Integrated cytological and transcriptomic analyses reveal the molecular basis of spur variation in Impatiens uliginosa

Integrated cytological and transcriptomic analyses reveal the molecular basis of spur variation in Impatiens uliginosa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrated cytological and transcriptomic analyses reveal the molecular basis of spur variation in Impatiens uliginosa Yang Liu, Xiaoli Zhang, Bo Hui, Luqiu Zhao, Hongyao Liao, Zhijia Gu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8829929/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 12 You are reading this latest preprint version Abstract Background The spur, a key morphological structure in Impatiens uliginosa , plays important functional roles in attracting specific pollinators, enhancing pollination efficiency, and facilitating interspecific reproductive isolation. While its ecological and evolutionary significance has been extensively studied, the cytological and molecular mechanisms underlying spur morphogenesis, particularly in non-model plants, remain poorly understood. In this study, we used wild type (WT) bearing a single spur, two-spur mutant (2SM) and three-spur mutant (3SM) plants of I. uliginosa . By integrating morphometric analysis, cytological examination, and transcriptome sequencing, we identified candidate genes and hormonal regulatory networks associated with spur variation at the molecular level for the first time. This study provides new insights into the molecular basis of spur formation in I. uliginosa and the genus Impatiens more broadly. Results The developmental dynamics of spurs was analyzed both in WT and mutant I. uliginosa , indicating that spur growth follows a typical sigmoidal curve, with lateral spur being significantly shorter than the main spur. The cellular development mechanisms were similar between the two spur types: initial spur formation was predominantly driven by cell division, whereas subsequent elongation primarily depended on cell expansion, with the formation of internal cellular protrusions regulated by anisotropic cell growth. Transcriptome sequencing of spurs at the early stage yielded 32.69 Gb of high-quality data, from which 42,721 unigenes were assembled. Functional annotation against the NR, Swiss-Prot, Pfam, COG, GO, and KEGG databases resulted in the annotation of 24,031 genes. Differential expression analysis identified 8,592 differentially expressed genes (DEGs), which were enriched in 371 GO terms and 111 KEGG pathways. Notably, the “plant hormone signal transduction” pathway showed the highest enrichment in the mutant spurs. A total of 955 transcription factors (TFs) belonging to 34 families, including MYB, AP2/ERF, and TCP, were identified. Through screening and qRT-PCR validation, eight of the 10 candidate genes, such as LTP , GRP5 , PDF1 , and GID2 , were confirmed to be involved in spur variation. Conclusions Our study elucidates the morphological and cellular developmental mechanisms of the spurs in WT and mutant I. uliginosa , and identifies a series of candidate genes associated with spur variation, including cell cycle, cell division, cell elongation, and plant hormones. The findings provide valuable data and resources for further unraveling the molecular mechanisms underlying spur variation in Impatiens species. Impatiens uliginosa Variant spur Spur development Spur morphology Transcriptome Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Introduction The spur—a tubular, saccate, or similarly shaped extension formed by the backward or lateral elongation of petals or sepals—is an evolutionary innovation that serves as an important taxonomic trait in plants. It is widespread across multiple families, such as Papaveraceae (e.g., Corydalis , Fumaria ), Ranunculaceae (e.g., Consolida , Delphinium , Aquilegia , Urophysa ), Balsaminaceae ( Impatiens ), Violaceae (e.g., Afrohybanthus , Viola ), and Orchidaceae (e.g., Calanthe , Dendrobium , Neottianthe , Orchis ) [1, 2]. Typically involved in nectar secretion or storage, or merely displaying a nectariferous appearance, the spur reduces nectar evaporation, enhances pollinator specificity and efficiency, and plays a key role in pollination ecology, thereby constituting a central component of the pollination syndrome hypothesis [3]. Studies show that spur morphology results from the coordinated processes of cell division and elongation. In Centranthus ruber , epidermal cell division contributes to roughly 30% of the final spur length during early development, after which anisotropic elongation drives subsequent growth [4, 5]. Aquilegia follows a comparable two-phase “division-then-elongation” pattern: localized cell division forms the nectary cup and stops once the spur attains a length of about 1 mm; thereafter, nearly all further elongation results from highly oriented cell extension, with the duration of this elongation phase shaping interspecific differences [6–8]. By contrast, spur length differences between sister species of Linaria stem primarily from variations in cell division rates and initial cell numbers, with no significant differences observed in mature cell anisotropy—suggesting that distinct developmental mechanisms can evolve across lineages [9]. Investigating spur cytology in I. uliginosa , Li et al. ( 2024 ) reported that early development is dominated by cell division, which initiates spur outgrowth from the labellum, whereas later development is marked by cell elongation that ultimately determines final spur length [10]. The coordinated development of the floral spur is governed by hormonal signaling and core TFs. Transcriptomic analysis of I. uliginosa indicates that pathways involved in plant hormone signal transduction are consistently enriched throughout spur development, where genes associated with the cell cycle, division, and elongation modulate spur morphology through auxin, brassinosteroid, and other hormonal pathways [11]. In Aquilegia, auxin-related genes ( ARF8 , SAUR , YUCCA6 ) and brassinosteroid signaling genes ( TCP4 , CYP71A ) exhibit early differential expression and have been co-opted for novel roles: AqTCP4 suppresses distal cell division, facilitating localized invagination around the nectary to initiate spur formation, while cooperative regulatory modules containing ARF and STY further refine hormone-directed cell proliferation and expansion [8, 12]. Concurrently, TFs including KNOX , POP , and bHLH constitute an evolutionarily conserved yet flexible regulatory network. In Antirrhinum majus, for instance, a transposon insertion in HIRZINA/INVAGINATA —a KNOX-class gene—can induce spur-like outgrowths on typically spurless petals. Orthologs of this gene are highly expressed in Linaria vulgaris and Dactylorhiza fuchsii , and their ectopic expression in tobacco similarly produces saccate structures [13–15]. In contrast, spur development in Papaveraceae proceeds independently of STM -like KNOX genes, suggesting alternative genetic mechanisms [16]. In Aquilegia , the C₂H₂ zinc-finger TF POPOVICH functions as a master regulator of spur presence or absence, with downstream effectors such as ARF6/8 controlling variation in spur length [17]. Meanwhile, in Tropaeolum majus , lineage-specific duplications of TCP3/4L and STM genes, combined with early differential expression of bHLH027/046/082/083/092, drive cell division and spur morphogenesis [18]. Collectively, these findings demonstrate that hormone signaling and TF networks interact in lineage-specific manners to establish the molecular foundation of spur development. Although floral spurs have been studied extensively, prior work has concentrated largely on developmental and pollination biology, leaving variation in spur morphology relatively unexplored. During field surveys of Impatiens populations, we observed mutants of I. uliginosa that produces flowers with two or three spurs. This study used WT, 2SM and 3SM of I. uliginosa as experimental materials. We analyzed growth dynamics, internal spur cellular structure, transcriptomic profiles, and validate candidate genes using qPCR to investigate the mechanisms underlying spur number variation. Our findings are expected to provide foundational insights and a theoretical framework for floral trait modification and cultivar development in Impatiens . Results Spur development of I. uliginosa Based on morphological traits and growth curves, the development of both WT and mutant floral spurs can be categorized into three distinct stages: early stage, middle stage, and anthesis stage (Fig. 1 A–C). The primary spurs of WT I. uliginosa , along with the main spur of the 2SM and 3SM, displayed highly similar growth dynamics, completing development within 22–23 days and following a characteristic sigmoidal (“S”-shaped) growth curve. Rapid elongation commenced during the early stage, slowed by the middle stage, and plateaued after anthesis stage, with final spur lengths ranging from 18 to 23 mm. Lateral spur in the 2SM and 3SM initiated growth 6–8 days after the primary spurs. While their growth patterns were similar to those of the primary spurs, lateral spur were significantly shorter. The lateral spur of the 2SM reached approximately 5 mm, whereas those of the 3SM attained about 10 mm (Fig. 1 D–F). The development of spurs in I. uliginosa involves both cell division and anisotropic cell elongation To elucidate the mechanisms driving spur variation in WT and mutant I. uliginosa , we analyzed the internal cellular architecture of spurs across three critical developmental phases: early stage, middle stage, and anthesis stage. Across these stages, the cellular morphology of all spur types transitioned progressively from a smooth to a papilla-like form (Fig. 2 ). During the early stage, internal cells of WT spurs were flattened and texturally uniform. Similarly, the main and lateral spur of both 2SM and 3SM showed relatively smooth internal cells, with no marked morphological differences detectable among the spur types at this phase (Fig. 2 : OS, TwSm, TwSs, ThSm, ThSs). Upon progression to the middle stage, WT spur cells started to bulge. In the 2SM, main and lateral spur presented distinct layered and wrinkled textures. In contrast, the main and lateral spur of the 3SM developed irregularly spaced protrusions and a generally looser cellular organization (Fig. 2 : MOS, MTwSm, MTwSs, MThSm, MThSs). By the anthesis stage, protrusions on WT spur had become more rounded and densely arranged. The structural complexity increased further in both mutant types, characterized by an interwoven network of wrinkles and protrusions (Fig. 2 : AOS, ATwSm, ATwSs, AThSm, AThSs). Based on scanning electron microscopy (SEM) images, we conducted statistical analyses of cell length, cell width, and mean anisotropy within the floral spurs (Fig. 3 ). The results showed that from the early stage to anthesis stage, both cell length and cell width increased significantly in WT and mutant spurs (P < 0.05) (Fig. 3 A, B). However, at the same developmental stage, significant differences were observed in cell length, cell width, and anisotropy among different spur types. Specifically, at the early stage, significant differences were detected in spur length, spur width, and anisotropy among different types (P < 0.05); at the middle stage, significant differences were found in spur length and anisotropy among different types (P < 0.05), whereas no significant difference in cell width was observed between OS and TwSm; at full bloom, significant differences in spur length were detected among OS, TwSm, and TwSs (P < 0.05), and significant differences were also observed in cell width and cell anisotropy among different types (P < 0.05). As the floral spurs of both WT and mutant I. uliginosa were fully developed at anthesis stage, we performed histological examinations and quantified their internal cell number and cell area (Fig. 4 , Table 1 ). All five spur types across WT and mutant plants contained the same cell types, with densely arranged nucleated cells and vascular bundle tissues potentially interspersed among the epidermal cells (Fig. 4 ). The main spur of both WT and mutants possessed a significantly greater number of cells than mutant lateral spur, suggesting that cell number may be a primary factor underlying the length disparity between main and lateral spur. In contrast, the internal cell area showed no significant differences among the different spur types (Table 1 ). By integrating scanning electron microscopy with histological analysis, we conclude that the cellular development mechanisms of floral spurs in WT and mutant I. uliginosa are fundamentally similar: early spur formation depends primarily on cell division, whereas later elongation is driven mainly by cell expansion. The formation of internal papillary cell structures and the morphological variation among spurs appear to be regulated principally by cellular anisotropy. Table 1 Statistical data on spur cell morphology in WT and mutant I. uliginosa Sample ID Number of cells Cell area/µm² OS 446 a 13031.4 a TwSm 463 a 12507.25 a TwSs 254 b 11722.33 a ThSm 424 a 11719.86 a ThSs 254 b 10498.86 a Note: Based on wax section images of the middle outer epidermis of the nectar spur from WT and mutant I. uliginosa at anthesis stage, the sample area, cell count, and average cell anisotropy data were calculated at 400x magnification. Transcriptome RNA sequencing and De Novo assembly Transcriptomic analysis was performed on floral spurs at the early stage, specifically focusing on five distinct parts: WT spurs (OS), main spur of the 2SM (TwSm), lateral spur of the 2SM (TwSs), main spur of the 3SM (ThSm), and lateral spur of the 3SM (ThSs). A total of 230,870,736 raw reads were obtained, amounting to approximately 32.69 Gb of sequencing data. After quality control, each sample yielded approximately 6.11 Gb of high-quality clean data, consisting of roughly 40 million valid reads. The Q30 scores of the reads ranged from 94.36% to 94.50%, and the GC content exceeded 43.49% in all cases. The sequencing alignment coverage for all five samples was above 87.13% (Table 2 ). Table 2 Summary of sequencing data of I. uliginosa transcriptome Attributes OS TwSm TwSs ThSm ThSs Raw reads 47,634,206 43,413,662 46,311,164 46,773,286 46,738,418 Clean reads 45,455,420 41,650,530 44,664,562 45,210,178 45,028,386 Clean bases 6,685,598,669 6,109,376,886 6,594,787,929 6,669,815,808 6,635,220,780 Q30(%) 94.36 94.48 94.46 94.50 94.42 GC content(%) 43.97 43.69 43.86 43.49 43.67 Total mapped 19,931,464 (87.70%) 18,173,957 (87.27%) 19,485,808 (87.25%) 19,695,150 (87.13%) 19,731,553 (87.64%) Following de novo assembly using high-quality sequencing data, 81,512 transcripts with a combined length of 104.9 Mb were obtained after optimization and filtering. The transcripts ranged in length from 201 bp to 13,223 bp, with a mean length of 1,287.05 bp and an N50 of 1,910 bp. Subsequently, 42,721 unigenes (total length 46.9 Mb) were generated. Their length range matched that of the transcripts, with a mean length of 1,097.13 bp and an identical N50 of 1,910 bp (Table 3 ). Unigene length distribution analysis indicated that 83.59% were 200–2,000 bp long, 15% were 2,000–4,000 bp, and an additional 998 unigenes (2.34%) exceeded 4,000 bp (Fig. 5 ). Table 3 Statistics of transcriptome assembly Attributes unigenes transcripts Total number 42,721 81,512 Total base 46,870,323 104,909,795 Largest length (bp) 13,223 13,223 Smallest length (bp) 201 201 Average length (bp) 1,097.13 1,287.05 N50 length (bp) 1,910 1,910 Functional annotation Of the 42,721 assembled unigenes, 24,031 (56.25%) were successfully annotated based on alignments against six major databases (Table 4 , Table S1 ). Among these annotated unigenes, 23,818 (55.75% of the total) matched entries in the NR database; of these, 47.93% (11,417 unigenes) exhibited a sequence similarity > 80%, and 18,828 (79.05%) showed high homology (E-value < 1 × 10⁻³⁰). The five species with sequences most similar to those of I. uliginosa were Camellia sinensis (5,058 matches; 21.24%), Actinidia chinensis (2,780; 11.67%), Nyssa sinensis (2,575; 10.81%), Actinidia rufa (968; 4.06%), and Vitis vinifera (682; 2.86%) (Fig. S1 ). Furthermore, 18,989 (44.32%), 18,326 (42.90%), 21,780 (50.98%), 20,568 (48.14%), and 10,405 (24.36%) unigenes were aligned to the Swiss-Prot, Pfam, COG, GO, and KEGG databases, respectively. Table 4 Functional annotation of I. uliginosa unigenes Database Number of unigenes Percentage NR 23,818 55.75% Swiss-Prot 18,935 44.32% Pfam 18,326 42.90% COG 21,780 50.98% GO 20,568 48.14% KEGG 10,405 24.36% Total 24,031 56.25% A total of 21,780 unigenes were assigned to 23 COG categories, with 11,249 lacking functional annotation. Among the annotated unigenes, the largest COG category was "Posttranslational modification, protein turnover, chaperones," followed by "Signal transduction mechanisms" and "Transcription" (Fig. 6 A). Functional annotations were retrieved from the GO database for 20,568 unigenes and distributed across the three major GO classes: Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). Within the BP class, "macromolecule metabolic process" and "cellular macromolecule metabolic process" contained