Comparative Analysis of Ratoon-Competent and Ratoon-Deficient Sugarcane by Hormonal and Transcriptome Profiling

Abstract

The ratooning capacity of sugarcane cultivars represents a crucial agronomic trait that significantly influences the sustainability of crop yields. This study elucidates the physiological and molecular mechanisms underlying the sugarcane ratooning ability observed in ratoon-competent GuiTang 29 (GT29) and ratoon-deficient Badila cultivars following stem excision. Through integrated hormonal profiling and transcriptome analysis, we identified significant differences in hormone levels and gene expression patterns. The quantification of 15 endogenous hormones via HPLC revealed marked reductions in zeatin (ZA) and zeatin riboside (ZR) in both cultivars. Additionally, GT29 exhibited notable reductions in gibberellins (GA3 and GA5) and strigolactone (5-DS) post-stem-excision, while Badila displayed stable or distinct hormonal changes. Additionally, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses indicated that hormone signal transduction, MAPK signaling pathways, phenylpropanoid biosynthesis, flavonoid biosynthesis, and other metabolic pathways were significantly enriched in both GT29 and Badila, with a particularly higher enrichment of plant hormone signal transduction in GT29. Furthermore, several differentially expressed genes (DEGs) had different expression patterns between GT29 and Badila, including the cytokinin receptor B-ARR and transcription factor A-ARR, gibberellin pathway components GID1 and DELLA, and AUX/IAA and SAUR in the auxin pathway. The real-time quantitative PCR (qRT-PCR) validation of 12 DEGs corroborated the RNA-seq data, further supporting the reliability of the transcriptomic analysis. This study delineates a clear molecular framework distinguishing ratoon competence, offers novel insights into the molecular basis of perennial regeneration and provides reliable candidate genes for functional marker development in sugarcane breeding.

1. Introduction

Sugarcane (Saccharum spp. hybrids), a globally vital C4 crop, supplies approximately 80% of commercial sugar and 40% of bioethanol production [1]. The ratooning process in sugarcane fundamentally depends on the activation of tiller buds at basal stem nodes, where dormant buds develop into productive tillers that ultimately form sucrose-accumulating stalks. This bud sprouting mechanism directly determines ratoon longevity and yield potential. Globally, ratoon sugarcane occupies over 75% of the total cultivation area, typically maintained through production cycles of 2 to 7 years. In contrast, China’s ratoon systems cover 60% of the sugarcane area but persist for only 1–2 years, significantly constraining industrial efficiency and cost competitiveness [2]. Therefore, developing ratoon-competent cultivars through genomic breeding has become a strategic priority for sustainable sugarcane production.

In sugarcane production systems, ratooning ability serves as a key agronomic trait for evaluating the regeneration capacity and economic viability of sugarcane varieties. Varieties that exhibit ratoon competency demonstrate a higher underground bud germination rate and vigorous tillering capacity, which facilitate the rapid establishment of a robust seedling population and ensure uniform field stand development [3]. Under appropriate cultivation and management practices, the yield of ratoon crops from such varieties can approach or even surpass that of plant cane, reflecting their high yield stability and productive potential [4]. From an economic perspective, ratoon-competent varieties substantially reduce production costs by minimizing the need for repeated plowing, seed cane procurement, and replanting [5]. In contrast, varieties with ratoon-deficient performance often experience low shoot emergence rates, necessitating replanting, which increases input costs and results in reduced yields in successive ratoon crops. This decline in productivity hampers the sustainability of sugarcane production systems [6]. Therefore, the development and adoption of ratoon-competent sugarcane cultivars can significantly enhance regenerative capacity, improve yield performance, lower production expenditures, and promote the sustainable development of the sugarcane industry.

The dynamics of basal bud sprouting govern both primary crop yield and ratoon regeneration. The activation of underground buds establishes new growth points, while residual stem buds compensate for yield losses related to lodging. Crucially, the sprouting vigor of these latent buds directly determines the effective stalk numbers. Understanding the molecular drivers of bud activation thus holds key implications for yield enhancement [7]. Previous studies have established correlations between ratoon competence and tillering capacity [8,9], with distinct phytohormonal profiles observed in ratoon-competent versus -deficient cultivars [10]. In rice, key tillering-related genes including OsMAX1a, OsMAX1b, OsMAX1c, OsMAX1d, and OsMAX1e [11] have been characterized, providing foundational genetic resources. Previous studies have identified quantitative trait loci (QTLs) and candidate genes associated with tillering and ratoon ability in sugarcane through genetic maps constructed using high-density SNP microarrays [12]. Li et al. employed transcriptomics and metabolomics approaches to investigate the molecular basis underlying superior sugarcane traits [13]. Additionally, genome-wide association studies (GWASs) and genomic prediction have identified candidate genes and molecular markers linked to ratoon ability, offering preliminary data to support genome-assisted breeding efforts aimed at improving ratoon traits in sugarcane [14]. Through transcriptomic and phytohormonal analyses of 23 rice varieties, Henry et al demonstrated that auxin plays a pivotal role in the formation of perennial roots [15]. However, the molecular mechanisms underlying basal bud activation in sugarcane ratoons remain poorly elucidated.

