Genome-wide-association-analysis-based identification of genetic loci and candidate genes associated with seedling emergence rate in super-sweet corn (Zea mays L.)

Abstract

Seed germination is the initial and critical stage of plant growth and development. This study conducted a genome-wide association study (GWAS) on seedling emergence rate (SER) and seedling length (SL) in super-sweet corn using a panel of 201 Chinese core germplasm accessions as the association population. Through multi-model GWAS integration of BLINK, FarmCPU and MLM approaches applied to 56 K maize SNP array data, 12, 18 and 19 significant SNPs were detected by each model respectively. Venn diagram analysis revealed significant overlaps among different models, ultimately identifying eight SNPs significantly associated with SER and two SNPs related to SL. Through functional annotation and literature analysis, 15 candidate genes were identified from an initial set of 49 annotated genes. Among these, 10 genes regulated SER, and 5 were linked to SL control, primarily functioning in protein biosynthesis, energy transduction, and cell division regulation. Allelic effect analysis demonstrated that the A/A and T/T genotypes represent key superior allelic variants regulating emergence rate. Through marker-assisted selection, pyramiding these favorable alleles can significantly enhance the seedling emergence rate in super-sweet corn. This study provides valuable genetic loci for marker-assisted breeding in super-sweet corn and offers an important reference for fine-mapping genes associated with seedling emergence rate characteristics.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08228-z.

Introduction

Sweet corn is an important economic crop with versatile uses such as grain, fruit, vegetable, and feed. As an endosperm mutant regulated by single or multiple recessive genes, it is also known as fruit corn or vegetable corn [1]. In recent years, the planting area of sweet corn has expanded significantly in response to growing market demand. Currently, the sweet corn sold in the market is primarily the super-sweet corn controlled by the sh2 endosperm mutant gene [2, 3]. Super-sweet corn contains a recessive mutant allele in the endosperm that disrupts starch biosynthesis in developing kernels. This mutation leads to endosperm collapse and reduced kernel weight at maturity. Compared to regular corn, it exhibits lower seed vigor and slower germination speed, among other differences [4, 5]. Therefore, developing novel super-sweet corn varieties with rapid and uniform seedling emergence through genetic improvement has become a key research priority in current sweet corn breeding programs.

Seed vigor refers to the ability of seeds to germinate and develop into healthy seedlings, serving as a critical indicator of seed quality [6, 7]. That has the most pronounced impact during the crop germination and seedling establishment stage, directly influencing metrics such as germination rate and seedling emergence rate. Existing studies have demonstrated significant genotypic dependence of seed vigor in maize, exhibiting distinct variations not only among different cultivars but also across various germplasm materials within the same variety [8, 9]. This variation primarily stems from differences in the internal genetic regulatory networks of seeds, involving multiple physiological mechanisms including material metabolism, hormonal regulation, and stress response [10]. Studies on seedling emergence have recently focused on mesocotyl elongation-mediated regulation of rapid seedling emergence [11], phytohormone regulation of seed germination [12, 13], and molecular mechanisms governing hypocotyl growth [14]. However, fundamental research on maize emergence rate remains relatively scarce.

GWAS is a method based on linkage disequilibrium (LD) that statistically analyzes the obtained genotype data and corresponding phenotype data for identifying significantly associated loci and candidate genes. Currently, GWAS has been widely applied in genetic screening of major crops such as maize [15, 16] wheat, rice [17], and rapeseed [18]. In the genetic dissection of maize germination and early seedling traits, GWAS and related multi-omics integrative analyses have been widely validated as effective tools. For instance, He et al. identified two candidate genes that positively regulate low-temperature germination ability using QTL-Seq and transcriptome analysis [19]; Zhang et al. identified a key gene, ZmKCH5, which regulates seed germination and root growth under salt stress by integrating GWAS with weighted gene co-expression network analysis (WGCNA) [20] ; based on germination-related phenotypes from 304 maize inbred lines, Xu et al. combined GWAS with transcriptome sequencing and further verified through qRT-PCR that ZmBARK1 is a key gene controlling low-temperature tolerance during maize germination [21].