the highest number of genes, with 4,303 and 3,310 assignments, respectively. In the CC class, 6,947 unigenes were assigned to "integral component of membrane." At the MF level, the most prominent subcategories were "nucleic acid binding" and "anion binding" (Fig. 6 B), encompassing 4,044 and 3,851 unigenes, respectively. Additionally, 10,405 unigenes were mapped to 19 pathways across five major KEGG categories. The most representative pathways included "Carbohydrate metabolism," "Translation," "Folding, sorting and degradation," and "Transport and catabolism" (Fig. 6 C). Diferential gene expression To identify potential regulators of floral spur variation in I. uliginosa , we analyzed DEGs among different spurs at the early stage. In total, 8,592 DEGs were detected across the six pairwise comparisons (Fig. 7 A). Comparisons of the OS with TwSm and ThSm revealed 1,355 and 1,618 DEGs, respectively. Among these, 713 and 1,030 were upregulated, while 642 and 588 were downregulated. A total of 2,150 DEGs were identified between TwSm and TwSs, consisting of 1,022 upregulated and 1,128 downregulated genes. Similarly, 1,798 DEGs were found between ThSm and ThSs, with 410 upregulated and 1,388 downregulated. Only four DEGs were co-expressed across all these six groups (Fig. 7 B). Enrichment analysis of DEGs To explore the potential functions of DEGs in floral spur variation of I. uliginosa , we conducted functional enrichment analysis. All 8,592 DEGs mapped to 371 Gene Ontology (GO) terms, with 30 terms significantly enriched ( P < 0.05; Table S2). The most significantly enriched GO terms were "DNA-binding transcription factor activity," "photosystem," and "photosynthesis, light harvesting" (Fig. 8 A). The enrichment profiles of DEGs varied notably among different spur types. In comparisons of main spur—specifically, OS vs TwSm and OS vs ThSm—"DNA-binding transcription factor activity" was a prominent enriched biological process. Conversely, the comparison between TwSm vs ThSm was primarily enriched for "cytochrome-c oxidase activity" and "oxidoreductase activity, acting on a heme group of donors." Among lateral spur, the comparison between TwSs vs ThSs was predominantly enriched for "photosystem II." Within the 2SM, TwSm vs TwSs showed enrichment of "DNA-binding transcription factor activity," while the corresponding comparison in the 3SM (ThSm vs ThSs) was mainly enriched for "plastid thylakoid membrane" and "chloroplast thylakoid membrane." Remarkably, "DNA-binding transcription factor activity" was consistently enriched across all pairwise comparisons (OS vs TwSm, OS vs ThSm, TwSm vs ThSm, TwSm vs TwSs, ThSm vs ThSs, and TwSs vs ThSs), implying its crucial role in spur differentiation (Fig. S2). KEGG enrichment analysis revealed that all 8,592 unigenes were assigned to 111 KEGG pathways, with 11 pathways showing significant enrichment (P < 0.05) (Fig. 8 B, Table S3). Further analysis of DEGs across different spurs indicated that "Plant-pathogen interaction," "MAPK signaling pathway - plant," "Plant hormone signal transduction," "Phenylpropanoid biosynthesis," and "Monoterpenoid biosynthesis" were consistently enriched across various spur types. Specifically, "Plant-pathogen interaction," "Plant hormone signal transduction," and "MAPK signaling pathway - plant" were highly prominent in comparisons involving main spur (OS vs TwSm, OS vs ThSm, TwSm vs ThSm). In contrast, "Photosynthesis - antenna proteins," "Plant hormone signal transduction," "DNA replication," and "Plant-pathogen interaction" were notably enriched in the comparison of lateral spur (TwSs vs ThSs). Pathways such as "Starch and sucrose metabolism," "Oxidative phosphorylation," "Phenylpropanoid biosynthesis," "Monoterpenoid biosynthesis," "Tryptophan metabolism," "Brassinosteroid biosynthesis," and "Carbon fixation in photosynthetic organisms" were exclusively enriched in the 2SM comparison (TwSm vs TwSs). Meanwhile, "Photosynthesis," "Photosynthesis – antenna proteins," "Plant hormone signal transduction," "Carbon fixation in photosynthetic organisms," and "Glucosinolate biosynthesis" were uniquely enriched in the 3SM comparison (ThSm vs ThSs) (Fig. S3). Identification of TFs TFs were predicted by analyzing domain information within the transcriptome. A total of 955 unigenes were annotated as TFs, belonging to 34 distinct families (Fig. 9 ). Among these, the MYB family contained the most members (148 genes), followed by AP2/ERF (111), C2C2 (92), bHLH (79), and WRKY (53). All 955 transcription factor genes were differentially expressed across different floral spurs, with expression levels varying by up to 56-fold (Table S4). Members of the MYB and bHLH families exhibited higher expression in the main spur of 2SM and 3SM (Fig. 10 A, D). The AP2/ERF family showed elevated expression in the WT single spur and the main spur of mutant plants (Fig. 10 B). High expression of the C2C2 family was specific to the main spur of the 3SM, while its expression was lower in the lateral spur of both the 2SM and 3SM (Fig. 10 C). The WRKY family was highly expressed only in the WT spur and the lateral spur of the 2SM (Fig. 10 E). Lastly, the NAC family showed relatively high expression in both the 2SM (TwSm/TwSs) and 3SM (ThSm/ThSs) mutants (Fig. 10 F). Candidate genes involved in spur development Through comprehensive transcriptome analysis integrating gene function, enrichment, expression levels, differential expression patterns, and relevant studies in other species, we identified 20 preliminary candidate genes for floral spur variation in I. uliginosa (Table S5). This set includes two hormone-related genes: GID2 (TRINITY_DN8041_c0_g1), which is linked to plant hormone signal transduction, and SOB5 (TRINITY_DN4278_c0_g1), involved in cytokinin biosynthesis. Both genes were upregulated in mutants relative to the WT. Two TFs, NAC2 (TRINITY_DN8903_c0_g1) and SOC1 (TRINITY_DN2881_c0_g1), together with a highly expressed lipid transport and metabolism gene, LTP (TRINITY_DN11611_c0_g1), showed higher expression in main spur compared to lateral spur across WT and mutant plants. In contrast, the TF TCP4 (TRINITY_DN11420_c0_g1) was downregulated in main spur relative to lateral spur—notably, TCP4 has previously been shown to regulate spur development in Aquilegia and Linaria . Additionally, a highly expressed cell wall structure gene, GRP5 (TRINITY_DN8368_c0_g1), and an extremely highly expressed plant defensin gene, PDF1 (TRINITY_DN486_c0_g1; TPM > 2000), were also expressed at lower levels in main spur than in lateral spur. qRT-PCR validation of the candidate genes To validate the transcriptome data, eight candidate genes were selected for qRT-PCR analysis (Fig. 11 , Table S6). The results revealed differential expression patterns among the genotypes. Specifically, GID2 expression was significantly higher in the main spurs than in lateral spurs ( P < 0.05). SOB5 expression was significantly lower in the WT compared to the TwSm and TwSs mutants ( P < 0.05), while TwSm exhibited significantly higher expression than ThSm and ThSs (P < 0.05). Expression of NAC2 was elevated across all mutant types relative to the WT, with TwSm, TwSs, and ThSm showing significantly higher levels than OS ( P < 0.05). Conversely, transcript levels of LTP and TCP4 were significantly higher in the WT than in the mutants ( P < 0.05). GRP5 expression was uniquely and significantly elevated in the TwSs compared to all other tissues ( P < 0.05). While SOC1 expression was generally higher in the WT than in the mutants, it was specifically and significantly greater than in TwSm, TwSs, and ThSs. Finally, among the mutants, PDF1 expression was significantly higher in lateral spurs than in the main spurs ( P < 0.05). Discussion The floral spur is a hollow tubular structure that typically functions as a nectar reservoir. It represents both an evolutionary innovation and an important taxonomic trait in plants. Numerous plant taxa exhibit distinctive spurs, including Aquilegia and Consolida (Ranunculaceae), Linaria (Scrophulariaceae), and certain Cymbidium species (Orchidaceae). Spurs enhance pollination and reproductive success, contribute to morphological diversification [19], and may influence mechanisms related to biological invasion [20, 21]. Consequently, the floral spur is considered a key innovation trait, and its development and function have garnered considerable research interest. All species within the genus Impatiens bear floral spurs; their extensive interspecific morphological variation makes them an ideal model for studying spur biology. Earlier cytological studies showed that during early development, the spur of I. uliginosa displays prominent cell clustering and active division compared to other regions of the labellum [11]. Before spur differentiation, the spur primordium initiates intense cell division, which establishes the basis for subsequent morphogenesis and drives the outward extension of the spur from the labellum [10]. In the present study, we observed a significant increase in cell number in the spurs of both WT and mutant I. uliginosa from the early stage to middle stage, followed by marked cell elongation from early stage to anthesis stage. This developmental pattern closely parallels that reported in Aquilegia and Centranthus ruber , wherein spur growth involves phases of cell division and anisotropic elongation, with elongation being the principal driver of spur elongation [5,6]. Differentially expressed genes between the spur and the limb likely participate in regulating spur development [10,11]. Comparative transcriptomic analysis of five floral spur types in I. uliginosa at the early stage systematically uncovered key molecular mechanisms of spur morphogenesis. We identified 8,592 DEGs, of which four were differentially expressed in every pairwise comparison, implying their potential function as core regulators of spur development. Enrichment analysis revealed significant DEG enrichment in pathways including “Plant-pathogen interaction,” “MAPK signaling pathway – plant,” “Plant hormone signal transduction,” “Phenylpropanoid biosynthesis,” and “Monoterpenoid biosynthesis” in comparisons between WT and mutant spurs. These results indicate that early spur development involves active cell division, growth, and specific secondary metabolism—consistent with earlier reports in I. uliginosa , Aquilegia , and Linaria [8, 9, 11, 22]. In comparisons between the WT spur and mutant main spur, pathways such as “Plant-pathogen interaction,” “Plant hormone signal transduction,” and “MAPK signaling pathway – plant” remained consistently enriched, suggesting that main spur sustain vigorous elongation alongside basic immune and signaling networks. By contrast, DEGs from lateral spur were prominently enriched in “Photosynthesis – antenna proteins,” “Plant hormone signal transduction,” “DNA replication,” and “Plant-pathogen interaction,” indicating that lateral spur exhibit active cell division, rapid elongation, and heightened photosynthetic capacity. Notably, “Plant hormone signal transduction” was significantly enriched across all spur-type comparisons. This underscores the plant hormone signaling network as a central regulatory hub coordinating morphological variation in I. uliginosa spurs, likely by integrating developmental and defense signals to ultimately shape distinct spur morphologies. TFs act as central hubs within gene regulatory networks and are pivotal in shaping complex plant traits. In I. uliginosa , we identified 955 differentially expressed TFs representing 34 families. The expression patterns of key families-including AP2/ERF, MYB, bHLH, WRKY, NAC, and C2C2 were closely associated with the developmental states of the WT single spur and the main spur of 2SM and 3SM, offering important insights into the molecular mechanisms underlying the morphogenesis and variation of this specialized floral organ. The AP2/ERF family functions as a global regulator of plant development and stress responses, integrating multiple signals via its “identity determination-cell differentiation–senescence regulation” network [23, 24]. In this work, AP2/ERF genes were consistently highly expressed in the WT single spur and the main spur of mutants, strongly implicating this family in the early fate determination and initial differentiation of spur primordia. This pattern likely reflects the established role of AP2/ERF members in regulating cell-cycle genes and influencing proliferation and differentiation. Therefore, we propose that AP2/ERF TFs may serve as key “initiators” and “coordinators” of early spur development, potentially responding to upstream signals and activating downstream genes involved in cell division and differentiation, thereby establishing the cellular basis for spur outgrowth and elongation. Both MYB and bHLH families exhibited specific high expression in the main spur of 2SM and 3SM. The MYB family, with its dual “activation–repression” regulatory mode, acts as a hub for spatiotemporal precision in organ development [25], whereas bHLH factors are critical for processes such as organogenesis and pigment deposition [26]. Their co-expression in mutant main spur suggests they may form a functional module that co-regulates developmental pathways either absent or less active in WT plants. For instance, they could jointly activate target genes associated with cell elongation, directional growth, or specific metabolic pathways, thereby driving distinct morphological or growth dynamics in mutant main spur. This differential expression provides direct molecular evidence of diverging developmental mechanisms between WT and mutant spurs. Expression patterns of the WRKY and NAC families point to their potential roles in generating spur morphological diversity. WRKY genes were specifically highly expressed in the lateral spur of the WT and the 2SM. As signaling hubs that integrate environmental and endogenous cues [27, 28], their specific expression in lateral spur implies that lateral spur initiation or development may require the reception and integration of a unique set of signals, which could be essential for the formation of additional (lateral) spurs in mutants. In contrast, NAC genes were highly expressed in all mutant spurs. NAC factors participate in multiple processes, including organ boundary establishment, morphogenesis, and senescence [29, 30]. Their broad up-regulation suggests that the entire developmental program of mutant spurs—from primordium boundary setting to final morphogenesis—may be substantially reshaped by the NAC regulatory network, thereby stabilizing the multi-spur architecture. This study further revealed that the C2C2 family displayed significant expression differences among spurs while maintaining generally high expression levels. This observation aligns closely with reports in Aquilegia, where C2C2 members are involved in regulating spur presence/absence [17], strongly supporting an ancient, conserved core regulatory role for this family in angiosperm spur development. The widespread yet differential expression of C2C2 genes suggests that distinct members may fine-tune spur initiation, polar growth, and final morphology—possibly by establishing tissue polarity, modulating hormone (e.g., auxin) gradients, or responding to specific developmental signals—thereby representing key targets for understanding spur evolution and development. To identify potential regulators of floral spur variation in I. uliginosa , 10 candidate genes were screened, eight of which were validated using qRT-PCR. This validation confirmed the reliability of the transcriptome data. All eight genes exhibited significantly differential expression across distinct spur types in both WT and mutant plants.The GID2 gene, which participates in the gibberellin (GA) signaling pathway, functions in plant growth and development by coordinating with other GID family proteins to regulate cell division and elongation [31]. In this study, GID2 expression was significantly higher in the main spur compared to lateral spur, suggesting it may contribute to spur elongation and could partially account for the greater length of main spur relative to lateral spur. Both GRP5 and PDF1 are implicated in growth, defense, and stress responses. GRP5 encodes a glycine-rich protein likely involved in stress response and cell-wall maintenance [32]. PDF1 , a membrane component, shows preferential expression during Gossypium hirsutum fiber initiation and early elongation, where GbPDF1 interacts with GhMYB25-like to modulate H-O homeostasis, ethylene signaling, and pectin biosynthesis [33]. Here, GRP5 expression was significantly elevated in TwSs, whereas PDF1 was consistently lower in main spur than in lateral spur across all mutant backgrounds-a pattern indicating their potential roles in spur morphological variation. The SOB5 gene, involved in cytokinin biosynthesis, belongs to a family of plant-specific small proteins. Its overexpression elevates endogenous cytokinin levels by upregulating AtIPT3/7 , significantly altering growth and hormonal homeostasis in Arabidopsis [34]. In I. uliginosa , SOB5 expression was lower in the WT than in mutants, and significantly lower compared to 2SM, implying