PEER-REVIEW ARTICLE PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6108 Chemical Composition of Dalbergia tonkinensis Heartwood Extracts and Comparison with Dalbergia odorifera Zhongmian Wei,a Jia Luo,b Ruoke Ma,b and Yunlin Fu b,* Dalbergia tonkinensis is very similar to Dalbergia odorifera in material, texture, and other macroscopic characteristics, and a comparison of the commonalities and differences in the chemical composition of the extractives may help to distinguish the two. However, the chemical composition of Dalbergia tonkinensis heartwood is unknown. The total flavonoid content of Dalbergia tonkinensis heartwood was determined to range between 49.67 to 101.95 mg, which indicated that Dalbergia tonkinensis is as equally rich in flavonoids as Dalbergia odorifera. Thirty- one flavonoids were identified via ultra-performance liquid chromatography-mass spectrometry in Dalbergia tonkinensis. Among them, the contents of 15 medicinal active flavonoids with were determined, and the liquiritigenin, naringenin, formononetin, pinocembrin, and biochanin A contents were found to be high. The 12 volatile compositions of Dalbergia tonkinensis heartwood identified via gas chromatography- mass spectrometry were very similar to those of Dalbergia odorifera and were dominated by trans-nerolidol, caryophyllene oxide, and eudesmol. The relationship between the chemical composition of Dalbergia tonkinensis and Dalbergia odorifera heartwood extracts was determined via a principal component analysis, and the results indicated that there was no significant difference in chemical composition between the two. This suggested that Dalbergia tonkinensis could be used as a potential substitute for Dalbergia odorifera. DOI: 10.15376/biores.17.4.6108-6122 Keywords: Dalbergia tonkinensis; Dalbergia odorifera; Heartwood; Chemical composition Contact information: a: Nanning Arboretum, Guangxi Zhuang Autonomous Region, Nanning 530000 China; wzm123908@163.com; b: College of Forestry, Guangxi University, Nanning 530004 China; 18284905882@163.com (J. L); maruoke@163.com (R.M.); *Corresponding author: fylin@126.com INTRODUCTION Dalbergia tonkinensis Prain (D. tonkinensis), commonly known as Vietnamese “Huanghuali”, is a semi-deciduous tree of the genus Dalbergia in the family Papilionaceae (Wu et al. 2013). It is native to Vietnam, primarily in the regions of Hanoi, Quang Binh, and Ninh Binh (Ma et al. 2016). The wood is similar to that of Dalbergia odorifera T. Chen (D. odorifera), with a hard texture, beautiful grain, strong aroma, and abundant “ghost eyes” (Qin et al. 2022). It is native to Hainan Province, China, and has since been introduced to Guangxi, Guangdong, Fujian, and Yunnan (Zhao et al. 2020). The heartwood of D. odorifera is highly valued for its applications and can be used to make high-end furniture; as such, it is the most expensive wood among the 29 species of mahogany in the Chinese standard GB/T18107 (2017). The wood color, aroma, structure, density, and other PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6109 wood characteristics of D. tonkinensis do not considerably differ from those of D. odorifera and is the most similar wood to D. odorifera on the market. In terms of its anatomical properties, that there are no major differences in the arrangement of wood rays, distribution of ducts, and other microstructure between the two (Liu et al. 2021). Heartwood is rich in a large number of extractive components, which can be chemically divided into volatile and non-volatile components. The volatile components of wood are primarily terpenoids, which are the main source of aroma for the heartwood (Jones et al. 2011; Moniodis et al. 2015; Tasnim et al. 2020). A comparison of the volatile components of D. odorifera and D. tonkinensis using gas chromatography-mass spectrometry (GC/MS) revealed a few differences in composition, but the formation of the chemical composition of the extracts is strongly influenced by the environment in which the trees are grown (Yang et al. 2016). Furthermore, the available research material is often taken from processed products of unknown origin, and the differences between the two are still not sufficiently studied. The non-volatile compound components of mahogany heartwood extractives are primarily flavonoids. The flavonoid compounds not only affect the color of the wood, but are also mostly active in traditional Chinese medicine (Ninh 2017; Shao et al. 2019). The flavonoid compounds in D. odorifera heartwood have rich biological activities, e.g., anti-tumour, anti-inflammatory, anti-oxidant, anti-bacterial, anti- thrombotic, and anti-angiogenic, and have the ability to treat blood disorders, anemia, swelling, necrosis, and rheumatism (Li et al. 2014; Choi et al. 2017; Fan et al. 2017). The flavonoid content of D. odorifera heartwood extract has been previously investigated by Ma et al. (2022), but the effects of D. tonkinensis have not been reported. In addition, the flavonoid composition of D. tonkinensis is unknown, which also limits the application of D. tonkinensis. In this study, ultra-performance liquid