Untargeted metabolites profiling of volatile components of Chinese Antique Lotus (Nelumbo nucifera Gaertn.) using solid-phase microextraction (SPME) GC/MS

Plants and chemicals

Six Antique Lotus accessions were planted in the Gulian Pond of the China National Botanical Garden (South Garden), Institute of Botany, Chinese Academy of Sciences in Beijing, China. The following were selected: N. nucifera ‘Zhongnanhai Antique’ (ZNH), from Zhongnanhai, Xicheng District, Beijing City; N. nucifera ‘Kaifeng Antique’ (KF), from Kaifeng City, Henan Province; N. nucifera ‘Pulandian Antique’ (PLD) from Pulandian Village, Liaoning Province; N. nucifera ‘Liangshan Antique’ (LS) from Liangshan County, Shandong Province; N. nucifera ‘Zhangqiu Antique’ (ZQ) from Zhangqiu City, Shandong Province; and N. nucifera ‘Yuanmingyuan Antique’ (YMY) from Yuanmingyuan Ruins Park, Beijing City. Based on historical records, these lotuses could be dated back to 100 to 1,200 years ago (Priestley & Posthumus, 1982; Shen-Miller et al., 1995; Wang et al., 2023a). The petals, stamens, and receptacles (with pistils) were collected between 6:30 am and 7:00 am from the first-day blooms of each flower (Fig. 1, Fig. S1). A total of 1 g of fresh tissues were immediately weighed and placed into 20 mL screw-cap headspace vials after sampling, which were then sealed with aluminum caps and PTFE-silicone septa. Three biological replicates of each floral organ were collected.

The volatile compounds standards of α-pinene, caryophyllene, 1,4-dimethoxybenzene, pentadecane, and C7-C40 saturated alkane standard mixture, were purchased from Sigma-Aldrich (USA).

Headspace solid-phase microextraction (HS-SPME)

Prior to use, the DVB/CAR/PDMS extraction fiber (Supelco, Bellefonte, PA, USA) was conditioned at 250 °C for 22.5 min in the inlet of the gas chromatograph. Subsequently, the samples were equilibrated for 5 min in the autosampler (Agilent GC Sampler 80) set at 37 °C. The conditioned solid-phase microextraction (SPME) fiber was then placed into sealed headspace vials, with the fiber positioned approximately 1 cm above the sample. The headspace extraction process lasted for 40 min, after which the extraction fiber was introduced into the gas chromatograph inlet for desorption, a process that lasted 3 min, preparing the sample for gas chromatography-mass spectrometry (GC-MS) analysis.

Gas chromatography-mass spectrometry

Chromatographic conditions were optimized using an Agilent 7890A-5975C tandem triple quadrupole gas chromatography-mass spectrometer, equipped with an HP-5MS elastic quartz fiber capillary column (30 m × 250 μm × 0.25 μm) and an autosampler. High-purity helium was employed as the carrier gas at a flow rate of 1.0 mL/min and a shunt ratio of 10:1. The analysis followed a programmed heating procedure with an initial temperature of 55 °C for 3 min, then increasing to 170 °C at a rate of 3 °C/min. The total analysis duration was 41.33 min. A 1 min post-run was conducted at 230 °C.

Mass spectrometry was performed under the following conditions: electron ionization (EI) mode, ion source temperature at 230 °C, electron energy at 70 eV, quadrupole temperature at 150 °C, and transmission line temperature at 280 °C. The analysis was conducted in full scan mode, with a mass range of 30 to 600 amu.

Relative odor activity value analysis

In aroma analysis, the odor activity value (OAV) is commonly used to quantify the contribution of individual compounds to the overall aroma profile. To address the need for comparative evaluation of volatile compounds based on their relative concentrations, a modified metric, termed the relative odor activity value (rOAV), has been introduced (Liu et al., 2023). This parameter enables the assessment of the contribution of individual volatile compounds to the holistic aroma by accounting for their relative abundance within the volatile mixture. The calculation formula of rOAV is as follows: rOAV = C_i/OT_i. Where OT_i represents the threshold value of the compound in water; C_i represents the relative concentration of the volatile compound.