Given this, this study will focus on two sugarcane cultivars with significant differences in ratoon performance. Plant endogenous hormone and transcriptome sequencing analyses were conducted by truncating stems at the jointing stage, collecting tissues after 1 d, 3 d, and without truncation. From both physiological and transcriptomic perspectives, the characteristics of the experimental materials after truncation regarding changes in plant endogenous hormones, differential gene expression, and metabolic pathway response were analyzed. Subsequently, the key metabolic pathways and candidate genes affecting root performance were identified. The aim of this study was to analyze the mechanism of rapid sprouting of basal buds of ratoon-competent sugarcane through molecular and physiological studies, so as to lay a foundation for the cultivation of ratoon-competent sugarcane.

2. Materials and Methods

2.1. Materials

Two sugarcane cultivars, GuiTang29 (GT29) and Badila were selected as the research materials. The sugarcane cultivar GT29 possesses moderate drought tolerance, high sucrose content, robust ratooning ability, and strong potential for effective stem development [16]. In contrast, the cultivar Badila exhibits poor performance under drought stress, which may adversely impact its ratooning capacity [17]. Furthermore, ratoon yield in Badila significantly declined following disease infection, indicating its susceptibility and limited regeneration potential [18]. Both cultivars’ germplasm materials were obtained from the national sugarcane germplasm resources nursery at the Sugarcane Research Institute, Yunnan Academy of Agricultural Sciences, Kaiyuan, China.

2.2. Methods

Single-bud stems derived from the two sugarcane germplasm materials were cultivated in plastic pots with two stems per pot starting on 15 May 2024. All plants were maintained under standard greenhouse conditions. Following stem emergence, specimens exhibiting uniform growth and morphological normality were selected for stem excision. At 110 days post-planting, when the plants reached the jointing stage (Figure 1), experimental stem excision was performed. In the genotype GT29, which exhibits normal regenerative capacity, bud germination was observed within three days following stem decapitation. In contrast, the ratoon-deficient Badila exhibited no regenerative response. Basal tissues from the main stem stubble were collected at 0 (control), 1, and 3 days post-excision, immediately frozen in liquid nitrogen, and maintained in triplicate biological replicates. The samples were subsequently divided for parallel analyses: one batch underwent transcriptome sequencing at BGI Genomics (Shenzhen, China; accessed on 18 October 2024), while another batch was processed for phytohormone profiling at Suzhou Keming Biotechnology Co., Ltd. (Suzhou, China; accessed on 20 October 2024), with all specimens transported under cryogenic conditions.

2.3. Determination of Hormone Content

To quantify the endogenous hormone levels, a total of fifteen plant hormones were analyzed, including four cytokinins (kinetin, KT; 6-benzylaminopurine, 6-BA; zeatin, ZA; zeatin riboside, ZR), three auxins (indole-3-acetic acid, IAA; chlorinated IAA, CIAA; 1-naphthaleneacetic acid, NAA), five gibberellins (GA₁, GA₃, GA₄, GA₅, and GA₇), two forms of abscisic acid (ABA; conjugated ABA, C-ABA), and one strigolactone (5-deoxystrigol, 5-DS).

All hormone measurements were performed by Suzhou Keming Biotechnology Co., Ltd. (Suzhou, China; accessed on 25 October 2024). Briefly, approximately 0.1 g of fresh tissue was ground in liquid nitrogen and extracted using a methanol-based extraction buffer. Following centrifugation and filtration, the supernatants were analyzed using an Agilent 1100 high-performance liquid chromatography (HPLC) system equipped with a Kromasil C18 reverse-phase column (250 mm × 4.6 mm, 5 μm). The mobile phase consisted of solvent A (0.1% formic acid in water) and solvent B (acetonitrile) under gradient elution conditions. Hormones were detected using UV absorbance at the appropriate wavelengths for each compound. Quantification was performed using external calibration with authenticated hormone standards, and both three biological and three technical replicates were conducted for hormone detection. All hormone changes were analyzed to explore their regulatory roles in the germination of basal buds following stem truncation treatment.

2.4. RNA Extraction, Transcriptome Sequencing, and Annotation

Total RNA extraction was performed by BGI, and transcriptome sequencing was conducted using the BGI BGISEQ-500 platform (Shenzhen, China; Shenzhen BGI Co., Ltd., accessed on 18 October 2024). High-quality clean reads were obtained using the SOAPnuke v1.4.0 filtering software independently developed by Shenzhen Huada Gene Co., Ltd. (Shenzhen, China), the de novo assembly of clean reads was executed using Trinity v2.1.1 assembly software, and Unigenes were obtained by clustering deredundancy of the transcripts using Tgicl-2.1. The quality of the assembled transcripts was assessed using the single-copy direct orthology database BUSCO 5.6.0. The coding regions in the Unigenes were predicted using TransDecoder V3.0.1 software. The Simple Sequence Repeat (SSR) sites of the Unigenes were detected using the MISA V1.0 software. The assembled Unigenes were annotated using BLAST software (version 2.16.0+) in seven functional databases (KEGG, GO, NR, NT, SwissProt, Pfam, and KOG).

2.5. DEGs Screening and Enrichment Analysis

Gene expression levels for each sample were calculated using RNA-Seq by Expectation-Maximization (RSEM). Principal Component Analysis (PCA) and Pearson correlation analysis were performed using the princomp function in R statistical software version 4.4.3. Differentially expressed genes (DEGs) between the treated and control samples were identified at a false discovery rate (FDR) < 0.001 and a fold change ≥ 2.0. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted for classification and enrichment assessment.

