Integrated GC–MS and UPLC-Q-Orbitrap MS untargeted metabolomics for the comprehensive identification of metabolites and characterization of metabolic pathways in key tissues of Cinnamomum cassia

Abstract

Cinnamomum cassia, a commercially valuable aromatic medicinal plant, is widely used in the food, pharmaceutical and perfume industries. Here, we characterizd the metabolites and metabolic pathways in C. cassia bark, fruits and leaves using integrated GC–MS and UPLC-Q-Orbitrap MS techniques. A total of 71 volatile and 2882 non-volatile metabolites were identified, with terpenoids as the dominant class. Bark exhibited the highest volatile oil content (2.30%), which was dominated by trans-cinnamaldehyde (937 mg/g of oil). Tissue-specific biomarkers were screened from differential metabolites. Specifically, fruits accumulated higher levels of fatty acids, terpenoids and alkaloids relative to leaves, while bark was enriched in carbohydrates, phenylpropanoids, and terpenoids. Furthermore, a putative acetyl-CoA-centered inter-tissue metabolite exchange network was proposed, which might link photosynthate-derived acetyl-CoA from leaves to bark and fruits. This study advances our understanding of C. cassia metabolism and provides novel insights for its quality evaluation and comprehensive utilization of multi-tissue resources.

Highlights

• •First tissue-specific C. cassia phytochemical profile via metabolic pathway analysis.
• •Bark showed the highest volatile oil content, with terpenoids predominant by GC–MS.
• •Fruits accumulated fatty acids specifically; bark was dominated by phenylpropanoids.
• •The proposed acetyl-CoA-centered network highlights potential metabolic crosstalk.

1. Introduction

As a member of the Lauraceae family, Cinnamomum cassia Presl, widely recognized as Chinese cinnamon, is a commercially valuable culinary spice and a traditional medicinal herb with extensive cultivation across Southeast Asia, Indonesia, South America, and southern China (Guangxi, Guangdong, Yunnan, Fujian; >270,000 ha) (Geng et al., 2012; Wang et al., 2009). Over 160 chemical compounds have been isolated from C. cassia, including volatile components (e.g. cinnamaldehyde, sesquiterpenes) and non-volatile metabolites (e.g. flavonoids, alkaloids) (Zhang et al., 2019).This species exhibits diverse pharmacological effects, including anti-tumor, anti-inflammatory and analgesic, anti-diabetic, anti-obesity, anti-bacterial, anti-fungal, anti-viral properties, cardiovascular protective, cytoprotective, neuroprotective, immunoregulatory effects, and anti-tyrosinase activity (Gu et al., 2024; Zhang et al., 2019). C. cassia has been officially listed in the Pharmacopoeia of the People's Republic of China (ChP) since 1963. Notably, more than 500 formulations incorporating C. cassia used to treat inflammatory diseases, chronic gastrointestinal diseases, and gynecological disorders (Han et al., 2024).

Its bark(“Rougui”), the main harvested part, is rich in volatile oils. Besides, its (cinnamaldehyde ≥85% v/v). It exerts tonifying-fire and cold-dispelling effects clinically, and serves as a core raw material for food additives, daily chemicals and botanical pesticides (Chinese Pharmacopoeia Commission, 2025; Geng et al., 2011). Leaves (“Rougui Ye”) yield cinnamon oil (cinnamaldehyde, eugenol, cinnamyl acetate) with insecticidal, anti-inflammatory, and antioxidant properties (Zhang et al., 2019), while dried immature fruits (“Rougui Zi”) function as a food spice, dietary supplement, flavoring agent, and preservative, enriched in antimicrobial sesquiterpenoids (Guoruoluo et al., 2017; Han et al., 2024). Despite extensive characterization of C. cassia's bioactive components, the molecular and physiological mechanisms underlying the tissue-specific accumulation, inter-organ transport, and dynamic exchange of specialized metabolites (e.g., cinnamaldehyde) in C. cassia remain largely unknown.

Cinnamaldehyde, the official indicator component of C. cassia (Chinese Pharmacopoeia Commission, 2025; Neto et al., 2022), is synthesized via the phenylpropanoid pathway—one of the plant's most critical secondary metabolic pathways. Terpenoids represent the second-largest class of volatile organic compounds (VOCs) in C. cassia, biosynthesized through two conserved pathways: the cytoplasmic mevalonate (MVA) pathway and plastidial methylerythritol phosphate (MEP) pathway (Zhang et al., 2019). Acetyl coenzyme A (acetyl-CoA) acts as the initial substrate for the MVA pathway, with acetyl-CoA C-acetyltransferase (AACT) catalyzes the formation of acetoacetyl-CoA—a key step in terpenoid biosynthesis (Wang et al., 2017). Isopentenyl diphosphate (IPP) links the MVA and MEP pathways, mediating metabolic flux between these routes (Henry et al., 2018). Specifically, the MVA pathway ultimately gives rise to sesquiterpenes and triterpenes, whereas the MEP pathway is responsible for the biosynthesis of monoterpenes and diterpenes (Kempinski et al., 2019; Manina & Forlani, 2023). However, the tissue-specific differences in the metabolic pathways governing the biosynthesis of active ingredients in C. cassia remain to be fully elucidated.

