1. Introduction
The Orchidaceae family encompasses a vast array of germplasm resources, comprising more than 28,000 distinct species, which are classified into over 800 genera across five subfamilies. These diverse species predominantly inhabit tropical and subtropical regions [
1]. Some of the notable Orchidaceae species, including
Phalaenopsis,
Dendrobium,
Cymbidium,
Bletilla, and
Gastrodia plants, are cultivated for their significant aesthetic, cultural, and pharmaceutical value.
Dendrobium, the second-largest genus within the Orchidaceae family, typically thrives in warm, moist, and semi-shaded environments, often as epiphytes on tree trunks or rocks [
2].
Dendrobium species have attracted significant attention due to their abundant reserves of polysaccharides, alkaloids, flavonoids, and other bioactive compounds [
3]. These compounds have been widely utilized in healthcare due to their notable anti-aging, anti-oxidation, and anti-tumor properties [
3,
4].
Dendrobium sinense, a perennial epiphytic plant, is distributed in the mountainous areas of central and western Hainan Island, including the counties of Lingshui, Baoting, and Qiongzhong [
5].
D. sinense stands out as a precious and endemic species on Hainan Island, holding significant economic value. Morphologically, its flowers are white, solitary, and located at the apex of the stem. With several petals, they collectively present an aesthetically pleasing appearance, adding to the plant’s high ornamental value. Notably,
D. sinense has been extensively used as traditional ethnic medicine in Hainan [
6]. This species exhibits remarkable capabilities in synthesizing various bioactive compounds, including alkaloids, bibenzyls, flavonoids, phenanthrenes, polysaccharides, and other valuable constituents [
6,
7,
8]. Consequently, there has been an increasing interest in investigating the pharmacological effects and molecular mechanisms associated with
D. sinense. For instance, four bibenzyls and twelve phenolic compounds were isolated from the entire
D. sinense plant [
9]. Furthermore, the characterization of a
bibenzyl synthase (
BBS) genes in
D. sinense has yielded valuable insights into potential biosynthetic pathways [
5]. Metabolomics in
D. sinense unveiled a significant abundance of phenylpropanoids and polyketides, with many of these compounds responding to drought stress [
10]. Despite progress, our understanding of the regulatory mechanisms underlying the active ingredients in this species remains limited.
Polyketides constitute a substantial class of secondary metabolites present in plants that fulfill crucial roles in biotic and abiotic stress responses, signaling pathways, and tissue compositions [
11]. This category of natural compounds is widely recognized for its potential therapeutic applications, including antimicrobial, anticancer, antiparasitic, and anti-cholesterol properties [
12]. The biosynthesis of polyketides is governed by a group of enzymes known as polyketide synthases (PKS), which facilitate the sequential condensation of simple two-carbon acetate units with an acyl starter molecule [
13]. PKS is generally classified into three types based on protein domains and catalytic mechanisms [
14]. Type I PKS is characterized by multiple catalytic modules and is primarily responsible for catalyzing the biosynthesis of compounds such as erythromycin, sirolimus, and rifamycin, typically found in fungal or bacterial systems [
15]. Type II PKS, comprising two ketosynthase units (α- and β-ketosynthase), is specifically involved in the production of bacterial aromatic polyketides, such as anthracyclines and tetracyclic compounds [
16]. Type III PKS features a relatively simpler structure but plays a significant role in generating a diverse array of polyketide compounds, such as chalcones, pyrones, acridones, phloroglucinols, stilbenes, and bibenzyls [
12].
In plants, type III PKS emerges as a pivotal enzyme that partakes in the biosynthesis of secondary metabolites [
11]. This enzyme family comprises a range of enzymes including chalcone synthase (CHS), stilbene synthase, acridone synthase, stilbenecarboxylate synthase, and BBS [
15]. Notably, plant type III PKSs facilitate cyclization reactions of polyketides, which can be classified into three main types: Claisen cyclization, aldol cyclization, and lactonization [
11]. According to our previous next-generation sequencing analysis, the CHS type predominantly prevails within the type III PKS family in
D. sinense [
5]. CHS primarily assumes responsibility for catalyzing the stepwise condensation of
p-coumaric coenzyme A, with three acetate units derived from malonyl coenzyme A, initiating the formation of chalcones [
13]. Subsequently, the enzyme chalcone flavanone isomerase (CHI) facilitates the isomerization of chalcone, leading to the formation of flavanone [
17]. From these core intermediates, the pathway diverges into various secondary routes, each guiding distinct categories of flavonoids (
Figure 1). For instance, dihydroflavonol 4-reductase contributes to flavanols synthesis, flavanone 3-β-hydroxylase to dihydroflavonols, flavonol synthase to flavonols, and anthocianin synthase to anthocyanins [
18]. Hence, the sequential condensation step catalyzed by CHS is the rate-limiting process in the biosynthesis of all plant flavonoids.