Collectively, these studies demonstrate that GWAS is effective in identifying significantly associated loci and trait-related candidate genes. Nevertheless, GWAS focusing on seedling emergence rate-related traits in super-sweet corn remains notably underrepresented in the scientific literature. Therefore, this study utilized a panel of 201 accessions of super-sweet corn inbred lines as the core germplasm and conducted a GWAS based on the maize 56 K SNP array, and the research systematically dissected the genetic basis regulating SER and SL, identified major-effect SNPs and candidate genes, to provide a theoretical foundation for breeding super-sweet maize germplasm with rapid and uniform seedling emergence.

Materials and methods

Plant materials

The plant materials used in this study were 201 super-sweet corn inbred lines provided by the Maize Innovation Team of Anhui Science And Technology University, all of which originated from China.

Experimental procedures

Preparation of germination substrate

Previous studies have demonstrated that agar-based media exhibit superior moisture retention capacity in germination substrates compared to conventional artificial matrices [22, 23].

A germination chamber measuring 19 cm (L) × 13 cm (W) × 8 cm (H) was employed as the containment vessel for the agar-based germination substrate.

Based on the established optimal concentration (10 g/L) [22] for agar-based germination substrates, an appropriate quantity of agar powder was dissolved in distilled water through controlled heating until fully homogenized.

Following attainment of optical clarity through heating, 300 mL of the solution was aseptically aliquoted into each germination chamber and maintained undisturbed under ambient conditions (28 ± 1 °C) until the agar medium completely gelled and equilibrated thermally.

Seedling emergence test

From each inbred line of super-sweet corn, 50 healthy seeds with no significant size differences were randomly selected for a standard seedling emergence test. The experiment was set up with three replicates.

Germination conditions: Temperature 28 ± 1 °C, humidity 60%, 12-hour light/12-hour dark cycle (Fig. 1).

Fig. 1.

Schematic diagram of SER in super-sweet corn seeds

Germination and emergence criteria: Radicle and plumule elongation ≥ 1/2 seed length is considered germination. Seedling growth ≥ 5 cm was considered successful emergence. Observations were recorded at 12-hour intervals up to the final time point of 240 h (10 days), with germination and seedling emergence counts documented at each interval. Emergence time (ET), corresponding radicle emergence time (RT).

The coefficient 5 in this formula represents the standardized morphogenetic progression threshold (unit: cm) from radicle emergence to complete cotyledon expansion. SER is defined as the time required for seedling growth per centimeter, which is an inverse trait; thus, a lower SER value indicates faster emergence speed and higher efficiency.

Phenotypic trait analysis

This study performed a comprehensive analysis of phenotypic data for SER and SL traits using a panel of 201 super-sweet corn germplasm accessions. During data processing, Excel 2019 was used for data cleaning and standardization. Statistical analysis was performed using SPSS 24.0 to calculate mean, standard deviation (SD), and coefficient of variation (CV) for each trait. One-way analysis of variance (ANOVA) was performed using GraphPad Prism 9 software for significance testing [24, 25]. For visualization, normality was assessed by constructing distribution plots using the ggplot2 package in R. The broad-sense heritability (h²) for each trait was calculated as: h² = σ²g/(σ²g + σ²e/n) where σ²g is the genetic variance, σ²e is the error variance, and n is the number of experimental replications [26].

Genome-wide association analysis

Based on established research findings [27], the current investigation methodically performed the subsequent analytical procedures. The initial quality control of raw SNP data was performed using Plink 1.90 software, with exclusion criteria set at a missing rate > 0.2 or minor allele frequency (MAF) < 0.05, resulting in the retention of 36,747 high-quality SNP markers for subsequent analyses. The GWAS was conducted using three mixed linear model approaches (BLINK, FarmCPU, and MLM), incorporating principal component analysis (PCA) results and the kinship matrix (K) as covariates to effectively minimize false-positive associations in the study findings. The significance threshold was set at -log10(P) ≥ 3.6 (P < 1.8 × 10⁻⁴) to identify significant SNP loci associated with SER and SL. Manhattan and Q-Q plots visualizing the GWAS results were generated using the CMplot package in R. Population linkage disequilibrium (LD) decay analysis was performed across 200 kb genomic intervals using PopLDdecay (v3.42) to characterize population-level linkage disequilibrium patterns [28].

