Evolutionary and molecular insights into the terpene synthase gene (TPS) family in the medicinal plant Bidens alba

Abstract

Terpenoids are important secondary metabolites in plants, valued for their pharmacological activities and economic significance. Terpene synthases (TPS) serve as the key enzymes in terpenoid biosynthesis. To explore the TPS gene family and their potential functions in Bidens alba, this study conducted a comprehensive analysis of TPS genes based on whole-genome data using bioinformatics approaches. A total of 84 BaTPS genes were identified and unevenly distributed across 37 chromosome-scale scaffolds. Phylogenetic analysis classified these genes into five subfamilies: TPS-a, TPS-b, TPS-c, TPS-e/f, and TPS-g. Among them, the TPS-a and TPS-b subfamilies contained the largest number of members, while the TPS-g showed significant expansion. Further investigation revealed that whole-genome duplication and segmental duplication were the main drivers of BaTPS gene family expansion, with purifying selection playing a predominant role in their evolution. Gene structure analysis showed that BaTPS genes exhibit highly conserved intron-exon organization. Motif analysis identified several conserved domains, including the characteristic “DDxxD” motif, along with the “RDR,” “RRX(8)W,” and “NSE/DTE” motifs. Promoter region analysis uncovered numerous cis-acting elements, particularly those responsive to light, methyl jasmonate (MeJA), and abscisic acid (ABA). Expression profiling using RNA-seq and RT-qPCR demonstrated tissue-specific expression patterns of BaTPS genes, with BaTPS02, BaTPS16, BaTPS41, BaTPS47, BaTPS53 and BaTPS60 highly expressed in floral organs. Subcellular localization showed that BaTPS16 and BaTPS41 are localized in cytoplasm. Collectively, these findings suggest that B. alba floral organs may be rich in diverse terpenoids. This study lays theoretical foundation for future functional analyses of the TPS gene family in B. alba and supports the development of its medicinally active compounds.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-55330-1.

Introduction

Terpenoids constitute one of the largest and most structurally diverse groups of specialized metabolites in plants. They play indispensable roles in growth regulation, environmental adaptation, and plant defense, while also contributing to aroma, pigmentation, and pharmaceutically valuable compounds¹. Due to their wide biological relevance and substantial industrial potential, terpenoids have become a major focus in plant metabolic biology, natural product chemistry, and synthetic biology.

In plants, terpenoids are mainly synthesized through the cytosolic mevalonic acid (MVA) pathway and the plastidial methylerythritol phosphate (MEP) pathway. These pathways generate the C5 precursors isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), which are subsequently converted into geranyl diphosphate (GPP), farnesyl diphosphate (FPP), and geranylgeranyl diphosphate (GGPP)². Terpene synthases (TPSs) then catalyze the formation of a wide variety of terpenoid structures using these prenyl diphosphates as substrates³. Notably, substrate specificity varies among species; for example, several TPS genes in the mature fruits of Prunus persica can utilize both GPP and FPP as substrates to synthesize 15 monoterpenes and 14 sesquiterpenes⁴.

TPS proteins typically contain two conserved domains: an N-terminal domain (Pfam PF01397) featuring the “RRX8W” motif and a C-terminal domain (Pfam PF03936) that provides the metal-binding residues required for catalysis. The C-terminal region includes three subdomains (α, β, and γ), among which the α-subdomain contains the conserved “DDxxD” motif that drives ionization-initiated reactions, whereas the β/γ-subdomains harbor the “DxDD” motif essential for protonation-initiated reactions. These conserved motifs and domain architectures form the basis for TPS classification and for predicting catalytic properties. The remarkable diversity of terpenoid structures is largely generated by the TPS family, which catalyzes the key cyclization and rearrangement reactions that convert linear prenyl diphosphate precursors into mono-, sesqui-, and diterpenes with highly variable skeletons⁵. In angiosperms, TPS genes are typically classified into several clades—including TPS-a, TPS-b, TPS-c, TPS-e/f, TPS-g, and others—based on conserved motifs, domain architecture, and phylogenetic relationships⁶. Each clade exhibits distinct catalytic characteristics: TPS-a is commonly associated with sesquiterpene formation; TPS-b participates largely in monoterpene biosynthesis; TPS-c genes encode copalyl diphosphate synthases; and TPS-e/f members are involved in diterpene biosynthesis, including the production of ent-kaurene, a key precursor of gibberellins^7,8. These clade-specific functional features underline the evolutionary diversification of the TPS family and explain its central role in shaping terpenoid metabolic complexity.

With increasing availability of plant genome databases, terpene synthase (TPS) genes have been identified in a growing number of plant species, including Arabidopsis thaliana⁹, Camellia sinensis¹⁰, Daucus carota¹¹, Eucalyptus globulus and E.grandis¹², Malus domestica¹³, Solanum lycopersicum¹⁴, Selaginella moellendorffii¹⁵, Chimonanthus praecox and C. salicifolius¹⁶, Populus trichocarpa¹⁷, and Vitis vinifera¹⁸. In addition, a recent pangenomic analysis across 19 Asteraceae species identified a total of 1,714 TPS genes, further highlighting the remarkable expansion of this gene family within the family¹⁹. These comparative studies have revealed that whole-genome duplication (WGD), tandem duplication, and segmental duplication events are major contributors to TPS family expansion, often giving rise to lineage-specific subfamilies with specialized ecological or biochemical functions. Nevertheless, despite shared catalytic cores, TPS genes often exhibit pronounced species-specific roles, reflecting adaptation to ecological niches, specialized pollinator interactions, or defense strategies. Consequently, the regulatory networks that shape TPS gene evolution, their functional divergence, and their contributions to tissue-specific terpenoid biosynthesis remain insufficiently understood in many taxonomic groups, particularly in medicinal plants.