a possible function in variant spur formation. The LTP , SOC1 , and TCP4 genes each contribute to distinct aspects of plant development. LTP participates in wax or cutin deposition within expanding epidermal cells and secretory tissue cell walls. SOC1 , a MADS-box protein closely tied to floral development, is widely studied in flowering and fruit ripening. TCP4 has been established as a regulator of spur development in Aquilegia and Linaria [8, 9, 35, 36]. All three genes displayed a consistent expression pattern in I. uliginosa , with higher transcript levels in the WT than in mutants, suggesting they may function similarly in the molecular regulation of spur variation. Furthermore, the TF NAC2 showed elevated expression in mutants relative to the WT, with particularly high levels in the main spurs of 2SM and 3SM. Given its known roles in leaf senescence, floral development, lateral root formation, and secondary wall thickening [37–39], NAC2 may also be involved in regulating lateral spur formation in I. uliginosa . Conclusions This study investigated the developmental dynamics of floral spurs in WT and mutant I. uliginosa . Both spur types followed a typical sigmoidal growth curve, with lateral spur significantly shorter than main spur. Rapid spur development commenced at the early stage, decelerated by early flowering, and stabilized by anthesis stage. Cytological analyses revealed similar cellular development mechanisms in WT and mutant spurs: early spur formation depends mainly on cell division, whereas later length increase relies predominantly on cell elongation; the formation of internal cell protrusions is largely driven by cellular anisotropy. We further conducted transcriptome sequencing of spurs at the early stage in both genotypes. Cluster analysis and functional enrichment of 8,592 differentially expressed genes identified candidate genes associated with spur variation. Our results indicate that hormones play a pivotal role in spur variation, whereas genes involved in cell elongation, division, and the cell cycle are among the most critical factors shaping spur morphology. This study offers the first molecular insights into spur variation in the genus Impatiens , providing valuable information and a theoretical foundation for understanding spur diversity in dicotyledonous plants. Methods Plant materials Plants bearing a single floral spur are considered the WT of I. uliginosa , whereas those with two or three spurs are classified as mutants. Seeds of both WT and mutant I. uliginosa were collected from the area surrounding Kunming Laoyu River Wetland Park,located on the eastern shore of Dianchi Lake along East Huanhu Road, Kunming City, Yunnan Province.This site is a plateau lake-type wetland, with a central geographic location of approximately 24.83° N,102.81° E and an elevation of about 1890 m. Then we cultivated these seeds in the greenhouse at Southwest Forestry University under controlled conditions of 18–25 ℃ with a photo period of 11–13 h of light per day. Observation of spur growth dynamics Following the methodology described by Li et al. ( 2024 ), buds in which floral spurs had not yet initiated were randomly selected from 30 healthy I. uliginosa plants and monitored daily, resulting in 30 biological replicates. The length of each spur was measured daily at 9:00 AM, with three technical replicates per measurement. Measurements were conducted consecutively for 15 to 25 days, depending on the developmental progression of the plants. Growth curves were plotted based on the observational data. Three key developmental stages were identified through analysis of the spur growth curves in both WT and mutant I. uliginosa . Scanning electron microscopy and histological examination Following the methodology of Li et al. ( 2024 ), tissue samples were fixed in FAA (a mixture of 50% ethanol, glacial acetic acid, and 38% formaldehyde at an 18:1:1 ratio). Subsequent to dehydration through an ethanol gradient, the samples were critical-point dried with carbon dioxide using an EMS 850 dryer (Hatfield, PA, USA) and then imaged with a Zeiss Sigma 300 scanning electron microscope (Oberkochen, Germany). Following another ethanol gradient dehydration, the tissues were cleared in xylene, embedded in paraffin, and sectioned. Sections of 8 µm thickness were stained with 0.01% Safranin O and 50% Fast Green before being imaged under a Leica DM750 optical microscope (Wetzlar, Germany). RNA sequencing and de novo assembly of transcriptome Following the methodology of Li et al. ( 2022 ), nectary spurs from three developmental stages of both WT and mutant I. uliginosa were collected. Immediately upon excision, the spurs were preserved in liquid nitrogen. Each sample represented a pool of at least three biological replicates. Total RNA was extracted using Plant RNA Purification Reagent for plant tissue (Invitrogen, Carlsbad, CA, USA), followed by genomic DNA removal with DNase I (Takara). Subsequently, RNA-seq libraries were prepared using the TruSeq™ RNA Sample Preparation Kit (Illumina, San Diego, CA, USA). After quantification with a TBS380 fluorometer using Picogreen, the libraries were sequenced in a single lane on an Illumina HiSeq X Ten or NovaSeq 6000 platform (Illumina, San Diego, CA, USA) to generate 2 × 150 bp paired-end reads. The raw paired-end sequencing data were trimmed and quality controlled using SeqPrep ( https://github.com/jstjohn/SeqPrep ) and Sickle ( https://github.com/najoshi/sickle ) with default parameters. Clean data were de novo assembled using Trinity ( http://trinityrnaseq.sourceforge.net/ ), and homologous clustering of the assembled transcripts was performed, designating the longest transcript in each cluster as a unigene [40]. The assembled sequences were further filtered for optimization using TransRate ( http://hibberdlab.com/transrate/ ), redundant and similar sequences were removed using CD-HIT ( http://weizhongli-lab.org/cd-hit/ ), and the completeness of the transcriptome assembly was evaluated using the Benchmarking Universal Single-Copy Orthologs tool (BUSCO, http://busco.ezlab.org ). Functional annotation All assembled transcripts were functionally annotated by querying the NR, Swiss-Prot, Pfam, and COG databases. The BLASTX software was used to identify proteins with the highest similarity to the given transcript sequences, with a cutoff E-value typically set to less than 1.0×10⁻⁵. GO annotations for the unique assembled transcripts were obtained using the BLAST2GO program ( http://www.blast2go.com/b2ghome ) [41] to describe biological processes, molecular functions, and cellular components. Metabolic pathway analysis was conducted using the KEGG database ( http://www.genome.jp/kegg/ ) [42]. Diferential expression analysis and functional enrichment The expression levels of genes and transcripts were calculated using the TPM (transcripts per million reads) / FPKM (fragments per kilobase of transcript per million reads) method, and the quantification of gene and transcript abundance was performed using RSEM ( http://deweylab.biostat.wisc.edu/rsem/ ) [43]. Differential expression analysis was conducted using DESeq2 [44] / DEGseq [45] / EdgeR [46], with a threshold of Q value ≤ 0.05, |log2FC| > 1, and genes with Q value ≤ 0.05 (DESeq2 or EdgeR) / Q value ≤ 0.001 (DEGseq) being considered significantly differentially expressed. GO and KEGG functional enrichment analyses were performed on differentially expressed genes. GO functional enrichment analysis was carried out using Goatools ( https://github.com/tanghaibao/GOatools ), while KEGG pathway analysis was performed using KOBAS ( http://kobas.cbi.pku.edu.cn/home.do ) [47]. Fisher's exact test was utilized for calculations, and P-values were corrected using the Bonferroni and BH (FDR) methods. The threshold for corrected P-values was set at 0.05. Identifcation of TFs and cluster analysis Through HMMER analysis, the domain information of transcripts was compared with the PlantTFDB ( http://planttfdb.cbi.pku.edu.cn/ ) database to obtain homologous TF information for gene TF prediction and family analysis. Hierarchical clustering was conducted on differentially expressed TFs. qRT-PCR analysis of gene expression Total RNA was extracted from five flower distances of both WT and mutant Daphne retusa using the E.Z.N.A.® Plant RNA Kit from Omega. The first-strand cDNA was synthesized as a template using the EasyScript® One-Step gDNA Removal SuperMix from TransGen. qRT-PCR amplification was performed using the Roche LightCycler®480 II real-time quantitative PCR detection system, in combination with the Hieff® qPCR SYBR® Green Premix from Yisense. Details of the qRT-PCR amplification primers are provided in Supplementary Data Table S7, with IuActin serving as the housekeeping gene. The fluorescence quantitative PCR detection utilized a three-step method, with three technical replicates for each sample. The amplification protocol was as follows: pre-denaturation at 95°C for 5 minutes, followed by 40 cycles (denaturation at 95°C for 10 seconds, annealing at 60°C for 20 seconds, and extension at 72°C for 20 seconds). Gene expression levels were ultimately calculated using 2 −ΔΔCt method. Abbreviations NR NCBI Non-Redundant Protein Sequence Database Pfam Protein Families COG Clusters of Orthologous Groups of proteins GO Gene Ontology KEGG Kyoto Encyclopedia of Genes and Genomes. Declarations Acknowledgements We thank Shanghai Majorbio Bio-pharm Technology Co.,Ltd. for its help in sequencing. The data were analyzed through the free online platform of Majorbio Cloud Platform (www.majorbio.com). Authors’ contributions YL, XLZ and BH were responsible for the experimental design. YL, XLZ and BH carried out sample collection, experiments, data analysis and article writing. LQZ, HYL and Zhijia Gu participated in the experiment. MJH and HQH supervised the research and revised the manuscript. All authors read and approved the final manuscript. Funding This work was supported by the National Natural Science Foundation of China (grant NO. 32560389), Yunnan Fundamental Research Projects (grant NO. 202501AS070052), the Key Project of Yunnan Provincial Agricultural Joint Special Program (grant NO. 202301BD070001-011), and the Project of High-level Introduction talents in Yunnan Province. Data availability Sequence data that support the findings of this study have been deposited in the National Center for Biotechnology Information (NCBI). The project access number is PRJNA1422964 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1422964). Author details 1 College of Landscape Architecture and Horticulture Sciences, Yunnan Key Laboratory of Landscape Plant Resource Cultivation and Application, Yunnan Province Engineering Research Center for Functional Flower Resources and Industrialization, Research and Development Center of Landscape Plants and Horticulture Flowers, Southwest Forestry University, Kunming 650224, China 2 Key Laboratory for Plant Biodiversity and Biogeography of East Asia, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650201, China Ethics approval and consent to participate Not applicable Consent for publication Not applicable. Availability of data and materials The data and materials that support the findings of this study are available in the text and public data Competing interests The authors declare no competing interests. References Figueiredo ACS, Pais MS. Ultrastructural aspects of the nectary spur of Limodorum abortivum (L) Sw. (Orchidaceae). Ann Bot. 1992;70:325–31. Pacini E, Nepi M, Vesprini JL. Nectar biodiversity: a short review. Plant Syst Evol. 2003;238:7–21. Stang M, Klinkhamer PG, Meijden VDE. Asymmetric specialization and extinction risk in plant-flower visitor webs: a matter of morphology or abundance? Oecologia. 2007;151(3):442–53. Tsai T, Diggle PK, Frye HA, Jones CS. Contrasting lengths of Pelargonium floral nectar tubes result from late differences in rate and duration of growth. Ann Bot. 2018;121(3):549–60. Mack JK, Davis AR. The relationship between cell division and elongation during development of the nectar-yielding petal spur in Centranthus ruber (Valerianaceae). Ann Bot. 2015;115(4):641–9. Puzey JR, Gerbode SJ, Hodges SA, Kramer EM, Mahadevan L. Evolution of spur-length diversity in Aquilegia petals is achieved solely through cell-shape anisotropy. Proc. Biol. Sci. 2012; 279(1733): 1640-5. Sharma B, Yant L, Hodges SA, Kramer EM. Understanding the development and evolution of novel floral form in Aquilegia . Curr Opin Plant Biol. 2014; 1722–27. Yant L, Collani S, Puzey J, Levy C, Kramer EM et al. Molecular basis for three-dimensional elaboration of the Aquilegia petal spur. Proc. Biol. Sci. 2015; 282(1803): 20142778. Cullen E, Fernández-Mazuecos M, Glover BJ. Evolution of nectar spur length in a clade of Linaria reflects changes in cell division rather than in cell expansion[J]. Ann Botany. 2018;122(5):801–9. Li Y, Huang LW, Li YX, Zhang YD, Meng DC, Wei CM, et al. The cellular and molecular basis of the spur development in Impatiens uliginosa [J]. Hortic Res. 2024;11(3):uhae015–015. Li Y, Wei CM, Li XY, Meng DC, Gu ZJ, Su PQ, et al. De novo transcriptome sequencing of Impatiens uliginosa and the analysis of candidate genes related to spur development[J]. BMC Plant Biol. 2022;22(1):553–553. Zhou ZL, Duan YW, Luo Y, Yang YP, Zhang ZQ. Cell number explains the intraspecific spur-length variation in an Aquilegia species[J]. Plant Divers. 2019;41(5):307–14. Golz JF, Keck EJ, Hudson A. Spontaneous Mutations in KNOX Genes Give Rise to a Novel Floral Structure. Antirrhinum[J] Curr Biology. 2002;12(7):515–22. Box MS, Dodsworth S, Rudall PJ, Bateman RM, Glover BJ. Characterization of Linaria KNOX genes suggests a role in petal-spur development.[J]. Plant journal: cell Mol biology. 2011;68(4):703–14. Box MS, Dodsworth S, Rudall PJ, Bateman RM, Glover BJ. Flower-specific KNOX phenotype in the orchid Dactylorhiza fuchsii [J]. J Exp Bot. 2012;63(13):4811–9. Damerval C, Citerne H, Le GM, Domenichini S, Dutheil J, Ronse de CL, et al. Asymmetric morphogenetic cues along the transverse plane: shift from disymmetry to zygomorphy in the flower of Fumarioideae[J]. Am J Bot. 2013;100(2):391–402. Ballerini ES, Min Y, Edwards MB, Kramer EM, Hodges SA. POPOVICH, encoding a C2H2 zinc-finger transcription factor, plays a central role in the development of a key innovation, floral nectar spurs, in Aquilegia[J]. Proc Natl Acad Sci USA. 2020;117(36):22552–60. Li XY, Huang H, Zhang ZQ. Characterization and expression analysis of bHLH transcription factors reveal their putative regulatory effects on nectar spur development in Aquilegia species[J]. Gene, 2022, 852147057–147057. Michael B, Štěpán J. Pollinator-induced twisting of flowers sidesteps floral architecture constraints[J]. Curr Biol. 2014;24(17):R795–795. Vervoort A, Cawoy V, Jacquemart AL. Comparative Reproductive Biology in Co-Occurring Invasive and Native Impatiens Species[J]. Int J Plant Sci. 2011;172:366–77. Chupp AD, Battaglia LL, Schauber EM, et al. Orchid–pollinator interactions and potential vulnerability to biological invasion[J]. Aob Plants. 2015;7:99. Conway SJ, WalcherChevillet CL, Barbour KS, Kramer EM, et al. Brassinosteroids regulate petal spur length in Aquilegia by controlling cell elongation[J]. Ann Botany. 2021;128(7):931–42. Liu CC, Wang FF, Li RH, Zhu Y, Zhang CL, He YH. Marigold ( Tagetes erecta ) MADS-Box Genes: A Systematic Analysis and Their Implications for Floral Organ Development[J]. Agronomy. 2024;14(9):1889–1889. Doody E, Moyroud E. Evolution of petal patterning: blooming floral diversity at the microscale[J]. New Phytol. 2025;247(6):2538–56. Jae UH, Seonghwan K, Heejeong S, June MK. MYB Transcription Factors in Plant Developmental Plasticity[J]. J Plant Biology. 2025;68(6):1–14. Ning M, Li QY, Wang Y, Li QM, Tao YH, Zhang F, et al. Alternative splicing drives the functional diversification of a bHLH transcription factor in the control of growth and drought tolerance in rice[J]. Sci Bull. 2024;70(2):153–6. Zhao LR, Liu YW, Zhu Y, Chen SD, Du Y, Deng LY, et al. Transcription factor OsWRKY11 induces rice heading at low concentrations but inhibits rice heading at high concentrations[J]. J Integr Plant Biol. 2024;66(7):1385–407. Chen XJ, Zhang TL, Wang H, Zhao WS, Guo ZJ. Transcription factor WRKY complexes in plant signaling pathways[J]. Phytopathol Res. 2025;7(1):54–54. Xiong HY, He HD, Chang Y, Miao BB, Liu ZW, Wang QQ, et al. Multiple roles of NAC transcription factors in plant development and stress responses[J]. J Integr Plant Biol. 2025;67(3):510–38. Ma CD, Zhou PY, Ma YF, Tan W, Huang X, Segbo S, et al. A NAC family gene PmNAC32 associated with photoperiod promotes flower induction in Prunus mume [J]. Hortic Res. 2025;12(9):uhaf157. Xu RH, An JY, Wang YJ, Zhao XB, Li JR, Zhong TH, et al. OsGRF11 regulates cell division and elongation by interacting with GF14e and affecting gibberellin signaling[J]. Plant physiology; 2025. Yang HJ, Lee YS, Choi IS. Comparison of physicochemical properties and antioxidant activities of fermented soybean-based red pepper paste, Gochujang, prepared with five different red pepper ( Capsicum annuum L.) varieties[J]. J Food Sci Technol. 