Transcriptomic, phenotypic and physiological analyses identify candidate regulators of Dendrobium stigma maturation and pollen-stigma interaction

Abstract

Background

Dendrobium species are prized for their ornamental and economic value, yet their low hybrid compatibility remains a critical challenge, potentially linked to stigma recognition mechanisms. The stigma, as the primary site for pollen interaction, undergoes dynamic developmental changes that lay the foundation for successful pollination. However, the molecular mechanisms governing stigma maturation and subsequent pollen recognition are poorly characterised. This study systematically investigated the transcriptomic, phenotypic, and physiological dynamics of Dendrobium stigma development to establish a mechanistic basis for understanding pollen-stigma recognition.

Results

Morphological observations of the four developmental stages of stigma maturation (day 1 post-anthesis [CK], early anthesis [T1], mid-anthesis [T2], and late anthesis [T3]) and investigations into the optimal receptivity window revealed that stigmatic mucilage secretion and viscosity peaked during the mid-anthesis phase (T2). Concurrently, stigma receptivity was highest at this stage, as evidenced by robust pollen tube elongation and ovarian enlargement. Enzyme activities linked to stigma-pollen recognition—esterase, superoxide dismutase (SOD), and peroxidase (POD)—demonstrated stage-specific dynamics, with peak activity occurring during mid-anthesis (T2). Transcriptome analysis revealed that a total of 13,209 (T1), 12,557 (T2), and 18,593 (T3) Differentially expressed genes (DEGs) were identified, predominantly enriched in plant hormone signal transduction, MAPK signaling, and plant-pathogen interaction pathways. Transcription factors (TFs) and KIN-class genes exhibited coordinated regulation of these pathways. During the peak pollination period (T2 stage), a total of 3,069 stage-specific differentially expressed genes (DEGs) were identified that were not observed in other developmental stages. Among these unique DEGs, one was found to encode superoxide dismutase (SOD) enzymes, while seventeen encoded esterases. Reactive oxygen species (ROS) levels peaked at T2, potentially implicating their involvement in MAPK signaling through Ca²⁺-CNGCs-CDPK-Rboh cascades, though this proposed mechanism requires experimental validation. O-GlcNAc glycosylation was detected in the stigma transcriptome for the first time, potentially linked to kinase activity and immune responses.

Conclusions

Developmental and physiological profilings suggest that mid-anthesis represents a critical period for active pollen-stigma recognition processes, potentially driven by dynamic biochemical and transcriptional adaptations in the stigma. TFs, KIN-class genes, and ROS interactively regulate stigma maturation and pollen recognition through hormone signaling, MAPK cascades, and pathogen-like defense pathways. Key candidates, including MPK6, MYC2, ERF1, and Pti1/5, were hypothesized as critical regulators. However, functional validation of these genes and pathways remains absent, and the study’s reliance on unpollinated stigmas limits insights into pollination-specific responses. Further experimental validation is required to confirm mechanistic interactions.

Introduction

Dendrobium species are renowned for their profuse flowering, extensive range of flower colours, diverse sizes and shapes, year-round availability and longevity [1]. These orchids are among the world’s four most significant foreign orchids, along with Cattleya, Phalaenopsis and Vinelandia. The excellent ornamental characteristics of Dendrobium serve to determine its considerable economic value and extensive market potential, with this species occupying a significant position within the global flower market [2, 3]. Consequently, the breeding of Dendrobium has become a primary area of research however, species of the genus Dendrobium exhibit the lowest affinity for distant hybridisation among orchids. In a cross-pollination study conducted in the Himalayas involving 37 Dendrobium species, the resulting rate of successful cross-pollination was only 8.97% [4]. An intergroup cross-pollination test was conducted on 11 varieties of Dendrobium with high ornamental value. The results demonstrated that the intergroup cross-fruitfulness of Dendrobium in Callista and Latouria was 6.6% [5, 6]. The majority of current studies on hybrid affinity in Dendrobium are phenotypic in nature, with self-incompatibility being the predominant molecular phenomenon [7, 8].

Plant hybrid incompatibility can be classified into two categories: prefertilisation and postfertilisation disorders. The success of cross-pollination is contingent upon a multitude of factors, including pollen viability, receptivity of stigma, species affinity and environmental conditions. The stigma is situated at the apex of the pistil and serves as a crucial reproductive site, facilitating the recognition and reception of pollen and its subsequent germination. The ability of pollen and stigma cells to recognise and transmit signals is a prerequisite for determining whether a plant is capable of undergoing fertilisation. The stigma selectively accepts pollen with which it is compatible and rejects that which is not [9]. During this process, the pollen and the pistil engage in a continuous exchange of substances and signals, ensuring the optimal unfolding of the pollination process [10]. Consequently, the development of the stigma frequently has a direct effect on normal pollination and fertilisation processes, thereby exerting a significant influence on the reproductive biology of plants [11]. The pollination affinity of Dendrobium can be determined by observing the growth status of the pollen tube after pollination [12, 13]. The study of Dendrobium self-infertility revealed four main cases [14]: (1) the pollen tube does not germinate; (2) the pollen tube stops growing in 1/3 of the stylar channel; (3) the pollen tube stops growing at the top of the stylar; and (4) the pollen tube stops growing in the upper 1/3 of the ovary [7, 15, 16].

Stigmas are classified into two main categories: dry and wet. Dendrobium is a wet stigma, characterised by an exterior covered by epidermal cells that produce surface secretions. This secretion varies in quantity from none to some, and then to less, in accordance with the time of opening of the flowers. The timing of flower opening significantly affects the outcomes of pollination and fruiting [17, 18]. The secretions in question are primarily composed of proteins, lipids, polysaccharides, pigments, and enzymes (esterases and peroxidases) [19]. These compounds are released by the dissolution of epidermal cells and affect stigma‒pollen interactions. Therefore, determining the optimal pollination period and exploring the changes in secretions that occur during this period are fundamental to clarify the role of pollen recognition. However, studies on esterases have focused primarily on plants with dry stigma types. For example, EXL4, which has esterase activity was identified on the surface of the pollen shell and stigma in Arabidopsis thaliana [20], and esterase activity was observed at the pollen‒stigma junction in the family Cruciferae. This evidence suggests that these esterases play crucial roles in hydrating pollen on dry stigmas, and can disrupt the continuous keratinisation of the cuticle to the extent that pollen can reach the stigma [21]. No studies have been conducted at the molecular level regarding esterases in wet stigmas. However, it has been postulated that owing to the discontinuity of the cuticle in wet stigmas, their esterases lack the capacity to recognise pollen entering the stigma [22].

Peroxidase is a protein that is found in the organelle peroxisomes. Its primary function is to facilitate the breakdown of excess reactive oxygen species (ROS), such as H₂O₂, to protect cells from oxidative damage. The increased peroxidase activity observed at stigma maturity can be attributed to the accumulation of elevated levels of ROS in the stigma. ROS, pivotal components of stigma secretion, plays crucial roles in pollen recognition [23]. ROS exist in various forms within the plant body. These include monomeric oxygen, the superoxide anion, hydrogen peroxide and the hydroxyl radical [23]. It has been demonstrated that ROS may be involved in pollen and stigma interactions [24–26]. The ROS produced in plant cells play a dual role, they can play a defensive role, but excessive concentrations can also result in the production of harmful phenomena. To prevent the excessive accumulation of ROS, plants employ superoxide dismutase (SOD) and ROS scavenging enzymes to remove them [23]. SOD facilitates the decomposition of superoxide into molecular oxygen and H₂O₂, representing the initial line of cellular defense against ROS. Additionally, they are a vital component of the peroxisome [27]. It can therefore be surmised that SOD may also influence pollen recognition. Consequently, the activities of SOD and esterases were selected for observation to ascertain their involvement in the process of change during the stigma developmental stage.

The transcriptome represents the link between the genome and protein synthesis. It is subject to regulation by both endogenous and exogenous factors, which clearly display temporal and spatial characteristics. The collection of RNA molecules expressed by a single cell under different growth conditions and periods of growth is distinct. A number of studies have been conducted on the transcriptomes of plant stigmas and styles. These include the transcriptome sequencing of stigmas from tobacco and Arabidopsis thaliana to reveal genes unique to wet and dry stigma development [9], the analysis of pollination-compatible and incompatible stigma transcriptomes to explore unigenes during pollination [28], and the use of the transcriptome to identify genes critical for color differentiation of melon stigmas [29]. There are preliminary studies on the role of stigma-pollen recognition, the small peptide PCP-B in pollen binds to receptor kinases to reduce ROS production [30], which promotes hydrated pollen germination, and the Ca²⁺ signaling pathway regulates small G proteins through receptor kinases to increase the polarity of pollen tubes [31]. The utilisation of RNA-seq enables the regulation of genes exhibiting disparate expression patterns between normal samples and those of interest, facilitating a rapid and comprehensive overview of genes that play pivotal roles in plant traits of interest.

