Metabolomics reveals differences in metabolite profiles between wild and cultivated Tetrastigmae Radix

2.2.1 Chemicals and sample preparation

The experiment utilized a set of standards, as follows: lysine, histidine, serine, aspartic acid, threonine, glutamic acid, proline, valine, tyrosine, isoleucine, leucine, phenylalanine, epigallocatechin, catechin, epicatechin, orientin, isoorientin, vitexin-2″-O-rhamnoside, aromadendrin, hyperoside, isoquercitrin, quercitrin, astragalin, nicotiflorin, narcissin, quercetin, kaempferin, luteolin, kaempferol, isorhamnetin, apigenin, cytidine, uracil, hypoxanthine, adenosine, uridine, 2′-deoxyadenosine, guanosine, inosine, 2′-deoxyguanosine, 2′-deoxyinosine, gallic acid, protocatechuic acid, 3,4-dihydroxybenzaldehyde, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, caffeic acid, polydatin, piceatannol, resveratrol, procyanidin B2, procyanidin B1. Information regarding the standards can be found in Table S2.

The dried TR was pulverized and subsequently sifted through a 65-mesh sieve, resulting in the production of a dried TR powder. A precise amount of 5 mL of ethanol with a concentration of 60 % was added, the container was sealed, and its weight was measured. Ultrasonic treatment was conducted at a power of 500 W and a frequency of 40 kHz for 50 min, followed by a cooling period. To compensate for the weight loss, a solution containing 60 % ethanol should be prepared, thoroughly mixed, and subsequently filtered. The filtrate underwent centrifugation at a speed of 12,000 revolutions per minute for 10 min, and subsequently, the supernatant was collected. The solution was passed through a microporous filter membrane with a pore size of 0.22 μm and collected in a vial containing the liquid phase. Subsequently, the samples were injected into a UFLC-Triple TOF-MS/MS instrument for analysis. Equal quantities of extracts from each sample were combined and employed as quality control (QC) samples.

2.2.2 Chromatography and mass spectrometry conditions

Samples were subjected to analysis using an Ultra-Fast Liquid Chromatography (UFLC) system manufactured by Shimadzu, located in Kyoto, Japan. Chromatographic column: Agilent ZORBAX SB-C₁₈ column (4.6 mm × 250 mm, 5 μm); mobile phase: methanol (A) - water (containing 0.4 % formic acid, B); gradient elution: 0 min, 5 % A; 6 min, 21 % A; 20 min, 35 % A; 26 min. 45 % A; 30 min, 60 % A; 34 min, 82 % A; 36 min, 90 % A; 40 min, 90 % A; 43 min, 5 % A; 46 min, 5 % A; column temperature of 30 ℃, flow rate of 0.9 mL/min, and injection volume of 10 μL.

Mass spectrometry data were obtained using an AB Sciex Triple TOF TM 5600 system-MS/MS (AB SCIEX, Framingham, MA, USA) equipped with an electrospray ionization (ESI) source. ESI data were acquired using both positive and negative ion modes. The mass scan range was set to m/z 50–1500. The temperature of the ion source (TEM) was maintained at 550 ℃. In negative ion mode, the spray voltage (IS) was set to −4500 V, while in positive ion mode, it was set to 4500 V. The declustering potential (DP) was set to −100 V in negative ion mode and 100 V in positive ion mode. The mass sweep range was set to 50 to 1500 m/z. The curtain gas pressure (CUR) was maintained at 40 psi, while the atomizing gas pressure (GS1) and auxiliary gas pressure (GS2) were both set to 60 psi.

2.2.3 Identification of metabolite profiling and data analysis

A metabolite database of TR was built by conducting a comprehensive search across various databases, including China National Knowledge Infrastructure (CNKI), HMDB, ZINC, ChemSpider, and PubMed. SciFinder was utilized to validate the chemical name, molecular formula, and structure of the metabolites. The identification of metabolites in the samples was achieved through a combination of methods. These methods involve matching retention times, precise measurement of masses against established standards and databases, cross-referencing metabolites and databases, and the guidelines outlined in relevant literature about structural modifications in metabolites. UFLC-Triple TOF-MS/MS data were analyzed using PeakView 1.2 (AB Sciex) and MarkerView 1.2.1 software (AB Sciex).

2.5.1 Chemicals and sample preparation

Since catechin, procyanidin B2, astragalin, and kaempferin were located in the pathway that was enriched according to KEGG enrichment analyses, and these metabolites in TR have been found to possess significant pharmacological activity (Ding et al., 2019; Fu et al., 2015; Lee et al., 2017; Sun et al., 2015). Therefore, in order to further explore the differences between the two ecotypes, a quantitative analysis was conducted on four metabolites. Detailed information regarding the four standards can be found in Table S2.