Bidens alba, a species in the genus Bidens (family Asteraceae), belongs to a diverse group comprising approximately 280 species²⁰. It is a herbaceous plant native to North and Central America and widely distributed in subtropical and tropical regions²¹. B. alba is rich in terpenoids compounds, including monoterpenes and sesquiterpenes, which exhibit notable anti-inflammatory properties^22,23. Additionally, protein extracts from B. alba have been shown to induce apoptosis in the colon cancer cell line SW480 by generating reactive oxygen species and reducing intracellular glutathione levels, highlighting its therapeutic potential for treating wounds, inflammation, hypertension, jaundice, and diabetes²⁴. Despite its promising pharmacological value, the functional roles of TPS genes in the growth and development of different tissues—including flowers, roots, stems, and leaves—of B. alba remain largely unexplored.

In this study, we performed a genome-wide analysis to systematically identify TPS genes in B. alba and investigate their evolutionary and functional characteristics. We examined chromosomal distribution, conserved motif composition, gene structure variation, duplication patterns, and selective constraints using integrated evolutionary and comparative genomics approaches. Additionally, expression profiling across multiple tissues was used to clarify potential functional divergence among different BaTPS members and to highlight candidates potentially involved in tissue-specific terpenoid biosynthesis. Together, these results provide new insights into the evolution and functional specialization of the TPS family in B. alba and lay a theoretical foundation for future biochemical validation, functional genomics studies, and metabolic engineering aimed at enhancing valuable terpenoid production in this medicinally important species.

Materials and methods

Genome identification of BaTPS gene family members

The whole-genome data of B. alba²⁵, including genome sequences and annotation files, were downloaded from the Zenodo repository (https://zenodo.org/records/10160015). TPS gene sequences from (A) thaliana²⁶ were retrieved from the TAIR database (https://www.arabidopsis.org/). Hidden Markov Model (HMM) profiles for the TPS domains (PF01397, PF03936) were obtained from the Pfam database (http://pfam.xfam.org/)²⁷. Identification of TPS family members in (B) alba was performed using a combination of TBtools²⁸ and BLASTP²⁹. First, the Simple HMM Search module in TBtools was used to scan the B. alba protein sequences against the TPS domain HMM profiles, applying an E-value cutoff of < 1 × 10⁻⁵. This generated two initial sequence sets, which were merged to form a unified candidate set. Next, TPS protein sequences from (A) thaliana served as a reference library for BLASTP alignment against the (B) alba proteome, also with an E-value threshold of < 1 × 10⁻⁵. Results from both approaches were combined to yield a comprehensive list of candidate TPS proteins. These candidates were further validated for conserved domains using the NCBI Conserved Domain Database (CDD, https://www.ncbi.nlm.nih.gov/cdd/) and the SMART database (https://smart.embl.de/). Redundant sequences and those lacking complete conserved domains at either the N- or C-terminus were removed to finalize the set of BaTPS proteins.

Physical and chemical property analysis

The physicochemical properties of BaTPS family members were analyzed using the ExPASy ProtParam tool (https://web.expasy.org/)³⁰. Parameters assessed included amino acid length (AA), molecular weight (MW), theoretical isoelectric point (pI), instability index (II), aliphatic index (AI), and grand average of hydropathicity (GRAVY). Using the WoLF PSORT (https://wolfpsort.hgc.jp/)³¹ and DeepLoc − 2.1 (https://services.healthtech.dtu.dk/)³² tools to predict the subcellular localization, with the highest score prediction results. The secondary structures of the proteins were predicted using SOPMA (http://bioinformatics.psb.ugent.be/webtools/plantcare/)³³. Three-dimensional structure prediction of proteins was carried out using the online analysis software SWISS-MODEL (https://swissmodel.expasy.org/).

Phylogenetic and collinearity analysis

Multiple sequence alignment of TPS protein sequences from B. alba, Oryza sativa and (A) thaliana was performed using Molecular Evolutionary Genetics Analysis version 12 (MEGA12) software (https://www.megasoftware.net). A phylogenetic tree was constructed using the Neighbor-Joining (NJ) method with 1,000 bootstrap replicates, and all other parameters set to default. The resulting evolutionary tree was visualized and refined using the iTOL platform (https://itol.embl.de/). Chromosomal scale scaffold location data for BaTPS genes were extracted from the (B) alba GFF3 annotation file using TBtools version 2.0 (https://github.com/CJ-Chen/TBtools). Gene duplication events within the BaTPS gene family were identified using MCScanX, with a significance threshold of E-value < 1 × 10⁻⁵. The chromosomal distribution and collinear relationships of BaTPS genes were visualized using the Advanced Circos tool in TBtools. Ka/Ks ratios for collinear gene pairs were calculated using the Simple Ka/Ks Calculator (NG) module. A Ka/Ks ratio > 1 suggests positive selection, a ratio = 1 indicates neutral evolution, and a ratio < 1 reflects purifying selection .

Analysis of conserved motifs and gene structure

Conserved motif analysis of BaTPS genes was performed using the MEME Suite (https://meme-suite.org/meme/doc/meme.html), resulting in the identification of 20 distinct motifs. Gene structure analysis, including exon-intron organization, was carried out using the Gene Structure View (Advanced) tool in TBtools. The results of conserved motif analysis, conserved domain identification, and gene structure organization were subsequently visualized using TBtools to provide an integrated view of the structural features of BaTPS genes.

Analysis of cis-acting elements

The 2000 bp upstream sequences of the BaTPS genes were extracted using TBtools software to analyze potential regulatory regions. Cis-acting elements within these promoter regions were predicted using the PlantCARE database, and the results were visualized with TBtools for intuitive presentation.