As a hybrid of Dendrobium with a highly unaffiliated lineage, the genes and molecular mechanisms involved in stigma development and stigma‒pollen recognition remain unclear. Accordingly, the present study employed physiological analysis and pollen tube fluorescence detection to determine the optimal pollination period for Dendrobium. Additionally, RNA-seq was used to sequence the transcriptome of Dendrobium winged calyx stigmas at various developmental stages, with the objective of elucidating the relationship between the affinity pathway and the developmental stage, as well as the expression of the relevant unigenes throughout the process. Furthermore, this study aimed to investigate the influence of esterases and SOD on the functions of Dendrobium stigma‒pollen recognition and associated unigenes, thereby establishing a novel foundation for investigating Dendrobium affinity.

Materials and methods

Study site and plant materials

The experiments were conducted in the Arboretum of Southwest Forestry University (25°06′N, 102°45′E), Kunming, Yunnan Province, China. This region is situated in a subtropical plateau monsoon climate zone at an elevation of 1,964 m above sea level, characterised by a temperate climate. The area exhibits the following climatic parameters: annual frost-free period of approximately 240 days, mean annual temperature of 16.5 °C, mean annual precipitation of 1,035 mm, and mean annual relative humidity of 67%. The dominant soil type is classified as red soil. Within the greenhouse, temperature ranged from 18 °C to 37 °C, and relative humidity fluctuated between 22% and 47% during the experimental period.

Dendrobium cariniferum plants were obtained from an unrecorded vendor at the Kunming Dounan Flower Market (24°09′N, 102°78′E), China, in April 2020. Although supplier details (e.g., batch numbers, geographic origin) were unavailable, species identity was morphologically confirmed by matching floral and vegetative traits to published descriptions [32]. A voucher specimen was deposited in the Herbarium of Southwest Forestry University (SWFC), Kunming, China (Accession number: SWFC1000240521). Plants were acclimatized in a greenhouse at Kunming (25°06′N, 102°45′E) under controlled conditions: humus-rich substrate (pH 6.2–6.5), 25 ± 2 °C daytime temperature, 12-h photoperiod, and 70–80% relative humidity. Only uniformly healthy individuals (30–40 cm height, no visible lesions) were utilized for subsequent experiments.

Morphological dynamics and pollination receptivity assessment during stigma development

The dynamic secretory patterns of stigmatic cavity mucilage in Dendrobium cariniferum were systematically monitored from the day of anthesis (Day 1, defined by complete sepal expansion) through initial petal wilting (developmental endpoint). Five biological replicates (each consisting of 6 flowers) were established for longitudinal observation with concurrent documentation of stigmatic secretory volume and mucilage viscosity. Secretory volume was categorised into five discrete grades (0–4) following a standardised protocol: Grade 0 denoted absence of detectable secretions; Grade 1 indicated basal cavity moistening without substantial mucilage accumulation; Grade 2 corresponded to mucilage occupying ≤ 50% of the cavity volume; Grade 3 represented mucilage filling 51–75% of the cavity space; and Grade 4 signified complete cavity saturation with no observable air pockets. Mucilage viscosity was evaluated using a three-tiered system (0–2): Non-adhesive interactions (Grade 0) were defined by immediate pollen mass detachment upon mechanical contact, while partial adhesion (Grade 1) required ≥ 50% pollen retention following vertical touch assays. Specimens demonstrating full adhesion (Grade 2) maintained pollen integrity throughout 90° inversion testing, confirming optimal viscoelastic properties.

Stigma receptivity assessment was conducted through direct phenological evaluation: pollen tube growth trajectories and ovarian development were systematically monitored post-pollination. A developmental stage was deemed receptive if and only if dual criteria were satisfied: Pollen tubes exhibited uninterrupted growth to the ovarium (verified by aniline blue fluorescence microscopy); Significant ovarian enlargement was observed. Dendrobium cariniferum is a self-compatible species. To eliminate confounding effects of heterogeneous pollen viability on stigma receptivity assessment, all controlled pollinations were performed using day-10 post-anthesis pollen grains. Flowers at different developmental stages were randomly pollinated between 9:00–10:00 AM. Stigma samples were collected at designated post-pollination intervals (0 h, 4 h, 16 h, 24 h, 48 h, and 72 h). Pollinated stigmas were placed in FAA fixative (70% alcohol, glacial acetic acid, and formalin at a 90:5:5 ratio) at 4 °C for 24 h. The stigmas were removed from the FAA mixture, washed three times in distilled water, soaked in a 2 mol/L NaOH solution, and boiled in water for 30 min under a gentle flame. The mixture was then softened and washed two to three times, washed with buffer with a pH value of 7.0 for 50 min, subsequently placed in 0.1% aqueous aniline blue and stained for one day. After being stained and washed a total of four to five times in distilled water, the stigmas were excised with a razor blade and tweezers and mounted on slides with a minimal amount of cover slipping. Observations were conducted using an Olympus DP71 inverted fluorescence microscope with excitation/emission wavelengths set at 350/450 nm. Digital images were acquired under standardized exposure parameters.

Enzymatic activity profiling across stigma developmental stages in Dendrobium cariniferum

SOD activity was quantified via a commercially available SOD activity kit (Product No. D799594-0100) and the esterase activity was quantified using a commercially available Plant esterase (esterase) ELISA Kit (Product No. BY-Z17668) from China BAIYI Biology Corporation. Stigmas of similar size and synchronized flowering time (5–10 stigmas per sample) were collected and immediately preserved on ice in a cooler during transport to the laboratory. Upon arrival, the stigmas were weighed, homogenized in PBS (pH 7.4) at a ratio of 1:10 (w/v), and snap-frozen in liquid nitrogen for storage at −80 °C until use. All experimental procedures were strictly performed according to the manufacturer’s instructions provided with the assay kit, with three technical replicates per sample.

The peroxidase activity in Dendrobium stigmas was assessed histochemically using the benzidine-hydrogen peroxide method [33], wherein peroxidase catalysis of H₂O₂ generates hydroxyl radicals that oxidise benzidine into a blue chromogenic product, with concomitant oxygen bubble formation serving as a secondary activity indicator. Five stigmas were randomly sampled at 10:00 AM and immediately transferred to Petri dishes containing freshly prepared reaction solution (1% benzidine: 3% hydrogen peroxide: distilled water = 4:11:22, v/v). Qualitative Assessment Protocol: Development of intense azure coloration in the reaction medium within 2 min indicated high POD activity. Intensive bubbling at the stigma-reagent interface was correlated with enzymatic vigour.

Sample collection, RNA extraction, library construction and sequencing

The unpollinated stigma of Dendrobium cariniferum was selected as the material. Stats were collected during different development phase. All the samples were rapidly frozen in liquid nitrogen and stored at − 80 °C until further analysis.

Total RNA was isolated from Dendrobium cariniferum stigmas using Trizol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s protocol. RNA integrity was verified via an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA), and checked using RNase free agarose gel electrophoresis. After total RNA was extracted, eukaryotic mRNA was enriched by polyA selection using Oligo(dT) beads. To improve transcriptome purity, ribosomal RNA (rRNA) was depleted using the Ribo-Zero™ Magnetic Kit (Epicentre, Madison, WI, USA) following the manufacturer’s protocol for plant samples. Then the enriched mRNA was fragmented into short fragments using fragmentation buffer and reverse transcribed into cDNA with random primers. First-strand cDNA was synthesized from fragmented mRNA using random hexamer primers and SuperScript™ IV Reverse Transcriptase (Invitrogen) at 50 °C for 30 min. Second-strand cDNA was synthesized by DNA polymerase I, RNase H, dNTP and buffer. Then the cDNA fragments were purified with QiaQuick PCR extraction kit (Qiagen, Venlo, The Netherlands), end repaired, A base added, and ligated to Illumina sequencing adapters. The ligation products were size selected by agarose gel electrophoresis, PCR amplified, and sequenced using Illumina novaseq 6000 by Gene Denovo Biotechnology Co. (Guangzhou, China).

Filtering of clean reads and denovo assembly

Reads obtained from the sequencing machines included raw reads containing adapters or low quality bases which would affect the following assembly and analysis. Thus, to get high quality clean reads, reads were further filtered by fastp [34] (version 0.18.0). The parameters were as follows: (1) removing reads containing adapters; (2) removing reads containing more than 10% of unknown nucleotides (N); (3) removing low quality reads containing more than 50% of low quality (Q-value ≤ 20) bases.

Transcriptome denovo assembly was carried out with short reads assembling program– Trinity [35].

Trinity is a modular method and software package which combines three components: Inchworm, Chrysalis and Butterfly. Firstly, Inchworm assembles reads by a greedy kmer based approach, resulting in a collection of linear contigs. Next, Chrysalis clusters related contigs that correspond to portions of alternatively spliced transcripts or otherwise unique portions of paralogous genes, and then builds a de Bruijn graphs for each cluster of related contigs. Finally, Butterfly analyzes the paths taken by reads and read pairings in the context of the corresponding de Bruijn graph, and outputs one linear sequence for each alternatively spliced isoform and transcripts derived from paralogous genes.