To systematically identify the type III family of D. sinense PKS (DsPKS), the previous data obtained from second- and third-generation sequencing were reanalyzed. The identification of type III DsPKSs was based on conserved domain analysis, followed by phylogenetic classification into distinct clusters. Multiple protein sequence alignments revealed that the DsCHS3 gene exhibited two alternative splice forms. Prokaryotic expression plasmids of DsCHS3-1 and DsCHS3-2 were successfully constructed based on gene cloning techniques. Subsequently, the specific DsCHS3-1 and DsCHS3-2 fusion proteins were purified through GST affinity chromatography. To assess their enzymatic activity, in vitro enzyme activity assays were performed utilizing recombinant DsCHS3-1 and DsCHS3-2 proteins. To validate the functional differences in enzyme activity in vivo, we conducted complementation experiments by introducing transgenic DsCHS3-1 and DsCHS3-2 constructs into Atchs (AT5G13930) mutants. This study successfully identified the two alternative splice forms of DsCHS3 in D. sinense and elucidated their distinct activity profiles. These findings provide a foundation for a deeper comprehension of the biosynthetic mechanism underlying flavonoid production.
2. Materials and Methods
2.1. Identification of Type III PKS Genes and Phylogenetic Tree Analysis
The identification of type III
DsPKS genes in
D. sinense involved a comprehensive analysis of Illumina and PacBio sequencing data (NCBI BioProject: PRJNA723915) [
5,
10]. A screening approach was implemented based on the presence of the conserved ketoacyl synthase (KAS) domain within the candidate sequences. Based on the PacBio sequencing data under drought stress [
10], the digital expressions of
DsPKS genes were calculated by Fragments Per Kilobase Million (FPKM) using the TBtools v1.09876 [
19].
For comparative analysis, homologous protein sequences of type III PKS from
Arabidopsis thaliana (AtPKS),
Phalaenopsis equestris (PePKS), and
Dendrobium catenatum (DcPKS) were acquired from TAIR10, NCBI ASM126359v1, and NCBI ASM160598v2, respectively. Subsequently, all the protein sequences belonging to the type III PKS family were aligned using ClustalW with default options [
20], ensuring the accurate alignment of the conserved regions. To construct a phylogenetic tree, the aligned sequences were processed using MEGA 7.0 software, utilizing the parameters described in the previous study [
21].
2.2. Analysis of Conserved Domains
The protein sequences of type III PKSs were compared by Jalview (
www.jalview.org/ accessed on 24 July 2023). To validate the alternative splicing types of
DsCHS3-1 and
DsCHS3-2 genes, DNA was extracted from
D. sinense using a plant tissue DNA extraction kit (Tiangen, Beijing, China). PCR was conducted with primers (AS-F 5’-GCATCTCTCC TTGTCAGTGAT-3’ and AS-R 5’-CGGGTTCTCCTTGAGGATTT-3’) designed to target distinct positions in the
DsCHS3-1 and
DsCHS3-2 genes. Subsequently, the PCR products were sequenced (Sangon, Guangzhou, China) to verify the presence of alternative splicing events.
The three-dimensional structure of the protein was predicted using trRosetta server (
yanglab.nankai.edu.cn/trRosetta/help/index.html accessed on 24 July 2023) to observe the difference between DsCHS3-1 and DsCHS3-2 proteins. A homologous 3D model was established using five templates to generate five homologous models, which were sorted based on the TM-score (a confidence score for estimating prediction model quality). The protein’s three-dimensional structure was imported into PyMOL 2.5.4 for visualization, and the two structures were superimposed using the align module to compare the differences between the two protein structures.