Metabolomics has driven significant advances in traditional Chinese medicine (TCM) research, with gas chromatography–mass spectrometry (GC–MS) and ultra-performance liquid chromatography-quadrupole-Orbitrap mass spectrometry (UPLC-Q-Orbitrap MS) emerging as powerful tools for metabolite profiling due to their high sensitivity and specificity (Wang, Yang, et al., 2023; Xie et al., 2025). These techniques have been successfully applied to identify biomarkers and characterize metabolic pathways in medicinal plants, e.g., Clausena lansium (Fan et al., 2020), Panax quinquefolius (Pang et al., 2023), and citrus peels (Wang et al., 2024), as well as to distinguish metabolite profiles in different forms of Forsythiae Fructus (Xie et al., 2025).

Here, we employed GC–MS and UPLC-Q-Orbitrap MS to analyze volatile and non-volatile metabolites in C. cassia leaves, fruits, and bark. Metabolomic profiling and pathway analyses were used to characterize the plant's phytochemical landscape and metabolic network interactions. We report tissue-specific metabolite distributions, construct a comprehensive map of metabolite allocation and inter-tissue exchange mechanisms, and provide a foundation for future investigations into C. cassia metabolite biosynthesis and targeted regulatory strategies.

2. Materials and methods

2.1. Plant materials and chemicals

Tissues of C. cassia (fruits, bark, and leaves) were collected in April 2025 from the Germplasm Resource Planting Base of Liangtian Village, Tonghe Town, Pingnan County, Guigang City, Guangxi Zhuang Autonomous Region, China (110°27′58″E, 23°47′01″N). The samples were authenticated by Associate Professor Zhonghua Dai of the Guangxi University of Chinese Medicine. The voucher specimens were deposited in the University herbarium (No. 005382). For GC–MS analysis, samples were air-dried indoors under cool, dry, and well-ventilated conditions, then ground into powder, sealed, and stored at room temperature for essential oil extraction. Aliquots of the same samples were immediately frozen in liquid nitrogen after collection and stored at −80 °C for subsequent UPLC-Q-Orbitrap MS analysis. All samples were preserved in our laboratory.

trans-Cinnamaldehyde (HPLC ≥98%, No. D28HB204416) was purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). Ethyl acetate (No. C1808097) was obtained from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China) and anhydrous sodium sulfate (No. 20201215) from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). LC-MS-grade methanol, acetonitrile, and 2-propanol were supplied by CNW Technologies (Shanghai, China). LC-MS-grade formic acid was purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.2. Determination of VOCs in key tissues using GC–MS

Powdered C. cassia fruits, bark, and leaves (50 g each, accurately weighed) were subjected to steam distillation for volatile extraction. The resultant oils were dehydrated over anhydrous sodium sulfate overnight, dissolved in ethyl acetate, and made up to 5 mL in a volumetric flask. Following filtration through a 0.22 μm organic membrane, the filtrates were transferred into glass vial for GC–MS analysis (GCMS-TQ8050NX, Shimadzu Corporation, Kyoto, Japan).

Analyses were performed on a SH-I-5Sil MS capillary column (30 m × 0.25 mm i.d. × 0.25 μm film thickness). A 1 μL aliquot was injected in split mode with a split ratio of 60:1. The injection port, transfer line, and ion source temperatures were adjusted to 230 °C, 280 °C, and 230 °C, respectively. The GC oven temperature program was optimized as follows: initial hold at 90 °C for 3 min; ramped to 135 °C at 10 °C min⁻¹ (3 min hold); increased to 170 °C at 10 °C min⁻¹ (4 min hold); and finally elevated to 220 °C at 15 °C min⁻¹ (4 min hold). Ultra-high-purity helium (99.999%) was used as the carrier gas at a constant flow rate of 1.0 mL min⁻¹. Electron ionization (EI) was applied at 70 eV, with a mass scanning range of m/z 30–550 and a 3-min solvent delay to avoid solvent interference.

trans-Cinnamaldehyde quantification was carried out by the external calibration curve method. All experiments were designed with three independent biological replicates per tissue type, and each biological replicate was analyzed in triplicate (technical replicates), resulting in a total of n = 9 measurements per tissue for statistical robustness.

2.3. Analysis of non-volatile compounds in key tissues using UPLC-Q-Orbitrap MS

Under low-temperature conditions, 25 mg of samples were accurately weighed into Eppendorf (EP) tubes, followed by extraction with 500 μL of chilled solvent (methanol: acetonitrile: water, 2:2:1, v/v) spiked with a mixture of isotope-labeled internal standard. The suspension was vortex-mixed for 30 s, bead-beaten at 35 Hz for 4 min, and sonicated in an ice-water bath for 5 min. After three homogenization cycles, protein precipitation was achieved by incubation at −40 °C for 60 min. The extracts were clarified by centrifugation at 13800 ×g (4 °C, 15 min; R = 8.6 cm), and the resulting supernatants were transferred to injection vials for UPLC-Q-Orbitrap MS analysis. Three independent biological replicates were prepared for each tissue.

