (brs) 107.4 5.92 (d, 2.2) 11a 160.3 159.6 12a 103.4 6.41 (t, 2.1) 102.2 6.10 (t, 2.1) 13a 159.6 159.6 14a 110.0 6.12 (brs) 107.4 5.92 (d, 2.2) 1b 130.0 131.5 2b 131.2 6.92 (d, 8.6) 126.8 6.51 (m) 3b 116.1 6.60 (d, 8.6) 127.9 4b 157.9 159.6 5b 116.1 6.60 (d, 8.6) 110.0 6.56 (d, 8.3) 6b 131.2 6.92 (d, 8.6) 130.2 6.91 (dd, 8.3, 1.3) 7b 132.2 6.25 (d, 12.0) 131.8 6.08 (d, 12.2) 8b 126.4 6.07 (d, 12.0) 126.3 5.96 (d, 12.2) 9b 137.5 137.6 10b 123.2 123.3 11b 162.6 162.6 12b 98.3 6.54 (d, 1.7) 98.1 6.47 (d, 2.1) 13b 160.4 160.0 14b 110.6 6.55 (d, 1.7) 110.8 6.48 (d, 2.1) 13b-Glc 1′ 102.4 4.73 (d, 7.1) 102.4 4.72 (d, 7.2) 2′ 74.8 3.40–3.50 (m) 74.9 3.30–3.50 (m) 3′ 77.9 3.40–3.50 (m) 77.8 3.30–3.50 (m) 4′ 71.1 3.40–3.50 (m) 71.0 3.30–3.50 (m) 5′ 78.0 3.40–3.50 (m) 77.8 3.30–3.50 (m) 6′ 62.3 3.81 (dd, 12.1, 2.2) 62.2 3.78 (dd, 12.2, 2.3) 11a-Glc 1′′ 102.6 4.76 (d, 7.1) 2′′ 74.8 3.40–3.50 (m) 3′′ 77.8 3.40–3.50 (m) 4′′ 71.1 3.40–3.50 (m) 5′′ 78.0 3.40–3.50 (m) 6′′ 62.2 3.84 (dd, 12.1, 2.2) 1c 132.6 2c 127.9 6.60 (d, 8.6) 3c 116.2 6.55 (d, 8.6) 4c 158.0 5c 116.2 6.55 (d, 8.6) 6c 127.9 6.60 (d, 8.6) 7c 92.5 5.44 (d, 5.8) 8c 52.9 4.22 (d, 5.8) 9c 142.3 10c 120.4 11c 162.5 12c 96.8 6.30 (t, 2.1) 13c 160.3 14c 107.5 6.13 (d, 2.1) 1d 134.2 2d 128.0 7.12 (d, 8.4) 3d 116.5 6.77 (d, 8.4) 4d 158.5 5d 116.5 6.77 (d, 8.4) 6d 128.0 7.12 (d, 8.4) 7d 95.0 5.30 (d, 5.2) 8d 57.8 4.26 (d, 5.2) 9d 147.6 10d 107.2 5.97 (brs) 11d 159.9 12d 102.5 6.05 (t, 2.1) 13d 159.9 14d 107.2 5.97 (brs)
Table 3 Calibration data and percent of recovery rates (Rec, high, medium, and low spike) for three major compounds in I. lactea var. chinensis seeds, including regression equation, correlation coefficient (R
Compound Regression Equation R2 LOD LOQ Rec 1 Rec 2 Rec 3 y = 73.868 0.9999 0.0009 0.0028 95.9 96.4 97.6 y = 17.892 1.0000 0.0040 0.0122 90.4 92.7 94.9 y = 42.076 0.9999 0.0024 0.0073 100.9 100.5 100.9
Table 4 Intra- and Inter-day precision of analysis and the contents of three major compounds in I. lactea var. chinensis seeds.
Compound Intra-Day ( Inter-Day ( Contents Day 1 Day 2 Day 3 21.19 (2.1) 21.56 (0.5) 21.60 (0.4) 21.45 (1.1) 2.32 ± 0.06 10.76 (2.8) 10.83 (0.7) 10.89 (0.3) 10.83 (0.6) 4.95 ± 0.14 8.39 (0.3) 8.48 (0.1) 8.50 (0.1) 8.46 (0.7) 1.64 ± 0.01
Conceptualization, H.W.K.; Data curation, H.W.K. and S.S.K.; Formal analysis, H.W.K., S.S.K., B.R., E.P., and H.-S.C.; Supervision, S.H.S. and Y.-W.C.; Writing–original draft, H.W.K.; Writing–review and editing, K.B.K., J.H., W.K.J., W.K.O., J.K. and Y.-W.C. All authors have read and agreed to the published version of the manuscript.
This work was supported by the National Research Council of Science and Technology (NST) grant by the Korean government (Ministry of Science, ICT, and Future Planning) (No. K17850, K18850, G16230, G17290, and CRC-15-04-KIST) and a grant from the National Research Foundation of Korea (NRF) funded by the Korean government (MSIT) (NRF-2019R1A2C2009053).
Authors declare no conflict of interest.
This article is dedicated to the memory of our esteemed colleague Sang Hyun Sung, a good friend, inspiring mentor, and talented scientist who prematurely passed away on July 24th, 2018.
Figures S1–S7: NMR, sugar analysis data, and ECD spectra of compounds 1 and 2. Figure S8: HPLC-UV chromatograms of reference compounds. Figure S9: UV absorbance spectra for peak specificity. Figure S10: Inhibitory effect of isolated compounds (
By Hyun Woo Kim; Soo Sung Kim; Kyo Bin Kang; Byeol Ryu; Eunjin Park; Jungmoo Huh; Won Kyung Jeon; Hee-Sung Chae; Won Keun Oh; Jinwoong Kim; Sang Hyun Sung; Young-Won Chin and Francesco Cacciola
Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author