2.3. Cloning of DsCHS3-1 and DsCHS3-2 Genes
Total RNA was extracted using the RNA Easy Plant Tissue Kit (Taingen, Beijing, China), followed by reverse transcription using the High-Capacity cDNA Archive Kit (Thermo Fisher Scientific, Shanghai, China). The DsCHS3-1 and DsCHS3-2 genes were amplified by PCR using cDNA as the template and the following primers: DsCHS3-1-F 5′-TCAGTGATAATTAAGAAGGAAACCA-3′ and DsCHS3-1-R 5′-CAGCACACTTTATAGTTCCACCC-3′; DsCHS3-2-F 5′-CTCAAGGAAAAATTCAAAAGAATGT-3′ and DsCHS3-2-R 5’-GCGAGTTGAAGATGAACAGGTAG-3’. The PCR products were cloned into T vectors of PCloneEZ-TOPO (Solarbio, Beijing, China) and sequenced (Sangon, Guangzhou, China).
2.4. Prokaryotic Expression
The DsCHS3-1 and DsCHS3-2 genes were both cloned from PCloneEZ-TOPO. Consequently, they share the same forward primer, 5’-GGATAACAATTCCCCTCTAGATGTAAAACGACGGCCAGT-3’. Both DsCHS3-1 and DsCHS3-2 genes have identical 3’ terminal sequences, leading to the utilization of the same reverse primer, 5’-GCAAGCTTGTCGACGGAGCTCTTGGGTACACTGTGGAG-3’. Employing the ClonExpress II One Step Cloning Kit C112-01 (Vazyme, Nanjing, China), the PCR products were integrated into linear pET28a vectors, with double digestion of XbaI (BioLabs, Beijng, China) and SacI (BioLabs, Beijng, China), yielding the fusion vectors of DsCHS3-1-HisTag and DsCHS3-2-HisTag.
Subsequently, the fusion vectors were transferred into competent
Escherichia coli BL21 (DE3) cells. The bacterial culture was incubated at 37 °C and 180 rpm until the optical density reached 0.4–0.6. Following this, the recombinant DsCHS-HisTag protein was induced using isopropyl-β-D-thiogalactopyranoside (IPTG) under different conditions to optimize protein solubility. The optimal IPTG concentration was determined by assessing protein solubility at different temperatures, including 23 °C for 16 h, 37 °C for 6 h, and 15 °C for 24 h [
22]. According to the selected IPTG concentration and induction temperature, a large-scale bacterial culture was performed. The harvested bacterial solution was centrifuged at 4 °C for 20 min, and the supernatant was carefully decanted. The pellet containing
E. coli lysate was re-suspended in 150 mL of a 10 mmol/L imidazole solution. Ultrasonication was performed using an ultrasonic instrument on an ice-water mixture for 20 min. The lysate was then centrifuged at 4 °C for 20 min, and the resulting supernatant underwent purification using Ni-IDA resin (Novagen, Madison, WI, USA). The purified protein was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and visualized via Coomassie brilliant blue staining. Additionally, the concentration of the purified protein was determined using a BCA protein concentration detection kit (Bioshorp, Beijing, China).
2.5. Enzyme Activity Assays
For the kinetic analysis, enzymatic reactions were conducted using a reaction mixture consisting of 0.5 mM malonyl-CoA (Yuanye, Shanghai, China), 50 mM Hepes buffer pH 7.0 (Bioshorp, Beijing, China), and 10 μg of the CHS-HisTag fusion protein. The concentration of the p-coumaroyl-CoA substrate (Yuanye, Shanghai, China) was varied across the range of 0 to 5 mM (0.0, 0.5, 1.0, 3.0, and 5.0 mM). The reaction volume was adjusted to 50 μL using ddH2O. The enzymatic reactions were carried out at 37 °C for 30 min, and subsequently terminated by the addition of an equal volume of methanol. Following centrifugation at 12,000 rpm for 10 min, the supernatant was collected and filtered through a 0.22 μm filter membrane. A 20 μL portion of the filtered solution was subjected to analysis by high-performance liquid chromatography (HPLC).