Chromatographic separation was carried out on a Vanquish UHPLC (Thermo Fisher Scientific, Waltham, MA, USA) coupled to an Orbitrap Exploris 120 mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA), equipped with a Phenomenex Kinetex C₁₈ column (2.1 mm × 50 mm, 2.6 μm). The mobile phase consisted of eluent A (0.01% v/v acetic acid in water) and eluent B (isopropanol–acetonitrile, 1:1, v/v); the autosampler was maintained at 4 °C, with an injection volume of 2 μL per sample.

Mass spectrometry data were acquired in information-dependent acquisition (IDA) mode using the Orbitrap Exploris 120 mass spectrometer, controlled by Xcalibur software (Thermo Fisher Scientific). The system performed real-time full-scan MS evaluation for on-the-fly initiation of MS/MS scans. The electrospray ionization (ESI) source settings were as follows: sheath gas at 50 arbitrary units (AU), auxiliary gas at 15 AU, capillary temperature at 320 °C; resolution was at 60000 for survey scans and 15,000 for tandem spectra.; stepped normalized collision energies (NCE) of 20%, 30%, and 40%, and spray voltages of 3.8 kV (positive) and −3.4 kV (negative).

2.4. Data preprocessing and annotation

For GC–MS data, compound identification was conducted by matching mass spectra against the NIST20, NIST20S, and AROMA-5MS libraries. Compounds detected in all three tissues (fruits, bark, and leaves) were defined as common metabolites, while tissue-specific metabolites were those exclusively identified in a single tissue.

For UPLC-Q-Orbitrap MS raw data, preprocessing was initiated by converting files to mzXML format using ProteoWizard software. Subsequent peak detection, extraction, alignment, and integration were performed via an in-house R script based on the XCMS package. Metabolite annotation was carried out through database matching against Biotree DB (v3.0) withR-based computational tools.

Multivariate statistical analysis, including principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA), was conducted in SIMCA 16.0.2 and MetaboAnalyst (http://www.metaboanalyst.ca/) to characterize metabolic differences among tissues. Differential metabolites were screened with the following thresholds: variable importance in projection (VIP) > 1 (from the OPLS-DA), fold change (FC) ≥ 2, and p-value <0.05. Metabolic pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.kegg.jp/).

2.5. Statistical analyses

One-way ANOVA and Student's t-test were employed to evaluate statistical differences using IBM SPSS Statistics 24.0. All figures were generated with GraphPad Prism 5.0. A double asterisk (**) denotes highly significant difference at p < 0.01.

3. Results

3.1. Analysis of the VOCs in the fruits, bark, and leaves of C. cassia using GC–MS

Fruits, bark, and leaves of C. cassia were collected, with their morphological traits documented (Fig. 1A). The leaves were long-ovate to sublanceolate and involute, with a green, glossy adaxial surface and a pale green, matte abaxial surface. Fruits were elliptical with a yellow-brown epidermis. Bark exhibited a grooved or rolled morphology, featuring a gray-brown, slightly rough outer surface and a reddish-brown, relatively smooth inner surface with fine longitudinal striations.

Fig. 1.

The tissue characteristics and volatile components analysis of C. cassia. (A) The phenotypic characteristics of plants and their fruits, bark and leaves. (B) Specific compounds and common compounds in fruits, bark and leaves. (C) GC–MS chromatograms of samples. The content of volatile oils (D) and trans-cinnamaldehyde (E) of three tissues. Error bars represent 95% confidence intervals. Values marked with “**” indicate statistically significant differences (p < 0.01).

VOCs in the three tissues were analyzed and quantified via GC–MS. A total of 71 volatile compounds were identified, including 24 fruit-specific, 10 bark-specific, 20 leaf-specific, and 9 compounds common to all tissues (Fig. 1B). As detailed in Supplementary Table 1, terpenoids dominated the tissue-specific volatile profiles. Notably, 90.00% (9/10) of bark-specific volatile compounds were terpenoids.

The nine common VOCs across all tissues were trans-cinnamaldehyde, copaene, β-caryophyllene, γ-muurolene, α-guaiene, α-muurolene, γ-cadinene, (−)-isoledene, and α-bisabolene. Fruit-specific compounds included 2-undecanone, (−)-aristolene, lauric acid, ylangene, and (+)-aromandendrene, while bark-specific VOCs comprised cubenene, τ-muurolol, di-epi-1,10-cubenol, β-Bisabolol and (+)-Cyclosativene. Leaf-specific constituents featured cinnamyl acetate, o-methoxycinnamic aldehyde, 2-Methoxyphenylacetone, o-anisaldehyde, and benzaldehyde. Additionally, four compounds (α-calacorene, (+)-sativene, β-elemene, and (±)-β-copaene) were shared by fruits and bark, whereas cis-cinnamaldehyde, α-humulene, benzenepropanal, and α-bisabolol, were common to bark and leaves.