Functional annotation and unigene expression analysis

Basic annotation of unigenes includes protein functional annotation, pathway annotation, COG/KOG functional annotation and Gene Ontology (GO) annotation. To annotate the unigenes, we used BLASTx program (http://www.ncbi.nlm.nih.gov/BLAST/) with an E-value threshold of 1e-5 to NCBI non-redundant protein (Nr) database (http://www.ncbi.nlm.nih.gov), the Swiss-Prot protein database (http://www.expasy.ch/sprot), the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.jp/kegg), and the COG/KOG database (http://www.ncbi.nlm.nih.gov/COG). Protein functional annotations could then be obtained according to the best alignment results.

Protein coding sequences of unigenes were aligned via BLASTp to Plant TFdb (http://planttfdb.cbi.pku.edu.cn/) to predicte TF families.

Protein coding sequences of unigenes were aligned by BLASTp to R-Gene database PRGdb (http://prgdb.crg.eu/wiki/Main_Page) to predict R genes. The presence and location of signal peptide cleavage sites in amino acid sequences were predicted by SignalP 4.1 Server (http://www.cbs.dtu.dk/services/SignalP-4.1/). Mucin type GalNAc O-glycosylation sites in mammalian proteins were predicted by NetOGlyc 4.0 Server (https://services.healthtech.dtu.dk/service.php?NetOGlyc-4.0). Arginine and lysine propeptide cleavage sites in eukaryotic protein sequences were predicted by ProP 1.0 Server (http://www.cbs.dtu.dk/services/ProP/).

The unigene expression was calculated and normalized to RPKM (Reads Per kb per Million reads) [36]. The formula is as follows:

Let RPKM to be the expression of Unigene A. C is the number of reads that are uniquely mapped to Unigene A. N is the total number of reads that are uniquely mapped to all unigenes. L is the length (base number) of Unigene A.

Differentially expressed genes (DEGs) between stages were identified using DESeq2 (v1.38.3) with thresholds:|log2(fold change)| ≥ 1 and adjusted p-value (FDR) < 0.05.

Statistical analysis

All experiments were performed in triplicate with three independent biological replicates (n = 3). Physiological assays (enzyme activity) statistical analyses were performed as follows: Differences among developmental stages (CK, T1-T3) were assessed by one-way ANOVA followed by Tukey’s HSD post-hoc test (α = 0.05). Data normality was verified by Shapiro-Wilk test, and homogeneity of variance by Levene’s test. Non-parametric data were analyzed using Kruskal-Wallis test.

Results

Morphological dynamics across stigma developmental stages and identification of the optimal receptivity window

The secretory dynamics of stigmatic cavity mucilage in Dendrobium cariniferum exhibited distinct developmental phase dependency (Fig. 1). At first day post-anthesis (DPA), no secretions were detectable (Secretory Volume Grade 0). Mucilage production commenced at 5 DPA (Fig. 1a), manifesting as translucent gelatinous deposits at the basal cavity (Grade 1), though exhibiting low viscosity (Viscosity Grade 1). By 10 DPA (Fig. 1b), secretions intensified to Grade 3 (51–75% cavity occupancy) with concomitant viscosity elevation (Grade 2), evidenced by sustained pollen adhesion under inversion assays. At 20 DPA (Fig. 1c), a marked reduction in cavity capacity (≤ 30% residual volume, Grade 1) coincided with mucilage solidification and complete loss of adhesiveness (Viscosity Grade 0), as quantified in Table 1. Based on these histochemical profiles, developmental phases were designated as follows: 5, 10, and 20 DPA corresponding to early anthesis (T1), mid-anthesis (T2), and late anthesis (T3) stages, respectively.

Fig. 1.

Changes of stigma exudate with developmental stage of Dendrobium cariniferum. a: stigma morphology at 5 days post-anthesis (DPA); (b): Stigma morphology at 10 DPA; (c): stigma morphology at 20 DPA

Table 1.

Changes of stigma maturity with developmental stage of Dendrobium cariniferum

Stigma Maturity Index		Different Developmental Stages
		CK	T1	T2	T3
stigmatic secretory volume	-		+	+++	+
mucilage viscosity	-		+	++	+

“-” represents Grade 0, “+” represents Grade 1, “++” represents Grade 2, “+++” represents Grade 3. CK: first day post-anthesis (DPA); T1: 5 DPA (Early Anthesis Phase); T2: 10 DPA (Mid-Anthesis Phase); T3: 20 DPA (Late Anthesis Phase)

To investigate the optimal receptivity window at different developmental stages, self-pollination was performed on stigmas across distinct phases. The results revealed that at mid-anthesis phase (10 DPA), pollen tube growth rates exceeded those observed in other developmental stages. Pollen tubes exhibited bright fluorescence without dark zones, accompanied by ovary enlargement, indicating that mid-anthesis stigmas were receptive to compatible pollen. Observations of pollen tube dynamics following pollination on 1 DPA (CK) are illustrated in Fig. 2. No pollen tube germination was observed 24 h post-pollination (Fig. 2a). Limited pollen germination occurred at 48 h (Fig. 2b), with short, newly emerged pollen tubes. By 72 h, partial pollen tube growth was detected (Fig. 2c), though most tubes exhibited aberrant growth patterns, including twisting (Fig. 2d), and failed to form cohesive pollen tube bundles. The pollinated results for stigmas of early anthesis phase (5 DPA) are shown in Fig. 3. Sparse pollen germination was observed 24 h post-pollination, with limited elongation by 48 h. Gradual pollen tube extension was noted at 72 h, though growth rates remained slow, despite the gradual formation of pollen tube bundles. Late anthesis stigmas (20 DPA) displayed pollination outcomes similar to early anthesis stigmas but with further reduced growth rates (Fig. 4). For mid-anthesis phase stigmas (10 DPA), no pollen tube germination was observed within 2–4 h post-pollination (Fig. 5b). Scattered pollen tube emergence from pollen masses was detected between 4 and 16 h (Fig. 5c). By 48 h, directional pollen tube bundles had formed (Fig. 5e). At 72 h, pollen tubes were observed extending into the ovary, with visible sperm nuclei (Fig. 5f). Significant ovary enlargement was evident by 100 h post-pollination (Fig. 5g–i). Unsampled pollinated stigmas successfully developed into swollen capsules.

Fig. 2.

Pollen tube growth dynamics at different time intervals after pollination on the stigmas of the first day post-anthesis. a: Stigma at 24 h after pollination; (b): Stigma at 48 h after pollination; (c): Stigma at 72 h after pollination; (d): Pollen tube morphology at 72 h after pollination. P: Pollen mass. PT: Pollen tube; SN: Sperm nucleus. Red arrows indicate twisted growth of pollen tubes

Fig. 3.

Pollen tube growth dynamics at different time intervals after pollination on the stigmas of early anthesis phase (5 DPA). a: Stigma at 24 h after pollination; (b): Stigma at 48 h after pollination; (c): Stigma at 72 h after pollination; (d): Pollen tube morphology at 100 h after pollination. P: Pollen mass; PT: Pollen tube

Fig. 4.

Pollen tube growth dynamics at different time intervals after pollination on the stigmas of the late anthesis stage (20 DPA). a: Stigma at 24 h after pollination; (b): Stigma at 48 h after pollination; (c): Stigma at 72 h after pollination; (d): Stigma at 100 h after pollination. P: Pollen mass; PT: Pollen tube

Fig. 5.

Pollen tube growth dynamics and ovary developmental at different time intervals after pollination on the stigmas of the mid-anthesis stage (10 DPA). a: Unpollinated stigma; (b): Stigma at 4 h after pollination; (c): Stigma at 16 h after pollination; (d): Stigma at 24 h after pollination; (e): Stigma at 48 h after pollination; (f): Stigma at 72 h after pollination; (g): Unpollinated ovary; (h): Ovary at 120 DPA; (i): Ovary at 190 DPA. ST: stigma; P: pollen mass; PTB: pollen tube bundle; O: ovary. The region outlined by the white box corresponds to the ovary

Changes of enzyme activity in stigma across developmental stages

To evaluate the biochemical basis of stigma maturation, we measured esterase, superoxide dismutase (SOD) and peroxidase (POD) activities in stigmas harvested at three distinct developmental stages: T1 (early anthesis phase, 5 DPA), T2 (mid-anthesis phase, 10DPA), and T3 (late anthesis phase, 20 DPA). These enzymes were selected for their roles in cuticle remodeling (esterase), oxidative stress regulation (SOD and POD), both critical for pollen adhesion and compatibility. The results demonstrated a biphasic pattern, with an initial increase followed by a decrease in enzyme activity (Fig. 6). As the stigma matured, there both esterase and SOD activities increased. The highest levels were observed at T2, which coincided with the optimal pollination period. With stigma senescence, t the activities of esterase and SOD decreased. The stigma POD activity of Dendrobium was identified through the use of a benzidine and hydrogen peroxide method. The peroxidase (POD) activity in stigmas can be visually assessed through staining intensity and the presence of oxygen bubbles generated during the enzymatic reaction. The stigma of the early anthesis phase (T1) exhibited few bubbles, the presence of bubbles in the stigma was evident, accompanied by a notable intensification of color changes (Fig. 7a). By the mid-anthesis anthesis (T2), the stigma was observed to contain many bubbles, accompanied by a pronounced alteration in color. The content of H₂O₂ was also markedly elevated (Fig. 7b). By late anthesis phase (T3), the stigma had become less viscous and had taken on a bluer hue (Fig. 7c). The catalase activity exhibited a fluctuating pattern with respect to the progression of flowering. Initially, it manifested as weak, then strengthened, and subsequently weakened.