2.6. HPLC Analysis
The samples were detected by LC-100 pump (Wufeng, Shanghai, China) with HC-C18 (18 μm, 4.6 × 250 mm, Agilent, Santa Clara, CA, USA). Chromatographic conditions: column temperature, 30 °C; sample volume, 20 μL, solution A, 0.1% phosphate water; solvent B, 100% acetonitrile; flow rate, 1 mL/min. Setting the mobile phase gradient: 0 min, solution A, 70%, solution B, 30%; 6 min, solution A, 55%, solvent B, 45%; 40 min solution A, 20%, solvent B, 80%. Naringin chalcone (Yuanye, Shanghai, China) was used as the standard substance.
2.7. Eukaryotic Expression
The
DsCHS3-1 and
DsCHS3-2 genes were cloned from PCloneEZ-TOPO and share the same forward (5′-GAGAACACGGGGGACTCTAGATGTAAAACGACGGCCAGT-3′) and reverse (5′-ACGATCGGGAAAATTCGAGCTCCAGGAAACAGCTATGACC-3′) primers. The PCR products were integrated into linear pBI121 vectors (after double digestion of
XbaI and
SacI) using the ClonExpress II One Step Cloning Kit C112-01 (Vazyme, Nanjing, China). The T-DNA insertion mutants of
Atchs (AT5G13930, SALK_076535C) of Col-0 was bought from AraShare (
www.arashare.cn/index/ accessed on 24 July 2023). The homozygotes were screened out using the three-primer method. Transgenic plants were generated using
Agrobacterium tumefaciens strain EHA105 (WEIDI, Shanghai, China) via the inflorescence dip method [
23]. After further kanamycin (Solarbio, Beijing, China) resistance screening and PCR verification, stable genetic lines were obtained.
2.8. Determination of Total Flavonoid Content
A slightly modified version of the previously described sodium nitrite–aluminum nitrate colorimetric method was employed to determine total flavonoids [
24]. The flavonoid content in the leaves of one-month-old
Arabidopsis was detected. A total of 5 g samples were thoroughly ground and extracted using 50 mL anhydrous ethanol at 90 °C for 2 h. Following centrifugation at 12,000 rpm for 10 min, the supernatant of the extract was transferred to a new centrifuge tube. After rotary evaporation and concentration, 2 mL anhydrous ethanol was mixed, and a 1 mL sample was added into a 25 mL volumetric flask and the volume was fixed with anhydrous ethanol. For each 10 mL sample, 30% anhydrous ethanol was added to 12.5 mL, and 5% sodium nitrite solution was added to 0.75 mL before being shaken and then left to stand for 5 min. Then, 10% aluminum nitrate solution was added to 0.75 mL before being shaken and left to stand for 5 min. Then, 1 mol/L NaOH solution was added to 10 mL, mixed with 30% ethanol volume, and was left to stand for 10 min. Rutin (Yuanye, Shanghai, China) was utilized for generating standard curves. The wavelength of each sample was measured by a spectrophotometer (METASH, Shanghai, China) at 510 nm. The ANOVA analysis (
p < 0.01) was conducted using SPSS statistical software (version 19).
4. Discussion
Dendrobium sinense, a native orchid species in China, is gaining increasing attention due to its remarkable aesthetic appeal and valuable medicinal properties [
5,
7]. This species has been recognized as a traditional Chinese herbal medicine, revered for its antioxidant, antibacterial, and antitumor capabilities, which can be attributed to the presence of various bioactive compounds, including alkaloids, flavonoids, and bibenzyls [
6,
9]. Despite its significance, limited research has been conducted on the regulatory mechanisms underlying the biosynthesis of these bioactive compounds in
D. sinense. Our previously comprehensive transcriptomic analysis offers a significant opportunity to elucidate the intricate secondary metabolic processes occurring in
D. sinense and serves as a valuable resource for exploring gene families associated with these processes [
5,
10].