Volatile oil contents in bark (2.30%) and leaves (1.11%) were 4.51-fold and 2.18-fold higher than that in fruits (0.51%), respectively, with statistically significant differences among the three tissues (p < 0.01) (Fig. 1 C, D). As the primary bioactive compound and the official indicator of C. cassia specified in the ChP, trans-cinnamaldehyde content was quantified accurately using a standard calibration curve. The curve exhibited excellent linearity over the concentration range of 0.52–2.09 mg/mL, with a correlation coefficient (R²) of 0.9996 (Fig. S1). GC–MS analysis showed that trans-cinnamaldehyde accumulated at the highest level in bark (937.35 ± 42.10 mg/g), followed by leaves (772.22 ± 53.57 mg/g), with significant differences observed among fruits, bark and leaves (p < 0.01)(Fig. 1E).

3.2. Metabolite profiling of non-volatile components by UPLC-Q-Orbitrap MS

To comprehensively characterize the non-volatile metabolic profiles of C. cassia, UPLC-Q-Orbitrap MS-based metabolomics was performed to quantify metabolite abundance across fruits, bark, and leaves. PCA was used to reduce data dimensionality and visually assess inter-group variability. The PCA score plot revealed distinct metabolic profiles among fruits, bark, and leaves, indicating clear tissue-specific clustering (Fig. 2A). A total of 2882 metabolites were detected, which were hierarchically classified into 8 superclasses, 59 classes, and 260 subclasses (Fig. 2B). Among these, shikimates and phenylpropanoids accounted for the largest proportion (28.07%), followed by terpenoids (16.17%), fatty acids (8.81%), and alkaloids (6.59%) (Fig. 2C). To further clarify tissue-specific metabolic variations, differential metabolite analysis was conducted. Specifically, 1659 differential metabolites between fruits and leaves (711 upregulated, 948 downregulated) (Fig. 2D); 1533 between bark and leaves (770 upregulated, 763 downregulated) (Fig. 2E); and 1601 between fruits and bark (918 upregulated, 683 downregulated) (Fig. 2F).

Fig. 2.

Metabolite characteristics of three tissues of C. cassia. (A) Principal component analysis (PCA) of metabolites in different tissues. (B) Donut Plot of metabolite classification and proportion. Different colour blocks represent different classification categories. (C) The distribution of metabolites in eight superclasses. The percentage indicates the proportion of metabolites of a specific type relative to the total number of identified metabolites. Volcano plot of differential metabolites in the fruit-leaf group (D), bark-leaf group (E), fruit-bark group (F). Significantly upregulated metabolites are colored red, significantly downregulated metabolites are shown in blue, and metabolites without significant differences are depicted in gray. (G-M) The heatmap of metabolites. (G) flavonoids; (H) alkaloids; (I) fatty acids; (J) other phenylpropanoids except flavonoids; (K) carbohydrates; (L) terpenoids; (M) amino acids and peptides. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.3. Differential metabolites screening via multivariate statistical analysis

Multivariate statistical analysis confirmed significant differences in metabolite profiles among fruits, bark, and leaves. By integrating data from positive (POS) and negative (NEG) ion modes, 78, 77, and 107 differential metabolites were identified in leaf–fruit, leaf–bark, and fruit–bark comparisons, respectively (Table S2–S4). Venn analysis of these three pairwise comparisons (fruit vs leaf, leaf vs bark, fruit vs bark) revealed 13 metabolites exhibited significant differences in both the fruit–leaf and bark–leaf comparisons (Fig. S2). Notably, these 13 metabolites showed no significant differences between fruits and bark, suggesting their potential as metabolic biomarkers specific to C. cassia leaves. For bark-specific metabolic biomarkers, 32 differential metabolites were identified, including procyanidin B2, epicatechin and catechin. Additionally, oxolinic acid was detected across all three tissues (Table 1). Overall, differential metabolites were mainly classified into seven categories, with distinct tissue-specific accumulation patterns: phenylpropanoids predominated in bark, alkaloids and fatty acids were mainly accumulated in fruits, and carbohydrates and flavonoids were concentrated in leaves (Fig. 2G–M).

Table 1.

The specific and common DEM compounds of different tissues detected by UPLC-MS/MS in C. cassia.

No.	Leaf DEM compounds (13)a	Fruit DEM compounds (0)b	Bark DEM compounds (32)c	Common compounds (1)d
1	l-Glutamine		Epicatechin	Oxolinic acid
2	Hygric acid		Catechin
3	Pipecolic acid		Procyanidin B2
4	Orotic acid		Scopolin
5	Guaiazulene		2-Oxo-4-phenylbutyric acid
6	12-Hydroxydodecanoic acid		Benzoic acid
7	Kinetin		Sinapaldehyde
8	Synephrine		p-Methoxycinnamaldehyde
9	Zearalanone		Hydroxyphenyllactic acid
10	p-Tolualdehyde		3,4-Dihydroxyhydrocinnamic acid
11	Elemicin		Pinellic acid
12	Cinnamtannin D2		Eriodictyol
13	N-Formylmethionine		Lonicerin
14			Scopoletin
15			Arctigenin
16			Esculin
17			Umbelliferone
18			Canavanine
19			1-[4-hydroxy-3-(3-methylbut-2-enyl)phenyl]ethanone
20			Procyanidin_tetramer
21			Galloyl glucose
22			Licoisoflavone A
23			Nicotiflorin
24			3,4-Dihydrocoumarin
25			Caffeic acid
26			Arbutin
27			Daidzin
28			Patulin
29			Genipin
30			Paeonol
31			Acetovanillone
32			NPPB

^aSignificantly differentially expressed in leaf compared with fruit and bark.