Fig. 6.

Changes of enzyme activity during different developmental stage in the stigma of Dendrobium cariniferum. a: esterase activity in the stigmas during different developmental stages; (b): SOD activity in the stigmas during different developmental stages. T1: the early anthesis phase (5 DPA); T2: the mid-anthesis phase (10 DPA); T3: the late anthesis phase (20 DPA)

Fig. 7.

Changes of peroxidase activity during different developmental stage in the stigma of Dendrobium cariniferum

Esterase, SOD and POD activities exhibited stage-dependent dynamics that correlated with mucilage secretion profiles (Fig. 1; Table 1). When the stigma exhibited peak mucilage secretion and highest viscosity, the enzymatic activities (esterase, SOD, and POD) also reached their maximum levels. This pattern parallels the dynamics of stigma receptivity: the mid-flowering phase (T2) represents the optimal pollination window, during which the activities of esterase, superoxide dismutase (SOD), and peroxidase (POD) were all significantly elevated compared to other stages. These findings suggest that esterase and SOD may play a role in pollen-stigma recognition.

Comprehensive evaluation of stigma transcriptome data quality

The secretions and maturity of the stigma significantly influence pollination in Dendrobium orchids. The stigma-pollen recognition mechanism is highly complex, and investigating the mechanisms underlying stigma maturation serves as a prerequisite for understanding pollen-stigma interactions. Based on morphological and physiological studies, four key developmental stages were identified: CK, T1, T2, and T3 (where CK represents the first day of flowering with no visible secretion, and T1-T3 correspond to distinct secretory phases during anthesis: T1 = early anthesis phase, T2 = mid-anthesis phase, and T3 = late anthesis phase). Transcriptomic analysis was subsequently performed across these four stages.

A total of 38,833,171 raw reads and 384,146,047 clean reads were obtained from the four developmental stages of Dendrobium stigma via transcriptome sequencing. The mean number of raw reads for all the tissue samples was 647,219,525, and the Q20 for the original degree segment was 97.95%. Following cleaning, the number of reads was 6,402,434,125, the Q20 was 98.12%, and the GC content was 45.57% (Table 2). The GC percentage is 40.16%, the average length is 751 base pairs. Assembly robustness was evaluated using the following key indicators: N50 Statistic: Unigenes were sorted by length in descending order, and the N50 value represents the shortest contig length at which 50% of the total assembled sequence is covered. A higher N50 with fewer contigs indicates superior assembly continuity. The GC percentage is 40.16%, the average length is 751 base pairs, and the N50 value is 14,493. The N50 length is 1469 base pairs. A total of 105,065 unigenes were obtained from the transcripts. The assembly quality indicators conform to the standards of the plant transcriptome [37], ensuring suitability for high-confidence functional annotation of Dendrobium stigma genes.

Table 2.

Raw data from illumina RNA sequencing

Sample	Clean Reads	BF-Q20(%)	BF-GC Content (%)	Clean Reads	AF-Q20(%)	AF-GC Content (%)
CK1	6,387,869,400	97.91	46.00	6,328,803,411	98.07	45.93
CK2	6,081,212,400	97.91	46.02	5,985,161,195	98.12	45.90
CK3	6,656,101,200	97.88	45.97	6,600,055,873	98.04	45.91
T1	6,846,745,800	98.05	45.29	6,798,261,070	98.19	45.24
T2	6,378,624,900	98.01	45.41	6,311,658,719	98.15	45.32
T3	6,482,617,800	97.95	45.28	6,390,664,484	98.15	45.16

CK indicates the stigma of the first day on which the flowers bloom. T1-T3 correspond to secretion-active phases (early, mid, and late anthesis), while CK1-CK3 represent non-secretory phases at equivalent developmental time points. BF means before the filter, and AF means after the filter. Q20% is the proportion of nucleotides with a quality value greater than 20. The GC percentage is the proportion of guanine and cytosine nucleotides among total nucleotides

Stage-specific differential gene expression and functional enrichment (GO/KEGG) in stigmas

Compared to the control group (CK), the numbers of differentially expressed genes (DEGs) in the T1, T2, and T3 stigmas were 13,209 (8,560 upregulated and 4,649 downregulated), 12,557 (6,330 upregulated and 6,227 downregulated), and 18,593 (10,395 upregulated and 8,198 downregulated), respectively (Fig. 8). Further screening for stage-specific DEGs within the T groups revealed: 3,938 unique DEGs in T1 (2,895 upregulated and 1,043 downregulated); 3,069 unique DEGs in T2 (1,761 upregulated and 1,308 downregulated); 8,785 unique DEGs in T3 (5,695 upregulated and 3,090 downregulated) (Fig. 9). There is a total of 4,202 DEGs among them.

Fig. 8.

Results of DEGs in different treatments. CK represents the first day of flowering with no visible secretion, and T1-T3 correspond to distinct secretory phases during anthesis: T1 = early anthesis phase, T2 = mid-anthesis phase, and T3 = late anthesis phase

Fig. 9.

Venn diagram of differentially expressed genes (DEGs) in stigmas across developmental stages. CK represents the first day of flowering with no visible secretion, and T1-T3 correspond to distinct secretory phases during anthesis: T1 = early anthesis phase, T2 = mid-anthesis phase, and T3 = late anthesis phase

On the basis of the annotation results from the NR database, the different control group unigenes were categorised into GO units. The number of upregulated genes was significantly greater than the number of downregulated genes in the GO units of CK-T1 and CK-T3 (Fig. 10a, c). In contrast, the downregulated genes were more pronounced in the gene ontology annotation of CK-T2 (Fig. 10b). To gain deeper insight into the distinctions between the transcriptomes, we examined the top three GO terms and observed that while the terms were identical in the Biological Process (BP) category, they differed in the Cellular Component (CC) and Molecular Function (MF) categories (Table 3). In CK1-T1, the top three terms within CC were membrane, intrinsic component of the cellular periphery, and outer encapsulated structure. In CK2-T2, the top three terms were “membrane”, “component of membrane”, and “cellular periphery”. In CK3-T3, the top three terms were the cell wall, catalytic complex, and vacuolar part terms, respectively. In CK1-T1, the top three MFs were kinase activity, protein binding, phosphotransferase activity, and the alcohol group as a receptor. In CK2-T2, the top three terms were oxidoreductase activity, substrate-specific transporter activity, and transcription factor activity. In CK3-T3, the top three genes were involved in oxidoreductase activity, kinase activity, and protein binding.

Fig. 10.

Gene ontology annotations of the assembled unigenes in CK1-T1(a), CK2-T2(b) and CK3-T3(c)

Table 3.

Top 3 of gene ontology annotations in different developmental stages

Gene Ontology	Sample	GO term top3
Cellular Component	CK1-T1	membrane	cell periphery	external encapsulating struture
	CK2-T2	membrane	intrinsic component of membrane	cell periphery
	CK3-T3	cell wall	catalytic complex	Vacuolar part
Molecular Function	CK1-T1	kinase activity	protein binding	phosphotransferase activity alcohol group as acceptor
	CK2-T2	oxidoreductase activity	substrate-specific transporter activity	nucleic acid binding transcription factor activity
	CK3-T3	oxidoreductase activity	kinase activity	Protein binding
Biological Process	CK1-T1	single-organism process	single-organism cellular process	response to stimulus
	CK2-T2	single-organism process	single-organism cellular process	response to stimulus
	CK3-T3	single- organism process	single-organism cellular process	response to stimulus

The KEGG enrichment analysis revealed distinct stage-specific pathway activation patterns during stigma development (Fig. 11; Table 4). Notably, the plant-pathogen interaction pathway showed significantly higher enrichment during the mid-anthesis phase (T2) compared to early (T1) and late (T3) stages. This observation may reflect T2’s role as the presumed peak receptivity window, where the dual requirements of pollen recognition (a process sharing mechanistic similarities with immune responses) and microbial defense could collectively drive pathway activation. In parallel, T2 exhibited coordinated upregulation of: Metabolic pathways, potentially supporting energy demands for pollen tube guidance and ROS homeostasis. MAPK signaling, which may mediate oxidative stress adaptation and signal transduction. Ethylene/jasmonate hormone pathways, possibly regulating pollen acceptance while delaying senescence onset. In contrast, early development (T1) was associated with pathways linked to structural establishment (e.g., cell wall modification, auxin/gibberellin-mediated cell expansion), while the late phase (T3) displayed a shift toward catabolic processes and senescence-related pathways (e.g., ABA/ethylene-driven nutrient recycling).

Fig. 11.

KEGG annotations of the assembled unigenes in CK1-T1(a), CK2-T2(b) and CK3-T3(c)

Table 4.