2.6. Real-Time Quantitative PCR Analysis

Samples of GT29 and Badila collected at different treatment time points were finely ground in a pre-chilled mortar using liquid nitrogen. Approximately 100 mg of the resulting powder from each sample was transferred into a 1.5 mL microcentrifuge tube for total RNA extraction. The extracted total RNA was then reverse-transcribed into first-strand cDNA, which was utilized for real-time quantitative PCR (qRT-PCR; accessed on 9 January 2025). A total of 12 DEGs were selected for qRT-PCR detection to verify the reliability of the transcriptome data. Using 18S rRNA as the reference gene, specific primers were designed based on the transcriptome sequences (

Table 1. Primer sequences for qRT-PCR.

Gene ID	Forward Sequence (5′–3′)	Reverse Sequence (5′–3′)
18S rRNA	CAACCATAAACGATGCCGA	AGCCTTGCGACCATACTCC
CL11787.Contig4	CTTGGTCTTCCAGCAGGA	ACGGAGGTGTTCCTGAAG
CL15393.Contig2	GCAAGAGGAAATACGACCC	CCCATCAGGCAAGTAACG
Unigene11582	TGGTTTGGGATTGAGCAA	CGGCACAGTAGTATGGTC
CL2511.Contig5	GAAGCCTCTGGACTACGA	ATCTTCTTGCCCTCCTTCT
Unigene97471	ATGCTGGCTTAC CTTCTC	TTCCGAATCCCTGTATGAC
CL24.Contig3	TCCTCTCCTCCTCGTACT	TCGGCTCCTCCAATACAA
CL13067.Contig31	TTCAGACAAGCGTAGCCA	TGAAGGTGTGGTCAGCAT
CL11106.Contig1	TGTCTTGCTGGTGTCGTT	TCTTGAGGCGAGTTCTGAG
CL6710.Contig7	TGGATACACCAGGGCAAA	CACACTTCTCGGCAGTTG
CL16936.Contig6	CGGATTGACGAGAATGTGTA	AAGATGCGTTGGTGTAAGTAT
CL10318.Contig30	TCCAGACATCCAGTTCAG	CTTGTTGTTCACCACCAA
CL11630.Contig1	AGAAGACCCGTAGCAAAG	GCTTTCCAACTCCACTCT

). qRT-PCR reactions were performed on an analytikjena-qTOWER2.2 fluorescence quantitative PCR instrument using the DBI Bioscience 2 SYBR^® Green Master Mix (DBI) fluorescence quantification kit (Karlsruhe, Germany). Three biological and three technical replicates were performed using the 2^−ΔΔCt method [16].

2.7. Statistical Analysis of the Data

The data were processed by the Microsoft Excel 2019 software, data variance analysis and the LSD method significance test were performed using SPSS21 software, and mapping by OriginPro2021 software.

3. Results

3.1. Characteristic Analysis of Hormone Content

3.1.1. Characteristics of CTK Content

Phytohormonal profiling revealed differential cytokinin dynamics in response to stem excision (Figure 2, Table S1). Among the four quantified cytokinins, ZA and ZR exhibited consistent depletion patterns in both ratoon-competent GT29 and -deficient Badila cultivars at 1 and 3 days after treatment, suggesting their universal regulatory role in post-excision tiller initiation. In Badila, KT content showed no significant increase during the two post-truncation periods. However, in ratoon-competent germplasm Gui-tang29, KT content exhibited a significant elevation at 1 d after truncation, followed by a subsequent increase at 3 d post-truncation. These observations suggest that KT might be associated with the rapid growth mechanism in ratoon-competent germplasms. Contrastingly, 6-BA showed no significant variations in either cultivar, excluding its direct participation in early tillering regulation.

3.1.2. Characteristics of IAA Content

Auxin serves as the primary phytohormone regulating both the rate and directionality of plant growth [19]. This regulatory molecule facilitates cell elongation through cell wall relaxation while coordinating overall plant development [20,21]. We quantified three forms of auxin—CIAA, IAA, and NAA—following stem excision in the ratoon-competent GT29 and -deficient Badila cultivars (Figure 3, Table S2). CIAA levels remained stable in both cultivars throughout the observation period, suggesting minimal involvement in the differentiation of ratoon ability. Notably, IAA dynamics differed substantially between two cultivars; Badila displayed an initial decrease 1 day post-stem-excision followed by a significant elevation at day 3, while GT29 exhibited sustained increases from day 1 onward. These temporal patterns suggested superior IAA response kinetics in the ratoon-competent genotype, potentially facilitating rapid basal bud germination. NAA concentrations showed no significant variation in Badila but demonstrated a transient elevation at day 1 in GT29 before returning to baseline levels, indicating a possible involvement in the early regeneration processes of ratoon-competent cultivars.

3.1.3. Characteristics of GA Content

The variations in five GA species were quantified following treatment (Figure 4, Table S3). Badila exhibited significant reductions in GA1 content, while GT29 maintained stable GA1 levels throughout the observation period. This stability potentially facilitated basal stem bud germination in GT29. Notably, GA3 and GA5 concentrations showed significant decreases exclusively in GT29 post-truncation, while Badila demonstrated no statistically significant alterations. Divergent temporal patterns emerged in GA4 dynamics; Badila displayed an initial decline followed by recovery, whereas GT29 exhibited sustained reduction. GA7 profiles differed markedly between the two cultivars, with Badila showing significant depletion while GT29 demonstrated a transient increase preceding subsequent decline.