Candidate gene analysis

Candidate genes identified within 200 kb upstream and downstream regions of genome-wide significant SNPs were functionally annotated using both the maize-specific MaizeGDB database (https://maizegdb.org/) and NCBI databases (http://www.ncbi.nlm.nih.gov/). Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis using the OmicShare online analysis platform. Transcriptome data covering the entire maize developmental cycle were retrieved from MaizeGDB [29]. Expressed genes were selected using an FPKM ≥ 1 threshold [30], and after log2(FPKM + 1) normalization, gene expression heatmaps were generated using the Pheatmap package [31]. Protein-protein interaction (PPI) networks of candidate gene-encoded proteins were constructed using the STRING v12.0 database (https://cn.string-db.org/), with subsequent network visualization and analysis performed in Cytoscape v3.10.2 software [32].

Linkage disequilibrium analysis

Genotype data spanning the 200-kb flanking regions of significant SNPs were extracted for linkage disequilibrium analysis [33]. The linkage disequilibrium (LD) region was defined with a threshold of r² > 0.8 [34]. The LD heatmap package in R was used to plot the LD heatmap [35]. By integrating phenotypic data with genotype data, the allelic effects of significant SNPs were assessed, and the analysis results were visualized using the ggplot2 package.

Genotype verification

Based on the GWAS analysis results, inbred lines 3176 and YN11, which carried a higher number of favorable alleles, were selected to construct an F1 hybrid combination (3176×YN11). PCR combined with polyacrylamide gel electrophoresis (PAGE) was employed to validate the incorporation of the target favorable alleles in the hybrid [36]. Seedling leaves were collected at the three-leaf stage, and genomic DNA was extracted using the CTAB method [37]. Based on the significantly associated SNP loci within the candidate genes, two pairs of allele-specific primers were designed using Primer Premier 6.0 software [38].

The total volume of the PCR reaction system was 20 µL, consisting of 10 µL of 2× Taq PCR MasterMix (Sai Pu Biotechnology, Nanjing), 1 µL of genomic DNA template, 1 µL each of forward and reverse primers, and 7 µL of ddH₂O. The PCR amplification protocol was as follows: initial denaturation at 94 °C for 5 min; followed by 35 cycles of denaturation at 94 °C for 40 s, annealing at 60 °C for 35 s, and extension at 72 °C for 45 s; a final extension at 72 °C for 5 min; and holding at 4 °C for storage. The PCR amplification products were separated by 10% non-denaturing PAGE and visualized by silver staining [39]. By comparing the banding patterns of the F₁ hybrid with those of its parental inbred lines, the presence of the target favorable alleles in the hybrid was confirmed, thereby verifying their incorporation.

Results

Phenotypic analysis of Super-Sweet corn

The results of phenotypic trait analysis indicated, for the SER trait, the values ranged from 9.6 to 40.8 h/cm, with a mean of 19.386, a standard deviation of 0.63, and a coefficient of variation of 3.285%. For the SL trait, the phenotypic variation ranged from 4.8 to 20.58 cm/plant, with a mean of 14.001 and a coefficient of variation of 3.735%. Both SER and SL exhibited approximately normal distributions, with an extremely high correlation coefficient between them (Fig. 2). Moreover, these traits demonstrated high heritability, with estimates of 0.95 for SER and 0.87 for SL. The results indicate that SER and SL show strong genetic control and exhibit quantitative trait characteristics, making them suitable for association mapping (Table 1).

Fig. 2.

SER and SL Frequency Distributions and Scatter Plot with Correlation Coefficients. *The phenotypic frequency distributions of each trait are shown above the diagonal. The scatter plots and correlation coefficients between traits are illustrated in the regions below and above the diagonal. The red line in the scatter plots represents the correlation trend

Table 1.