The top 3 enriched KEGG terms in different developmental stages

Sample	TOP 5 of KEGG Enrichment
CK1-T1	Metabolic pathways	Biosynthesis of secondary metabolism	plant hormone signal transduction	plant-pathogen interaction	MAPK signaling pathway-plant
CK2-T2	Metabolic pathways	Biosynthesis of secondary metabolites	plant-pathogen interaction	plant hormone signal transduction	MAPK signaling pathway-plant
CK3-T3	Metabolic pathways	Biosynthesis of secondary metabolites	plant hormone signal transduction	plant-pathogen interaction	MAPK signaling pathway-plant

Integrated analysis of stigma transcriptomes across developmental stages: transcription factors, R-gene database screening, and protein O-GlcNAc glycosylation prediction

The predicted protein sequences were compared with the corresponding transcription factor database (plant TFdb). In the comparative transcriptome analysis between control (CK) and treatment (T) groups, a total of 1,154 transcription factors (TFs) were successfully annotated. The three transcription factor (TF) classification families with the highest number of unigenes sequences were ERF, bHLH, and C2H2 (Fig. 12). The results of the enrichment analysis of the transcription factors revealed that the primary focus was on plant hormone signal transduction, the MAPK signalling pathway in plants, and the plant-pathogen interaction pathway. MYBP, EREBP, HD-ZIP and HSFF were identified as key players. An enrichment analysis of the top three families: revealed that the ERF transcription factor family is involved primarily in plant‒pathogen interactions, the MAPK signalling pathway, and the ribosomal pathway (Fig. 13a). bHLH transcription factors are involved in the MAPK signaling pathway in plants, phytohormone signalling and circadian rhythms (Fig. 13b). C2H2 is also involved in the MAPK signalling pathway in plants, phytohormone signalling and circadian rhythm, whereas C2H2 is involved in Spiceosome and protein processing in the plasma reticulum (Fig. 13c).

Fig. 12.

The TF families in advance annotation of unigenes

Fig. 13.

Pathway enrichment statistics for transcription factor families. (a): ERF; (b): bHLH; (c): C2H2

Transcriptomic analysis revealed distinct stage-specific transcription factor (TF) profiles across treatment groups. In T1, we identified 125 unique TFs (Table 5), with the most abundant families being bHLH (basic helix-loop-helix), GRAS (Gibberellic Acid Insensitive, Repressor of GAI, SCARECROW), and C2H2 zinc finger proteins (Fig. 14a). These TFs showed significant enrichment in plant hormone signal transduction (ko04075), MAPK signaling pathway - plant (ko04016), and plant-pathogen interaction (ko04626) pathways. The T2 group exhibited 46 exclusive TFs (Table 5), predominantly from the NAC (NAM, ATAF1/2, CUC2), MYB (v-myb avian myeloblastosis viral oncogene homolog), and ERF (Ethylene Responsive Factor) families (Fig. 14b), with primary association to plant hormone signaling pathways. Most notably, T3 contained 120 stage-specific TFs (Table 5), dominated by bZIP (basic leucine zipper), ERF, and NAC family members (Fig. 14c). These demonstrated coordinated enrichment across three key regulatory pathways: plant hormone signaling, circadian rhythm - plant (ko04712), and MAPK-mediated signal transduction.

Table 5.

Quantification of stage-specific transcription factors (TFs), pathogenesis-related genes (PRGs), and O-GlcNAc-modified proteins across stigma developmental stages

Treatment	DEGs	uniqueDEGs	TFs	PGR	O-GlcNAc
CK1-T1	13,209	3938	125	67	1268
CK2-T2	12,557	3069	46	46	934
CK3-T3	18,593	3090	120	139	2328

Fig. 14.

Analysis of transcription factor families in stigma transcriptomes across developmental stages. (a): T1 stage (early anthesis phase); (b): T2 stage (peak anthesis phase); (c) T3 stage (late anthesis phase)

The predicted protein sequences were annotated with the corresponding R-Gene database (PRGdb) API. In the comparative transcriptome analysis between control (CK) and treatment (T) groups, the results demonstrated that the top five resistance gene (R-gene) classes were Protein Kinase (KIN), Receptor-Like Kinase (RLK), Nucleotide-Binding Site Leucine-Rich Repeat (N), Receptor-Like Protein (RLP), and Leucine-Rich Repeat (L) (Fig. 15). The KEGG enrichment analysis of the KIN family indicated that the unigenes were predominantly associated with plant-pathogen interactions, the mitogen-activated protein kinase (MAPK) signalling pathway and phytohormone signalling.

Fig. 15.

The PRG class in advance annotation of unigenes

Analysis of stage-specific differentially expressed genes revealed distinct patterns across developmental stages. In the T1-exclusive gene set, we identified 130 pathogenesis-related genes (PRGs), including 67 kinases (KINs) (Fig. 16a; Table 5), with predominant enrichment in the plant-pathogen interaction (KEGG pathway ko04626) and MAPK signalling pathways (ko04016). The T2-specific analysis yielded 46 PRGs (20 KINs) (Fig. 16b; Table 5), showing primary association with plant hormone signal transduction (ko04075) and MAPK signalling. Notably, the T3 stage contained 139 PRGs (64 KINs) (Fig. 16c; Table 5), with coordinated enrichment across three key pathways: plant-pathogen interaction, plant hormone signalling, and MAPK cascade regulation.

Fig. 16.

Analysis of pathogenesis-related genes (PRGs) among stage-specific differentially expressed genes (DEGs) during stigma development

Protein O-GlcNAc glycosylation plays a significant role in cellular processes such as growth, development and the onset and progression of disease. In the comparative transcriptome analysis between control (CK) and treatment (T) groups, the O-GlcNAc glycosylation site of 44,404 unigenes was predicted, and this site was identified in 19,755 (44.5%) of the unigenes. A KEGG enrichment analysis was conducted on the unigenes with more than 10 locus points, revealing a concentration in the ribosome, plant hormone signal transduction, and RNA transport pathways.

Analysis of stage-specific differentially expressed genes with predicted glycosylation sites revealed distinct pathway enrichments: In the T1-exclusive gene set, we identified 1,268 glycosylation sites (Table 5), predominantly enriched in metabolic pathways (ko01100), ribosome biogenesis (ko03010), and plant-pathogen interaction (ko04626). The T2-specific analysis predicted 934 glycosylation sites (Table 5), with primary associations to endocytosis (ko04144), photosynthesis (ko00195), and metabolic pathways. Most notably, the T3 stage exhibited 2,328 (Table 5) predicted glycosylation sites, showing coordinated enrichment in ribosomal pathways, plant hormone signal transduction (ko04075), and metabolic regulation.

SOD- and esterase-encoding DEGs link to stigma enzyme activity

A search for unigenes revealed the presence of two distinct SOD types, namely CuZn-SOD (SOD1) and Mn-SOD (SOD2). Additionally, unigenes regulating the expression of SOD were identified. The transcriptome of CK1-T1 revealed the presence of three unigenes encoding SOD, while CK2-T2 presented six unigenes, and CK3-T3 presented four unigenes (Table 6). In total, eight genes encoding SOD genes were identified throughout development, and the expression of these unigenes was upregulated. The measurements of SOD activity revealed the highest levels at T2, and the transcriptome results were in accordance with the activity measurements, indicating that the six genes encoding SOD were upregulated at T2 to achieve maximal activity.

Table 6.

SOD-encoding differentially expressed genes (DEGs) across stigma developmental stages (log2 fold change)

Sample	DEGs
CK1-T1	Unigenes0101660 (1.2), Unigenes0015254 (1.3), Unigenes0093808 (1.3)
CK2-T2	Unigenes 0003334 (1.3), Unigenes0015254 (1.3), Unigenes0101660 (1.7), Unigenes0093808 (1.1), Unigenes0114401 (2.2), Unigenes0070192 (4.0)
CK3-T3	Unigenes0093808 (1.1), Unigenes0101660 (1.1), Unigenes0104401 (1.1), Unigenes0070192 (1.4)

A search for unigenes with esterase activity identified 29 enzymes with esterase activity (Table 7), which are involved in 27 different pathways. The primary focus was on energy metabolism pathways, including glycoysis/gluconeogenesis, pentose and glucronate interconversions, starch and sucrose metabolism, and fatty acid biosynthesis. Additionally, amino acid metabolism pathways, such as valine, leucine and isoleucine degradation, and histidine metabolism, are included. The upregulation of pectinesterase expression in the pentose and glucuronate interconversions pathway is essential for the promotion of the conversion of 1,4-alpha-D-galacturonide to D-Galacturonate. Its activity increased in the physiological analysis, and its expression was significant at the molecular level. Additionally, we identified enolase-phosphatase E1 with esterase activity, which may affects ethylene synthesis and, consequently, phytohormone signalling.

Table 7.