3.1.4. Characteristics of ABA Content

The ABA quantification results following stem excision treatments revealed distinct dynamic patterns between the two cultivars (Figure 5, Table S4). In Badila, CABA levels significantly decreased after 3 days of treatment compared to the untreated controls. Conversely, GT29 exhibited a transient 1.8-fold increase in CABA levels at 1 day post-treatment, followed by a return to baseline levels by day 3. While ABA concentrations remained stable in Badila throughout the experimental period, GT29 displayed an initial 40% ABA reduction at day 1, with partial recovery observed at day 3. These findings suggest that moderated ABA levels may facilitate growth recovery and tiller initiation in perennial root systems with vigorous regenerative capacity.

3.1.5. Characteristics of 5-DS Content

We analyzed 5-DS content changes following stem truncation in the two sugarcane cultivars. Badila showed no significant alteration in 5-DS levels, while GT29 exhibited a remarkable 68.3% reduction from 105.88 μg·g⁻¹ FW to 33.51 μg·g⁻¹ FW (Figure 6, Table S5). This differential response suggests that 5-DS may serve as a key regulatory factor in ratoon-competent-capacity germplasm.

3.2. Transcriptome Analysis

Sequencing generated 188.42 Gb of high-quality clean reads using the BGISEQ-500 platform, with an average of 10.47 Gb per sample. The clean reads exhibited mean base quality scores of Q20 = 97.10% and Q30 = 92.48% (

Table 2. Transcriptome sequencing data statistics.

Sample ID	Total Clean Bases (Gb)	Q20 Content (%)	Q30 Content (%)	Mean Length (bp)	N50 (bp)	GC Content (%)
GT29-0 d1	10.48	97.04	92.34	1099	1678	48.43
GT29-0 d2	10.56	97.02	92.23	1104	1695	48.32
GT29-0 d3	10.60	97.05	92.47	1069	1643	49.02
GT29-1 d1	10.50	97.19	92.72	1083	1653	48.47
GT29-1 d2	10.54	97.16	92.59	1093	1687	48.35
GT29-1 d3	10.53	97.20	92.74	1062	1624	48.69
GT29-3 d1	10.47	97.25	92.85	1084	1683	48.36
GT29-3 d2	10.53	97.26	92.83	1062	1625	48.71
GT29-3 d3	10.51	97.20	92.69	1100	1694	48.43
Badila-0 d1	10.55	97.05	92.37	1203	1848	48.57
Badila-0 d2	9.59	97.09	92.45	1203	1825	48.68
Badila-0 d3	10.54	97.09	92.42	1203	1823	48.59
Badila-1 d1	10.56	97.16	92.57	1198	1828	48.34
Badila-1 d2	10.46	97.00	92.30	1204	1825	48.33
Badila-1 d3	10.51	97.00	92.28	1143	1758	48.52
Badila-3 d1	10.47	97.00	92.28	1178	1799	48.55
Badila-3 d2	10.52	97.00	92.27	1172	1780	48.73
Badila-3 d3	10.50	97.00	92.26	1200	1837	48.36
Mean	10.47	97.10	92.48	1137	1739	48.53

), demonstrating satisfactory data quality for downstream analyses. The subsequent functional annotation of Unigenes across seven databases successfully annotated a total of 198,985 Unigenes.

3.3. DEGs Screening

DESeq2 analysis was performed using ratoon-competent GT29 and ratoon-deficient Badila as controls. DEGs were identified with thresholds of FDR < 0.001 and |log2FC| ≥ 2.0. In the ratoon-deficient cultivar Badila, 3192 (1301 upregulated and 1891 downregulated) and 5049 (2182 upregulated and 2867 downregulated) DEGs were detected at 1 day and 3 days post-truncation, respectively. The ratoon-competent cultivar GT29 exhibited substantially higher DEG numbers, with 8318 DEGs at 1 day post-truncation, including 1926 upregulated and 6392 downregulated genes, and 6618 DEGs at 3 days post-truncation, consisting of 2330 upregulated and 4288 downregulated genes (Figure 7A). Notably, GT29 demonstrated significantly greater transcriptional reprogramming than Badila, particularly in downregulated genes at 1 day post-truncation. Venn analysis revealed 1522 shared DEGs between the two cultivars after 1 day, with 1670 Badila-specific and 6796 GT29-specific DEGs. At 3 days post-truncation, 1424 shared DEGs were identified, alongside 3625 Badila-specific and 5194 GT29-specific DEGs (Figure 7B).

3.4. GO Analysis of DEGs

Comparative transcriptomic analysis of GT29 and Badila identified both shared and unique DEGs, which were subsequently subjected to GO enrichment analysis (Figure 8). The shared DEGs demonstrated significant enrichment (Q value < 0.05) across 287 functional categories, comprising 173 biological processes, 96 molecular functions, and 18 cellular components. Among the top 25 enriched GO terms ranked by significance, 21 were related to biological processes and 4 to molecular functions (Figure 8A). Notably, the most significantly enriched biological processes included the cellulose metabolic process (GO: 0030243), cell wall biosynthesis (GO: 0042546), cell wall organization (GO: 0071555), and polysaccharide metabolic process (GO: 0005976). Prominent molecular functions comprised nucleic-acid-binding transcription factor activity (GO: 0001071), sequence-specific DNA-binding transcription factor activity (GO: 0003700), and cellulose synthase activities (GO: 0016759 and GO: 0016760) (Figure 8A). These findings collectively indicate that the shared DEGs predominantly participate in cell wall-related metabolic processes and transcriptional regulation mechanisms.