Gene expression and metabolic analysis

The transcriptomic and metabolomic data utilized in this study were derived from previous work²³. The transcriptome samples encompassed roots, stems, leaves, and flowers of B. alba. Fresh tissues of B. alba were originally collected from the campus of Fujian Agriculture and Forestry University, Fujian Province, China. The plant material was formally identified by Professor Yousry A. El-Kassaby. These data were subsequently leveraged to analyze the expression patterns of BaTPS genes. Using TBtools software to generate a heatmap of gene expression levels, with FPKM values > 0.5. Based on the analysis of gene expression patterns, 12 genes with high expression in roots, stems, leaves, and flowers were selected for RT-qPCR validation. Total RNA from the samples was extracted using a polysaccharide and polyphenol total RNA extraction kit, and RNA was reverse transcribed into cDNA using ABScript Neo RT Master Mix for qPCR with gDNA Remover. RT-qPCR primers were designed using Primer Premier 6 (Table S1). GAPDH was used as the internal reference gene, and RT-qPCR experiments were conducted using 2X Universal SYBR Green Fast qPCR Mix, with three biological replicates for each sample. The amplification conditions were as follows: pre-denaturation at 95 °C for 3 min; denaturation at 95 °C for 10 s, annealing at 60 °C for 35 s, for 45 cycles. The relative expression levels of the target genes were calculated using the 2^−ΔΔCt method, and data analysis was performed using SPSS 25.0, with graphs generated using Origin 2019.

Subcellular localization analysis of BaTPS16 and BaTPS41

To investigate the subcellular localization of BaTPS16 and BaTPS41 proteins, the full-length coding sequences (CDS) of both genes, excluding the stop codon, were amplified and cloned into the pMDC202 vector containing a 35 S promoter-driven GFP expression cassette. Cloning was performed using the ClonExpress^® Ultra One Step Cloning Kit (Vazyme, China), and KpnI and XbaI were selected as the restriction enzyme sites for vector construction. The resulting recombinant vectors (35 S::BaTPS16-GFP and 35 S::BaTPS41-GFP) were introduced into Agrobacterium tumefaciens strain GV3101, which was cultured at 25 °C for 2 days. The transformed A. tumefaciens cells were subsequently infiltrated into tobacco leaves, followed by an additional 2-day incubation under the same conditions. The empty pMDC202 vector was used as a negative control. After incubation, GFP fluorescence signals were visualized using a Nikon C2 confocal laser scanning microscope (Nikon, Japan).

Results

Identification and analysis of the BaTPS family

A total of 84 terpene synthase (TPS) gene family members were identified from the whole genome of B. alba and designated BaTPS01 to BaTPS84 based on their chromosomal positions (Table 1). Physicochemical analysis revealed that the BaTPS proteins range from 282 to 834 amino acids in length, with predicted molecular weights spanning 33.04 to 96.81 kDa. The isoelectric points (pI) vary from 4.75 to 7.57, with an average of 5.58. Notably, only BaTPS23 has a pI above 7, indicating that the vast majority of BaTPS proteins are acidic and may preferentially precipitate in acidic environments.

Table 1.

Analysis of the TPS family Member Information and Secondary Structure of B. alba.