2018;55(2):792–801. Zhao G, Le Y, Sun M, Xu JW, Qin Y, Men S, et al. A dominant negative mutation of GhMYB25-like alters cotton fiber initiation, reducing lint and fuzz[J]. The Plant cell; 2024. Zhang JY, Wrage EL, Vankova R, Malbeck J, Neff MM. Over-expression of SOB5 suggests the involvement of a novel plant protein in cytokinin-mediated development[J]. Plant journal: cell Mol biology. 2006;46(5):834–48. Lu YH, Guo LS, Wei ZY, Li YJ, Zhang Y, Zheng JY et al. Pan-genomic insights into LTP gene family evolution across diploid cotton species[J]. Front Plant Sci, 2025, 16. Han X, Wang D, Song GQ. Expression of a maize SOC1 gene enhances soybean yield potential through modulating plant growth and flowering[J]. Sci Rep. 2021;11:12758. Ghafari RF, Vaziri A, Hassanpour AM, Talesh SS, Olfati J. Effects of humic acid on antioxidant defense system and senescence-related genes expression in petals of LA-Hybrid Lily[J]. J Plant Biochem Biotechnol. 2021;31(2):1–8. Tang N, Wu PY, Cao ZY, Liu YN, Zhang X, Lou J, et al. A NAC transcription factor ZaNAC93 confers floral initiation, fruit development, and prickle formation in Zanthoxylum armatum [J]. Plant physiology and biochemistry: PPB; 2023. pp. 201107813–107813. Pereira SS, Guimarães FCM, Carvalho JFC, Stolf-Moreira R, Oliveira MCN, Rolla AAP, et al. Transcription factors expressed in soybean roots under drought stress[J]. Genet Mol research: GMR. 2011;10(4):3689–701. Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome[J]. Nat Biotechnol. 2011;29:644–52. Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005;21:3674–6. Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27–30. Li B, Dewey CN. RSEM: accurate transcript quantification from RNASeq data with or without a reference genome. BMC Bioinformatics. 2011;12:323. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. Wang L, Feng Z, Wang X, Wang X, Zhang X. DEGseq: an R package for identifying differentially expressed genes from RNA-seq data. Bioinfor matics. 2009;26:136–8. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40. Xie C, Mao X, Huang J, Ding Y, Wu J, Dong S, et al. KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Res. 2011;39:W316–22. Additional Declarations No competing interests reported. Supplementary Files Supplementarymaterial.rar Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 30 Mar, 2026 Reviews received at journal 26 Mar, 2026 Reviewers agreed at journal 19 Mar, 2026 Reviews received at journal 16 Mar, 2026 Reviewers agreed at journal 25 Feb, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviewers invited by journal 24 Feb, 2026 Editor assigned by journal 24 Feb, 2026 Editor invited by journal 15 Feb, 2026 Submission checks completed at journal 14 Feb, 2026 First submitted to journal 14 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8829929","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":597773397,"identity":"542d932d-a521-44fc-b76e-f852a5e5c8f8","order_by":0,"name":"Yang Liu","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Liu","suffix":""},{"id":597773399,"identity":"1bc54f7e-d624-4df8-8630-a91066973e9f","order_by":1,"name":"Xiaoli Zhang","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoli","middleName":"","lastName":"Zhang","suffix":""},{"id":597773401,"identity":"92b1900b-3311-4440-a930-e8f01ec9232f","order_by":2,"name":"Bo Hui","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Hui","suffix":""},{"id":597773403,"identity":"9fe82afc-4424-4d8e-9c0a-2ff1042cdb15","order_by":3,"name":"Luqiu Zhao","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Luqiu","middleName":"","lastName":"Zhao","suffix":""},{"id":597773405,"identity":"89095e5b-586d-43b2-b748-18d9c35b7061","order_by":4,"name":"Hongyao Liao","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Hongyao","middleName":"","lastName":"Liao","suffix":""},{"id":597773407,"identity":"d6ed766c-ffdb-4491-b2c1-a7e77d2c5b48","order_by":5,"name":"Zhijia Gu","email":"","orcid":"","institution":"Key Laboratory for Plant Biodiversity and Biogeography of East Asia, Kunming Institute of Botany, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Zhijia","middleName":"","lastName":"Gu","suffix":""},{"id":597773409,"identity":"406f27b5-3b2b-4194-a55d-5fd196e9c1c9","order_by":6,"name":"Meijuan Huang","email":"","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":false,"prefix":"","firstName":"Meijuan","middleName":"","lastName":"Huang","suffix":""},{"id":597773410,"identity":"2bd0870a-e3e4-445e-959a-281f9ccc10d4","order_by":7,"name":"Haiquan Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCklEQVRIiWNgGAWjYBACAwbGBigDBCoYGNhBfB7itZwBqj5AUAsyg7GNCC3m7Icbb/PU3GEwl0h+9vDrvLrEHokExgdv2xjkzXFosexJbLbmOfaMwXJGmrmx7LbDIC3MhnPbGAx3NuBw2IHENmketsMMBjcSzKQltx3I3S+RwCbN28aQYHAAh5bzD4Fa/oG0pH+TlpxTlwu0hf03Xi03gLbwtoG05JhJfmxgBmlhY8av5WGz5dw+oJYzb8qkGY4dru/hedgsOeechOEGnA5Lf3jjzTegluPp2yR/1NQZ87AnH/zwpsxGHpctICABxPUNQIIZEh3gyJXArR5ZlvEHXnWjYBSMglEwUgEAIeNdjLFQJn8AAAAASUVORK5CYII=","orcid":"","institution":"Southwest Forestry University","correspondingAuthor":true,"prefix":"","firstName":"Haiquan","middleName":"","lastName":"Huang","suffix":""}],"badges":[],"createdAt":"2026-02-09 11:39:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8829929/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8829929/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103557301,"identity":"eabd2387-d49d-4082-9cc3-2876d21ed9e4","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":279470,"visible":true,"origin":"","legend":"\u003cp\u003eSpur development of\u003cem\u003e I. uliginosa\u003c/em\u003e. Average growth curve of WT (\u003cstrong\u003eA\u003c/strong\u003e), average growth curve of the main spur and lateral spur in 2SM (\u003cstrong\u003eB\u003c/strong\u003e), average growth curve of the main spur and lateral spur in 3SM (\u003cstrong\u003eC\u003c/strong\u003e)\u003cem\u003e. \u003c/em\u003eThe red dashed line indicates the boundary between the three development stages. D/E/F Spur in the early stage, showing a straight appearance (left); Spur in the middle stage, producing an inward curve (middle); Spur at anthesis stage (right). Scale bar = 1 cm.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/6678f1dab6cf4150e08c841d.png"},{"id":103557303,"identity":"be060218-df86-4484-9d0e-45403907c76b","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":697445,"visible":true,"origin":"","legend":"\u003cp\u003eInternal structure of the nectar spur in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e at different developmental stages. Scale bar = 20 μm. a-e mean early stage, f-j mean middle stage, k-o mean anthesis stage. \u003cstrong\u003eOS\u003c/strong\u003e, WT; \u003cstrong\u003eTwSm\u003c/strong\u003e, main spur of 2SM; \u003cstrong\u003eTwSs\u003c/strong\u003e, lateral spur of 2SM; \u003cstrong\u003eThSm\u003c/strong\u003e, main spur of 3SM; \u003cstrong\u003eThSs\u003c/strong\u003e, lateral spur of 3SM.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/c152f4c3a0cffb106a59789b.png"},{"id":103557306,"identity":"dfeb11b8-e2bf-4be9-b9b0-4042be54bc21","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":130788,"visible":true,"origin":"","legend":"\u003cp\u003eStatistical analysis of cell length, cell width and average cell anisotropy of WT and mutant \u003cem\u003eI. uliginosa.\u003c/em\u003e \u003cstrong\u003eOS\u003c/strong\u003e, WT; \u003cstrong\u003eTwSm\u003c/strong\u003e, main spur of 2SM; \u003cstrong\u003eTwSs\u003c/strong\u003e, lateral spur of 2SM; \u003cstrong\u003eThSm\u003c/strong\u003e, main spur of 3SM; \u003cstrong\u003eThSs\u003c/strong\u003e, lateral spur of 3SM.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/8f30c431f3d129bfef84fd5d.png"},{"id":104398393,"identity":"63705c7f-35c4-4eb5-9097-f5e122550eb3","added_by":"auto","created_at":"2026-03-11 12:02:09","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":632598,"visible":true,"origin":"","legend":"\u003cp\u003eEpidermis of the spur in the WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e at anthesis stage. \u003cstrong\u003eOS\u003c/strong\u003e, WT; \u003cstrong\u003eTwSm\u003c/strong\u003e, main spur of 2SM; \u003cstrong\u003eTwSs\u003c/strong\u003e, lateral spur of 2SM; \u003cstrong\u003eThSm\u003c/strong\u003e, main spur of 3SM; \u003cstrong\u003eThSs\u003c/strong\u003e, lateral spur of 3SM. Scale bar = 100 μm.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/713962efbb4ba925f5750dae.png"},{"id":104398261,"identity":"c9a1492f-2a88-4a04-a411-7d4ec3b905eb","added_by":"auto","created_at":"2026-03-11 12:01:01","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":32018,"visible":true,"origin":"","legend":"\u003cp\u003eLength distribution of unigenes\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/02aaac185d5ff9a7a640c025.png"},{"id":103557307,"identity":"05acc20f-1893-4ab5-92a0-ca7b18b14b94","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":2930030,"visible":true,"origin":"","legend":"\u003cp\u003eCOG classification (A), main GO categories (B) and KEGG metabolic pathway (C) of unigenes in \u003cem\u003eI. uliginosa\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/342aac9e946502b40da0dc9b.png"},{"id":104398556,"identity":"a72cfd98-0aef-4274-9866-150fb28c2584","added_by":"auto","created_at":"2026-03-11 12:02:51","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":98623,"visible":true,"origin":"","legend":"\u003cp\u003eDifferentially expressed genes (\u003cstrong\u003eA\u003c/strong\u003e) in different developmental stages and tissues Venn diagram of DEGs (\u003cstrong\u003eB\u003c/strong\u003e)\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/368bb2a4703cefd96e9b1d3b.png"},{"id":103557312,"identity":"10912770-bdf3-494b-a202-6aa72f9fc052","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":194493,"visible":true,"origin":"","legend":"\u003cp\u003eThe top 20 enriched GO terms of DEGs (A) and the top 20 enriched KEGG pathways of DEGs (B).\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/97afe7b6ee6a1dd013fdd4e4.png"},{"id":104398449,"identity":"afca7504-a024-49d7-9fbc-0ff197773add","added_by":"auto","created_at":"2026-03-11 12:02:24","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":83957,"visible":true,"origin":"","legend":"\u003cp\u003eFamilies of TFs identified in the \u003cem\u003eI. uliginosa\u003c/em\u003etranscriptome\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/066be8de08c331111960a54b.png"},{"id":104397827,"identity":"cfac3ca4-3f33-47ec-8612-29006c6abd04","added_by":"auto","created_at":"2026-03-11 11:57:17","extension":"jpeg","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":1060049,"visible":true,"origin":"","legend":"\u003cp\u003eCluster heat maps of DEGs in TF families. MYB(\u003cstrong\u003eA\u003c/strong\u003e), AP2/ERF (\u003cstrong\u003eB\u003c/strong\u003e), C2C2 (\u003cstrong\u003eC\u003c/strong\u003e), bHLH (\u003cstrong\u003eD\u003c/strong\u003e), WRKY (\u003cstrong\u003eE\u003c/strong\u003e), and NAC(\u003cstrong\u003eF\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/3a4daa28a287cdb407f73d3f.jpeg"},{"id":103557309,"identity":"c927d0d2-d42d-42e2-a5f7-170107d538e4","added_by":"auto","created_at":"2026-02-27 04:37:26","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":155432,"visible":true,"origin":"","legend":"\u003cp\u003eExpression analysis of eight candidate genes in 5 tissues by qRT-PCR.\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/44ef8c5796c529e91efcbc38.png"},{"id":104407445,"identity":"6d02a958-b550-45a7-9c29-1718ceff493b","added_by":"auto","created_at":"2026-03-11 12:38:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7487719,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/361df5a7-beaf-4d05-91f2-c8ed27916a4a.pdf"},{"id":104398172,"identity":"4deb7da4-ad95-4af6-be7a-85df4d8504df","added_by":"auto","created_at":"2026-03-11 12:00:11","extension":"rar","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":4869451,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.rar","url":"https://assets-eu.researchsquare.com/files/rs-8829929/v1/5e9135b5fab281f7afdc4d56.rar"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrated cytological and transcriptomic analyses reveal the molecular basis of spur variation in Impatiens uliginosa","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe spur\u0026mdash;a tubular, saccate, or similarly shaped extension formed by the backward or lateral elongation of petals or sepals\u0026mdash;is an evolutionary innovation that serves as an important taxonomic trait in plants. It is widespread across multiple families, such as Papaveraceae (e.g., \u003cem\u003eCorydalis\u003c/em\u003e, \u003cem\u003eFumaria\u003c/em\u003e), Ranunculaceae (e.g., \u003cem\u003eConsolida\u003c/em\u003e, \u003cem\u003eDelphinium\u003c/em\u003e, \u003cem\u003eAquilegia\u003c/em\u003e, \u003cem\u003eUrophysa\u003c/em\u003e), Balsaminaceae (\u003cem\u003eImpatiens\u003c/em\u003e), Violaceae (e.g., \u003cem\u003eAfrohybanthus\u003c/em\u003e, \u003cem\u003eViola\u003c/em\u003e), and Orchidaceae (e.g., \u003cem\u003eCalanthe\u003c/em\u003e, \u003cem\u003eDendrobium\u003c/em\u003e, \u003cem\u003eNeottianthe\u003c/em\u003e, \u003cem\u003eOrchis\u003c/em\u003e) [1, 2]. Typically involved in nectar secretion or storage, or merely displaying a nectariferous appearance, the spur reduces nectar evaporation, enhances pollinator specificity and efficiency, and plays a key role in pollination ecology, thereby constituting a central component of the pollination syndrome hypothesis [3].\u003c/p\u003e \u003cp\u003eStudies show that spur morphology results from the coordinated processes of cell division and elongation. In \u003cem\u003eCentranthus ruber\u003c/em\u003e, epidermal cell division contributes to roughly 30% of the final spur length during early development, after which anisotropic elongation drives subsequent growth [4, 5]. \u003cem\u003eAquilegia\u003c/em\u003e follows a comparable two-phase \u0026ldquo;division-then-elongation\u0026rdquo; pattern: localized cell division forms the nectary cup and stops once the spur attains a length of about 1 mm; thereafter, nearly all further elongation results from highly oriented cell extension, with the duration of this elongation phase shaping interspecific differences [6\u0026ndash;8]. By contrast, spur length differences between sister species of \u003cem\u003eLinaria\u003c/em\u003e stem primarily from variations in cell division rates and initial cell numbers, with no significant differences observed in mature cell anisotropy\u0026mdash;suggesting that distinct developmental mechanisms can evolve across lineages [9]. Investigating spur cytology in \u003cem\u003eI. uliginosa\u003c/em\u003e, Li et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) reported that early development is dominated by cell division, which initiates spur outgrowth from the labellum, whereas later development is marked by cell elongation that ultimately determines final spur length [10].\u003c/p\u003e \u003cp\u003eThe coordinated development of the floral spur is governed by hormonal signaling and core TFs. Transcriptomic analysis of \u003cem\u003eI. uliginosa\u003c/em\u003e indicates that pathways involved in plant hormone signal transduction are consistently enriched throughout spur development, where genes associated with the cell cycle, division, and elongation modulate spur morphology through auxin, brassinosteroid, and other hormonal pathways [11]. In Aquilegia, auxin-related genes (\u003cem\u003eARF8\u003c/em\u003e, \u003cem\u003eSAUR\u003c/em\u003e, \u003cem\u003eYUCCA6\u003c/em\u003e) and brassinosteroid signaling genes (\u003cem\u003eTCP4\u003c/em\u003e, \u003cem\u003eCYP71A\u003c/em\u003e) exhibit early differential expression and have been co-opted for novel roles: \u003cem\u003eAqTCP4\u003c/em\u003e suppresses distal cell division, facilitating localized invagination around the nectary to initiate spur formation, while cooperative regulatory modules containing \u003cem\u003eARF\u003c/em\u003e and \u003cem\u003eSTY\u003c/em\u003e further refine hormone-directed cell proliferation and expansion [8, 12]. Concurrently, TFs including \u003cem\u003eKNOX\u003c/em\u003e, \u003cem\u003ePOP\u003c/em\u003e, and \u003cem\u003ebHLH\u003c/em\u003e constitute an evolutionarily conserved yet flexible regulatory network. In Antirrhinum majus, for instance, a transposon insertion in \u003cem\u003eHIRZINA/INVAGINATA\u003c/em\u003e\u0026mdash;a KNOX-class gene\u0026mdash;can induce spur-like outgrowths on typically spurless petals. Orthologs of this gene are highly expressed in \u003cem\u003eLinaria vulgaris\u003c/em\u003e and \u003cem\u003eDactylorhiza fuchsii\u003c/em\u003e, and their ectopic expression in tobacco similarly produces saccate structures [13\u0026ndash;15]. In contrast, spur development in Papaveraceae proceeds independently of \u003cem\u003eSTM\u003c/em\u003e-like \u003cem\u003eKNOX\u003c/em\u003e genes, suggesting alternative genetic mechanisms [16]. In \u003cem\u003eAquilegia\u003c/em\u003e, the C₂H₂ zinc-finger TF POPOVICH functions as a master regulator of spur presence or absence, with downstream effectors such as \u003cem\u003eARF6/8\u003c/em\u003e controlling variation in spur length [17]. Meanwhile, in \u003cem\u003eTropaeolum majus\u003c/em\u003e, lineage-specific duplications of \u003cem\u003eTCP3/4L\u003c/em\u003e and \u003cem\u003eSTM\u003c/em\u003e genes, combined with early differential expression of bHLH027/046/082/083/092, drive cell division and spur morphogenesis [18]. Collectively, these findings demonstrate that hormone signaling and TF networks interact in lineage-specific manners to establish the molecular foundation of spur development.