DEGs involved in the regulation of esterase expression during stigma development

esterases	DEGs ID	pathway
fructose-1,6-bisphosphatase I	Unigene0006614, Unigene0016513, Unigene0077868, Unigene0077868	glycoysis/gluconeogenesis pentose phosphate pathway carbon fixation in photosynthetic organisms
3‘(2’), 5’-bisphosphate nucleotidase	Unigene0004748, Unigene0061171	sulfur metabolism
3-hydroxyisobutyryl-CoA hydrolase	Unigene0000761, Unigene0019882, Unigene0021198, Unigene0027635, Unigene0027636, Unigene0046801 Unigene0046802, Unigene0053578, Unigene0053579, Unigene0053899, Unigene0065687, Unigene0068058 Unigene0086334, Unigene0096993	valine, leucine and isoleucine degradation β-alanine metabolism propanoate matabolism
5’-nucleotidase	Unigene0003416, Unigene0005043, Unigene0013574, Unigene0013859, Unigene0013861, Unigene0015852 Unigene0053926, Unigene0053927, Unigene0053928, Unigene0058501	purine metabolism pyrimidine metabolism nicotinate and nicotinamide metabolism
6-phosphogluconolactonase	Unigene0098261, Unigene0038027, Unigene0042088, Unigene0045406, Unigene0098261	pentose phosphate pathway
acyl-coenzyme A thioesterase 1/2/4	Unigene0098261, Unigene0038027, Unigene0042088, Unigene0045406, Unigene0098261	fatty acid elongation biosynthesis of unsaturated fatty acids
acylglycerol lipase	Unigene0005442, Unigene0012804, Unigene0013729 Unigene0015321, Unigene0015710, Unigene0023728, Unigene0044274, Unigene0044607, Unigene0046313 Unigene0048744, Unigene0053253, Unigene0053995, Unigene0060816, Unigene0064268, Unigene0069712 Unigene0075625, Unigene0075626, Unigene0076968, Unigene0087277, Unigene0088531, Unigene0091968 Unigene0100927, Unigene0100928	glycerolipid metabolism
chlorophyllase	Unigene0019192, Unigene0019196	porphyrin and chlorophyll metabolism
enolase-phosphatase E1	Unigene0049056, Unigene0059253, Unigene0059254, Unigene0086070	cysteine and methionine metabolism
fatty acyl-ACP thioesterase B	Unigene0011849, Unigene0037897, Unigene0063298, Unigene0081419	fatty acid biosynthesis
FMN hydrolase/5-amino-6-(5-phospho-D-ribitylamino) uracil phosphatase	Unigene0005992, Unigene0005993	riboflavin metabolism
fructose-1,6-bisphosphatase sedoheptulose-1,7-bisphosphatase	Unigene0006614, Unigene0010965, Unigene0016513, Unigene0077868, Unigene0104715	carbon fixation in photosynthetic organisms
imidazoleglycerol-phosphate dehydratase/histidinol-phosphatase	Unigene0001667, Unigene0075055	histidine metabolism
IMP and pyridine-specific 5’-nucleotidase	Unigene0053926, Unigene0053927, Unigene0053928, Unigene0058501	nicotinate and nicotinamide metabolism
inositol polyphosphate 1-phosphatase	Unigene0061171	glycerolipid metabolism phosphatid ylinositol signaling system
inositol-phosphate phosphatase/L-galactose 1-phosphate phosphatase	Unigene0067829, Unigene0001131	ascorbate and aldarate metabolism
lysophospholipase II	Unigene0007363, Unigene0044874, Unigene0067001, Unigene0073239	clycerophospholipid metabolism
medium-chain acyl-[acyl-carrier-protein] hydrolase	Unigene0087463	fatty acid biosynthesis
myo-inositol-1(or 4)-monophosphatase	Unigene0102684	glycerolipid metabolism phosphatid ylinositol signaling system
palmitoyl-protein thioesterase	Unigene0020521, Unigene0035089, Unigene0102175	fatty acid elongation
pectinesterase	Unigene0099473, Unigene0038027, Unigene0042088 Unigene0045406, Unigene0098261	pentose and glucronate interconversions
phosphatidate phosphatase	Unigene0072613	glycerolipid metabolism clycerophospholipid metabolism
phosphatidylinositol 3,5-bisphosphate 5-phosphatase	Unigene0034509, Unigene0034511, Unigene0034512 Unigene0034513, Unigene0042065, Unigene0042066 Unigene0047357, Unigene0060221, Unigene0103682	glycerolipid metabolism
phosphatidylinositol-3,4,5-trisphosphate 3-phosphatase and dual-specificity protein phosphatase PTEN	Unigene0049721, Unigene0049723, Unigene0051831 Unigene0104613	glycerolipid metabolism
phospholipase C	Unigene0005543, Unigene0015002, Unigene0018393 Unigene0030469, Unigene0030472, Unigene0034384 Unigene0036654, Unigene0055708, Unigene0091481 Unigene0099990	glycerolipid metabolism clycerophospholipid metabolism ether lipid metabolism
phospholipase D1/2	Unigene0009024, Unigene0024444, Unigene0027846 Unigene0033502, Unigene0033503, Unigene0038823, Unigene0039086, Unigene0050501, Unigene0059317 Unigene0067440, Unigene0090324, Unigene0103313 Unigene0103314, Unigene0103812	ether lipid metabolism
ribosome biogenesis GTPase/thiamine phosphate phosphatase	Unigene0034742	thiamine metabolism
secretory phospholipase A2	Unigene0091551	ether lipid metabolism ara chidonic acid metabolism linoleic acid metabolism α-linolenic acid metabolism
sucrose-6-phosphatase	Unigene0016569	starch and sucrose metabolism

Screening of T1-exclusive differentially expressed genes (DEGs) for esterase activity potential identified eight putative esterase-encoding genes. This number increased to seventeen in T2, while only one such gene was detected in T3 (Table 8). These molecular findings show complete concordance with the observed patterns of esterase activity fluctuations in corresponding physiological assays.

Table 8.

Stage-specific differentially expressed genes (DEGs) encoding esterase activity across stigma developmental phases

Sample	DEGs
CK1-T1	Unigene0015002, Unigene0015852, Unigene0042066, Unigene0042088 Unigene0053578, Unigene0053926, Unigene0053928, Unigene0055708
CK2-T2	Unigene0001667, Unigene0005543, Unigene0013402, Unigene0013574 Unigene0018393, Unigene0030472, Unigene0034384, Unigene0034509 Unigene0034512, Unigene0034513, Unigene0038027, Unigene0046801 Unigene0053899, Unigene0059253, Unigene0075626, Unigene0081419 Unigene0096993
CK3-T3	Unigene0046313

Stage-specific dynamics of defense and hormonal signalling pathways during stigma maturation

Our findings indicate that the unigenes of the KIN and TF families are associated primarily with the same pathway, suggesting the potential for a functional link between them. Accordingly, the three pathways involved were subjected to further analysis.

Two key factors related to plant‒pathogen interactions have been identified. The key factors are CNGCs and FLS2. Ca²⁺ entry into the cytoplasm is controlled by CNGCs, which in turn upregulate CDPK and Rboh expression, resulting in ROS accumulation and the subsequent triggering of a hypersensitive response (HR). FLS2 upregulation induces the expression of defense-related genes by upregulating the expression of MKK1/2, MKK4/5, MPK3/6 and other kinases and by upregulating the expression of the Pti1 transcription factor, which acts on DNA to trigger HR (Fig. 17). Additionally, FLS2 plays a pivotal role in the MAPK signalling pathway. In the event of pathogen infection, the upregulation of FLS2 induces the phosphorylation of various kinases (e.g., MPK1/2, MPK3/6, etc.) that act on DNA, thereby triggering an early defense response against the pathogen. It can therefore be hypothesised that FLS2 plays an important role in the defense of the plants.

Fig. 17.

Regulatory model of unigenes involved in plant-pathogen interaction pathways in the intermediate stage of Dendrobium stigma development after flowering. Note: Red Gradient: Upregulated genes (log2FC ≥ 1, FDR < 0.05), Intensity scale: Light red (1 ≤ log2FC < 6) → Dark crimson (log2FC ≥ 6); Green Gradient: Downregulated genes (log2FC ≤ −1, FDR < 0.05), Intensity scale: Pale green (−1 ≥ log2FC > −6), Deep forest green (log2FC ≤ −6). Solid Arrows: Experimentally validated protein interactions; Dashed Arrows: Co-expression predicted relationships

The plant hormone signal transduction pathway involves a number of different hormones, including auxin, cytokinin, gibberellin, abscisic acid (ABA), ethylene, brassinosteroid (BR), jasmonic acid (JA) and salicylic acid. Among these genes, gibberellin induces the upregulation of DELLA through the upregulation of GID1 expression and the upregulation of the transcription factor phytochrome-interacting factor 4 (PIF4), which is encoded by four unigenes and ultimately induces stem growth and germination (Fig. 18a). Additionally, other hormone signalling pathways are regulated by transcription factors, including those encoding ethylene and JA (Fig. 18b, c). In the ethylene signalling pathway, the expression of differentially expressed genes encoding MPK6 and ERF1/2 is significantly upregulated. The transcription factor ERF1/2 was encoded by three genes and acts on DNA to trigger senescence. In the JA signalling pathway, the expression of the JA receptor (JAR), the COI1 gene, the JAZ family of transcription factors, and the MYC2 transcription factor was upregulated. The MYC2 transcription factor was regulated by four DEGs (Fig. 18a). This process is regulated by four unigenes. The majority of the unigenes were upregulated during the course of hormone signalling, leading to the hypothesis that the highest hormone content would be observed at the intermediate stage after flowering.

Fig. 18.