^bSignificantly differentially expressed in fruit compared to leaf and bark.

^cSignificantly differentially expressed in bark compared with leaf and fruit.

^dCommon expressed in three types of tissue samples. DEM stands for differential expression of metabolites.

3.4. Metabolic pathway analysis of differential metabolites

To explore the metabolic pathways associated with differentially abundant metabolites in C. cassia, all identified differential metabolites were mapped to the KEGG database for functional annotation and pathway enrichment analysis. Distinct pathway enrichment patterns were observed among the three pairwise tissue comparison group. In leaf vs. fruit, differential metabolites were significantly enriched in key pathways includingamino acids biosynthesis, ABC transporters, carbon metabolism, and phenylpropanoid biosynthesis (Fig. S3A). In leaf vs. bark, the enriched pathways included phenylpropanoid biosynthesis, amino acids biosynthesis, ABC transporters, 2-oxocarboxylic acid metabolism, and cofactors biosynthesis (Fig. S3B). In the fruit vs. bark comparison, significant enrichment was detected in phenylpropanoid biosynthesis, ABC transporters, and amino acids biosynthesis (Fig. S3C).

Venn analysis results showed that a total of 96 metabolic pathways were significantly enriched across the three paired comparisons, 79 of which were common pathways (Fig. S3D and Table S5). Among these common pathways, four were differentially expressed in leaf tissues, including thiamine metabolism, ether lipid metabolism, phosphonate and phosphinate metabolism, and terpenoid backbone biosynthesis. Similarly, three pathways were differentially expressed in bark tissues: stilbenoid, diarylheptanoid and gingerol biosynthesis, betalain biosynthesis, and fatty acid metabolism. Compared with bark and leaf tissues, five pathways were specifically differentially expressed in fruit tissues, namely biosynthesis of unsaturated fatty acids, histidine metabolism, valine, leucine and isoleucine degradation, glutathione metabolism, and one carbon pool by folate (Fig. S3E). KEGG pathway annotation further indicated that differentially abundant metabolites from fruit vs. bark, leaf vs. bark, and leaf vs. fruit comparisons were predominantly enriched in metabolic pathways and the biosynthesis of secondary metabolites (Fig. S3F).

3.5. Integration of pathways and differential metabolites from various tissues

Differentially expressed metabolites (DEMs) identified from the three pairwise tissue comparisons were characterized, integrated, and mapped onto key primary and secondary metabolic pathways (Fig. 3, Fig. 4, Fig. 5). Specifically, comparative analysis of leaf vs. fruit revealed that 78 DEMs were significantly enriched in 90 metabolic pathways. Compared with leaves, fruits exhibited a marked accumulation of 12 fatty acids (e.g., phloionolic acid, dodecanoic acid, myristic acid), 4 alkaloids (e.g., stachydrine), and other bioactive compounds including kinetin (a cytokinin) and 8 amino acids/peptides (e.g., l-glutamine, 4-hydroxyproline, betaine). Additionally, 5 terpenoids (euscaphic acid, cannabichromeneine, guaiazulene, thymyl-acetate, iraldeine) were specifically enriched in fruits. Furthermore, 12 phenylpropanoids were more abundant in fruits, including 5 simple phenylpropanoids (e.g., 6-methoxy-alpha-pyrufuran, feruloyltyramine, elemicin), 5 flavonoids (e.g., cinnamtannin D2, cinnamtannin B1, procyanidin A2), and 2 coumarins (pimpinellin, bergaptol) (Fig. 3).

Fig. 3.

Analysis of metabolic pathways and metabolites in the leaf-fruit group of C. cassia. Metabolites labeled in red indicate |log₂ FC| > 2 in the fruit-to-leaf comparison, representing higher abundance in fruit. Metabolites labeled in black represent key pathway intermediates that were not detected in the metabolomic data. Blue text within boxes indicates metabolite classifications, and black text indicates pathway names. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4.

Analysis of metabolic pathways and metabolites in the leaf-bark group of C. cassia. Metabolites labeled in red indicate |log₂ FC| > 2 in the bark-to-leaf comparison, representing higher abundance in bark. Metabolites labeled in black represent key pathway intermediates that were not detected in the metabolomic data. Blue text within boxes indicates metabolite classifications, and black text indicates pathway names. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5.