Polyketides, a class of biologically active natural products in plants, are synthesized through the condensation of acyl-thioester units, such as malonyl-CoA and methylmalonyl-CoA, resulting in the formation of metabolites with a wide range of structures and biological activities [
14]. Plant type III PKSs have been widely studied since their initial discovery in parsley [
25], and they play a crucial role in the biosynthesis of various plant metabolites [
11]. Utilizing the comprehensive transcriptome data obtained from previous studies [
5,
10], a total of 10 type III
DsPKS genes were successfully identified in
D. sinense. The incorporation of multiple transcriptome data sources in this study has led to the identification of a greater number of type III
DsPKS genes compared to the previous report [
5]. These lengths of type III DsPKS proteins were consistent with a previous report that plant type III PKSs consist of approximately 400 amino acid [
11].
Despite the existence of over 20 different members of plant type III PKSs with diverse functions [
12], the type III PKS proteins from
D. sinense were classified into three distinct groups, namely CHS group, BBS group, and PKS group. Notably, the PKS group comprised the largest number of members among the identified type III PKS proteins in
D. sinense. This classification highlights the diversity and functional specialization within the type III PKS family in
D. sinense, further emphasizing the importance of these enzymes in the synthesis of various metabolites in this orchid species. Future investigations into the specific functions and regulatory mechanisms of these type III PKS genes in
D. sinense will provide valuable insights into the biosynthesis of bioactive compounds in this plant species and broaden our understanding of the unique metabolic pathways present in orchids.
CHS is a member of the type III PKS family, responsible for catalyzing the condensation reaction between one molecule of
p-coumaroyl-CoA and three molecules of malonyl-CoA. This enzymatic reaction leads to the formation of naringenin chalcone, which serves as a key precursor for the biosynthesis of a wide range of flavonoids, including flavones, flavonols, flavanones, and anthocyanins (
Figure 1) [
12]. The overall three-dimensional fold of plant type III PKSs is highly conserved, characterized by a catalytic core domain composed of the Cys–His–Asn triad [
26]. This conserved triad, depicted by a red box in
Figure 3a, plays a crucial role in decarboxylative condensation and cyclization reactions that give rise to the diverse array of products produced by PKS enzymes [
13]. Recent crystallographic and site-directed studies revealed that even small modification in the active site architecture can result in the functional diversity of the type III PKSs [
27]. In
D. sinense, the DsBBS and DsCHS proteins were found to possess specific amino acid residues at positions 202 (leucine and threonine) and 274 (alanine and phenylalanine), respectively (
Figure 3a). Our previous
in vitro enzyme assay confirmed that the DsBBS1 protein could catalyze the cyclization and aromatization of
p-coumaryol-CoA and malonyl-CoA to generate bibenzyls [
5]. In the present study, we have identified
DsCHS3 proteins as being involved in chalcone biosynthesis, further highlighting the functional diversity within the type III DsPKS family. It is plausible that these small modifications within the conserved domain contribute to the generation of functional diversity among type III DsPKS enzymes. To elucidate the relationship between amino acid residue changes and catalytic activity in type III PKS enzymes, additional investigations are warranted. Future studies could involve site-directed mutagenesis to introduce specific amino acid changes at key positions and subsequently assess the impact on catalytic efficiency.
Interestingly, upon conducting multiple sequence alignment, it was observed that the gene sequences of
DsCHS3-1 and
DsCHS3-2 were identical, except for a notable difference at the 5’ end, where the
DsCHS3-1 gene exhibited an additional 217-base segment. Thus, the full-length transcripts suggested the presence of alternative splice forms of the
DsCHS3 gene. Alternative splicing is a widespread process in cellular gene expression, where different combinations of exons within a pre-mRNA are selectively joined, giving rise to multiple mRNA isoforms [
28]. This dynamic mechanism substantially enhances proteomic diversity by generating distinct protein variants with unique functions or regulatory properties [
29]. Various types of alternative splicing events have been characterized, including intron retention, exon skipping, mutually exclusive exons, alternative 5’ or 3’ splicing sites, alternative first exons, and alternative last exons [
30,
31]. Although intron retention is the most prevalent form of alternative splicing in plants [
31], the alternative splicing of
DsCHS3 exhibited two distinct alternative first exons, as determined by Sanger sequencing. Based on the systematic analysis of alternative first exons in plants, the alternative splicing pattern observed in
DsCHS3 can be classified as type II of alternative first exons, where the first exon of one gene structure serves as an internal exon within an alternative gene structure [
32]. Usually, different promoters can initiate transcription at the beginning of a gene, particularly when multiple transcription start sites are present, leading to the inclusion of different first exons [
33]. Additionally, specific transcription factors are known to exert influence over the selection of alternative first exons [
32]. Further investigations are warranted to elucidate the underlying mechanisms governing the generation of alternative first exons in
DsCHS3, including the identification of regulatory elements, transcription factors, or other factors that contribute to the selection and usage of specific first exons.