Descriptive statistics analysis of SER and SL

Trait	Average ± SD	Skewness	Kurtosis	Range	CV(%)	h2
SER	19.386 ± 0.637	0.908	0.734	9.60–40.80	3.285	0.95
SL	14.001 ± 0.523	0.262	-0.615	4.80-20.58	3.735	0.87

*SER Seedling emergence rate (h/cm), SL Seedling length (/cm), SD Standard deviation, CV Coefficient of variation, h² Heritability

Genome-wide association analysis of SER and SL

GWAS analysis of SER and SL traits in super-sweet corn was conducted by integrating three statistical models. The BLINK model identified 12 significantly associated loci, with a PVE of 13.15%; The FarmCPU model detected 18 SNP loci significantly associated with SER and SL traits, with an average PVE of 12.09%. The MLM model identified 19 significant loci, with an average PVE of 12.48% (Fig. 3a-c, Supplementary Table S1). Further analysis identified 8 overlapping loci associated with SER and 2 loci linked to SL traits (Fig. 3d, e). Three loci on chromosomes 3 and 8 (Affx-90520409, Affx-90338401, and Affx-90640949) showed relatively higher PVE, with PVE ranging from 14.08% to 19.92%.

Fig. 3.

GWAS results for SER and SL traits. a BLINK model results, b FarmCPU model results, c MLM analysis results. *The Manhattan plot displays chromosomal positions along the horizontal axis and − log10(p-values) for each genetic marker on the vertical axis, with a black horizontal line indicating the genome-wide significance threshold and red data points highlighting statistically significant SNPs. In the Q-Q plot, the red diagonal line represents the expected theoretical distribution. d, e Venn diagram showing the number of overlapping significant SNP loci associated with SER and SL traits. *The Venn diagram illustrates the distribution of identified loci across different analytical models. Distinct colored circles represent the quantity of selected loci: pink for the BLINK model, blue for the FarmCPU model, and green for the MLM model

Candidate gene analysis

To further explore the key genes affecting the SER of super sweet corn seeds, a total of 294 genes were screened based on the significant SNP confidence intervals, among which 49 genes had clear functional annotations. GO and KEGG enrichment analyses revealed that these genes were significantly enriched in processes such as endoplasmic reticulum protein processing, steroid biosynthesis, pathogen interaction, and ribosome biogenesis. Additionally, these genes showed significant enrichment in pathways such as RNA degradation, glucose catabolism, protein degradation, and tryptophan metabolism (Fig. 3). Among them, genes in the Glycolysis/Gluconeogenesis and Pyruvate metabolism pathways may be closely related to seed germination. During seed germination, the starch stored in the endosperm is broken down into glucose, which can quickly provide energy for embryo growth via Glycolysis [40]. Gluconeogenesis, in turn, is involved in the conversion of non-carbohydrate substances into sugars and serves as a core metabolic pathway during maize germination [41]. Pyruvate metabolism serves as the central hub of cellular respiration and metabolism. By supporting cell division and elongation in the crop embryo, it plays a critical bridging and regulatory role during the germination process [42, 43]. These findings suggest that these genes hold significant potential for further investigation and exploration (Fig. 4).

Fig. 4.

The GO and KEGG Enrichment Analysis of Candidate Genes. a GO functional annotation analysis plot. *The X-axis corresponds to the number of involved genes, while the Y-axis reflects gene functional annotation information. b KEGG pathway enrichment analysis bubble chart. *The X-axis displays the gene enrichment ratio, and the Y-axis represents enriched metabolic pathways, bubble sizes show a positive correlation with gene counts, and warmer red colors indicate stronger statistical significance

Based on the analysis of gene functional expression profiles, this study identified a total of 15 candidate genes associated with seedling emergence rate and their orthologous genes in Arabidopsis thaliana (Table 2). Among the 10 genes associated with SER, four were found to regulate this trait by modulating root growth and development. These genes include GRMZM2G148723, GRMZM2G123540, GRMZM2G131275 and GRMZM2G027627. The genes GRMZM2G003853, GRMZM2G337766, and GRMZM2G417954 were found to influence seed emergence rate by modulating enzymatic biosynthesis in plants. GRMZM2G087804 and GRMZM2G026614 regulate seed germination and growth by modulating chloroplast development, while GRMZM2G146163 influences plant growth and development through targeted protein degradation. Among the five candidate genes regulating SL, three genes GRMZM2G080588, GRMZM2G080375, and GRMZM2G095265 were found to influence leaf growth and development through the regulation of enzyme biosynthesis. The genes GRMZM2G173124 and GRMZM2G112681 primarily regulate leaf development through protein synthesis-related mechanisms.