GO enrichment analysis of the ratoon-deficient Badila revealed significant associations across three categories (Figure 8B). The 25 most enriched GO terms comprised 7 biological processes, 6 molecular functions, and 12 cellular components. Notably, the predominant biological processes included photosynthesis (GO: 0015979), light reactions (GO: 0019684), photosynthetic light harvesting (GO: 0009765), and photosystem light harvesting (GO: 0009768). The molecular functions showing the highest enrichment included oxidoreductase activity (GO: 0016491), chlorophyll binding (GO: 0016168), tetrapyrrole binding (GO: 0046906), and adenine phosphoribosyltransferase activity (GO: 0003999). The cellular components with maximal enrichment included photosystem I (GO: 0009522), photosystem complexes (GO: 0009521), photosynthetic membranes (GO: 0034357), and thylakoid components (GO: 0044436). These findings suggest that the identified GO functions are closely associated with the response to stem excision in the ratoon-deficient Badila cultivar.

For the ratoon-competent GT29, GO analysis of its unique DEGs identified 600 enriched terms (Figure 8C). The top 25 terms included 15 biological processes, 2 cellular components, and 8 molecular functions. The biological processes with the highest enrichment factors involved abscisic-acid-mediated signaling (GO: 0009738), cellular responses to abscisic acid (GO: 0071215), and responses to acidic conditions (GO: 0071229). The enriched molecular functions comprised sequence-specific DNA-binding transcription factor activity (GO: 0003700), protein serine/threonine kinase activity (GO: 0004674), and MAP kinase activity (GO: 0004707). Cellular component enrichment was primarily observed in the cell periphery (GO: 0071944) and plasma membrane (GO: 0005886).

3.5. KEGG Pathway of DEGs

To investigate the functional classification and metabolic pathways associated with basal bud germination-related genes, we performed KEGG database annotation and enrichment analysis on the DEGs common to GT29 and Badila. The results identified 15 significantly enriched metabolic pathways (

Table 3. KEGG pathways of shared DEGs in GT29 and Badila.

Pathway ID	Pathway Name	DEGs	Rich Ratio	Q Value
ko04016	MAPK signaling pathway-plant	122	0.044	1.28 × 10−13
ko00940	Phenylpropanoid biosynthesis	91	0.048	2.14 × 10−12
ko00941	Flavonoid biosynthesis	39	0.076	1.01 × 10−10
ko04712	Circadian rhythm-plant	47	0.064	1.66 × 10−10
ko04626	Plant–pathogen interaction	155	0.034	4.55 × 10−9
ko00945	Stilbenoid, diarylheptanoid, and gingerol biosynthesis	21	0.110	8.23 × 10−9
ko01110	Biosynthesis of secondary metabolites	341	0.026	6.94 × 10−6
ko04075	Plant hormone signal transduction	91	0.033	9.66 × 10−5
ko00944	Flavone and flavonol biosynthesis	9	0.129	1.53 × 10−4
ko00500	Starch and sucrose metabolism	57	0.036	3.36 × 10−4
ko00904	Diterpenoid biosynthesis	14	0.062	2.26 × 10−3
ko00360	Phenylalanine metabolism	16	0.053	5.23 × 10−3
ko03015	mRNA surveillance pathway	92	0.028	9.82 × 10−3
ko00908	Zeatin biosynthesis	9	0.062	2.30 × 10−2
ko00430	Taurine and hypotaurine metabolism	10	0.057	2.56 × 10−2

). The top enriched pathways included MAPK signaling (ko04016), phenylalanine biosynthesis (ko00940), flavonoid biosynthesis (ko00941), plant circadian rhythm regulation (ko04712), plant–pathogen interaction (ko04626), and diarylheptanoid biosynthesis (ko00945). The pathways with the highest numbers of DEGs were secondary metabolite biosynthesis (ko01110), plant–pathogen interaction (ko04626), and the MAPK signaling pathway (ko04016) (Table 3).

The KEGG enrichment analysis of DEGs in ratoon-deficient Badila identified 15 significantly enriched pathways. The most prominently enriched pathways included photosynthesis antenna proteins (ko00196) and photosynthesis (ko00195). Core pathways such as secondary metabolite biosynthesis (ko01110) and general metabolic pathways (ko01100) encompassed the highest number of DEGs, with 421 and 733 associated genes, respectively (

Table 4. KEGG pathways of DEGs in Badila.