Alternative splicing is a widespread phenomenon in plants, playing a critical role in the regulation of gene expression and protein diversity across various biological processes, including development, tissue-specific gene expression, and environmental responses [
30,
34]. Importantly, alternative splicing has emerged as a key regulatory mechanism governing secondary metabolism [
31,
35]. Advancements in third-generation sequencing technology have revealed that alternative splicing contributes to the regulation of genes involved in flavonoid biosynthesis and their transcriptional regulators [
36,
37,
38]. Despite systematic investigations of alternative first exons being conducted in model plants of
Oryza sativa and
A. thaliana [
32,
39], our understanding of alternative first exons in
D. sinense remains limited. Through full-length transcriptome sequencing, the
DsCHS3 gene underwent different alternative splicing patterns, and
DsCHS3-1 was characterized as the main product of alternative splicing. Moreover, the expression profiles of
DsCHS3-1 and
DsCHS3-2 were tightly regulated in response to varying degrees of drought stress, suggesting a potential role for the alternative first exons of
DsCHS3 in the plant’s response to drought stress. It is plausible that different promoters and transcription factors associated with
DsCHS3-1 and
DsCHS3-2 may confer an additional layer of gene expression regulation [
39]. Elucidating the regulatory impact of alternative splicing on CHS activity holds significant importance as it can provide valuable insights into flavonoid biosynthesis and their functions in plant physiology and defense responses. In plants, CHS catalyzes the initial and rate-limiting step in flavonoid biosynthesis, thereby controlling the flow of substrates into the flavonoid pathway [
12]. Flavonoids are multifunctional compounds that play crucial roles in plant growth, development, defense against stresses, and plant-microbe interactions [
40]. Thus, alternative splicing serves as a regulatory mechanism to fine-tune CHS activity, which may be crucial for the control of flavonoid biosynthesis in response to drought stress.
The utilization of alternative first exons in gene transcription can result in the generation of alternative transcription start sites (ATGs), thereby giving rise to protein variants with distinct N-termini. This diversity in N-terminal sequences has been demonstrated to contribute to functional differences among protein isoforms [
41]. Interestingly, three-dimensional structure analysis revealed that the DsCHS3-1 protein possesses additional α-helices and β-sheets at the N-terminus compared to the DsCHS3-2 protein. Notably, these two extra β-sheets are positioned adjacent to the catalytic structural domain of the DsCHS3-1 protein. Consequently, it is conceivable that the unique structural characteristics of the DsCHS3-1 protein may impact its catalytic activity. To further investigate the functional implications of these structural differences,
in vitro enzyme assays were performed. Surprisingly, the
in vitro enzyme assay indicated that the catalytic efficiency of DsCHS3-2 activity was three times higher than that of the DsCHS3-1 activity. In addition, in-depth functional analyses were conducted through transgenic complementation tests
in vivo, which corroborated the findings from the enzyme assays. These tests demonstrated that the DsCHS3-2 protein exhibited higher catalytic activity. The comprehensive evaluation of enzyme activity both
in vitro and
in vivo strongly supports the notion that the alternative first exons can regulate the functional properties of the DsCHS3 proteins, as previously reported in related studies [
41,
42]. The observed structural differences between the DsCHS3-1 and DsCHS3-2 proteins, particularly in their N-terminal regions, suggest that alternative splicing can modulate the conformation and activity of the resulting protein isoforms. The presence of additional α-helices and β-sheets in the N-terminus of DsCHS3-1 may influence its interactions with substrates, cofactors, or regulatory molecules, potentially leading to altered enzymatic activity. Further investigations into the precise molecular mechanisms underlying these functional disparities are warranted to gain a deeper understanding of how alternative splicing modulates the activity and regulation of
DsCHS3 isoforms.