Analysis of metabolic pathways and metabolites in the fruit-bark group of C. cassia. Metabolites labeled in red indicate |log₂ FC| > 2 in the bark-to-fruit comparison, representing higher abundance in bark. Metabolites labeled in black represent key pathway intermediates that were not detected in the metabolomic data. Blue text within boxes indicates metabolite classifications, and black text indicates pathway names. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

To further elucidate the tissue-specific metabolic divergence across leaves, fruits, and bark, pairwise comparisons of leaf vs. bark and fruit vs. bark were performed. Comparative analysis of leaf vs. bark identified DEMs enriched in 88 metabolic pathways—consistent with the pathway enrichment observed in leaf vs. fruit, thereby reflecting shared metabolic networks underlying inter-tissue differentiation (Fig. 4). Notably, bark exhibited distinct metabolic accumulation patterns compared to both leaves and fruits. Specifically, relative to leaves, bark accumulated higher levels of metabolites across multiple categories, including 4 fatty acids (e.g., pinellic acid, 12-hydroxydodecanoic acid), 1 carbohydrates (melezitose), 5 amino acids and peptides (e.g., glutamine, canavanine), 7 terpenoids (e.g., genipin, retinoic acid, pulegone), 3 alkaloids (synephrine, kinetin, cycloDopa-5-O-glucoside), 9 flavonoids (e.g., procyanidin tetramer, procyanidin B2, cinnamtannin D2), 7 coumarins (e.g., esculin, daphnin), 3 lignans (arctigenin, p-methoxycinnamaldehyde, aschantin), and 11 simple phenylpropanoids (e.g., jionoside B1, elemicin, paeonol).

Subsequent comparison of fruit vs. bark identified 107 DEMs, which were also enriched in 88 metabolic pathways—further confirming that bark shares similar metabolic pathway involvement in its differentiation from both leaves and fruits (Fig. 5). Relative to fruits, bark accumulated a distinct set of metabolites at higher levels, including 6 fatty acids (e.g., jasmonic acid, azelaic acid), 11 carbohydrates (e.g., glucose 6-phosphate, glucose 1-phosphate, fructose 1-phosphate), 5 amino acids and peptides (e.g., glutamate, canavanine, aminohippuric acid), 4 terpenoids (e.g., zeaxanthin, genipin), 23 flavonoids (e.g., tangeretin, procyanidin tetramer—consistent with its accumulation in bark relative to leaves), 5 coumarins (e.g., scopolin, esculin), 3 lignans (arctigenin, aschantin, 1-acetoxypinoresinol), 18 simple phenylpropanoids (e.g., cinnamaldehyde, 3-phenylpropanoic acid, coniferaldehyde). The consistent enrichment of secondary metabolites (e.g., flavonoids, simple phenylpropanoids) in bark across both comparisons highlights its specialized role in defensive metabolism, contrasting sharply with the photosynthesis-associated primary metabolism of leaves and the nutrient-storage metabolism of fruits (Fig. 6).

Fig. 6.

Potential mechanisms for the synthesis and accumulation of metabolites in the fruit, bark, and leaf tissues of C. cassia. The fruit accumulates high levels of fatty acids, terpenoids, and alkaloids. The bark is characterized by high levels of carbohydrates, phenylpropanoids and terpenoids.

4. Discussion

C. cassia is extensively cultivated across numerous countries, including China, Southeast Asia, South America, and Indonesia. As an important natural source of volatile oils, this plant is rich in cinnamaldehyde—a core bioactive constituent renowned for its significant industrial and pharmaceutical value. In addition to VOCs, C. cassia also contains a diverse array of non-volatile compounds, among which flavonoids have been extensively characterized in previous studies (Gao et al., 2020). Owing to the broad applicability of its volatile and non-volatile compounds in food, cosmetics, fragrance, pesticide, pharmaceutical, and cleaning products, C. cassia serves not only as a daily condiment but also a critical pharmaceutical raw material (Han et al., 2024; Yao et al., 2024). While existing research has primarily focused on the chemical composition, pharmacological activities and biosynthetic mechanisms of key metabolites, in-depth studies on compound distribution and their inter-tissue exchange remain scarce. Here, we employed an integrated metabolomics approach combining integrated GC–MS and UPLC-Q-Orbitrap MS to comprehensively profile volatile and non-volatile compounds, as well as their associated metabolic pathways in C. cassia bark, fruits and leaves. Furthermore, we explored inter-tissue metabolite exchange to provide foundational insights for future metabolic network reconstruction. Notably, this integrated multi-omics approach (metabolomics combined with pathway enrichment analysis) facilitates a more systematic and holistic characterization of tissue-specific metabolic features compared to previous single-technique or single-tissue studies, representing a key advantage of our research design.