Regulatory model involved in plant hormone signalling transduction pathways in the intermediate stage of Dendrobium stigma development after flowering. Note: Red Gradient: Upregulated genes (log2FC ≥ 1, FDR < 0.05), Intensity scale: Light red (1 ≤ log2FC < 6) → Dark crimson (log2FC ≥ 6); Green Gradient: Downregulated genes (log2FC ≤ −1, FDR < 0.05), Intensity scale: Pale green (−1 ≥ log2FC > −6), Deep forest green (log2FC ≤ −6). Solid Arrows: Experimentally validated protein interactions; Dashed Arrows: Co-expression predicted relationships

The MAPK signalling pathway is induced by a range of biotic and abiotic stimuli, including pathogen attack, wounding and ozone exposure. Additionally, it is associated with hormonal responses (Fig. 19). The serine/threonine-protein kinase OXI1 has been identified as a crucial component of the oxidative burst-mediated signalling pathway in H₂O₂ upstream signalling, and plays a pivotal role in maintaining the equilibrium of reactive oxygen species. The expression of OXI1 decreased at T2, whereas the expression of MKK4/5 and MKK3/6 increased (Fig. 19a). Additionally, the transcription factors ERF1 and MYC2 were identified as regulators of winging responses in ethylene and JA transduction in hormone responses (Fig. 19b). Additionally, ABA was observed to upregulate CAT1 expression in conjunction with H₂O₂ (Fig. 19c). Furthermore, ROS were found to affect ethylene synthesis through MPK3/6 upregulation (Fig. 19d).

Fig. 19.

Regulatory model of unigenes involved in MAPK signalling pathways of T2. Note: Red Gradient: Upregulated genes (log2FC ≥ 1, FDR < 0.05), Intensity scale: Light red (1 ≤ log2FC < 6) → Dark crimson (log2FC ≥ 6); Green Gradient: Downregulated genes (log2FC ≤ −1, FDR < 0.05), Intensity scale: Pale green (−1 ≥ log2FC > −6), Deep forest green (log2FC ≤ −6). Solid Arrows: Experimentally validated protein interactions; Dashed Arrows: Co-expression predicted relationships

The KIN class, TFs and ROS are involved in stigma maturity

Our findings indicate that ROS play a pivotal role in the signalling processes associated with in the aforementioned pathways. Consequently, we conducted further analysis to ascertain whether there is a correlation between ROS and the KIN class and TF. Our results revealed that there is a mutual regulatory relationship among the three (Fig. 20a). ROS affect ethylene synthesis through MPK3/6. Ethylene, in turn, acts as an upstream signalling molecule, inducing the phosphorylation of CTR1 to form MKK9. This protein then continues to phosphorylate to form MPK3/9, which upregulates the expression of the transcription factor ERF1 in the DNA, thereby triggering the production of wounding responses. Furthermore, JA induces the phosphorylation of MKK3 and MPK6, resulting in the upregulation of the transcription factor MYC2, leading to wounding responses. The wounding response has been shown to upregulate the Ca²⁺ signalling pathway, which results in Ca²⁺ entry into the cytoplasm. This, in turn, causes CaM4 expression to be upregulated, which continues to upregulate MPK8 to act on DNA, causing RbohD to be upregulated and ROS to be generated via PAMP-triggered immunity (Fig. 20a). ROS are generated as a consequence of PAMP-triggered immunity. H₂O₂ produced by ROS induces the upregulation of SOD and CAT expression in the peroxisome, thereby preventing H₂O₂ production and subsequent cell death.

Fig. 20.

Hypothetical regulatory interactions among KIN-class unigenes, transcription factors (TFs), and reactive oxygen species (ROS) pathways in T2 (a) and T3 (b). Note: Red Gradient: Upregulated genes (log2FC ≥ 1, FDR < 0.05), Intensity scale: Light red (1 ≤ log2FC < 6) → Dark crimson (log2FC ≥ 6); Green Gradient: Downregulated genes (log2FC ≤ −1, FDR < 0.05), Intensity scale: Pale green (−1 ≥ log2FC > −6), Deep forest green (log2FC ≤ −6). Blue background fills: Kinase-encoding genes; Yellow background fills: Transcription factors (TFs) with DNA-binding domains. Solid Arrows: Experimentally validated protein interactions; Dashed Arrows: Co-expression predicted relationships

Further transcriptome analysis of the T3 senescence stage revealed that most of the DEGs in T2 were not expressed in T3. Kinases that were upregulated in T2, as well as transcription factors, were downregulated (Fig. 20b). For example, the kinase MPK6, which is important for both JA signalling and H₂O₂ and ROS transduction, was regulated by an upregulated differential gene in T2, whereas no expression was detected in T3. pti1 and pti5 were also not expressed. The expression of the differentially expressed genes involved in MYC2 and ERF1 was also reduced. We therefore speculate that the transcription factors MYC2, ERF1, Pti1/5 and the kinase MPK6 may play critical roles in pollen recognition.

Discussion

Pathways involved in stigma development and pollen recognition

From the first day of Dendrobium flower opening to the optimum pollination time of flower maturity and finally, flower extinction, the stigmas underwent various changes. In this study, the main pathways enriched with unigenes were carbon metabolism, secondary metabolism, RNA transport, plant hormone signalling, protein translation, endocytosis, and plant-pathogen interactions in the stigma development stage. These pathways play important roles in the maturation of stigma. In studies on the transcriptome of pear flower style development, unigenes were found to be involved mainly in the pathways of material and energy metabolism, plant hormone signalling, protein translation and stress resistance [38]. The results revealed that the stigma development of different species followed similar rules. One of the pathways most worthy of our attention is the pathways in which TFs and the KIN class are coenriched: plant hormone signal transduction, the MAPK signalling pathway-plant interaction and plant-pathogen interaction. A study of stigmas after pollination of Rhododendron revealed that many DEGs were enriched in several biological pathways, such as plant hormone signal transduction, flavonoid–compound biosynthesis, the MAPK signalling pathway and the plant‒pathogen interaction pathway [39]. In addition, other studies on the pollination compatibility of plants have shown that differentially expressed genes are enriched mainly in these pathways [40, 41]. These three pathways are significantly correlated with stigma-pollen recognition in this study, and the increased expression of related genes indicates that stigmas are preparing for pollination.

DEGs involved in the plant hormone signalling pathway

As regulatory substances, plant hormones play a very important role in the development of plant organs, apical dominance and tissue differentiation [42], and are also closely related to stigma development and cross compatibility [43–47]. Studies have shown that ABA may increase stigma resistance [48]. In this study, the transcription factor MYC2 of JA signalling pathway, which is related to plant defense, was identified [49]. OPCL1, ACX, MFP2 and other differential genes related to JA synthesis were upregulated, so that JA, the upstream signalling molecule in plant hormone signalling transduction, regulated the transcription factor MYC2 and induced the senescence stress response in plant. This process may be related to pollen recognition. Through analysis of the ethylene signal transduction pathway, we found that ethylene regulates the upregulation of ETR, MPK6, EIN3 and other differentially expressed genes, resulting in ubiquitin mediated proteolysis. Ubiquitin- mediated proteolysis is related to defense [50, 51], and it is one of the regulatory pathways with the highest correlation with the plant SI response. The SCF complex in this pathway plays an important role in the S-RNASSI system [52], which is important in inducing programmed cell death. It is speculated that one of the effects of plant hormone accumulation during stigma development in this study is to induce programmed death (PCD) to dissolve incompatible pollen tubes for subsequent pollination. It has been proven that plant hormones are related to PCD. For example, the balance of ABA and ethylene during endosperm development allows PCD to occur normally [53]. In studies on self-pollination, increased ABA and JA expression was shown to induce pollen tube PCD, which led to self-incompatibility [54].

DEGs involved in the MAPK signalling pathway

The synthesis and signal transduction of plant hormones (such as ethylene, JA, and ABA) involve the MAPK signalling pathway in plants [55]. The MAPK signalling pathway is crucial for plant growth and development. In this study, the main focus was on the signal transduction of H₂O₂, ethylene, JA, and ABA. Ethylene acts as an upstream signal, leading to the phosphorylation of CTR1, MKK9, and MPK3/6, as well as the upregulation of ERF1, which triggers defense responses. The MAPK signalling pathway also significantly impacts plant reproductive systems, such as the growth and development of floral organs such as ovules [56]. Additionally, key factors in the plant route of the MAPK signalling pathway, such as mitogen-activated protein kinases (MAPKs), enable Arabidopsis stigmas to accept compatible pollen [57]; these factors can also promote Arabidopsis pollen hydration and germination through exocytosis [58]. Reduced stigma receptivity may be due to a decrease in MAPK levels, which inhibits the MAPK–Exo70A1 interaction. Studies indicate that MAPK signalling is associated with self-incompatibility [54], possibly by inducing plant hormone biosynthesis leading to SI. Central serine/threonine protein kinases integrate signals from different hormones, resulting in corolla shortening to affect pollination [59].