Name	Gene ID	AAa	MWb	pIc	IId	AIe	GRAVYf	Subcellular location
		(aa)	(kDa)
BaTPS01	g13311	583	67.9	5.76	52.07	83.14	-0.433	Chloroplast
BaTPS02	g43385	578	67.7	5.32	49.36	87.68	-0.384	Cytoplasm
BaTPS03	g91903	282	33.04	5.96	48.15	85.07	-0.499	Cytoplasm
BaTPS04	g91904	421	48.62	6.25	49.59	87.05	-0.445	Chloroplast
BaTPS05	g91911	556	64.78	5.76	43.3	90.67	-0.336	Cytoplasm
BaTPS06	g92441	798	91.58	5.7	36.48	85.14	-0.294	Chloroplast
BaTPS07	g92979	464	53.9	5.7	42.51	90.15	-0.342	Cytoplasm
BaTPS08	g94104	556	65.32	5.34	45.22	89.23	-0.452	Cytoplasm
BaTPS09	g94496	567	66.18	5.17	53.01	91.31	-0.364	Cytoplasm
BaTPS10	g94501	571	66.76	5.24	54.35	91.84	-0.367	Cytoplasm
BaTPS11	g32185	816	92.5	5.71	37.66	92.93	-0.149	Chloroplast
BaTPS12	g57093	550	64.48	5.05	51.81	90.05	-0.322	Cytoplasm
BaTPS13	g45164	577	67.45	5.76	47.31	87.71	-0.336	Cytoplasm
BaTPS14	g28398	538	62.49	5.37	34.22	94.61	-0.267	Cytoplasm
BaTPS15	g21059	562	65.17	5.24	43.85	90.73	-0.282	Cytoplasm
BaTPS16	g21056	489	56.45	5.34	39.11	90.35	-0.29	Cytoplasm
BaTPS17	g76535	561	66.23	5.57	46.1	87.22	-0.467	Cytoplasm
BaTPS18	g155494	586	68.03	5.45	53.51	85.53	-0.386	Chloroplast
BaTPS19	g87071	834	96.81	5.87	44.59	86.61	-0.355	Cytoplasm
BaTPS20	g86964	602	71.25	5.57	52.99	95.32	-0.233	Cytoplasm
BaTPS21	g161023	519	59.52	5.32	39.5	95.8	-0.177	Cytoplasm
BaTPS22	g13420	548	63.48	5.27	49.38	90.55	-0.306	Cytoplasm
BaTPS23	g22378	765	88.59	7.57	42.29	83.62	-0.343	Chloroplast
BaTPS24	g3267	551	63.8	5.35	40.35	86.01	-0.315	Cytoplasm
BaTPS25	g32824	547	63.79	5.27	49.11	90.22	-0.327	Cytoplasm
BaTPS26	g103497	578	67.48	5.52	51.83	92.92	-0.291	Chloroplast
BaTPS27	g136081	583	67.95	6.01	50.7	83.14	-0.447	Chloroplast
BaTPS28	g130889	472	53.79	6.15	37.08	94.81	-0.25	Chloroplast
BaTPS29	g130890	445	50.42	5.78	34.22	96.43	-0.227	Chloroplast
BaTPS30	g79348	592	68.86	5.52	51.11	86.32	-0.375	Chloroplast
BaTPS31	g154521	549	64.37	5.23	51.25	88.23	-0.39	Cytoplasm
BaTPS32	g162522	587	68.72	6.1	47.98	87.75	-0.425	Chloroplast
BaTPS33	g163110	813	93.21	5.81	36.99	85.49	-0.281	Chloroplast
BaTPS34	g163674	548	63.62	5.45	41.35	91.08	-0.323	Cytoplasm
BaTPS35	g110150	556	65.29	5.51	45.01	89.93	-0.431	Cytoplasm
BaTPS36	g110155	514	60.59	5.61	45.29	88.37	-0.431	Cytoplasm
BaTPS37	g75735	571	66.76	5.24	54.35	91.84	-0.367	Cytoplasm
BaTPS38	g105052	535	60.76	5.35	40.72	89.07	-0.217	Cytoplasm
BaTPS39	g144878	544	62.55	5.46	44.03	91.38	-0.37	Cytoplasm
BaTPS40	g33024	548	64.2	5.31	48.64	90	-0.378	Cytoplasm
BaTPS41	g88203	589	68.31	5.55	44.07	89.59	-0.307	Cytoplasm
BaTPS42	g47591	822	94.1	5.42	45.46	90.34	-0.185	Chloroplast
BaTPS43	g151093	494	57.4	5.58	55.95	80.59	-0.465	Chloroplast
BaTPS44	g151094	588	67.93	5.47	51.46	94.54	-0.273	Chloroplast
BaTPS45	g157274	525	60.27	6.12	31.55	99.16	-0.222	Cytoplasm
BaTPS46	g98370	783	90.06	5.64	40.68	89.8	-0.244	Cytoplasm
BaTPS47	g97790	572	66.19	5.78	37.99	90.66	-0.359	Chloroplast
BaTPS48	g97787	425	49.86	5.42	40.48	90.33	-0.454	Cytoplasm
BaTPS49	g97785	487	55.81	6.27	34.2	94.7	-0.293	Chloroplast
BaTPS50	g82484	577	67.62	5.33	53.73	90.21	-0.334	Chloroplast
BaTPS51	g101339	613	71.33	5.7	42.89	89.71	-0.324	Chloroplast
BaTPS52	g131541	482	56.49	4.75	51.07	87.39	-0.305	Cytoplasm
BaTPS53	g131301	576	66.22	5.12	48.97	99.81	-0.234	Chloroplast
BaTPS54	g111474	583	66.8	5.71	40.4	94.49	-0.209	Chloroplast
BaTPS55	g80790	799	91.78	5.74	38.9	83.94	-0.304	Chloroplast
BaTPS56	g108490	589	68.15	5.45	41.41	87.11	-0.353	Chloroplast
BaTPS57	g108492	551	63.87	5.04	41.94	89.06	-0.361	Cytoplasm
BaTPS58	g108494	587	68.53	6.23	46.33	87.58	-0.418	Chloroplast
BaTPS59	g108499	584	67.85	6.19	47.23	87.65	-0.407	Chloroplast
BaTPS60	g24891	536	61.88	5.09	44.66	89.33	-0.331	Cytoplasm
BaTPS61	g81217	531	61.23	5.37	44.06	95.56	-0.248	Cytoplasm
BaTPS62	g81219	609	70.65	5.95	42.26	87.27	-0.308	Chloroplast
BaTPS63	g110861	561	66.23	5.52	46.08	87.22	-0.467	Cytoplasm
BaTPS64	g110752	783	89.64	5.32	44.9	88.75	-0.214	Chloroplast
BaTPS65	g110751	827	94.7	5.6	46.12	88.86	-0.246	Chloroplast
BaTPS66	g147233	550	65.14	5.18	53.93	90	-0.397	Cytoplasm
BaTPS67	g107194	601	69.65	6.48	37.01	101.06	-0.182	Chloroplast
BaTPS68	g68330	408	47.08	5.22	37.76	95.56	-0.31	Cytoplasm
BaTPS69	g134138	448	52.16	5.43	45.04	85.56	-0.397	Cytoplasm
BaTPS70	g9802	555	64.63	5.48	39.74	92.92	-0.264	Cytoplasm
BaTPS71	g8395	584	66.82	5.86	41.75	97.65	-0.195	Chloroplast
BaTPS72	g47214	799	91.78	5.74	38.9	83.94	-0.304	Chloroplast
BaTPS73	g2874	367	42.64	5.72	32.31	88.45	-0.422	Chloroplast
BaTPS74	g37352	773	87.59	5.72	35.95	92.19	-0.174	Chloroplast
BaTPS75	g75094	524	60.82	5.31	42.33	88.59	-0.338	Cytoplasm
BaTPS76	g74363	516	59.37	4.8	34.17	102.62	-0.1	Cytoplasm
BaTPS77	g73774	556	65.17	5.53	40.13	90.47	-0.388	Cytoplasm
BaTPS78	g5329	367	42.8	5.29	45.01	84.25	-0.279	Cytoplasm
BaTPS79	g127806	538	62.39	5.32	35.18	94.96	-0.237	Cytoplasm
BaTPS80	g35113	588	67.8	5.67	42.61	94.03	-0.243	Chloroplast
BaTPS81	g35114	400	46.52	5.52	53.58	81.7	-0.529	Cytoplasm
BaTPS82	g52958	834	96.76	5.82	44.46	86.61	-0.347	Cytoplasm
BaTPS83	g71866	572	66.06	5.78	38.26	90.33	-0.369	Chloroplast
BaTPS84	g71871	408	47.08	5.22	37.76	95.56	-0.31	Cytoplasm

Based on the instability index, 22 BaTPS proteins (26.19%) are classified as stable. All BaTPS proteins exhibit negative grand average of hydropathicity (GRAVY) values, suggesting they are hydrophilic. The aliphatic index values range from 80.59 to 102.62, reflecting a moderate to high degree of thermostability. Subcellular localization prediction revealed that the majority of BaTPS proteins are localized to the cytoplasm (48 proteins, accounting for 57.14% of the total), while the remaining 36 proteins are targeted to chloroplasts. These results suggest that BaTPS proteins are compartmentalized into distinct subcellular locations, implying their respective roles in the cytosolic (MVA) and plastidial (MEP) terpenoid biosynthesis pathways in B. alba.