\u003c/p\u003e \u003cp\u003eAlthough floral spurs have been studied extensively, prior work has concentrated largely on developmental and pollination biology, leaving variation in spur morphology relatively unexplored. During field surveys of \u003cem\u003eImpatiens\u003c/em\u003e populations, we observed mutants of \u003cem\u003eI. uliginosa\u003c/em\u003e that produces flowers with two or three spurs. This study used WT, 2SM and 3SM of \u003cem\u003eI. uliginosa\u003c/em\u003e as experimental materials. We analyzed growth dynamics, internal spur cellular structure, transcriptomic profiles, and validate candidate genes using qPCR to investigate the mechanisms underlying spur number variation. Our findings are expected to provide foundational insights and a theoretical framework for floral trait modification and cultivar development in \u003cem\u003eImpatiens\u003c/em\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eSpur development of\u003c/b\u003e \u003cb\u003eI. uliginosa\u003c/b\u003e\u003c/p\u003e \u003cp\u003eBased on morphological traits and growth curves, the development of both WT and mutant floral spurs can be categorized into three distinct stages: early stage, middle stage, and anthesis stage (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA\u0026ndash;C). The primary spurs of WT \u003cem\u003eI. uliginosa\u003c/em\u003e, along with the main spur of the 2SM and 3SM, displayed highly similar growth dynamics, completing development within 22\u0026ndash;23 days and following a characteristic sigmoidal (\u0026ldquo;S\u0026rdquo;-shaped) growth curve. Rapid elongation commenced during the early stage, slowed by the middle stage, and plateaued after anthesis stage, with final spur lengths ranging from 18 to 23 mm. Lateral spur in the 2SM and 3SM initiated growth 6\u0026ndash;8 days after the primary spurs. While their growth patterns were similar to those of the primary spurs, lateral spur were significantly shorter. The lateral spur of the 2SM reached approximately 5 mm, whereas those of the 3SM attained about 10 mm (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD\u0026ndash;F).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eThe development of spurs in\u003c/b\u003e \u003cb\u003eI. uliginosa\u003c/b\u003e \u003cb\u003einvolves both cell division and anisotropic cell elongation\u003c/b\u003e\u003c/p\u003e \u003cp\u003eTo elucidate the mechanisms driving spur variation in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e, we analyzed the internal cellular architecture of spurs across three critical developmental phases: early stage, middle stage, and anthesis stage. Across these stages, the cellular morphology of all spur types transitioned progressively from a smooth to a papilla-like form (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). During the early stage, internal cells of WT spurs were flattened and texturally uniform. Similarly, the main and lateral spur of both 2SM and 3SM showed relatively smooth internal cells, with no marked morphological differences detectable among the spur types at this phase (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e: OS, TwSm, TwSs, ThSm, ThSs). Upon progression to the middle stage, WT spur cells started to bulge. In the 2SM, main and lateral spur presented distinct layered and wrinkled textures. In contrast, the main and lateral spur of the 3SM developed irregularly spaced protrusions and a generally looser cellular organization (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e: MOS, MTwSm, MTwSs, MThSm, MThSs). By the anthesis stage, protrusions on WT spur had become more rounded and densely arranged. The structural complexity increased further in both mutant types, characterized by an interwoven network of wrinkles and protrusions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e: AOS, ATwSm, ATwSs, AThSm, AThSs).\u003c/p\u003e \u003cp\u003eBased on scanning electron microscopy (SEM) images, we conducted statistical analyses of cell length, cell width, and mean anisotropy within the floral spurs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The results showed that from the early stage to anthesis stage, both cell length and cell width increased significantly in WT and mutant spurs (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, B). However, at the same developmental stage, significant differences were observed in cell length, cell width, and anisotropy among different spur types. Specifically, at the early stage, significant differences were detected in spur length, spur width, and anisotropy among different types (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05); at the middle stage, significant differences were found in spur length and anisotropy among different types (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), whereas no significant difference in cell width was observed between OS and TwSm; at full bloom, significant differences in spur length were detected among OS, TwSm, and TwSs (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), and significant differences were also observed in cell width and cell anisotropy among different types (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003eAs the floral spurs of both WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e were fully developed at anthesis stage, we performed histological examinations and quantified their internal cell number and cell area (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). All five spur types across WT and mutant plants contained the same cell types, with densely arranged nucleated cells and vascular bundle tissues potentially interspersed among the epidermal cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The main spur of both WT and mutants possessed a significantly greater number of cells than mutant lateral spur, suggesting that cell number may be a primary factor underlying the length disparity between main and lateral spur. In contrast, the internal cell area showed no significant differences among the different spur types (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBy integrating scanning electron microscopy with histological analysis, we conclude that the cellular development mechanisms of floral spurs in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e are fundamentally similar: early spur formation depends primarily on cell division, whereas later elongation is driven mainly by cell expansion. The formation of internal papillary cell structures and the morphological variation among spurs appear to be regulated principally by cellular anisotropy.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStatistical data on spur cell morphology in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSample ID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of cells\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCell area/\u0026micro;m\u0026sup2;\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOS\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e446 a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13031.4 a\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTwSm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e463 a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12507.25 a\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTwSs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e254 b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11722.33 a\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThSm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e424 a\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11719.86 a\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThSs\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e254 b\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10498.86 a\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003eNote: Based on wax section images of the middle outer epidermis of the nectar spur from WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e at anthesis stage, the sample area, cell count, and average cell anisotropy data were calculated at 400x magnification.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eTranscriptome RNA sequencing and De Novo assembly\u003c/h2\u003e \u003cp\u003eTranscriptomic analysis was performed on floral spurs at the early stage, specifically focusing on five distinct parts: WT spurs (OS), main spur of the 2SM (TwSm), lateral spur of the 2SM (TwSs), main spur of the 3SM (ThSm), and lateral spur of the 3SM (ThSs). A total of 230,870,736 raw reads were obtained, amounting to approximately 32.69 Gb of sequencing data. After quality control, each sample yielded approximately 6.11 Gb of high-quality clean data, consisting of roughly 40\u0026nbsp;million valid reads. The Q30 scores of the reads ranged from 94.36% to 94.50%, and the GC content exceeded 43.49% in all cases. The sequencing alignment coverage for all five samples was above 87.13% (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of sequencing data of \u003cem\u003eI. uliginosa\u003c/em\u003e transcriptome\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAttributes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTwSm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTwSs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eThSm\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eThSs\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRaw reads\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47,634,206\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43,413,662\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46,311,164\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e46,773,286\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e46,738,418\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClean reads\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e45,455,420\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e41,650,530\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e44,664,562\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e45,210,178\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e45,028,386\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eClean bases\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6,685,598,669\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6,109,376,886\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6,594,787,929\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6,669,815,808\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6,635,220,780\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eQ30(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e94.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e94.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e94.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e94.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGC content(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e43.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e43.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e43.67\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal mapped\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19,931,464\u003c/p\u003e \u003cp\u003e(87.70%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18,173,957\u003c/p\u003e \u003cp\u003e(87.27%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19,485,808\u003c/p\u003e \u003cp\u003e(87.25%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e19,695,150\u003c/p\u003e \u003cp\u003e(87.13%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e19,731,553\u003c/p\u003e \u003cp\u003e(87.64%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFollowing de novo assembly using high-quality sequencing data, 81,512 transcripts with a combined length of 104.9 Mb were obtained after optimization and filtering. The transcripts ranged in length from 201 bp to 13,223 bp, with a mean length of 1,287.05 bp and an N50 of 1,910 bp. Subsequently, 42,721 unigenes (total length 46.9 Mb) were generated. Their length range matched that of the transcripts, with a mean length of 1,097.13 bp and an identical N50 of 1,910 bp (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Unigene length distribution analysis indicated that 83.59% were 200\u0026ndash;2,000 bp long, 15% were 2,000\u0026ndash;4,000 bp, and an additional 998 unigenes (2.34%) exceeded 4,000 bp (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStatistics of transcriptome assembly\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAttributes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eunigenes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003etranscripts\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal number\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42,721\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81,512\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal base\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46,870,323\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e104,909,795\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLargest length (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13,223\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13,223\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmallest length (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e201\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e201\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAverage length (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,097.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,287.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN50 length (bp)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1,910\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1,910\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eFunctional annotation\u003c/h3\u003e\n\u003cp\u003eOf the 42,721 assembled unigenes, 24,031 (56.25%) were successfully annotated based on alignments against six major databases (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Among these annotated unigenes, 23,818 (55.75% of the total) matched entries in the NR database; of these, 47.93% (11,417 unigenes) exhibited a sequence similarity\u0026thinsp;\u0026gt;\u0026thinsp;80%, and 18,828 (79.05%) showed high homology (E-value\u0026thinsp;\u0026lt;\u0026thinsp;1 \u0026times; 10⁻\u0026sup3;⁰). The five species with sequences most similar to those of \u003cem\u003eI. uliginosa\u003c/em\u003e were \u003cem\u003eCamellia sinensis\u003c/em\u003e (5,058 matches; 21.24%), \u003cem\u003eActinidia chinensis\u003c/em\u003e (2,780; 11.67%), \u003cem\u003eNyssa sinensis\u003c/em\u003e (2,575; 10.81%), \u003cem\u003eActinidia rufa\u003c/em\u003e (968; 4.06%), and \u003cem\u003eVitis vinifera\u003c/em\u003e (682; 2.86%) (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Furthermore, 18,989 (44.32%), 18,326 (42.90%), 21,780 (50.98%), 20,568 (48.14%), and 10,405 (24.36%) unigenes were aligned to the Swiss-Prot, Pfam, COG, GO, and KEGG databases, respectively.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFunctional annotation of \u003cem\u003eI. uliginosa\u003c/em\u003e unigenes\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDatabase\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of unigenes\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePercentage\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e23,818\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e55.75%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSwiss-Prot\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e18,935\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e44.32%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePfam\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e18,326\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e42.90%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e21,780\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e50.98%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGO\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e20,568\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e48.14%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKEGG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10,405\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24.36%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e24,031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e56.25%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eA total of 21,780 unigenes were assigned to 23 COG categories, with 11,249 lacking functional annotation. Among the annotated unigenes, the largest COG category was \"Posttranslational modification, protein turnover, chaperones,\" followed by \"Signal transduction mechanisms\" and \"Transcription\" (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Functional annotations were retrieved from the GO database for 20,568 unigenes and distributed across the three major GO classes: Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). Within the BP class, \"macromolecule metabolic process\" and \"cellular macromolecule metabolic process\" contained the highest number of genes, with 4,303 and 3,310 assignments, respectively. In the CC class, 6,947 unigenes were assigned to \"integral component of membrane.