DEGs involved in plant‒pathogen interactions

The entry of foreign pollen into the stigma triggers plant defense mechanisms, which have been found to share similarities with plant‒pathogen interactions [11, 60]. Numerous studies have demonstrated that pollinated stigmas are enriched in pathways related to stress and defense responses [61]. During plant‒pathogen interactions, two immune strategies can reduce pathogen damage: pattern-triggered immunity (PTI), triggered by pathogen-associated molecular patterns, and effector-triggered immunity (ETI) [62]. Both PTI and ETI involve similar reactions, such as the accumulation of ROS, the activation of mitogen-activated protein kinase (MAPK) signalling cascades, and the expression of defense genes. Compatible pollen entering the stigma induces PTI, during which the pollen tube grows normally, similar to a pathogen attack [63]. Kodera et al. suggested that the regulatory mechanism of the PTI pathway is complex, and potentially participates in pollen tube acceptance while simultaneously preventing the invasion of foreign pathogens into the plant body [64]. Research has shown that increased expression of FLS, a gene associated with pathogen defense, inhibits growth, similar to the defense mechanisms observed in SI responses [65]. In this study, we discovered that FLS2, which acts as a key factor in plant‒pathogen interactions, induces the expression of transcription factors that upregulate PTI and lead to defense-related gene induction.

Stage-specific coordination of KIN family transcription factors with plant-pathogen, MAPK, and hormone pathways during stigma maturation

Three pathways—plant hormone signal transduction, the MAPK signalling pathway-plant, and plant‒pathogen interactions—are involved in pollen‒stigma recognition. In this study, these three pathways were identified through the enrichment of transcription factors and KIN class genes. Therefore, the transcription factors and KIN class involved are likely to have a significant effect on regulating stigma recognition. Many transcription factors play regulatory roles during reproductive development. bHLH and MYB play crucial roles in the development of Papaya pistils [42, 59]. Similarly, studies on Liriodendron flower development, have demonstrated that bHLH and MYB influence anther development and female fertility by controlling the biosynthesis of flavonoids and lignin [11, 66]. Research has reported that bHLHs and MYBs have specific effects on the sterility of Japanese apricot pistils [60, 67]. Studies of pollen transcriptomes revealed that the transcription factor AtMYB120 plays a crucial role in sperm release [68], and that the protein encoded by AtMYB120 contains a domain with predicted kinase activity, which is essential for downstream kinase receptor functions. The transcription factor MYB30 can also trigger HR for plant defense [69]. In this study, various kinases such as pyruvate kinase, fructose kinase, BR-signalling kinase, and LRR receptor-like serine kinase, were found to play important roles in the development of Dendrobium stigma. Calcium-dependent protein kinase (CPK), protein-serine/threonine kinase, and 5’-AMP-activated protein kinase are important in the plant hormone signal transduction, the MAPK signalling pathway-plant, and plant‒pathogen interaction pathways, suggesting their critical roles in pollen‒stigma recognition. Other studies have confirmed the importance of kinases in pollen recognition: for example, receptor-like kinases play significant roles in pollen‒pistil interactions [70, 71]. Some kinases within the receptor-like cytoplasmic protein kinase family (RLCK) have also been proven to be essential for sexual reproduction [64].

The KIN class, TFs and ROS are involved in stigma maturity

In this study, H₂O₂ levels were highest during mid-flowering, and ROS appeared in pathways enriched with both the TF and KIN classes. H₂O₂ levels are often used as a reference for optimal pollination periods on the stigma. Coinciding their peak, the SOD enzyme activity was also the highest. In this study, it is hypothesised that during plant-pathogen interactions, Ca²⁺ enters the cytoplasm through CNGCs, leading to CDPK phosphorylation, the upregulation of Rboh expression, increased ROS production, the activation of SOD enzymes in peroxisomes, and the catalysis of excess ROS, with H₂O₂ participating in MAPK signalling. ROS have complex relationships with various kinases: ROS production cannot occur without these kinases [72], such as the receptor-like kinase FER; kinases are also regulated upstream by ROS [73]. Mutations in KINβγ, a component of SNF1-related protein kinases, result in reduced ROS levels in Arabidopsis pollen grains, ultimately causing failed hydration of the stigma [74]. This finding further demonstrates the connection between ROS and proteins with kinase activity. Currently, there are no direct reports linking TFs and ROS. However, in studies related to almond freezing tolerance, transcription factors (such as AP2/ERF, NAC, WRKY, and BZIP TFs) primarily increase the activity of antioxidant enzymes such as SOD to mediate ROS, and it is speculated that the bHLH13 and bHLH35 transcription factors regulate the involvement of SOD in JA signalling [75]. Transcriptomic data indicated upregulated SOD expression during peak stigma receptivity (T2), correlating with reduced H₂O₂ accumulation. While these trends suggest a potential role for SODs in ROS homeostasis, direct evidence of cellular protection was not assessed. Furthermore, the mechanistic interplay between SODs, ROS, and upstream regulators (kinases/TFs) remains to be elucidated.

Unigenes involved in glycosylation

In this study, O-GlcNAc glycosylation was detected in the stigma transcriptome of Dendrobium for the first time, and the pathway was enriched in ribosome, plant hormone signal transduction and splicesome. Glycosylation is a diverse posttranslational modification of proteins, and O-GlcNAc is a unique form of glycosylation. Glycosylation is a key cellular mechanism that regulates a variety of physiological and pathological functions [76]. In plants, glycosylation is related to protein folding and degradation, reacts to environmental stress [77], and plays an important role in many important pathways such as cell signalling and the immune response [78, 79]. Many key signalling pathways (e.g., PI3K/mTOR) and transcription factors (e.g., p53) are modified by O-GlcNAc, and site-specific O-GlcNAc glycosylation plays a unique molecular role in protein stability and function [80]. Notably, there is a close relationship between glycoylation and various kinases. Studies have shown that advanced glycoylation end products can activate mitogen-activated protein kinase (MAPK-), Janus kinase (JAC-), and mitogen-activated protein kinase/extracellular signalling to regulate the kinase MAPK/ERK signalling pathway [81].

While this study provides comprehensive transcriptional insights into stigma development, several limitations should be acknowledged. The absence of direct comparisons between pollinated and unpollinated stigmas restricts our ability to delineate developmental programmes from pollination-responsive mechanisms. Future time-course experiments with controlled pollination would resolve this temporal dynamic. Although we identified stage-specific candidate genes (e.g., FER-like kinases, ROS-related enzymes) through comparative transcriptomics, their precise roles in pollen-stigma recognition remain hypothetical. Functional validation through targeted genetic manipulation and in vitro binding assays are required to establish causal relationships. These limitations, however, define clear avenues for future research while not invalidating the foundational datasets and hypotheses generated herein.

Conclusion

This study provides the first transcriptomic characterization of Dendrobium stigma development, revealing dynamic gene expression patterns across three developmental stages (early, mid-, and late anthesis). Comparative analysis identified 13,209, 12,557, and 18,593 differentially expressed genes (DEGs) at early, middle, and late anthesis, respectively, with 3,938, 3,069, and 8,785 stage-exclusive DEGs. These results suggest phase-specific transcriptional regulation underlying stigma maturation.

KEGG enrichment analysis indicated that DEGs were predominantly associated with metabolic pathways, biosynthesis of secondary metabolites, ribosome biogenesis, and carbon metabolism, which may collectively support stigma development. Notably, transcription factors (TFs) from the ERF, bHLH, and C2H2 families were highly represented, and kinases such as pyruvate kinase and LRR receptor-like serine kinase showed stage-specific expression patterns, suggesting their potential roles in stigma maturation.

Co-enrichment of TFs and KIN-class genes in plant hormone signaling, MAPK cascades, and plant-pathogen interaction pathways implies a possible link between these pathways and pollen-stigma recognition. Reactive oxygen species (ROS) homeostasis emerged as a prominent feature, potentially mediating crosstalk between developmental and stress-response processes. Additionally, the detection of O-GlcNAc glycosylation in the stigma transcriptome, alongside its association with kinase activity, offers a novel avenue for exploring post-translational regulation in orchid reproduction.

While candidate TFs and kinases identified here require functional validation, and pollination-stage comparisons are needed to clarify recognition mechanisms, this study establishes the first transcriptomic framework for Dendrobium stigma development. Our findings highlight potential regulators for improving orchid hybridization, guiding future research on reproductive biology.

Abbreviations

BP

Biological Process

CC

Cellular Component

MF

Molecular Function

ROS

Reactive oxygen species

SOD

Superoxide dismutase

MAPK

Mitogen-activated protein kinase

HR

hypersensitive response

ABA

Abscisic acid

BR

Ethylene, brassinosteroid

JA

Jasmonic acid

PIF4

Phytochrome-interacting factor 4

PCD

Programmed death

HR

Hypersensitive response

DEGs

Differentially expressed genes

Authors’ contributions

RSX, RYH, HJX and LY conducted fluorescence observations of pollen tube growth, enzyme activity assays, and assessments of stigma receptivity. WQ analysed the transcriptome data and was a major contributor to the writing of the manuscript. All the authors read and approved the final manuscript.

Funding

This work was supported solely by institutional resources of Southwest Forestry University. No external funding was received.

Data availability

Raw data have been deposited to National Center for Biotechnology Information (NCBI) under the BioProject number PRJNA1232678.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.