BaTPS protein secondary structure analysis

Secondary structure prediction revealed that all BaTPS proteins are predominantly composed of α-helices and random coils, with relatively minor proportions of extended strands and β-turns (Table S2). α-Helices constitute the highest proportion, ranging from 55.84% (BaTPS33) to 75.65% (BaTPS79). Random coils account for 17.83% (BaTPS76) to 34.44% (BaTPS33) of the structures. In contrast, both extended strands and β-turns were present in lower amounts, comprising 0.71% (BaTPS03) to 8.56% (BaTPS46) and 2.61% (BaTPS04) to 6.73% (BaTPS38) of the structures, respectively.

Phylogenetic analysis of the BaTPS family

A phylogenetic tree was constructed using TPS protein sequences from (A) thaliana, O. sativa, and (B) alba (Fig. 1). The BaTPS genes were grouped into five subfamilies: TPS-a, TPS-b, TPS-c, TPS-e/f, and TPS-g. The TPS-a subfamily is the largest, comprising 32 members (38.10%), primarily associated with sesquiterpene and diterpene synthesis. TPS-b is the second largest subfamily, with 25 genes (29.76%) involved in the biosynthesis of monoterpenes, sesquiterpenes, and the terpenoid precursor isoprene. The TPS-c and TPS-e/f subfamilies, mainly responsible for encoding diterpene synthases, contain 5 and 8 members, respectively. The TPS-g subfamily, primarily associated with monoterpenoid biosynthesis, includes 14 BaTPS genes. These results indicate a diversification of TPS functions in B. alba, with a notable expansion in subfamilies linked to sesquiterpene and monoterpene synthesis.

Fig. 1.

Phylogenetic tree of TPS genes from (A) thaliana, O. sativa, and (B) alba. The TPS gene family is classified into five subfamilies: TPS-a, TPS-b, TPS-c, TPS-e/f, and TPS-g. TPS genes from B. alba are indicated by red circles, those from A. thaliana are marked with yellow circles, and those from O. sativa by green circles.

BaTPS chromosomal localization and synteny analysis

The chromosomal scale scaffold positions of the BaTPS genes were extracted based on the GFF3 annotation file, and an intraspecific synteny analysis was performed using the MCScanX tool. The results revealed that the 84 BaTPS genes are unevenly distributed across 37 chromosomes scale scaffold (Fig. 2; Table S3). Among them, A1_05 contains the highest number of BaTPS genes, with a total of 8; A2_05 follows with 6 genes; and both B1_05 and B1_09 harbor 5 genes each. In contrast, 17 chromosomes scale scaffold carry only a single BaTPS gene.The formation and expansion of gene families are largely attributed to gene duplication events³⁴. Analysis of gene duplication indicated that BaTPS members primarily expanded through whole-genome duplication (WGD) or segmental duplication events, involving 30 gene pairs (comprising 42 genes). Additionally, five tandem duplication events were identified, involving 10 genes (Fig. 2). In total, these duplication events encompassed 48 BaTPS genes, accounting for 57.17% of the gene family.

Fig. 2.

Collinearity relationships of BaTPS genes. From the innermost to the outermost layers, the figure displays collinearity links, gene density, and chromosomal distribution, respectively. Green lines indicate collinear gene pairs, while yellow highlights represent segmental duplication events.

Further analysis of selection pressure was conducted on the duplicated gene pairs (Table S4). The Ka values ranged from 0.00075481 to 0.34176834, and the Ks values varied from 0.00278681 to 1.19646224. The Ka/Ks ratios spanned from 0.04872042 to 1.65044406. Notably, only the gene pairs BaTPS08–BaTPS35 and BaTPS11–BaTPS38 exhibited Ka/Ks ratios greater than 1, suggesting that they were under positive selection. Overall, the results indicate that the BaTPS genes have predominantly been subject to purifying selection throughout their evolutionary history, helping to preserve the functional integrity of the gene family.

Analysis of BaTPS gene structure

Conserved protein motif analysis of BaTPS family members was performed using the MEME web server, with the number of motifs set to 20 (Fig. 3A). The number of motifs per BaTPS protein ranged from 7 to 16, with individual motif lengths varying between 21 and 50 amino acids (Fig. 3B). Among these, Motif 1 contains the highly conserved “DDxxD” domain (Fig. 3E), a signature residue involved in metal ion coordination and substrate catalysis³⁵, and is present in 67 BaTPS members. Motif 8 encodes the “RDR” motif, found in 72 BaTPS proteins, and is universally present in the TPS-c subfamily, which notably lacks Motif 1. Motif 14, which harbors the “RRX(8)W” motif, is specific to the TPS-a and TPS-b subfamilies. Motif 2 contains the “NSE/DTE” motif (Fig. 3E) and is found in all BaTPS proteins except BaTPS52. Additionally, Motifs 15, 16, and 19 are exclusively found in members of the TPS-c and TPS-e/f subfamilies. Analysis of conserved protein domains (Fig. 3C) revealed that TPS-c subfamily members possess only the PLN02279 superfamily domain, TPS-e/f members contain only the PLN02592 domain, and members of the TPS-a subfamily (except BaTPS46) exclusively contain the Terpene_cyclase_plant_C1 domain.

Fig. 3.

Conservative motif, domain structure, and gene structure of BaTPS. (A) Phylogenetic tree of BaTPS gene. (B) Conserved motif identified in BaTPS gene protein. (C) Conserved domain of BaTPS gene protein. (D) Structure of the BaTPS gene (Green blocks represent exons and the gray lines represent introns). (E) Sequence markers of motif 1, motif 2, motif 8, motif 14.