Quantification of volatile components

In accordance with the previously published methods, the major volatile compound standards were employed to carry out a relative quantification analysis of different types of compounds (Bao et al., 2023, 2020; Yang et al., 2022). Specifically, α-pinene was used for the quantification of monoterpenes, caryophyllene for sesquiterpenes, 1,4-dimethoxybenzene for benzenoids/phenylpropanoids, and pentadecane for fatty acids derivatives (ng/g/h). The α-pinene standard curve equation obtained was: y = 38,542,271.3137x + 16,116.2831 (R² = 0.9999). The caryophyllene standard curve equation obtained was: y = 18,282,242.3895x – 133,309.8386 (R² = 1.0000). The 1,4-dimethoxybenzene standard curve equation obtained was: y = 55,735,438.314x – 2,253,226.3245 (R² = 0.9997). The pentadecane standard curve equation obtained was: y = 74,027,300.1307x – 419,151.9155 (R² = 1.0000).

Data analysis

Structural volatile identification was based on retention time (RT), the NIST 14 mass spectral database (http://webbook.nist.gov/chemistry), and standard comparison using the Agilent Mass Hunter Qualitative Analysis B.07.00 chemistry workstation (https://www.agilent.com.cn). Kovats’ Retention Indices (RI) were determined using n-alkane standards (C7-C40), following the same heating procedure (Liao, Li & Lei, 2016). Statistical analysis was conducted using Excel 2016 (http://www.microsoft.com), and plots were generated with Origin 2022 (http://www.originlab.com) and TBtools (Chen et al., 2023). Significance analyses were performed using IBM SPSS 27 (http://www.ibm.com).

Identification of volatile components

To investigate the aroma characteristics of Antique Lotus flowers, the volatile compounds of three floral organs from six Antique Lotus were analyzed by HS-SPME/GC-MS. A total of 64 volatile compounds were identified, comprising 38 terpenoids (TPs), two benzenoids/phenylpropanoids (BPs), and 24 fatty acids derivatives (FADs) (Fig. 2A, Table S1). The TPs were further classified into 18 monoterpenes (MTs) and 20 sesquiterpenes (STs). Although the volatile compounds in different tissues of Antique Lotus vary greatly, 1,4-Dimethoxybenzene was the major components of the aromatic emissions of Antique Lotus (Tables S2–S4).

Volatile compounds from different floral tissues

Total ion chromatograph was established for three tissues (Fig. 2B, Fig. S2). The number of volatile compounds detected in different tissues also varies. The highest amount of volatile compounds was detected in the stamens, followed by the receptacle and petals, with 58, 48, and 47, respectively. A total of 32 VOCs were common across all parts, with 4, 8, and 5 unique to the petals, stamens, and receptacle, respectively (Fig. 2C). Although the amount of volatile compounds detected in petals was the lowest, the total content of volatile compounds in petals was the highest, except for KF, ranging from 6.95–11.45 µg/g/h FW. Followed by stamens, the total content was 5.18–9.29 µg/g/h FW. The lowest content was in the receptacle, which is 1.99–3.12 µg/g/h FW (Table 1, Tables S2–S4). Interestingly, there was no significant difference in the total volatile content between the stamens and petals in the ZNH, KF, and YMY. However, there was a significant difference in the total volatile content between the stamens and petals in the PLD, LS, and ZQ. Furthermore, there were significant differences in the content of volatile compounds between the stamens and receptacle.

The categorization of the volatile constituents revealed that the different lotus tissue types exhibited disparate compositional characteristics (Fig. 3). The volatile constituents of the petals were predominantly FADs, constituting 54–77% of the total. MTs, BPs, and STs were subsequently identified, in decreasing order of concentration. In contrast, the stamens and receptacle were dominated by BPs, comprising 36–69% of the total. It is noteworthy that KF and ZNH exhibited a higher content of STs than BPs in the stamens, in contrast to the other Antique Lotus. Additionally, the STs of PLD and ZQ accounted for less than 0.5% of the total.

Volatile compounds from different Antique Lotus

The number of volatile compounds detected varies among different Antique Lotus. The YMY had the highest number of volatile compounds, with 56 detected, followed by the ZNH, with 52 volatile compounds detected. Only 42 volatile compounds had detected in PLD (Fig. 4A). The number of volatile compounds detected in YMY petals is also the highest which was 37, while only 30 volatile compounds were detected in KF petals (Fig. 4B). In the stamens, the number of volatile compounds detected in YMY and ZNH was consistent and the highest among all Antique Lotus (Fig. 4C). In contrast to petals, KF detected the highest number of volatile compounds in the receptacle (Fig. 4D).