\" At the MF level, the most prominent subcategories were \"nucleic acid binding\" and \"anion binding\" (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB), encompassing 4,044 and 3,851 unigenes, respectively. Additionally, 10,405 unigenes were mapped to 19 pathways across five major KEGG categories. The most representative pathways included \"Carbohydrate metabolism,\" \"Translation,\" \"Folding, sorting and degradation,\" and \"Transport and catabolism\" (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eDiferential gene expression\u003c/h3\u003e\n\u003cp\u003eTo identify potential regulators of floral spur variation in \u003cem\u003eI. uliginosa\u003c/em\u003e, we analyzed DEGs among different spurs at the early stage. In total, 8,592 DEGs were detected across the six pairwise comparisons (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Comparisons of the OS with TwSm and ThSm revealed 1,355 and 1,618 DEGs, respectively. Among these, 713 and 1,030 were upregulated, while 642 and 588 were downregulated. A total of 2,150 DEGs were identified between TwSm and TwSs, consisting of 1,022 upregulated and 1,128 downregulated genes. Similarly, 1,798 DEGs were found between ThSm and ThSs, with 410 upregulated and 1,388 downregulated. Only four DEGs were co-expressed across all these six groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eEnrichment analysis of DEGs\u003c/h3\u003e\n\u003cp\u003eTo explore the potential functions of DEGs in floral spur variation of \u003cem\u003eI. uliginosa\u003c/em\u003e, we conducted functional enrichment analysis. All 8,592 DEGs mapped to 371 Gene Ontology (GO) terms, with 30 terms significantly enriched (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Table S2). The most significantly enriched GO terms were \"DNA-binding transcription factor activity,\" \"photosystem,\" and \"photosynthesis, light harvesting\" (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA). The enrichment profiles of DEGs varied notably among different spur types. In comparisons of main spur\u0026mdash;specifically, OS vs TwSm and OS vs ThSm\u0026mdash;\"DNA-binding transcription factor activity\" was a prominent enriched biological process. Conversely, the comparison between TwSm vs ThSm was primarily enriched for \"cytochrome-c oxidase activity\" and \"oxidoreductase activity, acting on a heme group of donors.\" Among lateral spur, the comparison between TwSs vs ThSs was predominantly enriched for \"photosystem II.\" Within the 2SM, TwSm vs TwSs showed enrichment of \"DNA-binding transcription factor activity,\" while the corresponding comparison in the 3SM (ThSm vs ThSs) was mainly enriched for \"plastid thylakoid membrane\" and \"chloroplast thylakoid membrane.\" Remarkably, \"DNA-binding transcription factor activity\" was consistently enriched across all pairwise comparisons (OS vs TwSm, OS vs ThSm, TwSm vs ThSm, TwSm vs TwSs, ThSm vs ThSs, and TwSs vs ThSs), implying its crucial role in spur differentiation (Fig. S2).\u003c/p\u003e \u003cp\u003eKEGG enrichment analysis revealed that all 8,592 unigenes were assigned to 111 KEGG pathways, with 11 pathways showing significant enrichment (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB, Table S3). Further analysis of DEGs across different spurs indicated that \"Plant-pathogen interaction,\" \"MAPK signaling pathway - plant,\" \"Plant hormone signal transduction,\" \"Phenylpropanoid biosynthesis,\" and \"Monoterpenoid biosynthesis\" were consistently enriched across various spur types. Specifically, \"Plant-pathogen interaction,\" \"Plant hormone signal transduction,\" and \"MAPK signaling pathway - plant\" were highly prominent in comparisons involving main spur (OS vs TwSm, OS vs ThSm, TwSm vs ThSm). In contrast, \"Photosynthesis - antenna proteins,\" \"Plant hormone signal transduction,\" \"DNA replication,\" and \"Plant-pathogen interaction\" were notably enriched in the comparison of lateral spur (TwSs vs ThSs). Pathways such as \"Starch and sucrose metabolism,\" \"Oxidative phosphorylation,\" \"Phenylpropanoid biosynthesis,\" \"Monoterpenoid biosynthesis,\" \"Tryptophan metabolism,\" \"Brassinosteroid biosynthesis,\" and \"Carbon fixation in photosynthetic organisms\" were exclusively enriched in the 2SM comparison (TwSm vs TwSs). Meanwhile, \"Photosynthesis,\" \"Photosynthesis \u0026ndash; antenna proteins,\" \"Plant hormone signal transduction,\" \"Carbon fixation in photosynthetic organisms,\" and \"Glucosinolate biosynthesis\" were uniquely enriched in the 3SM comparison (ThSm vs ThSs) (Fig. S3).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eIdentification of TFs\u003c/h3\u003e\n\u003cp\u003eTFs were predicted by analyzing domain information within the transcriptome. A total of 955 unigenes were annotated as TFs, belonging to 34 distinct families (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Among these, the MYB family contained the most members (148 genes), followed by AP2/ERF (111), C2C2 (92), bHLH (79), and WRKY (53). All 955 transcription factor genes were differentially expressed across different floral spurs, with expression levels varying by up to 56-fold (Table S4). Members of the MYB and bHLH families exhibited higher expression in the main spur of 2SM and 3SM (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eA, D). The AP2/ERF family showed elevated expression in the WT single spur and the main spur of mutant plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eB). High expression of the C2C2 family was specific to the main spur of the 3SM, while its expression was lower in the lateral spur of both the 2SM and 3SM (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eC). The WRKY family was highly expressed only in the WT spur and the lateral spur of the 2SM (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eE). Lastly, the NAC family showed relatively high expression in both the 2SM (TwSm/TwSs) and 3SM (ThSm/ThSs) mutants (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eCandidate genes involved in spur development\u003c/h2\u003e \u003cp\u003eThrough comprehensive transcriptome analysis integrating gene function, enrichment, expression levels, differential expression patterns, and relevant studies in other species, we identified 20 preliminary candidate genes for floral spur variation in \u003cem\u003eI. uliginosa\u003c/em\u003e (Table S5). This set includes two hormone-related genes: \u003cem\u003eGID2\u003c/em\u003e (TRINITY_DN8041_c0_g1), which is linked to plant hormone signal transduction, and \u003cem\u003eSOB5\u003c/em\u003e (TRINITY_DN4278_c0_g1), involved in cytokinin biosynthesis. Both genes were upregulated in mutants relative to the WT. Two TFs, \u003cem\u003eNAC2\u003c/em\u003e (TRINITY_DN8903_c0_g1) and \u003cem\u003eSOC1\u003c/em\u003e (TRINITY_DN2881_c0_g1), together with a highly expressed lipid transport and metabolism gene, \u003cem\u003eLTP\u003c/em\u003e (TRINITY_DN11611_c0_g1), showed higher expression in main spur compared to lateral spur across WT and mutant plants. In contrast, the TF \u003cem\u003eTCP4\u003c/em\u003e (TRINITY_DN11420_c0_g1) was downregulated in main spur relative to lateral spur\u0026mdash;notably, \u003cem\u003eTCP4\u003c/em\u003e has previously been shown to regulate spur development in \u003cem\u003eAquilegia\u003c/em\u003e and \u003cem\u003eLinaria\u003c/em\u003e. Additionally, a highly expressed cell wall structure gene, \u003cem\u003eGRP5\u003c/em\u003e (TRINITY_DN8368_c0_g1), and an extremely highly expressed plant defensin gene, \u003cem\u003ePDF1\u003c/em\u003e (TRINITY_DN486_c0_g1; TPM\u0026thinsp;\u0026gt;\u0026thinsp;2000), were also expressed at lower levels in main spur than in lateral spur.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eqRT-PCR validation of the candidate genes\u003c/h3\u003e\n\u003cp\u003eTo validate the transcriptome data, eight candidate genes were selected for qRT-PCR analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003e, Table S6). The results revealed differential expression patterns among the genotypes. Specifically, \u003cem\u003eGID2\u003c/em\u003e expression was significantly higher in the main spurs than in lateral spurs (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). \u003cem\u003eSOB5\u003c/em\u003e expression was significantly lower in the WT compared to the TwSm and TwSs mutants (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), while TwSm exhibited significantly higher expression than ThSm and ThSs (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Expression of \u003cem\u003eNAC2\u003c/em\u003e was elevated across all mutant types relative to the WT, with TwSm, TwSs, and ThSm showing significantly higher levels than OS (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Conversely, transcript levels of \u003cem\u003eLTP\u003c/em\u003e and \u003cem\u003eTCP4\u003c/em\u003e were significantly higher in the WT than in the mutants (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). \u003cem\u003eGRP5\u003c/em\u003e expression was uniquely and significantly elevated in the TwSs compared to all other tissues (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). While \u003cem\u003eSOC1\u003c/em\u003e expression was generally higher in the WT than in the mutants, it was specifically and significantly greater than in TwSm, TwSs, and ThSs. Finally, among the mutants, \u003cem\u003ePDF1\u003c/em\u003e expression was significantly higher in lateral spurs than in the main spurs (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe floral spur is a hollow tubular structure that typically functions as a nectar reservoir. It represents both an evolutionary innovation and an important taxonomic trait in plants. Numerous plant taxa exhibit distinctive spurs, including \u003cem\u003eAquilegia\u003c/em\u003e and \u003cem\u003eConsolida\u003c/em\u003e (Ranunculaceae), \u003cem\u003eLinaria\u003c/em\u003e (Scrophulariaceae), and certain \u003cem\u003eCymbidium\u003c/em\u003e species (Orchidaceae). Spurs enhance pollination and reproductive success, contribute to morphological diversification [19], and may influence mechanisms related to biological invasion [20, 21]. Consequently, the floral spur is considered a key innovation trait, and its development and function have garnered considerable research interest. All species within the genus \u003cem\u003eImpatiens\u003c/em\u003e bear floral spurs; their extensive interspecific morphological variation makes them an ideal model for studying spur biology. Earlier cytological studies showed that during early development, the spur of \u003cem\u003eI. uliginosa\u003c/em\u003e displays prominent cell clustering and active division compared to other regions of the labellum [11]. Before spur differentiation, the spur primordium initiates intense cell division, which establishes the basis for subsequent morphogenesis and drives the outward extension of the spur from the labellum [10]. In the present study, we observed a significant increase in cell number in the spurs of both WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e from the early stage to middle stage, followed by marked cell elongation from early stage to anthesis stage. This developmental pattern closely parallels that reported in \u003cem\u003eAquilegia\u003c/em\u003e and \u003cem\u003eCentranthus ruber\u003c/em\u003e, wherein spur growth involves phases of cell division and anisotropic elongation, with elongation being the principal driver of spur elongation [5,6]. Differentially expressed genes between the spur and the limb likely participate in regulating spur development [10,11].\u003c/p\u003e \u003cp\u003eComparative transcriptomic analysis of five floral spur types in \u003cem\u003eI. uliginosa\u003c/em\u003e at the early stage systematically uncovered key molecular mechanisms of spur morphogenesis. We identified 8,592 DEGs, of which four were differentially expressed in every pairwise comparison, implying their potential function as core regulators of spur development. Enrichment analysis revealed significant DEG enrichment in pathways including \u0026ldquo;Plant-pathogen interaction,\u0026rdquo; \u0026ldquo;MAPK signaling pathway \u0026ndash; plant,\u0026rdquo; \u0026ldquo;Plant hormone signal transduction,\u0026rdquo; \u0026ldquo;Phenylpropanoid biosynthesis,\u0026rdquo; and \u0026ldquo;Monoterpenoid biosynthesis\u0026rdquo; in comparisons between WT and mutant spurs. These results indicate that early spur development involves active cell division, growth, and specific secondary metabolism\u0026mdash;consistent with earlier reports in \u003cem\u003eI. uliginosa\u003c/em\u003e, \u003cem\u003eAquilegia\u003c/em\u003e, and \u003cem\u003eLinaria\u003c/em\u003e [8, 9, 11, 22]. In comparisons between the WT spur and mutant main spur, pathways such as \u0026ldquo;Plant-pathogen interaction,\u0026rdquo; \u0026ldquo;Plant hormone signal transduction,\u0026rdquo; and \u0026ldquo;MAPK signaling pathway \u0026ndash; plant\u0026rdquo; remained consistently enriched, suggesting that main spur sustain vigorous elongation alongside basic immune and signaling networks. By contrast, DEGs from lateral spur were prominently enriched in \u0026ldquo;Photosynthesis \u0026ndash; antenna proteins,\u0026rdquo; \u0026ldquo;Plant hormone signal transduction,\u0026rdquo; \u0026ldquo;DNA replication,\u0026rdquo; and \u0026ldquo;Plant-pathogen interaction,\u0026rdquo; indicating that lateral spur exhibit active cell division, rapid elongation, and heightened photosynthetic capacity. Notably, \u0026ldquo;Plant hormone signal transduction\u0026rdquo; was significantly enriched across all spur-type comparisons. This underscores the plant hormone signaling network as a central regulatory hub coordinating morphological variation in \u003cem\u003eI. uliginosa\u003c/em\u003e spurs, likely by integrating developmental and defense signals to ultimately shape distinct spur morphologies.\u003c/p\u003e \u003cp\u003eTFs act as central hubs within gene regulatory networks and are pivotal in shaping complex plant traits. In \u003cem\u003eI. uliginosa\u003c/em\u003e, we identified 955 differentially expressed TFs representing 34 families. The expression patterns of key families-including AP2/ERF, MYB, bHLH, WRKY, NAC, and C2C2 were closely associated with the developmental states of the WT single spur and the main spur of 2SM and 3SM, offering important insights into the molecular mechanisms underlying the morphogenesis and variation of this specialized floral organ. The AP2/ERF family functions as a global regulator of plant development and stress responses, integrating multiple signals via its \u0026ldquo;identity determination-cell differentiation\u0026ndash;senescence regulation\u0026rdquo; network [23, 24]. In this work, AP2/ERF genes were consistently highly expressed in the WT single spur and the main spur of mutants, strongly implicating this family in the early fate determination and initial differentiation of spur primordia. This pattern likely reflects the established role of AP2/ERF members in regulating cell-cycle genes and influencing proliferation and differentiation. Therefore, we propose that AP2/ERF TFs may serve as key \u0026ldquo;initiators\u0026rdquo; and \u0026ldquo;coordinators\u0026rdquo; of early spur development, potentially responding to upstream signals and activating downstream genes involved in cell division and differentiation, thereby establishing the cellular basis for spur outgrowth and elongation. Both MYB and bHLH families exhibited specific high expression in the main spur of 2SM and 3SM. The MYB family, with its dual \u0026ldquo;activation\u0026ndash;repression\u0026rdquo; regulatory mode, acts as a hub for spatiotemporal precision in organ development [25], whereas bHLH factors are critical for processes such as organogenesis and pigment deposition [26]. Their co-expression in mutant main spur suggests they may form a functional module that co-regulates developmental pathways either absent or less active in WT plants. For instance, they could jointly activate target genes associated with cell elongation, directional growth, or specific metabolic pathways, thereby driving distinct morphological or growth dynamics in mutant main spur. This differential expression provides direct molecular evidence of diverging developmental mechanisms between WT and mutant spurs. Expression patterns of the WRKY and NAC families point to their potential roles in generating spur morphological diversity. WRKY genes were specifically highly expressed in the lateral spur of the WT and the 2SM. As signaling hubs that integrate environmental and endogenous cues [27, 28], their specific expression in lateral spur implies that lateral spur initiation or development may require the reception and integration of a unique set of signals, which could be essential for the formation of additional (lateral) spurs in mutants. In contrast, NAC genes were highly expressed in all mutant spurs. NAC factors participate in multiple processes, including organ boundary establishment, morphogenesis, and senescence [29, 30]. Their broad up-regulation suggests that the entire developmental program of mutant spurs\u0026mdash;from primordium boundary setting to final morphogenesis\u0026mdash;may be substantially reshaped by the NAC regulatory network, thereby stabilizing the multi-spur architecture. This study further revealed that the C2C2 family displayed significant expression differences among spurs while maintaining generally high expression levels. This observation aligns closely with reports in Aquilegia, where C2C2 members are involved in regulating spur presence/absence [17], strongly supporting an ancient, conserved core regulatory role for this family in angiosperm spur development. The widespread yet differential expression of C2C2 genes suggests that distinct members may fine-tune spur initiation, polar growth, and final morphology\u0026mdash;possibly by establishing tissue polarity, modulating hormone (e.g., auxin) gradients, or responding to specific developmental signals\u0026mdash;thereby representing key targets for understanding spur evolution and development.