Pathway ID	Pathway Name	DEGs	Rich Ration	Q Value
ko00196	Photosynthesis antenna proteins	14	0.350	7.62 × 10−11
ko00195	Photosynthesis	28	0.118	1.61 × 10−9
ko04712	Circadian rhythm-plant	46	0.063	1.27 × 10−6
ko00941	Flavonoid biosynthesis	35	0.068	6.57 × 10−6
ko01110	Biosynthesis of secondary metabolites	421	0.032	7.81 × 10−6
ko01100	Metabolic pathways	733	0.029	3.26 × 10−5
ko00280	Valine, leucine, and isoleucine degradation	33	0.051	3.55 × 10−3
ko00640	Propanoate metabolism	21	0.062	3.55 × 10−3
ko03010	Ribosome	82	0.037	5.66 × 10−3
ko00940	Phenylpropanoid biosynthesis	72	0.038	7.15 × 10−3
ko00500	Starch and sucrose metabolism	61	0.038	1.45 × 10−2
ko00620	Pyruvate metabolism	42	0.041	2.09 × 10−2
ko00514	Other types of O-glycan biosynthesis	14	0.062	2.28 × 10−2
ko00945	Stilbenoid, diarylheptanoid, and gingerol biosynthesis	12	0.063	3.65 × 10−2
ko00905	Brassinosteroid biosynthesis	7	0.085	4.41 × 10−2

).

In the analysis of DEGs unique to the ratoon-competent GT29, 22 pathways demonstrated significant enrichment (

Table 5. KEGG pathways of DEGs in GT29.

Pathway ID	Pathway Name	DEGs	Rich Ratio	Q Value
ko04626	Plant–pathogen interaction	463	0.101	1.46 × 10−36
ko04016	MAPK signaling pathway-plant	293	0.106	1.83 × 10−25
ko04075	Plant hormone signal transduction	268	0.096	1.57 × 10−17
ko00940	Phenylpropanoid biosynthesis	171	0.091	3.92 × 10−9
ko00500	Starch and sucrose metabolism	149	0.094	4.90 ×10−9
ko00906	Carotenoid biosynthesis	54	0.124	5.84 × 10−7
ko00945	Stilbenoid, diarylheptanoid, and gingerol biosynthesis	28	0.147	4.26 × 10−5
ko00460	Cyanoamino acid metabolism	61	0.097	2.26 × 10−4
ko01110	Biosynthesis of secondary metabolites	827	0.062	2.54 × 10−4
ko00999	Biosynthesis of various plant secondary metabolites	55	0.099	2.54 × 10−4
ko00591	Linoleic acid metabolism	21	0.152	2.72 × 10−4
ko00515	Mannose type O-glycan biosynthesis	14	0.192	4.32 × 10−4
ko00250	Alanine, aspartate, and glutamate metabolism	54	0.094	1.11 × 10−3
ko00410	beta-Alanine metabolism	43	0.095	3.61 × 10−3
ko00941	Flavonoid biosynthesis	46	0.09	7.67 × 10−3
ko00520	Amino sugar and nucleotide sugar metabolism	113	0.074	7.67 × 10−3
ko04070	Phosphatidylinositol signaling system	66	0.082	7.67 × 10−3
ko00220	Arginine biosynthesis	35	0.095	9.78 × 10−3
ko04712	Circadian rhythm-plant	59	0.081	1.56 × 10−2
ko00360	Phenylalanine metabolism	29	0.095	2.00 × 10−2
ko00524	Neomycin, kanamycin, and gentamicin biosynthesis	10	0.149	2.25 × 10−2
ko00130	Ubiquinone and other terpenoid-quinone biosynthesis	34	0.089	2.37 × 10−2
ko00910	Nitrogen metabolism	36	0.088	2.50 × 10−2
ko00908	Zeatin biosynthesis	16	0.111	3.31 × 10−2
ko00901	Indole alkaloid biosynthesis	3	0.375	4.12 × 10−2
ko00965	Betalain biosynthesis	12	0.12	4.59 × 10−2

). The most prominently enriched pathways included plant–pathogen interactions (ko04626), MAPK signaling (ko04016), phytohormone signaling (ko04075), phenylpropanoid biosynthesis (ko00940), and starch/sucrose metabolism (ko00500). Notably, secondary metabolite biosynthesis (ko01110) and key biological processes involving plant–pathogen crosstalk (ko04626), MAPK cascade regulation (ko04016), and phytohormone-mediated responses (ko04075) were substantially represented among the enriched pathways (Table 5).

3.6. Characterization of Key Genes in Phytohormone Signal Transduction Pathways

Following the stem excision applied to the GT29 and Badila cultivars, we systematically analyzed the dynamic alterations in their endogenous phytohormone profiles. The results demonstrated a significant correlation between these hormonal fluctuations and the post-truncation developmental responses in both sugarcane varieties. KEGG pathway enrichment analysis revealed differential gene expression patterns, with GT29 exhibiting the substantial enrichment of DEGs in plant hormone signal transduction pathways (Table 5). Notably, this enrichment pattern was absent in Badila. Consequently, our investigation specifically focused on elucidating DEGs associated with major phytohormone signaling pathways, including IAA, CTK, GA, ABA, brassinosteroid (BR), jasmonic acid (JA), ethylene (ET), and salicylic acid (SA)-mediated regulatory networks (Figure 9).