The TR was subjected to hot air drying at a temperature of 60 °C, followed by pulverization using a powdering machine. Then 0.5 g of TR powder was weighed, screened with a 65-mesh sieve and transferred into a 50 mL centrifuge tube. Next, add 12.5 mL of 60 % ethanol to the tube. After subjecting the sample to 50 min of ultrasonication at a power of 500 W and a frequency of 40 kHz, the weight loss was compensated for using a 60 % ethanol solution. The filtrate was subsequently subjected to filtration and centrifugation at a speed of 12,000 revolutions per minute for 10 min. The supernatant underwent filtration using a microporous membrane with a pore size of 0.22 μm. Subsequently, the filtered solution was injected into a liquid phase vial to serve as the sample solution.

2.5.2 Chromatography and mass spectrometry conditions

The chromatographic analysis of TR was conducted using a SIL-20A XR system manufactured by Shimadzu in Kyoto, Japan. The separation was performed using the XBridge®C₁₈ column (4.6 mm × 100 mm, 3.5 µm) at a temperature of 30 ℃, with an injection volume of 2 µL. The mobile phase consisted of a 0.4 % formic acid water solution (A) and a methanol solution (B) with a flow rate of 0.8 mL/min. The gradient elution was performed as follows: 0 min, 7 % B; 4 min, 9 % B; 6 min, 21 % B; 10 min, 35 % B; 12 min, 38 % B; 16 min, 46 % B; 20 min, 64 % B; 21 min, 7 % B.

The AB SCIEX API5500 triple quadrupole linear ion trap tandem mass spectrometer, equipped with an electrospray ionization (ESI) source, was utilized for detection (Framingham, MA, USA). The ion source temperature was set at 550 ℃. The nebulizer gas flow, auxiliary gas flow, and curtain gas flow were 55 psi, 55 psi, and 40 psi respectively. In the negative mode, the spray voltage (IS) was set to −4500 V. Analytes were detected using the multiple-reaction mode (MRM). The method validation for the quantitative analysis of the four targeted metabolites is presented in Table S3.

3.1.1 Identification of flavonoids

Flavonoids are secondary metabolites produced in natural plants in response to environmental stress. Flavonoids possess potent antioxidant, anti-tumor, and other pharmacological activities that are advantageous for human health. Most of the flavonoids are derivatives of chromone, whose basic nucleus is 2-phenylchromone (C₆ − C₃ − C₆). This nucleus consists of three rings, namely A, B, and C. The classification of flavonoids is based on the attachment of the benzene ring to the middle C₃ portion, the position of the B ring attachment, and the oxidation degree of the C₃ portion. Flavonoids are the major active pharmaceutical metabolites found in TR. The flavonoids those were identified in this study predominantly consisted of flavonols, flavones, flavanols, flavanonols, flavanones, dihydrochalcones, isoflavones, and their respective glycosides.

Flavones and flavonols: The C ring of flavones and flavonols is not easy to open due to the stable presence of double bonds. In the negative ion mode, the C ring tends to lose neutral fragments such as CO, CO₂, and H₂O, or experiences RDA cleavage. For instance, compound 114 produced a fragment of m/z 273.0399 [M − H − CO]⁻ after the initial loss of a CO fragment (Pascale et al., 2020). Simultaneously, the C ring had RDA cleavage generating m/z 151.0032 [^1,3A]⁻ and m/z 149.0242 [^1,3B]⁻, with similar cleavage patterns including 115, 117, 120, and others (Śliwka-Kaszyńska et al., 2022; Soubeyrand et al., 2018). Furthermore, it has been observed that glycosides of flavones and flavonols undergo significant RDA cleavage following the removal of saccharides. This cleavage was characterized by a notable response of fragment m/z 151.0032 [^1,3A]⁻, such as compounds 93, 97, 104, 105, and etc (Tsimogiannis et al., 2007).

Flavanols: The C ring of flavanols is unstable and undergoes multiple cleavages in negative ion mode. For instance, the C rings of compounds 55, 63 were firstly broken between the 1-bond and the 2-bond, 3-bond, and the 4-bond, producing fragments such as m/z 167.0345 [^1,2A]⁻, 165.0193 [^1,4B]⁻, 151.0394 [^1,3B]⁻, and 137.0241 [^1,3A]⁻ (Fig. S1A); secondly, they lost B rings and produced m/z 179.0345 [M − H − B ring]⁻; in addition, they lost CO₂ and produced 245.0808 [M − H − CO₂]⁻ (Tsimogiannis et al., 2007). Due to these similar fragments of the two compounds, they were finally identified as 55 (catechin) and 63 (epicatechin) based on the retention times of the standards.

3.1.2 Identification of lipids

Lipids play a crucial role as metabolites in the growth, development, and response to environmental changes in plants. Moreover, it is imperative to note that this particular nutrient holds significant importance in fulfilling the nutritional requirements of the human body from external sources. There is a limited body of research on lipids in TR. Zeng (Zeng et al., 2017) conducted a study and successfully identified a total of 20 lipids in TR. In this study, 28 lipids were identified in CTR and WTR. These components primarily consisted of phospholipids, glycolipids, sphingolipids, fatty acids, and other miscellaneous lipid types.