The localization of bioactive constituents in plants is tightly linked to their tissue types, and anatomical differences directly influence metabolite composition (Chen et al., 2016). A recent study reported that C. cassia bark accumulated significantly higher levels of essential oils, cinnamaldehyde, flavonoids, and procyanidins compared to leaves and branches (Gao et al., 2023). However, metabolite profiles in fruits and comprehensive inter-tissue comparisons remain largely unexplored. Our GC–MS analysis confirmed the presence of abundant volatile compounds in leaves, bark, and fruits, with bark exhibiting the highest accumulation levels (Fig. 1), consistent with the findings of Gao et al. (2023). Cinnamaldehyde is recognized as the primary bioactive component of C. cassia (Neto et al., 2022). Our quantitative results showed its highest content in bark (937.35 ± 42.10 mg/g), which was 1.21-fold and 156.75-fold higher than that in leaves and fruits, respectively (Fig. 1E). While this finding aligns with previous studies (Chen et al., 2016; Gao et al., 2023), it conflicts with reports suggesting a gradient of increasing cinnamaldehyde from basal to apical plant parts (Hou et al., 2013; Wang et al., 2022). Resolving this discrepancy through further research is crucial for optimizingextraction technologies, establishing grade specifications, and improving quality control standards.

In addition to volatile compounds, C. cassia contains a diverse range of non-volatile compounds Chen et al., 2016 identified 8 compounds, including cinnamaldehyde and coumarin, via LC-qTOF-MS fingerprinting of 125 samples. Gao et al. (2023) detected small amounts of flavonoids, coumarins, and cinnamaldehyde in leaves and branches using UPLC-Q/TOP-MS, while Zhou et al. (2018) reported 19 bark-specific metabolites Wang, Chi, et al. (2023) identified 58 compounds in Cinnamomi ramulus (twigs) and Cinnamomi cortex (bark) via UPLC-Orbitrap-Exploris-120-MS/MS, categorized into 9 classes: flavonoids, phenylpropanoids and phenolic acids, coumarins, lignans, terpenoids, organic acids, and others. Using UPLC-Q-Orbitrap MS, in both positive and negative ion modes, we identified 78, 77, and 107 differential metabolites in leaves vs. bark, fruits vs. bark, and leaves vs. fruits comparisons, respectively (Table S2–S4). This represents a substantial expansion of the known C. cassia metabolome, particularly regarding tissue-specific metabolites. Venn analysis confirmed 13 leaf-specific and 32 bark–specific metabolites; epicatechin, procyanidin B2, procyanidin tetramer, and p-methoxycinnamaldehyde were proposed as bark-specific metabolic biomarkers (Table 1 and Fig. 3, Fig. 4, Fig. 5). This is consistent with Tanaka et al. (2013), who identified epicatechin and procyanidin B2 as marker for Chinese and Vietnamese cassia. Cassia tannin A, a proanthocyanidin tetramer, has also been confirmed as a bark-specific component (Killday et al., 2011). P-methoxycinnamaldehyde shows distinct distribution patterns (97.4% in bark, 64.0% in twigs, and 50.0% in shaved bark) (Chen et al. (2016), further verifying its bark-specific enrichment. These identified biomarkers can be incorporated into official pharmacopoeia standards for C. cassia to prevent adulteration with low-quality tissues (e.g., leaves, twigs) or counterfeit materials. In industrial production, rapid detection of these biomarkers can streamline the raw material screening process, reducing production costs and ensuring product safety.

Amino acid biosynthesis serves as a critical source of precursors for secondary metabolic production (Liras & Martín, 2023), while ABC transport pathway supplies energy for primary-to-secondary metabolism transition and metabolic flux allocation (Yazaki, 2006; Zhao et al., 2022). Our KEGG enrichment analysis revealed that among the top 20 enriched pathways, phenylpropanoid biosynthesis, ABC transporters, amino acid biosynthesiswere prominently represented, exhibiting distinct tissue-specific patterns: terpenoid backbone biosynthesis was enriched in leaves, whereas unsaturated fatty acids biosynthesis was enriched in fruit (Fig. S3). This tissue-specific pathway enrichment strongly correlates with the distribution of tissue-specific metabolites, indicating the presence of specialized metabolic networks for unique secondary metabolite biosynthesis and accumulation.

UPLC-Q-Orbitrap MS results revealed higher terpenoid contents in C. cassia fruits and bark compared to leaves. Specifically, five terpenoids (euscaphic acid, cannabichromeneine, guaiazulene, thymyl-acetate, and iraldeine) were detected exclusively in fruits, while seven terpenoids (e.g., genipin, retinoic acid, and pulegone) were highly abundant in bark (Fig. 3, Fig. 4, Fig. 5). Plant terpenoids are synthesized via two acetyl-CoA-derived pathways: the cytoplasmic MVA pathway (responsible for sesquiterpenoids, triterpenoids) and plastidial MEP pathway (responsible for monoterpenoids, diterpenoids) (Chen et al., 2025; Yao et al., 2024). Acetyl-CoA is a pivotal intermediate that serves as a carbon source for both terpenoid and fatty acid biosynthesis (Sha et al., 2025). It acts as a direct precursor for squalene and provides carbon skeletons for multiple terpenoid subtypes, thereby coordinating the biosynthesis of both primary and secondary metabolites. Enhanced pyruvate synthesis promotes acetyl-CoA accumulation, which in turn increases the content of squalene and acetyl-CoA-derived terpenoids (Huang et al., 2018), consistent with the principle that sufficient precursor supply amplifies terpenoid and fatty acid production (Guo et al., 2025). Our KEGG enrichment analysis identified unsaturated fatty acid biosynthesis in fruits and glutathione/fatty acid metabolism in bark; thus, we hypothesize that acetyl-CoA plays a central role in coordinating the biosynthesis and metabolism of terpenoids and fatty acids in C. cassia (e.g., euscaphic acid and 12-hydroxydodecanoic acid).