Further analysis of the gene structure characteristics of BaTPS genes revealed that TPS-c and TPS-e/f subfamily members tend to have more introns, ranging from 7 to 15, with only BaTPS38 containing fewer than 10 introns (Fig. 3D). In contrast, members of the TPS-a, TPS-b, and TPS-g subfamilies typically have fewer introns, ranging from 4 to 8. Moreover, several gene pairs, including BaTPS01–BaTPS27, BaTPS06–BaTPS33, BaTPS08–BaTPS35, BaTPS08–BaTPS77, and BaTPS60–BaTPS75, display highly conserved gene structures, with similar exon counts and nearly identical intron lengths and positions. These findings suggest a strong evolutionary conservation in gene architecture among closely related BaTPS genes.

Analysis of Cis-acting elements in the BaTPS promoter

Analysis of promoter cis-acting elements provides valuable insights into the tissue-specific expression and stress response mechanisms of BaTPS genes. Examination of the 2000 bp upstream promoter regions of the BaTPS coding sequences revealed that these promoters contain numerous cis-acting elements associated primarily with light responsiveness (890 elements), plant hormone responsiveness (692), stress responsiveness (419), and plant growth and development (145) (Fig. 4; Table S5). Among these categories, light-responsive elements were the most abundant, with the G-box element being the most prevalent across BaTPS promoters. This suggests that light is a major regulatory factor influencing BaTPS gene expression.

Fig. 4.

Cis-acting elements of the promoter of the BaTPS gene. (A) Distribution of cis-acting elements. (B) Statistics on the number of cis-acting elements with four different functions.

Within the hormone-responsive category, the ABRE element—associated with abscisic acid (ABA) signaling—was the most widely distributed, followed by the CGTCA-motif, which mediates responses to methyl jasmonate (MeJA). In terms of stress-related elements, the anaerobic response element (ARE) was the most frequently observed, indicating potential involvement in hypoxic stress responses. This was followed by the low-temperature response element (LTR), suggesting that some BaTPS genes may also play roles in cold stress tolerance.

For elements linked to growth and developmental processes, the O2-site—associated with zein metabolism regulation—was the most common, followed by the CAT-box, which is related to meristem-specific expression. Collectively, these findings indicate that BaTPS genes are regulated by complex networks involving light, hormone signaling (particularly ABA and MeJA), and various abiotic stresses. This highlights the potential roles of BaTPS genes in contributing to the growth, development, and environmental adaptability of B. alba.

BaTPS gene expression pattern and metabolite association analysis

To investigate the tissue-specific expression of BaTPS gene family members, transcriptome data (FPKM > 0.5) were analyzed. The results demonstrated distinct tissue-specific expression patterns of BaTPS genes in B. alba (Fig. 5A). Six genes were highly expressed exclusively in stems, with BaTPS06, BaTPS33, BaTPS55, and BaTPS72 all belonging to the TPS-c subfamily. Eighteen genes showed high expression specifically in flowers, predominantly from the TPS-a and TPS-g subfamilies, except for BaTPS45, which belongs to the TPS-b subfamily. Eight genes were uniquely highly expressed in leaves, and five genes were predominantly expressed in roots; both groups mainly comprised members of the TPS-a and TPS-b subfamilies. Expression levels of the TPS-e/f subfamily members varied considerably: BaTPS19 and BaTPS82 exhibited high expression in both flowers and leaves, whereas BaTPS11 and BaTPS74 were highly expressed in roots and stems.

Fig. 5.

BaTPS gene expression pattern and metabolite association analysis. (A) Expression profile of BaTPS gene. (B) The relative expression levels of the BaTPS gene in different tissues. (Different lowercase letters indicate statistically significant differences (P < 0.05) among various tissue types). (C) Terpenoid-related correlation network diagram.

.

To further validate the reliability of the transcriptome data, 12 BaTPS genes exhibiting high expression in stems, leaves, and flowers—identified through gene expression pattern analysis—were selected for RT-qPCR validation. The RT-qPCR results demonstrated that the relative expression patterns of these genes across different tissues largely matched the transcriptome data, confirming its accuracy (Fig. 5B). BaTPS01, a member of the TPS-b subfamily, showed high expression exclusively in stems and may be involved in monoterpenoid synthesis. BaTPS16 and BaTPS41, both belonging to the TPS-a subfamily, were highly expressed only in flowers, suggesting roles in sesquiterpenoid biosynthesis. BaTPS33, a TPS-c subfamily member, exhibited strong expression specifically in stems, indicating a potential function in diterpene synthase synthesis. BaTPS47, BaTPS53, and BaTPS60—all from the TPS-g subfamily—were highly expressed exclusively in flowers, with minimal or no expression detected in leaves, roots, or stems. These genes are likely associated with monoterpene synthase activity and the biosynthesis of volatile compounds. Subsequent correlation analysis between candidate BaTPS genes and terpenoid metabolites further supported the potential associations between these genes and terpenoid biosynthesis (Fig. 5C, Table S6). For example, BaTPS16 (g21056.t1) and BaTPS41 (g88203.t1), both belonging to the TPS-a subfamily, showed positive correlations with several putative sesquiterpenoid compounds, including Dendronobilin I (Zjjp092610), 4-(1,2,4a,5-tetramethyl-2,3,4,7,8,8a-hexahydronaphthalen-1-yl)butan-2-one (PD0619170), methyl 9-hydroxy-1-(5-hydroxy-6-methylheptan-2-yl)-4-methylspiro[4.5]dec-7-ene-8-carboxylate (PDP193819), and Balsamiferine E (Zmap010726). In addition, BaTPS47 (g97790.t1), BaTPS53 (g131301.t1, g131301.t2), and BaTPS60 (g24891.t1), which belong to the TPS-g subfamily, were positively correlated with several putative monoterpenoid-related compounds, including Blumenol C (Qmyp101440), 12,13-Dehydrogeranylgeraniol (Rlp00117), methyl 9-hydroxy-1-(5-hydroxy-6-methylheptan-2-yl)-4-methylspiro[4.5]dec-7-ene-8-carboxylate (PDP193819), and 5-hydroxy-1,4a-dimethyl-3,4,5,6,7,8-hexahydronaphthalen-2-one (Wasdp07799).