chromatography-mass spectrometry (UPLC/MS), and GC-MS were used to identify the characteristic flavonoid and volatile components of D. tonkinensis and to obtain detailed chemical information on the heartwood of D. tonkinensis. Secondly, from the perspective of the secondary metabolic components of the plants, the differences and similarities between the chemical composition from the heartwood extracts of D. tonkinensis and D. odorifera were compared on the basis of a previous studies on the composition of D. odorifera, so as to provide theoretical guidance for the rational utilization of D. tonkinensis. EXPERIMENTAL Chemical Reagents The HPLC-grade methanol and formic acid were obtained from Shanghai Macklin Biochemical Co., Ltd (Shanghai, China). Ultrapure water was prepared using a Milli-Q system (Millipore Corp., Bedford, MA). A standard solution of (C8 through C40) n-alkanes was purchased from Sigma Chemical Co., St. (Louis, MO). Nerolidol, formononetin, pinocembrin, prunetin, butein, isoliquiritigenin, liquiritigenin, butin, dalbergin, eriodictyol, naringenin, genistein, daidzein, tectorigenin, biochanin A, luteolin, Calycosin, and fisetin were purchased from Shanghai Yuanye Co., Ltd (Shanghai, China). PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6110 Material Sources The D. tonkinensis heartwood was collected from the Quang Binh (DT1), Quang Ninh (DT2), and Hanoi (DT3) Provinces in Vietnam. Three groups of D. odorifera samples were collected from the Guangxi University (DO1), Liangfengjiang Forest Park (DO2), and Pingxiang City (DO3), of the Guangxi Zhuang Autonomous Region, China. Four groups of D. odorifera were collected in Dongfang City (DO4), Haikou City (DO5), Anding County (DO6), and Ledong County (DO7) in Hainan Province, China. All samples were identified by the Guangxi University Quality Inspection Centre according to the macroscopic and microscopic anatomical structure of the wood. The samples were crushed by a grinder (SL-100, Zhejiang Yongkang Songqing Hardware Co.) and used for subsequent analysis. Sample Extraction First, 0.050 g of naturally dried D. tonkinensis heartwood powder was added to a grinding mouth bottle, and then 5 mL of an 80% methanol (v/v) solution was added, after which the ultrasonic (40 KHz) extraction was performed using an ultrasonic extractor (KQ2200, Kunshan Ultrasonic Instruments Co., Ltd.) for 1 h. Finally, the solution was filtered through a 0.22 µm microporous membrane and stored in a -4 °C refrigerator for subsequent analysis. Determination of the Total Flavonoid Content The total flavonoid content of the heartwood was determined using the method reported by Ma et al. (2022). Briefly, the rutin standard solution (100 ug/mL) was prepared; 0, 1, 2, 3, 4, and 5 mL of this reagent were removed in 10 mL volumetric flasks; thereafter, 0.3 mL of 5% NaNO2 solution and 0.3 mL of 10% AlCl3 solution were added respectively, following which the flasks were shaken well and left undisturbed. Finally, 4 mL of NaOH solution was added, the solutions were fixed, and their absorbance was measured using a UV spectrophotometer at a wavelength of 510 nm. Results were used to draw the standard working curve, as shown in Eq. 1, y = 0.0095x + 0.0041, R2=0.9997 (1) where y is the absorbance value of the sample and x (ug/mL) is the concentration of rutin. For total flavonoids content in naturally dried D. tonkinensis heartwood, sample solution (0.3 mL) was processed according to the steps of the rutin standard curve method. Total flavonoid content was expressed as the mg of rutin equivalent (RE) / g dry weight (DW). Characterization of the Flavonoid Components and Determination of the Characteristic Flavonoid Content The chemical composition of the flavonoids was identified via ultra-performance liquid chromatography in tandem with a quadrupole-electrostatic field orbit trap high resolution mass spectrometer (UPLC-Q-EXACTIVE-MS). The UPLC conditions were as follows: an ACQUITY UPLCBEHC18 (2.1 mm × 50 mm, 1.7 μm) column; the mobile phase: A was 0.1% formic acid water and B was methanol; the gradient elution conditions: 0 min to 4.0 min, 20% to 30% B; 4.0 min to 8.0 min, 30% to 40% B; 8.0 min to 17.0 min, 40% to 55% B, 17 min to 20 min, 55% to 70% B, 20.1 min to 23 min, 100% B and, 23.1 min to 26 min, 80% B; the flow rate was 0.3 mL·min-1, the column temperature was 30 °C, and the injection volume was 1 μL. PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6111 The high-resolution mass spectrometry conditions were as follows: an ESI ion source; a temperature of 300 °C, a transmission capillary temperature of 320 °C, sheath gas of 206 KPa, an auxiliary gas flow rate of 69 kPa, and a spray voltage of 3.0 kV in positive ion mode, a scan mode of Full MS and Full MS/dd-MS2, and a mass range of 100 to 1000 Da. The content of the characteristic flavonoid components was determined via the external standard method. The 15 (formononetin, pinocembrin, butein, isoliquiritigenin, liquiritigenin, butein, dalbergin, eriodictyol, naringenin, genistein, daidzein, tectorigenin, biochanin A, calycosin, and fisetin) flavonoid controls were precisely