\u003c/p\u003e \u003cp\u003eTo identify potential regulators of floral spur variation in \u003cem\u003eI. uliginosa\u003c/em\u003e, 10 candidate genes were screened, eight of which were validated using qRT-PCR. This validation confirmed the reliability of the transcriptome data. All eight genes exhibited significantly differential expression across distinct spur types in both WT and mutant plants.The GID2 gene, which participates in the gibberellin (GA) signaling pathway, functions in plant growth and development by coordinating with other \u003cem\u003eGID\u003c/em\u003e family proteins to regulate cell division and elongation [31]. In this study, \u003cem\u003eGID2\u003c/em\u003e expression was significantly higher in the main spur compared to lateral spur, suggesting it may contribute to spur elongation and could partially account for the greater length of main spur relative to lateral spur. Both \u003cem\u003eGRP5\u003c/em\u003e and \u003cem\u003ePDF1\u003c/em\u003e are implicated in growth, defense, and stress responses. \u003cem\u003eGRP5\u003c/em\u003e encodes a glycine-rich protein likely involved in stress response and cell-wall maintenance [32]. \u003cem\u003ePDF1\u003c/em\u003e, a membrane component, shows preferential expression during Gossypium hirsutum fiber initiation and early elongation, where \u003cem\u003eGbPDF1\u003c/em\u003e interacts with \u003cem\u003eGhMYB25-like\u003c/em\u003e to modulate H-O homeostasis, ethylene signaling, and pectin biosynthesis [33]. Here, \u003cem\u003eGRP5\u003c/em\u003e expression was significantly elevated in TwSs, whereas \u003cem\u003ePDF1\u003c/em\u003e was consistently lower in main spur than in lateral spur across all mutant backgrounds-a pattern indicating their potential roles in spur morphological variation. The \u003cem\u003eSOB5\u003c/em\u003e gene, involved in cytokinin biosynthesis, belongs to a family of plant-specific small proteins. Its overexpression elevates endogenous cytokinin levels by upregulating \u003cem\u003eAtIPT3/7\u003c/em\u003e, significantly altering growth and hormonal homeostasis in \u003cem\u003eArabidopsis\u003c/em\u003e [34]. In \u003cem\u003eI. uliginosa\u003c/em\u003e, \u003cem\u003eSOB5\u003c/em\u003e expression was lower in the WT than in mutants, and significantly lower compared to 2SM, implying a possible function in variant spur formation. The \u003cem\u003eLTP\u003c/em\u003e, \u003cem\u003eSOC1\u003c/em\u003e, and \u003cem\u003eTCP4\u003c/em\u003e genes each contribute to distinct aspects of plant development. \u003cem\u003eLTP\u003c/em\u003e participates in wax or cutin deposition within expanding epidermal cells and secretory tissue cell walls. \u003cem\u003eSOC1\u003c/em\u003e, a MADS-box protein closely tied to floral development, is widely studied in flowering and fruit ripening. \u003cem\u003eTCP4\u003c/em\u003e has been established as a regulator of spur development in \u003cem\u003eAquilegia\u003c/em\u003e and \u003cem\u003eLinaria\u003c/em\u003e [8, 9, 35, 36]. All three genes displayed a consistent expression pattern in \u003cem\u003eI. uliginosa\u003c/em\u003e, with higher transcript levels in the WT than in mutants, suggesting they may function similarly in the molecular regulation of spur variation. Furthermore, the TF \u003cem\u003eNAC2\u003c/em\u003e showed elevated expression in mutants relative to the WT, with particularly high levels in the main spurs of 2SM and 3SM. Given its known roles in leaf senescence, floral development, lateral root formation, and secondary wall thickening [37\u0026ndash;39], \u003cem\u003eNAC2\u003c/em\u003e may also be involved in regulating lateral spur formation in \u003cem\u003eI. uliginosa\u003c/em\u003e.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study investigated the developmental dynamics of floral spurs in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e. Both spur types followed a typical sigmoidal growth curve, with lateral spur significantly shorter than main spur. Rapid spur development commenced at the early stage, decelerated by early flowering, and stabilized by anthesis stage. Cytological analyses revealed similar cellular development mechanisms in WT and mutant spurs: early spur formation depends mainly on cell division, whereas later length increase relies predominantly on cell elongation; the formation of internal cell protrusions is largely driven by cellular anisotropy. We further conducted transcriptome sequencing of spurs at the early stage in both genotypes. Cluster analysis and functional enrichment of 8,592 differentially expressed genes identified candidate genes associated with spur variation. Our results indicate that hormones play a pivotal role in spur variation, whereas genes involved in cell elongation, division, and the cell cycle are among the most critical factors shaping spur morphology. This study offers the first molecular insights into spur variation in the genus \u003cem\u003eImpatiens\u003c/em\u003e, providing valuable information and a theoretical foundation for understanding spur diversity in dicotyledonous plants.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e "},{"header":"Methods","content":"\u003ch2\u003ePlant materials\u003c/h2\u003e\u003cp\u003ePlants bearing a single floral spur are considered the WT of \u003cem\u003eI. uliginosa\u003c/em\u003e, whereas those with two or three spurs are classified as mutants. Seeds of both WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e were collected from the area surrounding Kunming Laoyu River Wetland Park,located on the eastern shore of Dianchi Lake along East Huanhu Road, Kunming City, Yunnan Province.This site is a plateau lake-type wetland, with a central geographic location of approximately 24.83° N,102.81° E and an elevation of about 1890 m. Then we cultivated these seeds in the greenhouse at Southwest Forestry University under controlled conditions of 18–25 ℃ with a photo period of 11–13 h of light per day.\u003c/p\u003e\u003ch2\u003eObservation of spur growth dynamics\u003c/h2\u003e\u003cp\u003eFollowing the methodology described by Li et al. (\u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e), buds in which floral spurs had not yet initiated were randomly selected from 30 healthy \u003cem\u003eI. uliginosa\u003c/em\u003e plants and monitored daily, resulting in 30 biological replicates. The length of each spur was measured daily at 9:00 AM, with three technical replicates per measurement. Measurements were conducted consecutively for 15 to 25 days, depending on the developmental progression of the plants. Growth curves were plotted based on the observational data. Three key developmental stages were identified through analysis of the spur growth curves in both WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e.\u003c/p\u003e\u003ch2\u003eScanning electron microscopy and histological examination\u003c/h2\u003e\u003cp\u003eFollowing the methodology of Li et al. (\u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e), tissue samples were fixed in FAA (a mixture of 50% ethanol, glacial acetic acid, and 38% formaldehyde at an 18:1:1 ratio). Subsequent to dehydration through an ethanol gradient, the samples were critical-point dried with carbon dioxide using an EMS 850 dryer (Hatfield, PA, USA) and then imaged with a Zeiss Sigma 300 scanning electron microscope (Oberkochen, Germany). Following another ethanol gradient dehydration, the tissues were cleared in xylene, embedded in paraffin, and sectioned. Sections of 8 µm thickness were stained with 0.01% Safranin O and 50% Fast Green before being imaged under a Leica DM750 optical microscope (Wetzlar, Germany).\u003c/p\u003e\u003ch2\u003eRNA sequencing and de novo assembly of transcriptome\u003c/h2\u003e\u003cp\u003eFollowing the methodology of Li et al. (\u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e), nectary spurs from three developmental stages of both WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e were collected. Immediately upon excision, the spurs were preserved in liquid nitrogen. Each sample represented a pool of at least three biological replicates. Total RNA was extracted using Plant RNA Purification Reagent for plant tissue (Invitrogen, Carlsbad, CA, USA), followed by genomic DNA removal with DNase I (Takara). Subsequently, RNA-seq libraries were prepared using the TruSeq™ RNA Sample Preparation Kit (Illumina, San Diego, CA, USA). After quantification with a TBS380 fluorometer using Picogreen, the libraries were sequenced in a single lane on an Illumina HiSeq X Ten or NovaSeq 6000 platform (Illumina, San Diego, CA, USA) to generate 2 × 150 bp paired-end reads.\u003c/p\u003e\u003cp\u003eThe raw paired-end sequencing data were trimmed and quality controlled using SeqPrep (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/jstjohn/SeqPrep\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and Sickle (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/najoshi/sickle\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) with default parameters. Clean data were de novo assembled using Trinity (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://trinityrnaseq.sourceforge.net/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and homologous clustering of the assembled transcripts was performed, designating the longest transcript in each cluster as a unigene [40]. The assembled sequences were further filtered for optimization using TransRate (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://hibberdlab.com/transrate/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), redundant and similar sequences were removed using CD-HIT (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://weizhongli-lab.org/cd-hit/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and the completeness of the transcriptome assembly was evaluated using the Benchmarking Universal Single-Copy Orthologs tool (BUSCO, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://busco.ezlab.org\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\u003ch2\u003eFunctional annotation\u003c/h2\u003e\u003cp\u003eAll assembled transcripts were functionally annotated by querying the NR, Swiss-Prot, Pfam, and COG databases. The BLASTX software was used to identify proteins with the highest similarity to the given transcript sequences, with a cutoff E-value typically set to less than 1.0×10⁻⁵. GO annotations for the unique assembled transcripts were obtained using the BLAST2GO program (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.blast2go.com/b2ghome\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [41] to describe biological processes, molecular functions, and cellular components. Metabolic pathway analysis was conducted using the KEGG database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.genome.jp/kegg/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [42].\u003c/p\u003e\u003ch2\u003eDiferential expression analysis and functional enrichment\u003c/h2\u003e\u003cp\u003eThe expression levels of genes and transcripts were calculated using the TPM (transcripts per million reads) / FPKM (fragments per kilobase of transcript per million reads) method, and the quantification of gene and transcript abundance was performed using RSEM (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://deweylab.biostat.wisc.edu/rsem/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [43]. Differential expression analysis was conducted using DESeq2 [44] / DEGseq [45] / EdgeR [46], with a threshold of Q value ≤ 0.05, |log2FC| \u0026gt; 1, and genes with Q value ≤ 0.05 (DESeq2 or EdgeR) / Q value ≤ 0.001 (DEGseq) being considered significantly differentially expressed.\u003c/p\u003e\u003cp\u003eGO and KEGG functional enrichment analyses were performed on differentially expressed genes. GO functional enrichment analysis was carried out using Goatools (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/tanghaibao/GOatools\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), while KEGG pathway analysis was performed using KOBAS (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://kobas.cbi.pku.edu.cn/home.do\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [47]. Fisher's exact test was utilized for calculations, and P-values were corrected using the Bonferroni and BH (FDR) methods. The threshold for corrected P-values was set at 0.05.\u003c/p\u003e\u003ch2\u003eIdentifcation of TFs and cluster analysis\u003c/h2\u003e\u003cp\u003eThrough HMMER analysis, the domain information of transcripts was compared with the PlantTFDB (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://planttfdb.cbi.pku.edu.cn/\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) database to obtain homologous TF information for gene TF prediction and family analysis. Hierarchical clustering was conducted on differentially expressed TFs.\u003c/p\u003e\u003ch2\u003eqRT-PCR analysis of gene expression\u003c/h2\u003e\u003cp\u003eTotal RNA was extracted from five flower distances of both WT and mutant Daphne retusa using the E.Z.N.A.® Plant RNA Kit from Omega. The first-strand cDNA was synthesized as a template using the EasyScript® One-Step gDNA Removal SuperMix from TransGen. qRT-PCR amplification was performed using the Roche LightCycler®480 II real-time quantitative PCR detection system, in combination with the Hieff® qPCR SYBR® Green Premix from Yisense. Details of the qRT-PCR amplification primers are provided in Supplementary Data Table S7, with IuActin serving as the housekeeping gene. The fluorescence quantitative PCR detection utilized a three-step method, with three technical replicates for each sample. The amplification protocol was as follows: pre-denaturation at 95°C for 5 minutes, followed by 40 cycles (denaturation at 95°C for 10 seconds, annealing at 60°C for 20 seconds, and extension at 72°C for 20 seconds). Gene expression levels were ultimately calculated using 2\u003csup\u003e−ΔΔCt\u003c/sup\u003e method.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNCBI Non-Redundant Protein Sequence Database\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePfam\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eProtein Families\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCOG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eClusters of Orthologous Groups of proteins\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGene Ontology\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eKEGG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eKyoto Encyclopedia of Genes and Genomes.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Shanghai Majorbio Bio-pharm Technology Co.,Ltd. for its help in sequencing. The data were analyzed through the free online platform of Majorbio Cloud Platform (www.majorbio.com).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYL, XLZ and BH were responsible for the experimental design. YL, XLZ and BH carried out sample collection, experiments, data analysis and article writing. LQZ, HYL and Zhijia Gu participated in the experiment. MJH and HQH supervised the research and revised the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China (grant NO. 32560389), Yunnan Fundamental Research Projects (grant NO. 202501AS070052), the Key Project of Yunnan Provincial Agricultural Joint Special Program (grant NO. 202301BD070001-011), and the Project of High-level Introduction talents in Yunnan Province.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSequence data that support the findings of this study have been deposited in the National Center for Biotechnology Information (NCBI). The project access number is PRJNA1422964 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1422964).