Tertiary structure prediction and subcellular localization analysis of BaTPS16 and BaTPS41

Gene structure and expression analyses suggested that BaTPS16 and BaTPS41 may be involved in sesquiterpene biosynthesis in the floral organs of B.alba. Comparative three-dimensional structural modeling revealed high similarity to AtTPS21 and AtTPS11, respectively (Fig. 6A). BaTPS16 and BaTPS41 share sequence identities of 70.44% and 77.11% with germacrene-A synthase, with GMQE scores of 0.91 and 0.86, indicating high model reliability (Fig. 6B). Given that AtTPS21 and AtTPS11 are responsible for sesquiterpene synthesis in Arabidopsis floral tissues, we hypothesize that BaTPS16 and BaTPS41 may perform analogous functions.

Fig. 6.

Tertiary structure prediction and subcellular localization analysis. (A) The three-dimensional spatial structures of the proteins encoded by BaTPS16, BaTPS41, AtTPS21 and AtTPS11. (B) The ramachandran plots of BaTPS16 and BaTPS41. (C) Subcellular localization of BaTPS16 and BaTPS41 proteins in tobacco leaves. Scale bars represent 20 μm.

To determine their subcellular localization,, we performed transient expression assays in tobacco leaves via Agrobacterium-mediated transformation of GFP-fusion constructs (BaTPS16::GFP and BaTPS41::GFP). Confocal laser scanning microscopy revealed that the GFP signals were predominantly localized in the cytoplasm, with minimal overlap with chlorophyll autofluorescence (Fig. 6C), consistent with in silico localization predictions (Table S2). These results suggest that BaTPS16 and BaTPS41 encode cytoplasmic proteins that likely perform their biological functions within the cytoplasm.

Discussion

Terpenoids represent the most diverse and abundant class of plant secondary metabolites, playing crucial roles in stress responses as well as the regulation of plant growth and development. Additionally, they possess antioxidant, anti-inflammatory, anticancer, and tumor-suppressive properties^36,37. In this study, a total of 84 BaTPS genes (designated BaTPS01 to BaTPS84) were identified from the whole genome of B. alba. This number is notably higher than that reported in (A) thaliana (33) and Medicago truncatula (23)²⁶, and comparable to Camphora longepaniculata (86)³⁸, suggesting that (B) alba may have a more complex terpenoid metabolic pathway. BaTPS proteins range from 282 to 834 amino acids in length, with molecular masses between 33.04 and 96.81 kDa. These parameters overlap with those reported for TPS proteins in other species and align with the fundamental characteristics of the TPS family. Prediction of protein secondary structure indicated that α-helices and random coils constitute the predominant structural components in BaTPS proteins, collectively exceeding 85% of the total structure. In contrast, extended strands and β-turns represent minor components, collectively accounting for less than 15%. In addition, subcellular localization predictions indicate that BaTPS proteins are predominantly localized to the chloroplasts and cytoplasm. Of the 36 BaTPS genes localized in chloroplasts, most belong to the TPS-b (15 genes) and TPS-g (8 genes) subfamilies, likely encoding proteins with N-terminal transit peptides involved in the MEP (methylerythritol phosphate) pathway. These enzymes catalyze the synthesis of monoterpenes and diterpenes from GPP (geranyl pyrophosphate) and GGPP (geranylgeranyl pyrophosphate), respectively³⁹. The remaining BaTPS genes localized in the cytoplasm probably lack transit peptides and are mainly responsible for sesquiterpene synthesis from FPP (farnesyl pyrophosphate), derived from the MVA (mevalonate) pathway³⁹.

The terpene synthase (TPS) gene family is classified into seven subfamilies based on structural features⁷. A large number of TPS genes have previously been identified in Asteraceae plants, mainly concentrated in the TPS-a and TPS-b subfamilies. Their expansion is primarily driven by tandem and segmental duplications, and they are highly expressed in roots, stems, and leaves, participating in the biosynthesis of sesquiterpenes and monoterpenes, respectively¹⁹. This feature is also observed in B. alba. Phylogenetic analysis showed that B. alba TPS genes cluster into five subfamilies—TPS-a, TPS-b, TPS-c, TPS-e/f, and TPS-g—similar to most angiosperms. Like other plants such as S. lycopersicum, Rosa chinensis, and V. vinifera^14,18,35, B. alba exhibits the greatest number of members in the TPS-a and TPS-b subfamilies. However, compared with other reported Asteraceae species, B. alba shows a relatively higher number of TPS-g members. For example, TPS-g genes have been reported to be fewer in species such as Chrysanthemum nankingense⁴⁰, Mikania micrantha⁴¹, and Artemisia argyi⁴², where TPS-a and TPS-b subfamilies dominate the family composition. This suggests that B. alba may have experienced a relatively greater retention or expansion of TPS-g genes, although broader comparative analyses are required to confirm whether this represents a lineage-specific feature. Moreover, the TPS-g subfamily shows a notable expansion in B. alba compared with A. thaliana. gene family evolution and expansion are mainly driven by gene duplication events³⁴. Collinearity analysis revealed that whole-genome duplication and segmental duplication events are the primary forces behind the expansion of the BaTPS family, involving 42 genes, a pattern consistent with that observed in Triticum aestivum⁴³. In the TPS-a subfamily, all gene pairs originated from whole-genome or segmental duplication events. Selective pressure analysis showed that, except for the BaTPS08-BaTPS35 pair, all gene pairs have Ka/Ks ratios below 1, indicating predominant purifying selection that maintains gene function stability.