weighed, dissolved in methanol or dimethyl sulfoxide, and configured as a 1 mg/mL control master batch. The appropriate amount of the control master batch was pipetted into the injection bottle and diluted with methanol to make standard mixtures of known gradient concentrations. A standard working curve of the flavonoids was established by the ratio of the peak area to the sample concentration. Analysis of Volatile Components First, 0.1 mL of extract was diluted 10 times using methanol and dried overnight by adding 0.1 g of anhydrous sodium sulfate, and the assay sample was prepared by filtration using a 0.22 um filter membrane. The volatile components of the samples were detected using a Aglient 5975C GC/MSD gas chromatography-mass spectrometer with a TG-5MS column (30 m × 0.25 mm × 0.25 μm) with the following conditions: helium gas was used as the carrier gas, the programmed temperature rise had an initial temperature 60 ℃, increasing at a rate of 20 °C/min up to 170 °C, and then held for 1 min and then increasing at a rate of 5°C/min up to 220℃ and then held for 2 min; no split injection; the solvent delay time was 3 min; the injection volume was 1 μL; the flow rate was 1.0 mL/min. The mass spectrometry conditions were as follows: an EI ion source; an ionization energy of 70 eV; an ion source temperature of 230 °C; and a scan range of 50 to 500 amu. Identification of the volatile oil compounds was assigned by the comparison of their Kovats retention indices relative to the (C8 through C40) n-alkanes (Sigma, Chemical Co., St. Louis, MO) with those in the literature. Furthermore, the data obtained was confirmed by matching their recorded mass spectra with those stored in the NIST05 library and related literature. Differences in the Chemical Composition of Heartwood of D. tonkinensis and D. odorifera Compared via Principal Component Analysis The UPLC-MS data of the D. odorifera and D. tonkinensis samples from different sources were analyzed via PCA to reveal the differences and commonalities between them. Preprocessing of the raw UPLC-MS data for the peak extraction, deconvolution, peak alignment, and peak integration was performed using the Compound discover v3.1 software workstation (Thermo Fisher Scientific). All ions with missing values less than 80% of all the samples were filtered from the data sets; the remaining ions with the missing values were substituted with a minimum value. The data matrix was normalized by sum. The elution conditions of the chromatographic column were changed to mobile phase: A was 0.1% formic acid water and B was methanol; the gradient elution conditions: 0 min to 2.0 min, 10% to 20% B; 2.0 min to 12.0 min, 20% to 70% B; 12.0 min to 19.0 min, 70% to 90% B, 19 min to 22 min, 90% to 100% B; 22.0 min to 24.0 min, 100% to 95% B; and 24.1 min to 27.0 min, 10% B. The chemical composition information was collected in both PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6112 positive and negative ion mode, and the rest of the parameters were consistent with the settings in Section 2.5. Data Processing All experiments were carried out three times. In addition, PCA was performed using the SIMCA software package. The PCA model was generated by the NIPALS algorithm. The data was processed using SPSS 19.0 analysis, and the results were expressed as the mean ± standard deviation (mean ± sd). RESULTS AND DISCUSSION Total Flavonoid Content of D. tonkinensis Recent studies have confirmed the safety and potential anti-inflammatory, anti- tumor, antioxidant, and other biological activities of flavonoids in wood, especially in heartwood (Celedon and Bohlmann 2018). The total flavonoid content is related to the quality of the herb, and it affects the color of the wood. There are differences in the color of the heartwood from the three origins of D. tonkinensis, with the heartwood from the Hanoi origin being the darkest, chestnut brown, and the heartwood from Quang Binh and Quang Ninh being light brown. The total flavonoid content of D. tonkinensis from three origins ranged from 49.7 mg to 102.0 mg RE/g dry weight (Fig. 1), which indicated that the heartwood of D. tonkinensis is relatively rich in flavonoids and has potential for use in food and pharmaceutical applications. The large variation in the total flavonoids from different origins indicated that the geographical environment has a strong influence on the synthesis of flavonoids in the heartwood extracts. Fig. 1. Total flavonoid content of D. tonkinensis from different origins Identification of the Flavonoid Components of D. tonkinensis The XIC Manager module of Compound Discoverer 3.1 software was used for accurate relative molecular mass screening and analysis. In combination with the identified flavonoid components of D. odorifera, the mzcloud and Chemspider mass spectrometry databases were consulted to infer the possible chemical structures (as shown in Fig. 2). This analysis was based on the mass spectrometric cleavage patterns of the compounds and the mass spectra of the chemical controls, and was used to assign structural categories and Cultivar Sample F la v o n o id