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor details\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e1 College of Landscape Architecture and Horticulture Sciences, Yunnan Key Laboratory of Landscape Plant Resource Cultivation and Application, Yunnan Province Engineering Research Center for Functional Flower Resources and Industrialization, Research and Development Center of Landscape Plants and Horticulture Flowers, Southwest Forestry University, Kunming 650224, China\u003c/p\u003e\n\u003cp\u003e2 Key Laboratory for Plant Biodiversity and Biogeography of East Asia, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650201, China\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data and materials that support the findings of this study are available in the text and public data\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eFigueiredo ACS, Pais MS. Ultrastructural aspects of the nectary spur of \u003cem\u003eLimodorum abortivum\u003c/em\u003e (L) Sw. (Orchidaceae). Ann Bot. 1992;70:325\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePacini E, Nepi M, Vesprini JL. Nectar biodiversity: a short review. Plant Syst Evol. 2003;238:7\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStang M, Klinkhamer PG, Meijden VDE. Asymmetric specialization and extinction risk in plant-flower visitor webs: a matter of morphology or abundance? Oecologia. 2007;151(3):442\u0026ndash;53.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTsai T, Diggle PK, Frye HA, Jones CS. Contrasting lengths of Pelargonium floral nectar tubes result from late differences in rate and duration of growth. Ann Bot. 2018;121(3):549\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMack JK, Davis AR. The relationship between cell division and elongation during development of the nectar-yielding petal spur in \u003cem\u003eCentranthus ruber\u003c/em\u003e (Valerianaceae). Ann Bot. 2015;115(4):641\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePuzey JR, Gerbode SJ, Hodges SA, Kramer EM, Mahadevan L. Evolution of spur-length diversity in \u003cem\u003eAquilegia\u003c/em\u003e petals is achieved solely through cell-shape anisotropy. Proc. Biol. Sci. 2012; 279(1733): 1640-5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharma B, Yant L, Hodges SA, Kramer EM. Understanding the development and evolution of novel floral form in \u003cem\u003eAquilegia\u003c/em\u003e. Curr Opin Plant Biol. 2014; 1722\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYant L, Collani S, Puzey J, Levy C, Kramer EM et al. Molecular basis for three-dimensional elaboration of the \u003cem\u003eAquilegia\u003c/em\u003e petal spur. Proc. Biol. Sci. 2015; 282(1803): 20142778.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCullen E, Fern\u0026aacute;ndez-Mazuecos M, Glover BJ. Evolution of nectar spur length in a clade of \u003cem\u003eLinaria\u003c/em\u003e reflects changes in cell division rather than in cell expansion[J]. Ann Botany. 2018;122(5):801\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Huang LW, Li YX, Zhang YD, Meng DC, Wei CM, et al. The cellular and molecular basis of the spur development in \u003cem\u003eImpatiens uliginosa\u003c/em\u003e[J]. Hortic Res. 2024;11(3):uhae015\u0026ndash;015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Wei CM, Li XY, Meng DC, Gu ZJ, Su PQ, et al. De novo transcriptome sequencing of \u003cem\u003eImpatiens uliginosa\u003c/em\u003e and the analysis of candidate genes related to spur development[J]. BMC Plant Biol. 2022;22(1):553\u0026ndash;553.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou ZL, Duan YW, Luo Y, Yang YP, Zhang ZQ. Cell number explains the intraspecific spur-length variation in an \u003cem\u003eAquilegia\u003c/em\u003e species[J]. Plant Divers. 2019;41(5):307\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGolz JF, Keck EJ, Hudson A. Spontaneous Mutations in \u003cem\u003eKNOX\u003c/em\u003e Genes Give Rise to a Novel Floral Structure. Antirrhinum[J] Curr Biology. 2002;12(7):515\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBox MS, Dodsworth S, Rudall PJ, Bateman RM, Glover BJ. Characterization of Linaria \u003cem\u003eKNOX\u003c/em\u003e genes suggests a role in petal-spur development.[J]. Plant journal: cell Mol biology. 2011;68(4):703\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBox MS, Dodsworth S, Rudall PJ, Bateman RM, Glover BJ. Flower-specific \u003cem\u003eKNOX\u003c/em\u003e phenotype in the orchid \u003cem\u003eDactylorhiza fuchsii\u003c/em\u003e[J]. J Exp Bot. 2012;63(13):4811\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDamerval C, Citerne H, Le GM, Domenichini S, Dutheil J, Ronse de CL, et al. Asymmetric morphogenetic cues along the transverse plane: shift from disymmetry to zygomorphy in the flower of Fumarioideae[J]. Am J Bot. 2013;100(2):391\u0026ndash;402.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBallerini ES, Min Y, Edwards MB, Kramer EM, Hodges SA. POPOVICH, encoding a C2H2 zinc-finger transcription factor, plays a central role in the development of a key innovation, floral nectar spurs, in Aquilegia[J]. Proc Natl Acad Sci USA. 2020;117(36):22552\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi XY, Huang H, Zhang ZQ. Characterization and expression analysis of bHLH transcription factors reveal their putative regulatory effects on nectar spur development in \u003cem\u003eAquilegia\u003c/em\u003e species[J]. Gene, 2022, 852147057\u0026ndash;147057.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMichael B, Štěp\u0026aacute;n J. Pollinator-induced twisting of flowers sidesteps floral architecture constraints[J]. Curr Biol. 2014;24(17):R795\u0026ndash;795.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVervoort A, Cawoy V, Jacquemart AL. Comparative Reproductive Biology in Co-Occurring Invasive and Native \u003cem\u003eImpatiens\u003c/em\u003e Species[J]. Int J Plant Sci. 2011;172:366\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChupp AD, Battaglia LL, Schauber EM, et al. Orchid\u0026ndash;pollinator interactions and potential vulnerability to biological invasion[J]. Aob Plants. 2015;7:99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConway SJ, WalcherChevillet CL, Barbour KS, Kramer EM, et al. Brassinosteroids regulate petal spur length in \u003cem\u003eAquilegia\u003c/em\u003e by controlling cell elongation[J]. Ann Botany. 2021;128(7):931\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu CC, Wang FF, Li RH, Zhu Y, Zhang CL, He YH. Marigold (\u003cem\u003eTagetes erecta\u003c/em\u003e) MADS-Box Genes: A Systematic Analysis and Their Implications for Floral Organ Development[J]. Agronomy. 2024;14(9):1889\u0026ndash;1889.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoody E, Moyroud E. Evolution of petal patterning: blooming floral diversity at the microscale[J]. New Phytol. 2025;247(6):2538\u0026ndash;56.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJae UH, Seonghwan K, Heejeong S, June MK. MYB Transcription Factors in Plant Developmental Plasticity[J]. J Plant Biology. 2025;68(6):1\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNing M, Li QY, Wang Y, Li QM, Tao YH, Zhang F, et al. Alternative splicing drives the functional diversification of a bHLH transcription factor in the control of growth and drought tolerance in rice[J]. Sci Bull. 2024;70(2):153\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao LR, Liu YW, Zhu Y, Chen SD, Du Y, Deng LY, et al. Transcription factor OsWRKY11 induces rice heading at low concentrations but inhibits rice heading at high concentrations[J]. J Integr Plant Biol. 2024;66(7):1385\u0026ndash;407.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen XJ, Zhang TL, Wang H, Zhao WS, Guo ZJ. Transcription factor WRKY complexes in plant signaling pathways[J]. Phytopathol Res. 2025;7(1):54\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong HY, He HD, Chang Y, Miao BB, Liu ZW, Wang QQ, et al. Multiple roles of NAC transcription factors in plant development and stress responses[J]. J Integr Plant Biol. 2025;67(3):510\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa CD, Zhou PY, Ma YF, Tan W, Huang X, Segbo S, et al. A NAC family gene \u003cem\u003ePmNAC32\u003c/em\u003e associated with photoperiod promotes flower induction in \u003cem\u003ePrunus mume\u003c/em\u003e[J]. Hortic Res. 2025;12(9):uhaf157.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu RH, An JY, Wang YJ, Zhao XB, Li JR, Zhong TH, et al. OsGRF11 regulates cell division and elongation by interacting with GF14e and affecting gibberellin signaling[J]. Plant physiology; 2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang HJ, Lee YS, Choi IS. Comparison of physicochemical properties and antioxidant activities of fermented soybean-based red pepper paste, Gochujang, prepared with five different red pepper (\u003cem\u003eCapsicum annuum\u003c/em\u003e L.) varieties[J]. J Food Sci Technol. 2018;55(2):792\u0026ndash;801.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao G, Le Y, Sun M, Xu JW, Qin Y, Men S, et al. A dominant negative mutation of \u003cem\u003eGhMYB25-like\u003c/em\u003e alters cotton fiber initiation, reducing lint and fuzz[J]. The Plant cell; 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang JY, Wrage EL, Vankova R, Malbeck J, Neff MM. Over-expression of \u003cem\u003eSOB5\u003c/em\u003e suggests the involvement of a novel plant protein in cytokinin-mediated development[J]. Plant journal: cell Mol biology. 2006;46(5):834\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu YH, Guo LS, Wei ZY, Li YJ, Zhang Y, Zheng JY et al. Pan-genomic insights into LTP gene family evolution across diploid cotton species[J]. Front Plant Sci, 2025, 16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan X, Wang D, Song GQ. Expression of a maize \u003cem\u003eSOC1\u003c/em\u003e gene enhances soybean yield potential through modulating plant growth and flowering[J]. Sci Rep. 2021;11:12758.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGhafari RF, Vaziri A, Hassanpour AM, Talesh SS, Olfati J. Effects of humic acid on antioxidant defense system and senescence-related genes expression in petals of LA-Hybrid Lily[J]. J Plant Biochem Biotechnol. 2021;31(2):1\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang N, Wu PY, Cao ZY, Liu YN, Zhang X, Lou J, et al. A NAC transcription factor ZaNAC93 confers floral initiation, fruit development, and prickle formation in \u003cem\u003eZanthoxylum armatum\u003c/em\u003e[J]. Plant physiology and biochemistry: PPB; 2023. pp. 201107813\u0026ndash;107813.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePereira SS, Guimar\u0026atilde;es FCM, Carvalho JFC, Stolf-Moreira R, Oliveira MCN, Rolla AAP, et al. Transcription factors expressed in soybean roots under drought stress[J]. Genet Mol research: GMR. 2011;10(4):3689\u0026ndash;701.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome[J]. Nat Biotechnol. 2011;29:644\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConesa A, G\u0026ouml;tz S, Garc\u0026iacute;a-G\u0026oacute;mez JM, Terol J, Tal\u0026oacute;n M, Robles M. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005;21:3674\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi B, Dewey CN. RSEM: accurate transcript quantification from RNASeq data with or without a reference genome. BMC Bioinformatics. 2011;12:323.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLove MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang L, Feng Z, Wang X, Wang X, Zhang X. DEGseq: an R package for identifying differentially expressed genes from RNA-seq data. Bioinfor matics. 2009;26:136\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie C, Mao X, Huang J, Ding Y, Wu J, Dong S, et al. KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Res. 2011;39:W316\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-plant-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pbio","sideBox":"Learn more about [BMC Plant Biology](http://bmcplantbiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pbio/default.aspx","title":"BMC Plant Biology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Impatiens uliginosa, Variant spur, Spur development, Spur morphology, Transcriptome","lastPublishedDoi":"10.21203/rs.3.rs-8829929/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8829929/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe spur, a key morphological structure in \u003cem\u003eImpatiens uliginosa\u003c/em\u003e, plays important functional roles in attracting specific pollinators, enhancing pollination efficiency, and facilitating interspecific reproductive isolation. While its ecological and evolutionary significance has been extensively studied, the cytological and molecular mechanisms underlying spur morphogenesis, particularly in non-model plants, remain poorly understood. In this study, we used wild type (WT) bearing a single spur, two-spur mutant (2SM) and three-spur mutant (3SM) plants of \u003cem\u003eI. uliginosa\u003c/em\u003e. By integrating morphometric analysis, cytological examination, and transcriptome sequencing, we identified candidate genes and hormonal regulatory networks associated with spur variation at the molecular level for the first time. This study provides new insights into the molecular basis of spur formation in \u003cem\u003eI. uliginosa\u003c/em\u003e and the genus \u003cem\u003eImpatiens\u003c/em\u003e more broadly.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe developmental dynamics of spurs was analyzed both in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e, indicating that spur growth follows a typical sigmoidal curve, with lateral spur being significantly shorter than the main spur. The cellular development mechanisms were similar between the two spur types: initial spur formation was predominantly driven by cell division, whereas subsequent elongation primarily depended on cell expansion, with the formation of internal cellular protrusions regulated by anisotropic cell growth. Transcriptome sequencing of spurs at the early stage yielded 32.69 Gb of high-quality data, from which 42,721 unigenes were assembled. Functional annotation against the NR, Swiss-Prot, Pfam, COG, GO, and KEGG databases resulted in the annotation of 24,031 genes. Differential expression analysis identified 8,592 differentially expressed genes (DEGs), which were enriched in 371 GO terms and 111 KEGG pathways. Notably, the \u0026ldquo;plant hormone signal transduction\u0026rdquo; pathway showed the highest enrichment in the mutant spurs. A total of 955 transcription factors (TFs) belonging to 34 families, including MYB, AP2/ERF, and TCP, were identified. Through screening and qRT-PCR validation, eight of the 10 candidate genes, such as \u003cem\u003eLTP\u003c/em\u003e, \u003cem\u003eGRP5\u003c/em\u003e, \u003cem\u003ePDF1\u003c/em\u003e, and \u003cem\u003eGID2\u003c/em\u003e, were confirmed to be involved in spur variation.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eOur study elucidates the morphological and cellular developmental mechanisms of the spurs in WT and mutant \u003cem\u003eI. uliginosa\u003c/em\u003e, and identifies a series of candidate genes associated with spur variation, including cell cycle, cell division, cell elongation, and plant hormones. The findings provide valuable data and resources for further unraveling the molecular mechanisms underlying spur variation in \u003cem\u003eImpatiens\u003c/em\u003e species.\u003c/p\u003e","manuscriptTitle":"Integrated cytological and transcriptomic analyses reveal the molecular basis of spur variation in Impatiens uliginosa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-27 04:37:21","doi":"10.21203/rs.3.rs-8829929/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-30T10:09:19+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-26T09:36:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"27147200440204176904242947371226594402","date":"2026-03-19T07:03:32+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-16T16:50:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"76374841217291995840729442256939432833","date":"2026-02-25T11:53:05+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"12909353186504353841469224469173211187","date":"2026-02-25T01:42:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"273251853776398522416084958878965761575","date":"2026-02-24T20:30:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-24T20:07:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-24T11:37:39+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-16T04:30:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-14T08:27:25+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Plant Biology","date":"2026-02-14T08:23:22+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-plant-biology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pbio","sideBox":"Learn more about [BMC Plant Biology](http://bmcplantbiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pbio/default.aspx","title":"BMC Plant Biology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e34de735-8d41-4b78-88ba-278fcf82fd96","owner":[],"postedDate":"February 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-18T04:56:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-27 04:37:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8829929","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8829929","identity":"rs-8829929","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}