Regarding conserved domains, most BaTPS genes share highly conserved exon–intron structures, supporting the accuracy of subfamily classifications. The hallmark conserved domain in BaTPS proteins is the “DDXXD” motif (Motif 1), which facilitates metal ion binding. Other important motifs include the “RDR” motif (Motif 8), “RRX(8)W” motif (Motif 14), and “NSE/DTE” motif (Motif 2). Comparative analysis showed that the TPS-c and TPS-e/f subfamilies contain significantly more introns (7 to 15) than the other subfamilies, consistent with findings in E. robusta¹². Since exon–intron structures often reflect evolutionary history and functional divergence, these results suggest that TPS-c and TPS-e/f subfamilies may contribute substantially to the functional diversification of BaTPS genes⁴⁴. Additionally, gene pairs such as BaTPS01–BaTPS27, BaTPS06–BaTPS33, BaTPS08–BaTPS35, BaTPS08–BaTPS77, and BaTPS60–BaTPS75 display highly similar gene structures, implying shared functional characteristics. Promoter analysis identified diverse cis-acting elements in BaTPS genes, including light-responsive, plant hormone-responsive, stress-responsive, and growth-related elements. Light-responsive elements are the most abundant and present in all gene members. In grape berries, photoperiod regulates the biosynthesis of volatile terpenoids⁴⁴, and the abundance of light-responsive elements in BaTPS promoters suggests that terpenoid biosynthesis in B. alba is also influenced by light. Notably, MYB transcription factor binding sites were detected in BaTPS promoters. MYB factors are known to regulate terpenoid biosynthesis, contributing to the functional diversification of TPS genes in B. alba⁴⁵. Furthermore, several BaTPS genes (e.g., BaTPS02, BaTPS41, BaTPS47, BaTPS53) contain MYB binding sites involved in flavonoid biosynthesis regulation, indicating possible roles in flavonoid production. Flavonoids exhibit antioxidant activity and diverse pharmacological effects, including anti-inflammatory, antimicrobial, anticancer, and cancer-preventive properties⁴⁶.

RNA-seq analysis revealed tissue-specific expression patterns of BaTPS genes in B. alba, consistent with reports in R chinensis³⁵ and Citrus sinensis⁴⁷. Of the 84 BaTPS genes, 51 were expressed in at least one tissue (FPKM > 0.5), with most genes (except BaTPS49) showing high expression in only one or two tissues. RT-qPCR validation of 12 highly expressed BaTPS genes confirmed the reliability of transcriptome data. Notably, four TPS-a subfamily members (BaTPS02, BaTPS08, BaTPS16, and BaTPS41) exhibited flower-specific high expression. Previous studies have shown that AtTPS21 and AtTPS11 are primarily responsible for sesquiterpene synthesis in Arabidopsis, with AtTPS21 confirmed to function mainly in floral tissues⁴⁸. Accordingly, we compared their three-dimensional protein structures and found that BaTPS16 and BaTPS41 exhibit high similarity to AtTPS21 and AtTPS11. Furthermore, subcellular localization experiments confirmed that both BaTPS16 and BaTPS41 are localized in the cytoplasm, consistent with bioinformatic predictions. Together, these findings suggest that BaTPS16 and BaTPS41 likely participate in the synthesis of sesquiterpenoid compounds in the floral organs of B.alba. Additionally, three TPS-g members (BaTPS47, BaTPS53, BaTPS60), key genes in monoterpene biosynthesis, showed flower-specific expression. These findings suggest that the floral organs of B. alba are rich in diverse terpenoid compounds. Interestingly, BaTPS02, BaTPS41, BaTPS47, and BaTPS53 contain MYB-related cis-elements and show flower-preferential expression. Given that MYB transcription factors have been reported to regulate terpene biosynthesis genes in several plant species^49,50, these genes may be subject to MYB-mediated regulation in floral tissues. However, whether this reflects a regulatory relationship between terpenoid and flavonoid metabolism in B. alba requires further experimental validation. This provides valuable insights for future exploration of the plant’s medicinal potential.

Conclusion

In this study, we performed a comprehensive genome-wide identification and characterization of the TPS gene family in B. alba, identifying 84 BaTPS genes unevenly distributed across 37 chromosomes-scale scaffolds. Phylogenetic analysis classified these genes into five subfamilies (TPS-a, TPS-b, TPS-c, TPS-e/f, and TPS-g), with a notable expansion of the TPS-g subfamily compared to (A) thaliana. Genome duplication and segmental duplication were identified as the main drivers of this expansion. Most BaTPS genes are under purifying selection, indicating strong functional conservation, while conserved motifs such as DDxxD, RDR, RRX(8)W, and NSE/DTE further support their structural integrity. Promoter analysis revealed abundant cis-acting elements related to light responsiveness, hormone signaling, and abiotic stress responses, indicating that BaTPS genes contain diverse cis-regulatory elements potentially involved in environmental and developmental responses. Transcriptome and RT-qPCR data demonstrated distinct tissue-specific expression patterns, with several BaTPS genes highly expressed in floral organs, suggesting that flowers may serve as major sites for terpenoid biosynthesis in (B) alba. Overall, this study provides a comprehensive genomic framework for understanding the evolutionary patterns, structural characteristics, and regulatory mechanisms of the BaTPS gene family, and lays the groundwork for future research on terpenoid biosynthesis and metabolic engineering in B. alba.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, Z.W., X.X., J.W. and R.S.; methodology, Z.W., X.X., J.W. and R.S.; software, Z.W., X.X. and P.Z.; validation, Z.W., P.Z. and X.X.; formal analysis, Z.W.,and X.X.; investigation, Z.W., X.X., X.H. and C.Z.; writing—original draft preparation, Z.W.,and X.X.; writing—review and editing, J.W., R.S. and Y.K.; supervision, J.W. and R.S.; project administration, J.W. and R.S.; funding acquisition, J.W. and R.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Jing Wang, the Clinical Research Center for Traditional Chinese Medicine Anorectal and Perianal Wound Repair of Fujian Province (2022Y2011).

Data availability

The transcriptome data have been deposited in the the National Center for Biotechnology Information (NCBI, https://www.ncbi.nlm.nih.gov/), with the accession number PRJNA1336433.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval and consent to participate

Not applicable. All samples were collected with permission from Fujian Agriculture and Forestry University. Since this study did not involve any endangered or protected species, no ethical approval or consent was required.