s C o n te n t (m g /g ) PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6113 major fragment ion information (as shown in Table 1). To the best of our knowledge, this study is the first report on the flavonoid compositions of D. tonkinensis heartwood extracts. Fig. 2. The chemical structures of the compositions identified in Dalbergia odorifera extracts (Note: the number of the chemical components (1 through 28) corresponds to the compound numbers in Table 1) Flavonoids and Flavonols The mass spectral information and retention times of compound 2 (tR = 20.60), compound 20 (tR = 10.29), and compound 22 (tR = 13.76) were in general agreement with those of the Chrysin, Fisetin, and Luteolin standards control. The excimer ion peak of Chrysin was m/z 255.0648 [M+H]+ ( δ = -0.12 ppm; δ indicates the error with the theoretical molecular mass), which was cleaved by the inverse-Diels-Alder reaction (RDA) to yield flavonoid characteristic ions m/z 153.0181 [A1,3]+ and m/z 103.0545 [B1,3]+. Fisetin belongs to the flavonol group with an excimer ion peak of m/z 287.0544 [M+H]+ (δ = -0.43 ppm); the C ring is prone to lose carbon groups and C3 has hydroxyl substitution, which is prone to lose H2O molecules, thus giving rise to the characteristic ion m/z 241.0494 [M+H-CO- H2O]+, with a continued loss of 1 molecule of CO, forming m/z 213.0542 [M+H-CO-H2O- CO]+, RDA cleavage characteristic ion m/z 151. 0069 [A1,3]+. Compounds 8 (tR = 8.64), 9 (tR=12.99) and 11 (tR=16.22) are isomers with the molecular formula C15H10O5. Compound 8 yielded ionic fragments with m/z 253.0492 [M+H-H2O]+, m/z 255.0537 [M+H-H2O-CO]+, and m/z 197.0589 [M+H-H2O-2CO]+, which were identified as apigenin according to the mzcloud database. Compound 9 and compound 11 were cleaved via RDA to yield the characteristic ions m/z 153.0179 [A1,3]+ and m/z 137.0594 [A1,3]+, which were identified as Galangin and 3',4',7- trihydroxyisoflavone, respectively, according to the reference found in Meng et al. (2020) and the mzcloud database. Compound 21 (tR = 11.32) has an excimer ion peak of m/z 287.0546 [M+H]+ (δ = -1.04 ppm), presumably with the chemical formula C15H10O6. The loss of the CO group from the C ring forms multiple molecular fragments, including m/z 259.0956 [M+H-CO]+, m/z 241.0593 [M+H-CO-H2O]+, and m/z 213.0538[M+H-CO-H2O-CO]+. The characteristic ion m/z 153.0181 [A1,3]+ was produced after RDA cleavage. It was identified as kaempferol based on the database information. PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6114 Compound 26 (tR=9.74), with an excimer ion peak of m/z 301.0699 [M+H]+ (δ = - 1.27 ppm), yielded characteristic ion fragments of m/z 286.0462 [M+H-CH3]+, m/z 258.0513 [M+H-CO]+, m/z 230.1562 [M+H-2CO]+, m/z 212.0461[M+H-2CO-H2O]+, and m/z 184.0514 [M+H-2CO-H2O-4CO]+, which was identified as diosmetin according to the mzcloud mass spectrometry library. Dihydroflavone and Dihydroflavonol Pinocembrin has an excimer ion peak at m/z 257.0801 [M+H]+ (δ = -2.00 ppm), which upon cleavage via RDA yields the characteristic ion m/z 153.0181 [A1,3]+. Liquiritigenin has an excimer ion peak at m/z 257.0801 [M+H]+ (δ = -1.60 ppm), secondary fragment-specific ions at m/z 239.0691 [M+H-H2O]+ and m/z 211.0850 [M+H-H2O-CO]+ after cleavage and characteristic ions at m/z 137.0230 [A1,3]+ after RDA cleavage. Naringenin and butin excimer ion peaks both have m/z 273.0750[M+H]+. Naringenin yielded the characteristic ion 153.0180[A1,3]+ after RDA cleavage. Butin was cleaved via collision-induced cleavage to produce characteristic ion fragments including loss of 1 molecule of H2O to produce fragment ion m/z 255.0644 [M+H-H2O]+, loss of a carbon group of C ring to produce m/z 227.0698 [M+H-H2O-CO]+, and cleavage via RDA to produce characteristic ion m/z 137.0232 [A1,3]+. Compound 24 (tR = 10.83) and compound 25 (tR = 5.00) are isomers and compound 24 was checked and identified with the mass spectral cleavage information and retention time of the Eriodictyol control. The excimer ion peak of compound 25 was m/z 289.0707 [M+H]+ (δ=-2.92 ppm) with characteristic fragmentation information of m/z 271.0953 [M+H-H2O]+, m/z 243.0633 [M+H-H2O-CO]+, and m/z 215.0670 [M+H-H2O- 2CO]+; the C-ring RDA cleavage yielded characteristic ion m/z 153.0184, which was identified as fustin according to the mzcloud database. Compound 29 (tR = 15.68) and compound 30 (tR = 11.30) both have the molecular formula C16H14O6 and are isomers. The excimer ion peak of compound 29 (tR = 15.68) is m/z 303.0860 [M+H]+ (δ = -1.20 ppm) and the characteristic ions of the secondary fragments are m/z 285.0755 [M+H-H2O]+, m/z 270.0520 [M+H-H2O-CH3]+, and m/z 242.0545 [M+H-H2O-CH3-CO]+; RDA cleavage yielded the characteristic ion m/z 153.0387, which was tentatively identified as hesperetin according to the mzcloud database. The compound 30 excimer ion peak was m/z 303.0860 [M+H]+ (δ = -1.97 ppm), and the secondary fragment characteristic ions were m/z 285.0750 [M+H-H2O]+, m/z 269.0438 [M+H-H2O-O]+, and m/z 275.0904 [M+H-CO]+, which combined with the literature analysis (Liu et al. 2005) and data identified as homoeriodictyol. Compound 31 (tR = 5.99) was identified after comparison with the mass spectral information, retention time of taxifolin control. Isoflavones Compound 1 (tR = 11.78), compound 6 (tR = 17.56), compound 10 (tR = 14.22), compound 16 (tR = 20.41), compound 18 (tR = 12.80), and compound 27 (tR = 14.37) were consistent with the mass spectrometric cleavage information and retention times of the Daidzein, formononetin, genistein biochanin A, calycosin, and tectorgenin controls, respectively. Compound 17 (tR = 14.66) and compound 19 (tR = 20.12) have excimer ion peaks m/z 285.0753[M+H]+. The loss of the carboxyl group of the C ring of compound 17 produced m/z 257.0784 [M+H-CH3-CO]+ and the continued loss of carboxyl and oxygen produced fragment ion peaks m/z 229.0853 [M+H-CH3-2CO]+ and m/z 212.0539 [M+H- PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6115 CH3-2CO-O]+; RDA cleavage produced the characteristic ion m/z 137.0230. This was identified as 2'-hydroxyformononetin in combination with the mzcloud database. Compound 19 readily loses a methyl group to produce the stronger m/z 270.0515 [M+H- CH3]+, and the C ring loses a carboxyl group to produce the fragment ion peak m/z 270.0515 [M+H-CH3-CO]+, and then loses 1 molecule of H2O to form the ion [M+H-CH3- CO-H2O]+. Combined with the mass spectral library, it was inferred to be prunetin. Compound 23 (tR = 11.39) has an excimer ion peak value of m/z 285.0753[M+H]+ (δ = -2.03 ppm), presumably with the molecular formula C16H14O5. The C ring loses its carboxyl group to form m/z 259.0956[M+H-CO]+, which continues to lose CO and H2O, yielding m/z 259.0956[M+H-CO-H2O]+, m/z 213.0538[M+H-CO]+, which was tentatively identified as the compound vestitone in combination with references found in Liu et al. (2005) and databases. Compound 32 (tR = 12.65) has an excimer ion peak equal to m/z 317.1011[M+H]+ (δ = -1.49 ppm) with the presumed molecular formula C17H16O6, yielding characteristic secondary fragments of m/z 289.1065[M+H-CO]+ and m/z 257.0802[M+H-CO-CH3OH]+, 229.0855[M+H-CO-CH3OH-CO]+, 197.0595[M+H-CO-CH3OH-CO-CH3OH]+ and 179.0697[M+H-CO-CH3OH-CO-CH3OH-H2O]+, respectively. The compound was tentatively identified as homoferreirin. Chalcone and others Compound 3 (tR = 16.72), compound 7 (tR = 16.50), and compound 14 (tR = 13.98) were identified after comparison of the mass spectral information and retention time of the isoliquiritigenin, dalbergin, and butein controls, respectively. Compound 15 (tR = 11.92) has an excimer ion peak of m/z 317.1011 [M+H]+ (δ = -1.76 ppm), with a presumed molecular formula of C16H12O5. The secondary mass spectra with characteristic fragment ions m/z 270.0512 [M+H-CH3]+, m/z 242.0564 [M+H-CH3- CO]+, and m/z 214.0615[M+H-CH3-2CO]+, which was presumed to be melannin in combination with the literature as well as the database (Liu et al. 2005). Compound 28 (tR = 17.78) has an excimer ion peak m/z 301.1056 [M+H]+ (δ = -2.48 ppm) and fragment ions with m/z 273.1113 [M+H-CO]+ and m/z 273.1113 [M+H-CO-CH3OH]+. Combined with the reference found in Wang et al. (2014) and databases, the compound was tentatively identified as Melilotocarpan A. Content of Characteristic Flavonoid Components Flavonoids are important secondary metabolites of plants with various biological activities and are considered to be the primary active ingredients of Chinese herbal medicines (Ninh 2017). Dalbergia tonkinensis is rich in flavonoid components with various biological activities, and the flavonoid components in 15 heartwood with activities were selected as the characteristic substances to evaluate the medicinal quality of D. tonkinensis heartwood. The quantification method of flavonoids was established via the external standard method (the results are shown in Table 2). Naringenin, biochanin A, butein, and pinocembrin were associated with antioxidant activity; isoliquiritigenin, daidzein, formononetin, butein, and medicarpin were associated with anti-inflammatory activities; and isoliquiritigenin, formononetin, and liquiritigenin were associated with antibacterial activities. The above flavonoids were detected in the heartwood of D. tonkinensis from different origins, and the liquiritigenin, daidzein, naringenin, isoliquiritigenin, formononetin, pinocembrin, biochanin A content was high. PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6116 Table 1. Ultra-High Performance Liquid Chromatography-Quadrupole Time-of-Flight Mass Spectrometry (UHPLC/Q-TOF-MS) Data of the Compositions Identified in D. tonkinensis Heartwood Extracted via the Method: Water (80:20) No. Identification RT Molecular Formula Error (ppm) m/z Calculated m/z Experimental Main Fragments (m/z) 1 Daidzein* 11.78 C15H10O4 2.35 255.0651 255.0645 227.0698, 199.0749, 137.0230, 91.0545 2 Chrysin 20.61 C15H10O4 1.18 255.0651 255.0648 209.0595, 181.0654, 153.0181, 103.0545 3 Isoliquiritigenin* 16.72 C15H12O4 2.72 257.0808 257.0801 239.0693, 211.0746, 137.0230, 119.0491 4 Pinocembrin* 19.48 C15H12O4 2.72 257.0808 257.0801 153.0181 5 Liquiritigenin* 10.82 C15H12O4 2.72 257.0808 257.0801 239.0691, 211.0750, 137.0230, 119.0491 6 Formononetin* 17.56 C16H12O4 2.60 