Auxin exhibited the highest number of stem excision response genes (Figure 9). Comparative analysis between ratoon-competent GT29 and -deficient Badila stem excision at 1 d and 3 d revealed significant upregulation of the auxin transport carrier gene AUX1 in GT29. In contrast, Badila showed downregulation of five auxin-related genes (CL5293.Contig3_All, CL5293.Contig8_All, CL16525.Contig9_All, CL12953.Contig1_All, and CL12953.Contig2_All.). The auxin-responsive AUX/IAA genes (CL5293. Contig3_All and CL5293. Contig8_All) demonstrated 7.86-fold and 6.52-fold upregulation in GT29 at 1 d and 3 d post-stem-excision, respectively, while in Badila they displayed 4.68-fold and 4.46-fold downregulation. Notably, the SAUR gene (CL12953. Contig2_All) showed 6.1-fold and 4.2-fold upregulation in Gt29, and >2-fold downregulation in Badila at corresponding timepoints. CTK signaling components revealed distinct regulation patterns. The CTK receptor gene B-ARR (CL849.Contig34_All) exhibited 11.74-fold and 5.69-fold upregulation in GT29 at 1 d and 3 d, respectively, compared to 0.4-fold and 0.2-fold downregulation in Badila. The transcription factor A-ARR (CL6710.Contig9_All) showed >8-fold upregulation exclusively in GT29, potentially activating bud germination. GA signaling analysis revealed contrasting regulation. The GA receptor GID1 (CL24.Contig3_All) and transcription factor (CL5895.Contig17_All) showed >2.5-fold and >6.4-fold upregulation in GT29, respectively, while CL5643.Contig4_All was downregulated. ET signaling components showed moderate upregulation, with the ethylene-receptor-associated CTR1 gene (CL1774. Contig7_All) demonstrating >2.7-fold induction in GT29 and <1-fold in Badila. BR-related genes including BAK1, BRI1, and TCH4 were upregulated, with the BR receptor BRI1 (CL4940. Contig3_All) showing >8-fold induction in GT29. These components directly regulate brassinosteroid biosynthetic genes through promoter binding in feedback mechanisms.

3.7. qRT-PCR Validation

To validate the accuracy of transcriptome profiling, twelve DEGs were selected for qRT-PCR verification in both the GT29 and Badila cultivars under stem excision treatment. The analyzed genes included CL2511.Contig5_All, CL13067.Contig31_All, Unigene11582_All, CL24.Contig3_All, CL10318.Contig30_All, CL11787.Contig4_All, CL16936.Contig6_All, CL11106.Contig1_All, Unigene11582_All, CL6710.Contig7_All, CL11630.Contig1_All, and CL15393.Contig2_All. Notably, all the listed genes exhibited downregulation in Badila following stem excision treatment but upregulation in GT29. Unigene97471_All showed consistent downregulation in both cultivars. In contrast, CL2511.Contig5_All displayed transient upregulation followed by gradual downregulation post-truncation in both cultivars. The qRT-PCR results showed strong concordance with transcriptome data (Figure 10, Table S6), confirming the reliability of our sequencing analysis.

4. Discussion

This study systematically elucidated the molecular physiological mechanisms underlying ratoon differentiation in sugarcane through an integrated analysis of endogenous hormone dynamics and transcriptome regulatory networks. At the hormonal regulation level, the key signaling components of the CTK pathway, zeatin (ZA), and zeatin riboside (ZR) exhibited synchronous attenuation following stalk decapitation (Figure 2). This observation presents a distinct contrast to the classical paradigm established in rice, where elevated CTK levels promote tiller development [22,23]. Subsequent investigations demonstrated distinct CTK signaling patterns, with the CTK receptor gene B-ARR (CL849.Contig34_All) and transcription factor A-ARR (CL6710.Contig9_All) exhibiting genotype-specific regulation. Notably, GT29 showed the remarkable upregulation of these genes by 11.74-fold and 8.3-fold, respectively, whereas Badila displayed downregulation (Figure 9). This expression pattern inversely correlated with the CTK accumulation trends observed during wheat tiller development [24,25,26], presenting an apparent “low hormone concentration–high receptor activity” paradox. We hypothesize that this phenomenon reflects a compensatory mechanism in ratoon-competent cultivars, where enhanced signal transduction efficiency counterbalances reduced absolute phytohormone concentrations. The specific induction of A-ARR in GT29 merits particular attention due to its functional conservation with OsAAP1, a well-characterized promoter of axillary bud development in Oryza sativa [27], suggesting that this gene may serve as a core regulatory node for rooting ability.

In the GA signaling pathway, GID1 functions as a canonical GA receptor, with homologous GID1 genes characterized in model species including Arabidopsis thaliana and Gossypium spp. [28,29,30]. DELLA proteins represent crucial regulatory components in GA signal transduction, serving as growth-repressing factors through their inhibitory effects on plant development [31,32]. The established GA signaling mechanism involves GA-GID1 complex formation, which subsequently triggers the ubiquitin-26S proteasome-mediated degradation of DELLA proteins [33,34]. This degradation cascade alleviates DELLA-mediated growth suppression while inducing the transcriptional reprogramming of downstream targets, ultimately manifesting GA-mediated physiological responses. In the present study, transcriptomic analysis results revealed distinctive GA signaling dynamics in GT29, exhibiting a coordinated reduction in endogenous GA3/GA5 levels coupled with a sustained upregulation of the GID1 receptor (CL24.Contig3_All), showing a 3.8-fold increase. This apparent “low GA–high receptor” configuration pattern potentially enhanced GID1-mediated DELLA protein turnover, as evidenced by the 6.4-fold upregulation of CL5895.Contig17_All, a putative DELLA degradation component (Figure 4 and Figure 9). Such regulatory dynamics may override the conventional GA-mediated tillering suppression mechanism documented in Oryza sativa [35], suggesting species-specific adaptation in sugarcane. Notably, this GA signaling profile showed functional convergence with SL pathway alterations, where GT29 exhibited a 68% depletion of 5-DS compared to the stable levels in Badila (Figure 4 and Figure 6). We propose a novel GA-SL crosstalk mechanism operating through two complementary pathways: reduced 5-DS levels may attenuate its inhibitory effects on GA3 biosynthesis, thereby potentiating GID1-mediated signaling efficiency for axillary bud primordium activation; concurrent SL reduction directly relieves bud dormancy constraints. This multi-layered hormonal regulation aligns with emerging models of SL-GA interaction in monocot development [36,37].