The volatile content in petals differed between different Antique Lotus, with ZQ exhibiting the highest content, followed by YMY, ZNH, KF, PLD, and LS (Fig. 5A). There were significant differences between ZQ to KF, PLD, and LS. The highest volatile content in stamen was identified in YMY and KF followed by ZNH, LS, ZQ, and PLD, while no significant differences were observed among the last three lotuses. The volatile content in the receptacle of YMY was also the highest; however, there was no significant difference in the volatile content of the receptacle among the KF, PLD, ZQ, and YMY. The principal components extracted from the principal component analysis (PCA), PC1 and PC2, accounted for 37.2% and 18.7% of the variance, respectively, indicating that different tissues could not be clearly distinguished based on volatile composition (Fig. 5B). Additionally, the petals and stamens demonstrate a more pronounced dispersion in comparison to the receptacle, which suggests that there were significant variations in the aromatic characteristics of different Antique Lotus.

The clustered heat map of volatile compounds illustrates the distinctive characteristics of six Antique Lotus, revealing notable differences among their floral tissues (Fig. 5A). The samples were grouped into three principal categories: (1) the petals of all six Antique Lotus; (2) the stamens and receptacle of PLD and ZQ, as well as the receptacle of YMY; and (3) the stamens and receptacle of ZNH and ZQ, along with the stamens of YMY.

To better understand the dominant compounds affecting aromatic properties, volatile compounds with a content exceeding 500 ng/g/h FW were selected for further analysis. Based on the clustering results of three floral organs, the volatile concentrations showed certain trends among the Antique Lotus, dividing them into two groups: (1) ZNH, YMY, KF, and LS; and (2) PLD and ZQ. In the petals, 18 compounds meeting this criterion were identified across all Antique Lotus (Fig. 6B). The 18 compounds were classified into three groups. The first group included five compounds, comprising four FADs and one BPs, present in relatively high concentrations across all lotus. The second group consists of only one STs (C21, caryophyllene), which is highly present in the A sample group. The third group encompasses 12 compounds, nine MTs and three FADs, found in higher concentrations in the B sample group than in the A. In the stamens, a total of 11 compounds were identified across all varieties and categorized into three groups (Fig. 5C). The first group consists of one BP (C40, 1,4-Dimethoxybenzene), which has the highest content across all samples. The second group comprises eight compounds, including six MTs, one BP and one FAD, which have higher concentrations in A sample group than in B sample group. The third group includes two STs, C21 (Caryophyllene) and C23 (Humulene), which are only present in B sample group. In the receptacle, five compounds were identified and classified into four groups (Fig. 5D). The first group consisted of only one Ben, which was the most abundant compound across all samples. The second group comprises a single STs, which exhibited the highest concentrations in the ZNH, KF, and LS samples. Interestingly, PLD, ZQ, and YMY did not contain this component. The third group contains one MT, which (C11, Eucalyptol) was concentrated highly in ZQ and PLD. The fourth group consists of two MTs, which (C13, γ-Terpinene and C16, Terpinen-4-ol) were more highly concentrated in sample group A than in sample group B.

Analysis of relative odor activity value (rOAV) of Antique Lotus

To further investigate the overall contribution of volatile compounds from different tissues of Antique Lotus to floral fragrance, the rOAV was employed for quantitative analysis of volatile components. Volatile compounds with rOAV > 1 were considered as aroma-active compounds contributing to the overall aroma profile of the samples. Tables S5–S7 presents the statistical data of detected volatile compounds with rOAV > 1 across various tissues of Antique Lotus. In this experimental investigation, 19 volatile compounds were subjected to rOAV analysis, with eight compounds demonstrating rOAV exceeding 1 across diverse tissue types. Comprehensive analysis identified eucalyptol, α-pinene, and β-myrcene as predominant aroma contributors in Antique Lotus. The remaining five compounds exhibited differential contributions depending on cultivars and tissues (Fig. S3). Specifically, three aromatic components, such as eucalyptol, β-myrcene, and α-pinene displayed rOAV > 10 in petal tissues, indicating their critical roles in shaping petal fragrance (Table S5). Notably, α-terpinene, limonene, and caryophyllene showed cultivar-dependent contributions, with significant aromatic impacts observed only in select cultivars. Stamen tissues exhibited distinct patterns: while eucalyptol and β-myrcene maintained rOAV > 10 across cultivars (consistent with petal profiles), caryophyllene uniquely contributed (rOAV > 1) to ZNH, KF, LS, and YMY cultivars (Table S6). Receptacle tissues displayed exceptional rOAV > 100 for eucalyptol, highlighting its dominance in this tissue type (Table S7). Due to the highest rOAV value of eucalyptol, which serves as the primary aroma contributor in the Antique Lotus group, combined with the complementary effect of β-myrcene and α-pinene, these components collectively formed a camphor-like odor accompanied by subtle spicy, woody and grassy notes. In contrast, cultivars containing caryophyllene exhibited a distinct clove-like aroma profile.