269.0808 269.0801 254.0555, 253.0488, 226.0617, 197.0594, 137.0230 7 Dalbergin* 16.50 C16H12O4 2.60 269.0808 269.0801 254.0560, 253.0489, 226.0616, 198.0665 8 Apigenin 10.64 C15H10O5 3.32 271.0601 271.0592 253.0492, 225.0537, 197.0589 9 Galangin 12.99 C15H10O5 2.95 271.0601 271.0593 255.0642, 227.0695, 199.0748, 153.0179, 137.0230 10 Genistein* 14.22 C15H10O5 2.21 271.0601 271.0595 243.0650, 215.0697, 197.0595, 169.0646, 153.0180, 119.0492 11 3', 4', 7- Trihydroxyisoflavone 16.22 C15H10O5 1.84 271.0601 271.0596 243.1009, 197.0595, 137.0594, 121.0647 12 Naringenin* 13.36 C15H12O5 3.30 273.0757 273.0748 153.018 13 Butin* 8.64 C15H12O5 2.56 273.0757 273.0750 255.0644, 227.0698, 209.0594, 181.0644, 137.0232 14 Butein* 13.98 C15H12O5 1.46 273.0757 273.0753 255.0647, 227.0702, 181.0647, 137.0232, 135.0440, 15 Melannin 11.92 C16H12O5 3.16 285.0758 285.0749 270.0512, 269.0433, 242.0564, 214.0615, 213.0539 PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6117 16 Biochanin A* 20.41 C16H12O5 3.16 285.0758 285.0749 269.0442, 213.0540, 133.0645 17 2'-Hydroxyformononetin 14.66 C16H12O5 3.16 285.0758 285.0749 270.0510, 257.0784, 229.0853, 213.0539, 137.0230 18 Calycosin* 12.80 C16H12O5 2.81 285.0758 285.0750 270.0511, 269.0437, 253.0488, 213.0540, 197.0592, 169.0643, 153.0543, 137.0230 19 Prunetin* 20.12 C16H12O5 1.75 285.0758 285.0753 270.0515, 242.0569, 224.0465 20 Fisetin* 10.29 C15H10O6 2.09 287.0550 287.0544 241.0494, 213.0542, 185.0593, 157.0647, 137.0233, 121.0285 21 Kaempferol 11.32 C15H10O6 1.39 287.0550 287.0546 259.0956, 241.0593, 213.0538, 153.0181, 125.0597, 107.0493 22 Luteolin* 13.76 C15H10O6 1.74 287.0550 287.0545 23 Vestitone 11.39 C16H14O5 1.74 287.0914 287.0909 259.0956, 241.0491, 213.0538 24 Eriodictyol* 10.83 C15H12O6 2.42 289.0707 289.0700 271.0602, 153.0181, 135.0438, 107.0492 25 (-)-Fustin 5.00 C15H12O6 0.00 289.0707 289.0707 271.0953, 243.0633, 215.0670, 153.0184 26 Diosmetin 9.74 C16H12O6 2.66 301.0707 301.0699 286.0462, 258.0513, 230.1562, 212.0461, 184.0514 27 Tectorgenin* 14.37 C16H12O6 2.33 301.0707 301.0700 286.0465, 255.0642, 245.0802 28 Melilotocarpan A 17.78 C17H16O5 4.98 301.1071 301.1056 273.1113, 241.0847 29 (+/-)-Hesperetin 15.68 C16H14O6 0.99 303.0863 303.0860 285.0755, 270.0520, 242.0545, 151.0387, 123.0440 30 Homoeriodictyol 11.30 C16H14O6 0.33 303.0863 303.0864 285.0750, 275.0904, 269.0438 31 Taxifolin 5.99 C15H12O7 12.46 305.0655 305.0617 287.0536, 259.0596, 231.0650, 213.0545, 195.0290, 153.0181 32 Homoferreirin 12.65 C17H16O6 2.52 317.1019 317.1011 299.0912, 289.1065, 257.0802, 229.0855, 197.0595, 179.0697, 163.0386 Note: *Compounds marked with an asterisk indicate that the compound was identified after comparison with the standard PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6118 Table 2. Characteristic Flavonoid Content of the Three Origins of D. tonkinensis Compound Calibration caves R2 Quang Binh Quang Ninh Hanoi Butin y = 239306*x + 3274810 0.9980 1.642 ± 0.008 0.639 ± 0.006 0.636 ± 0.039 Liquiritigenin y = 441675*x + 8014080 0.9970 3.007 ± 0.046 2.426 ± 0.033 1.348 ± 0.023 Daidzein y = 53546.8*x + 468109 0.9984 2.467 ± 0.079 1.117 ± 0.031 0.524 ± 0.034 Fisetin y = 196400*x + 895704 0.9980 0.144 ± 0.007 0.055 ± 0.004 0.205 ± 0.009 Erodcyol y = 128325*x + 843979 0.9989 0.778 ± 0.006 0.298 ± 0.007 4.999 ± 0.176 Calycosin y = 740657*x + 14057900 0.9959 1.030 ± 0.022 0.344 ± 0.011 1.194 ± 0.064 Naringenin y = 237970*x + 1914080 0.9992 6.233 ± 0.033 3.317 ± 0.062 11.361 ± 0.369 Genistein y = 477227*x + 7308970 0.9973 0.492 ± 0.021 0.122 ± 0.005 0.323 ± 0.019 Butein y = 215067*x + 870351 0.9994 0.986 ± 0.006 0.262 ± 0.038 0.243 ± 0.007 Tectorigenin y = 326967*x + 3030180 0.9988 0.478 ± 0.012 0.690 ± 0.020 1.209 ± 0.080 Dalbergin y = 650801*x + 7549020 0.9985 2.452 ± 0.019 0.830 ± 0.008 0.988 ± 0.023 Isoliquiritigenin y = 522337*x + 4255890 0.9981 2.168 ± 0.029 1.867 ± 0.054 0.604 ± 0.024 Formononetin y = 629429*x + 5283250 0.9993 5.118 ± 0.146 3.345 ± 0.026 3.993 ± 0.181 Pinocembrin y = 406989*x + 5609210 0.9994 2.447 ± 0.025 1.564 ± 0.021 29.051 ± 0.998 Biochanin A y = 598135*x + 8860280 0.9993 2.094 ± 0.072 0.929 ± 0.015 5.426 ± 0.251 Analysis of the Volatile Components of D. tonkinensis The identification of the volatile components in D. tonkinensis extracts was carried out via gas chromatography-mass spectrometry, and a total of 12 substances were identified in Table 1 of the supporting material. The total ion flow diagrams of the three sources of D. tonkinensis were more consistent (as shown in Fig. 3), and the relative areas were calculated using peak area normalization for volatile components (as shown in Table 3). The volatile components of the heartwood material from the three different origins were determined to be trans-nerolidol (32.9% to 54.7%), caryophyllene oxide (13.3% to 24.4%), eudesmol (2.53% to 8.15%), bisabolol oxide (0.88% to 3.15%), and farnesyl acetone (1.46% to 2.96%). The quantitative working curve of trans-nerolidol was established using the external standard method (y = 600000x + 6000000, R2 = 0.999), and the trans-nerolidol peak areas of the heartwood from different sources were used