Transcriptional profiling of auxin signaling networks uncovered spatiotemporally specific regulatory features. The Badila genotype exhibited a transient peak in IAA accumulation, with a 1.5-fold increase at 3 days post-treatment, which was concurrent with the progressive downregulation of the AUX/IAA inhibitor (CL5293.Contig3/8_All) (Figure 3 and Figure 9). This coordinated regulation resulted in the hyperactivation of auxin signaling pathways and the subsequent suppression of lateral bud germination. In contrast, the GT29 genotype demonstrated the sustained upregulation of Aux/IAA transcripts during its early (1 d) and mid-phase (3 d) response to apical meristem removal. Notably, GT29 established a phytohormone signaling configuration that favored rapid tiller development through an immediate induction, showing a 6.1-fold increase in the expression of the SAUR family gene (CL12953.Contig2_All) (Figure 9). This regulatory mechanism exhibited functional conservation through SAUR-mediated cell elongation processes previously characterized in Arabidopsis root meristems, yet it represented a novel discovery in sugarcane ratooning systems [38,39]. Particularly significant was the 8-fold upregulation of BRI1 (CL4940.Contig3_All) observed in GT29 (Figure 9), suggesting that BR signaling may optimize auxin distribution patterns within axillary bud microenvironments by modulating polar auxin transport. This molecular mechanism provides a plausible explanation for the enhanced ratoon sprouting capacity observed in ratoon-competent sugarcane cultivars.

Transcriptome profiling further validated the specificity of phytohormone regulatory networks. Notably, the MAPK signaling pathway exhibited significant enrichment in GT29, where a CTR1 homolog gene (CL1774.Contig7_All) showed a 2.7-fold upregulation (Table 3 and Table 5, Figure 9), potentially mediating ET-MAPK crosstalk. The concurrent activation of the phenylpropanoid biosynthesis pathway not only supplied structural precursors for cell wall remodeling but also generated flavonoid derivatives that might fine-tune auxin spatial distribution through the modulation of PIN transporters.

The dual-layered regulatory model of hormone signaling established in this investigation comprehensively elucidates the molecular basis of ratoon bud differentiation in sugarcane GT29, achieving perennial growth dominance through four synergistic mechanisms: the rapid potentiation of CTK signaling sensitivity, establishment of GA-SL mutual inhibitory release mechanisms, optimization of IAA/BR spatial distribution networks, and coordinated activation of secondary-metabolism- and defense-related pathways. These findings not only refine the conventional paradigm of single-hormone dominance in tiller regulation but also identify crucial molecular targets for the marker-assisted breeding of novel sugarcane cultivars with enhanced sucrose accumulation and ratoon-competent traits. Future investigations should employ single-cell transcriptomics to resolve hormonal gradients in axillary meristem primordia and validate the functional conservation of core regulatory elements, thereby completing the holistic regulatory network governing perennial regeneration capacity.

5. Conclusions

In this study, the physiological and molecular mechanisms underlying ratooning ability in sugarcane were systematically elucidated through the integration of endogenous hormone profiling and transcriptomic analyses. The ratoon-competent cultivar GT29 exhibited a unique regulatory pattern characterized by simultaneous reductions in hormone concentrations (ZA, ZR, GA₃, GA₅, and 5-DS) and the significant upregulation of key signaling genes, including B-ARR, A-ARR, GID1, and BRI1. This “low hormone–high receptor activity” compensation model highlighted an adaptive mechanism in GT29 that enhanced signal transduction efficiency, thereby promoting the activation of dormant buds following stem excision. Notably, this study provides the evidence of a putative crosstalk between GA and strigolactone pathways in sugarcane, wherein reduced 5-DS levels may relieve suppression on GA biosynthesis, ultimately enhancing GID1-mediated DELLA protein turnover. Moreover, the coordinated transcriptional activation of auxin-responsive genes (AUX/IAA, SAUR) and BR receptors further shaped a favorable hormonal microenvironment for tiller emergence. GT29 also demonstrated significant enrichment of DEGs involved in MAPK signaling, phenylpropanoid biosynthesis, and hormone signal transduction, revealing a multihormonal and multi-layered regulatory network underlying its superior regenerative performance. In contrast, the ratoon-deficient cultivar Badila showed an attenuated hormonal response and a lack of key transcriptional reprogramming, suggesting an insufficient molecular response to decapitation. These findings delineate a clear molecular framework distinguishing ratoon competence and provide reliable candidate genes for functional marker development in breeding programs. Altogether, this study offers novel insights into the molecular basis of perennial regeneration in sugarcane and establishes a foundation for future work utilizing single-cell transcriptomics and functional genomics to refine the regulatory network controlling ratoon bud activation and longevity.