As the primary photosynthetic source organ, C. cassia leaves produce acetyl-CoA and other carbon skeletons via the Calvin cycle, glycolysis, and TCA cycle, which serve as key precursors for the biosynthesis of tissue-specific metabolites in fruits and bark. In plants, leaves export photosynthetic products (sugars, acetyl-CoA, secondary-metabolite precursors) to sink organs (bark and fruits) through phloem. Studies using radiolabeled glucosinolates in Arabidopsis have confirmed phloem loading and long-distance translocation of glucoside-type secondary metabolites (Chen et al., 2001). Recent reviews have further verified the phloem transport of lipophilic metabolites (e.g., terpenoids, fatty acids) is mediated by specific transport proteins (Lohaus, 2022). Thus, the enrichment of terpenoids, alkaloids and fatty acids in fruits, together with the accumulation of phenylpropanoids and terpenoids in bark, likely reflects source-to-sink translocation of metabolites from leaves to these storage tissues (Fig. 6). This observation not only supports the traditional use of bark as the primary medicinal part of C. cassia but also highlights fruits as a complementary source of bioactive compounds, providing valuable insights for the comprehensive utilization of C. cassia resources.

Phenylpropanoids comprise a diverse group of metabolites, including simple phenylpropanoids phenolic acids flavonoids, lignin, lignans, coumarins, and stilbenoids (Ninkuu et al., 2025). Cinnamaldehyde, a simple phenylpropanoids and major bioactive constituents of C. cassia (Chinese Pharmacopoeia Commission, 2025), was consistently shown by GC–MS and UPLC-Q-Orbitrap MS to accumulate at higher levels in bark than in fruits and leaves (Fig. 1, Fig. 5). This finding provides a biochemical basis for prioritizing bark as the primary medicinal material. We also detected abundant sugar phosphates (e.g., glucose-6-phosphate, fructose-6-phosphate) in bark, which act as precursors for secondary metabolism and regulate the synthesis, distribution, and activity of terpenoids, alkaloids, and phenylpropanoids through signaling cascades, metabolic channeling, and glycosylation (Jiang et al., 2022; Zhao et al., 2023), potentially serving as. “hub-regulation-modification” nodes in C. cassia's metabolic network. Our comprehensive metabolic profiling lays a solid foundation for C. cassia's industrial and medicinal development: tissue-specific metabolite signatures (e.g., high trans-cinnamaldehyde in bark) can be used as quality control markers; bark's high bioactive compound content guides extraction optimization; and identified metabolites/pathways inform phytotherapeutic formulation development. Future studies should validate the acetyl-CoA-centered transport network using ¹³C labeling techniques, functionally characterize key biosynthetic genes (e.g., via CRISPR-Cas9-mediated gene editing), and explore the comprehensive utilization of fruits and leaves beyond traditional bark-focused applications of C. cassia.

5. Conclusions

Intergrated GC–MS and UPLC-Q-Orbitrap MS analysis of C. cassia key tissues provided valuable insights into metabolite accumulation patterns and metabolic pathway dynamics. The identification of 71 volatile and 2882 non-volatile metabolites (predominantly terpenoids) clarifies the plant's phytochemical profile. Potential tissue-specific metabolic biomarkers were screened from differential metabolites, and mapping these metabolites to primary/secondary metabolic pathways confirmed inter-tissue metabolite translocation, indicating a sophisticated metabolic network coordinating tissue-specific biosynthesis and resource allocation. A putative acetyl-CoA-centered inter-tissue metabolite exchange network was proposed, linking leaf-derived acetyl-CoA to bioactive compound biosynthesis in bark and fruits. These findings lay a solid foundation for future studies on metabolite biosynthetic pathways and targeted regulatory strategies for this economically and medicinally valuable species.

CRediT authorship contribution statement

Shaochang Yao: Writing – review & editing, Validation, Methodology, Funding acquisition, Formal analysis, Conceptualization. Libing Long: Writing – original draft, Visualization, Data curation. Ying Zhu: Methodology, Data curation. Jing Chen: Data curation. Liujun Chen: Validation, Methodology, Formal analysis. Linshuang Li: Methodology, Data curation. Ding Huang: Writing – review & editing. Ruhong Ming: Writing – original draft, Supervision. Rongshao Huang: Writing – review & editing. Jian Xiao: Supervision, Resources, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material 1: Supplementary figures of trans-cinnamaldehyde standard curve and differential metabolites in different tissues of C. cassia.

Supplementary material 2: Supplementary tables for volatile oil components and differential metabolites in three different tissue comparisons of C. cassia.

Data availability

Data will be made available on request.