as the response values to measure the contents (as shown in Table 4). The volatile components were also detected in the heartwood of D. odorifera, which indicated a very high similarity in the composition of the two species (Liu et al. 2005; Yang et al. 2016; Ma et al. 2022). PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6119 Table 3. Chemical Composition of the Volatile Compounds in the Oil Extracts of in D. tonkinensis Heartwood from Different Origins RT Compositions Formula RI Quang Binh Quang Ninh Hanoi 10.2926 Geranyl acetone C13H22O 1467.4975 / / 0.28 ± 0.05 11.1397 Elemol C15H26O 1544.4132 2.05 ± 0.07 2.71 ± 0.05 1.26 ± 0.18 11.2858 trans-Nerolidol C15H26O 1556.8653 54.71 ± 1.54 43.90 ± 1.20 32.92 ± 0.87 * 5.57 ± 0.21 4.81 ± 0.14 6.66 ± 0.29 11.3393 Ledol C15H26O 1561.425 / 1.90 ± 0.10 1.99 ± 0.17 11.4513 Caryophyllene oxide C15H24O 1570.9708 18.69 ± 0.64 24.42 ± 0.64 13.33 ± 1.00 11.5584 Globulol C15H26O 1580.0989 / / 1.08 ± 0.46 12.2546 Eudesmol C15H24O 1634.3097 2.53 ± 0.12 4.24 ± 0.08 8.15 ± 0.23 12.4834 Bisabolol oxide C15H26O2 1651.2754 2.55 ± 0.07 3.15 ± 0.07 0.88 ± 0.04 12.6246 α-Bisabolol C15H26O 1662 0.88 ± 0.11 1.73 ± 0.06 0.84 ± 0.07 13.8223 trans-Coniferyl alcohol C10H12O3 1745.3204 2.60 ± 0.17 2.45 ± 0.08 / 16.6607 Farnesyl acetone C18H30O 1925.2319 2.56 ± 0.50 2.96 ± 0.05 1.46 ± 0.59 20.7114 Hexadecanamide C16H33NO 2160.4745 / 1.51 ± 0.79 / Note: *row indicates the amount of trans-nerolidol in the sample (the unit is mg/g) Fig. 3. Total ion chromatograms of the extracts from D. tonkinensis heartwood via GC-MS PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6120 Comparison of the Chemical Composition of the Heartwood Extract from D. tonkinensis and D. odorifera The heartwood of D. tonkinensis and D. odorifera contained the same flavonoids (as discussed) and volatile components (as discussed) and only differed in content to effectively distinguish between the two. Furthermore, the signals of all compounds were collected via HPLC-TOF-MS and compared using principal component analysis (PCA) methods. In the positive ion mode (as shown in Fig. 4), the cumulative contribution of the PCA model was 74%. The D. odorifera heartwood, i.e., the Guangxi group (Nanning-LFJ, Nanning-GXU, Pingxiang) and Hainan group (Anding, Dongfang, Haikou), were compared with D. tonkinensis, i.e., (D. odorifera heartwood (Ledong) and D. tonkinensis (Hanoi), which were distinguished from other origins by the higher total flavonoids, nerolides, and characteristic flavonoid contents in the heartwood of these two origins. This is presumably the result of intra-species variation rather than inter-species variation. Consistent results were obtained in the negative ion model. The above results indicated that there is no significant difference in the chemical composition between D. tonkinensis and D. odorifera, which suggested that D. tonkinensis can be used as a potential substitute for D. odorifera. Fig. 4. Principal component analysis of the constituents of the extracts of D. tonkinensis and D. odorifera CONCLUSIONS 1. Thirty-two flavonoids and 12 volatile components from D. tonkinensis heartwood were identified via GC-MS and HPLC-MS. The total flavonoid content was high, ranging from 49.7 mg to 102.0 mg RE/g dry weigh, which indicated that D. tonkinensis heartwood also contained more abundant flavonoids. Butin, liquiritigenin, daidzein, fisetin, erodcyol, calycosin, naringenin, genistein, butein, and tectorigen were used as characteristic flavonoids for the identification and evaluation index of D. tonkinensis. The volatile components of D. tonkinensis were found to be very similar to those of D. odorifera, with trans-Nerolidol and Caryophyllene oxide as the primary components. PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6121 2. By comparing the chemical composition of D. tonkinensis and D. odorifera heartwood via PCA, it was concluded that there was no significant difference between the two in terms of chemical composition, and D. tonkinensis could be used as a potential substitute for D. odorifera heartwood. In this study, the identification and more comprehensive evaluation methods of D. tonkinensis heartwood were established to provide a basis for its standardized utilization. ACKNOWLEDGEMENTS This research was supported by the National Natural Science Foundation of China (Grant No. 31870540). REFERENCES CITED Celedon, J. M., and Bohlmann, J. (2018). “An extended model of heartwood secondary metabolism informed by functional genomic,” Tree Physiology 38(3), 311-319. DOI:10.1093/treephys/tpx070 Choi, H. S., Park, J., Hwang, J. S., Ham, S. A., Yoo, T., Lee, W. J., Paek, K. S., Shin, H., Lee, C., and Seo, H. G. (2017). “A Dalbergia odorifera extract improves the survival of endotoxemia model mice by inhibiting HMGB1 release,” BMC Complementary and Alternative Medicine 17(1), 212. DOI: 10.1186/s12906-017-1725-0 Fan, Z. M., Wang, D. Y., Yang, J. M., Lin, Z. X., Lin, Y. X, Yang, A. L., Fan, H., Cao, M., Yuan, S. Y., Liu, Z., et al. (2017). “Dalbergia odorifera extract promotes angiogenesis through upregulation of VEGFRs and PI3K/MAPK signaling