Phospholipids: Phospholipids are organic substances consisting of glycerol, phosphoric acid, fatty acids, and choline. In the positive ion mode, phospholipids exhibited a significant response in the primary mass spectrometry, producing the quasi-molecular ion peak [M + H]⁺; in the secondary mass spectrometry, they were found to lose fragments such as H₂O, glycerol, fatty acid chains, phosphoric acid or choline. For instance, compound 150 lost H₂O to generate m/z 502.3288 [M + H − H₂O]⁺, and then lost glycerol with fatty acids to produce the featured fragments m/z 184.0735 (Fig. S1B). Finally, it was identified as linoleoyllysolecithin by database and literature search (Zeng et al., 2017).

Glycolipids: Glycolipids have pharmacological activities, including antibacterial, antiviral, and antihemolytic properties. All the glycolipids identified in this study were glyceroglycolipids. Primary mass spectrometry in positive ion mode showed a better response of the quasi-molecular ion peak [M + H]⁺ for compound 149. The mass spectrometry software calculated the molecular formula, which was presumed to be a lipid. In the secondary mass spectrometry (Fig. S1C), fragments such as m/z 515.3165 [M + H − Gal]⁺ and 319.1963 were found. It was identified as gingerglycolipid A based on a database and literature search (Zeng et al., 2017). In contrast, compound 142 was cleaved differently. In the negative ion mode, the quasi-molecular ion peaks of the compound were hardly observed in the primary mass spectrometry; the molecular formula of the compound was calculated by the mass spectrometry software and database, and it was presumed to be a combination of the compound and a molecule of formic acid; in the secondary mass spectrometry, fragments such as m/z 397.1356 [M − H − C₁₈H₃₁O₂]⁻ was observed. It was identified as gingerglycolipid B based on a database and literature (Zeng et al., 2017).

3.1.3 Identification of amino acids

Amino acids are organic substances that contain amino and carboxy. Many of them are the units of protein molecules, and some of them are called essential amino acids that the human body must have but cannot synthesize on its own. In this study, 15 amino acids were identified in CTR and WTR, including six essential amino acids (lysine, threonine, leucine, isoleucine, and valine). In the positive ion mode, amino acids generally lose carboxy and amino. For instance, compound 45 produced the quasi-molecular ion peak [M + H]⁺ in the primary mass spectra; in the secondary mass spectrometry, it lost carboxy first to produce 120.0817 [M + H − FA]⁺, and then lost amino to produce 103.0561 [M + H − FA − NH₃]⁺. It was finally identified as phenylalanine based on standards. Similar cleavage patterns were found for compounds 23, 35, and 36.

3.1.4 Identification of nucleosides

Nucleosides are a part of nucleotides, and they have a positive impact on the body's immune system, nervous system, and cardiovascular system. Nucleoside is a glycoside formed by the condensation of a nucleobase with a saccharide. Eleven nucleosides were identified in this study. In the positive ion mode, the primary mass spectrometry of nucleoside produced the quasi-molecular ion peak [M + H]⁺ or [2 M + H]⁺; in the secondary mass spectrometry, the nucleoside lost its nucleobase, ribose, or deoxyribose. For instance, compound 32 firstly lost ribose to produce m/z 113.0361 [M + H − Rib]⁺, and then lost amino to produce m/z 96.0112 [M + H − Rib − NH₃]⁺, which was finally identified as uridine by the standard. In the same case, compound 33 firstly lost deoxyribose to produce m/z 136.0622 [M + H − Deoxyrib]⁺, and then lost amino to produce m/z 119.0365 [M + H − Deoxyrib − NH₃]⁺ (Chen et al., 2021). The compound 33 was identified as 2′-deoxyadenosine by the standard.

3.1.5 Identification of stilbenes

Stilbene is a general term for monomers and their polymers with a 1,2-stilbene backbone. Stilbenes are found in edible or medicinal plants, such as grapes, and possess antibacterial, antioxidant, antitumor, and antidiabetic properties. In this study, four stilbenes were identified in CTR and WTR, including polydatin, piceatannol, oxyresveratrol, and resveratrol (Bai et al., 2019). In the negative ion mode, it was found that the primary mass spectrometry of stilbenes generated the quasi-molecular ion peak [M − H]⁻, and C₂H₂O was lost in the secondary mass spectrometry. For instance, compound 102 initially lost one molecule of C₂H₂O in the negative ion mode, generating the m/z 185.0607 [M − H − C₂H₂O]⁻; then it lost another molecule of C₂H₂O to produce m/z 143.0503 [M − H − 2C₂H₂O]⁻. Finally, it was identified as resveratrol by the standard.