pathways,” Journal of Ethnopharmacology 204, 132-141. DOI:10.1016/j.jep.2017.04.006 GB/T 18107-2017. (2017). “hongmu[mahogany]”, Standards Press of China, Beijing. Jones, C. G., Moniodis, J., Zulak, K. G., Scaffidi, A., Plummer, J. A., Ghisalberti, E. L., Barbour, E. L., and Bohlmann, J. (2011). “Sandalwood fragrance biosynthesis involves sesquiterpene synthases of both the terpene dynthase (TPS)-a and TPS-b subfamilies, including santalene synthases,” Journal of Biological Chemistry 286(20), 7445-17454. DOI:10.1074/jbc.A111.231787 Li, X., Wu, L. H, Liu, W., Jin, Y .C., Chen, Q., Wang, L. L., Fan, X. H., Li, Z., and Cheng, Y.Y. (2014). “A network pharmacology study of Chinese medicine QiShenYiQi to reveal its underlying multi-compound, multi-target, multi-pathway mode of action,” PLOS One 9(5), e95004. DOI: 10.1371/journal.pone.0095004 Liu, R. X., Sun, J. H., Bi, K. S., and Guo, D. A. (2005). “Identification and determination of major flavonoids in rat serum by HPLC-UV and HPLC-MS methods following oral administration of Dalbergia odorifera extract,” Journal of Chromatography B 829(1-2), 35-44. DOI:10.1016/j.jchromb.2005.09.025 Liu, H., Shi, F. J., Li, W. S., Wei, Z. M., Ma, R. K., Liu, Z. G., Li, Y. J., and Fu, Y. L. (2021). “Distinguishability of Dalbergia odorifera and Dalbergia tonkinensis in wood anatomic properties and extraction chemical compositions,” Scientia Silvae Sinicae 57(2), 103-114. DOI: 10.11707/j.1001-7488.20210211 Ma, R. K., Luo, J., Qiao, M. J., Fu, Y. L., Zhu, P. C., Wei, P. L., and Liu, Z. G. (2022). “Chemical composition of extracts from Dalbergia odorifera heartwood and its correlation with color,” lndustrial Crops and Products 180, article no. 114728. PEER-REVIEWED ARTICLE bioresources.com Wei et al. (2022). “Dalbergia odorifera extractives,” BioResources 17(4), 6108-6122. 6122 DOI:10.1016/j.indcrop.2022.114728 Ma, Y., Yang, Z., Xu, D., Liu, X., and Zhang, N. (2016). “Variations in morphology and sowing quality of Dalbergia tonkinensis seeds from different provenances in Vietnam,” Journal of South China Agricultural University 37(3), 107-114. DOI: 10.7671/j.issn.1001-411X.2016.03.017 Meng, X. W., Cai, D. J., Zhu, Q., Zhang, H. G., Wang, M. F., Liu, Y., Tang, F. R., Shao, F., Chen, F. Y., and Liu, R. H. (2020). “Rapid identification and analysis of chemical components in heartwood of Dalbergia cochinchinensis by UPLC-Q-TOF-MS/MS,” Chinese Journal of Experimental Traditional Medical Formulae 26(10), 143-156. DOI: 10.13422/j.cnki.syfjx.20200813 Moniodis, J., Jones, C. G., Barbour, E. L., Plummer, J. A., Ghisalberti, E. L., and Bohlmann, J. (2015). “The transcriptome of sesquiterpenoid biosynthesis in heartwood xylem of Western Australian sandalwood (Santalum spicatum),” Phytochemistry 113, 79-86. DOI:10.1016/j.phytochem.2014.12.009 Ninh, T. S. (2017). “A review on the medicinal plant Dalbergia odorifera species: Phytochemistry and biological activity,” Evidence-Based Complementary and Alternative Medicine 2017, 1-27. DOI: 10.1155/2017/7142370 Qin, M., Zhu, C. J., Yang, J. B., Vatanparast, M., Schley, R., Lai, Q., Zhang, D.-Y., Tu, T.Y., Klitgård, B. B., Li, S. J., et al. (2022). “Comparative analysis of complete plastid genome reveals powerful barcode regions for identifying wood of Dalbergia odorifera and D. tonkinensis (Leguminosae),” Journal of Systematics and Evolution 60(1), 73-84. DOI: 10.1111/jse.12598 Shao, F., Zhang, L., Guo, J., Liu, X., Ma, W., Wilson, I. W., and Qiu, D. (2019). “A comparative metabolomics analysis of the components of heartwood and sapwood in Taxus chinensis (Pilger) Rehd,” Scientific Reports 9, article no. 17647. DOI:10.1038/s41598-019-53839-2 Tasnim, S., Gries, R., and Mattsson, J. (2020). “Identification of three monofunctional diterpene synthases with specific enzyme activities expressed during heartwood formation in western redcedar (Thuja plicata) Trees,” Plants 9(8), article no. 1018. DOI:10.3390/plants9081018 Wang, H., Dong, W. H., Zuo, W. J., Wang, H., Zhong, W. M., and Daia, H. F. (2014). “Three new phenolic compounds from Dalbergia odorifera,” Journal of Asian Natural Products Research 9, 168-173. DOI:10.1080/10286020.2014.968559 Wu, S. Q., Li, D. Z., Lu, W., Yang, Z. Y., and Huang, G. N. (2013). “Dalbergia spp. morphological characteristics and introduction botanical name research preliminary report,” Tropical Forest 41(3), 25-26. Yang, L., Fang, C. R., Zhang, J., Yu, H. X., Wang, Z., Zhu, J. L., Yang, W. M., and Chai, Z. L. (2016). “The identification of Dalbergia odorifera and D. tonkinensis by gas chromatography mass spectrometry,” Journal of Nanjing Forestry University (Natural Sciences Edition) 40, 97-103. DOI: 10.3969/j.issn.1000-2006.2016.01.016 Zhao, X., Zhang, S. H., Liu, D., Yang, M. H., and Wei, J. H. (2020). “Analysis of flavonoids in Dalbergia odorifera by ultra-performance liquid chromatography with tandem mass spectrometry,” Molecules 25(2), 389. DOI: 10.3390/molecules25020389 Article submitted: June 20, 2022; Peer review completed: August 28, 2022; Revised version received and accepted: September 6, 2022; Published: September 14, 2022. DOI: 10.15376/biores.17.4.6108-6122