Table 2.

Candidate genes for SER and SL of Super-sweet corn seeds

Trait	SNP	Chr	PVE	Candidate Gene	Gene Annotation	Module
SER	Affx-91,247,800	1	11.59%	GRMZM2G337766	PEP1 receptor 1	M1
	Affx-90,520,409	3	14.71%	GRMZM2G087804	ATGLK1, GLK1, GPRI	M2
				GRMZM2G148723	WPP domain-interacting protein 1	/
				GRMZM2G123540	ELD1 protein (ABI8, ELD1, KOB1)	M3
	Affx-90,761,633	7	13.32%	GRMZM2G417954	NCED9	M4
				GRMZM2G026614	chloroplast RNA-binding protein 33	M5
	Affx-91,394,330	7	12.46%	GRMZM2G131275	RHO guanyl-nucleotide exchange factor 3	/
	Affx-90,338,401	8	19.92%	GRMZM2G146163	(J20) DNAJ-like 20	M6
				GRMZM2G003853	(HYD1) C-8 7 sterol isomerase	M7
	Affx-115,333,800	9	11.65%	GRMZM2G027627	IAR4 (THDP-binding)	M8
SL	Affx-90,790,712	8	12.08%	GRMZM2G173124	ZF-Cx8Cx5Cx3H	/
				GRMZM2G080588	ataurora1 (AtAUR1, AUR1)	/
				GRMZM2G080375	(PFK3) phosphofructokinase 3	M9
				GRMZM2G112681	NAC domain containing protein 75	M10
	Affx-90,640,949	8	14.08%	GRMZM2G095265	Senescence-associated gene 12	M11

* M1 to M11 represent proteins that form significant protein interaction networks in the STRING database (interaction score > 0.4), while ‘/’ indicates proteins that do not form significant interaction networks (interaction score < 0.4)

To further elucidate the expression characteristics of the candidate genes, dynamic expression pattern analysis across various developmental stages was performed using RNA-seq data (Fig. 5). The results revealed significant differential expression patterns of these candidate genes across developmental stages, with GRMZM2G417954 showing peak expression in root tissues at 7 days post-sowing. GRMZM2G087804, GRMZM2G337766, and GRMZM2G080375 exhibited significant expression during leaf growth and development, suggesting that these candidate genes play pivotal roles in SER and SL, providing crucial clues to the regulatory mechanisms of seedling emergence in super-sweet corn.

Fig. 5.

Dynamic expression characteristics of candidate genes. *The scale represents the relative expression level of the gene. The X-axis represents different developmental stages and tissues, while the Y-axis represents the candidate genes. The gradient from blue to red represents expression levels from low to high. S1-S13 represent root development at 3–7 days after sowing DAS and leaf development at 6 DAS

Construction of the regulatory network for candidate genes

The STRING database was employed to construct the protein-protein interaction networks for 15 candidate genes. of these and their interacting partners, 11 genes were in-volved in various functional proteins either by forming an independent cluster or through their regulatory roles in species-specific biological networks (Fig. 6). The candidate gene GRMZM2G087804 (M2), identified as GLK13, regulates chloroplast biogenesis and modulates photomorphogenic signaling during seedling development [44, 45]. GRMZM2G146163 (M6) encodes B6U349, a chaperone protein that acts as a co-chaperone for HSP70 and plays a crucial role in plant growth regulation [46]. GRMZM2G112681 (M10), encoding NAC075 transcription factor, modulates ROS homeostasis through direct binding to the CAT2 promoter, thereby inhibiting senescence progression and preserving chlorophyll content in leaves [47].

Fig. 6.

Protein–protein interaction network among different candidate genes. *The black lines represent the functional interactions between proteins. M1 to M14 represent the proteins encoded by the candidate genes. The circles represent proteins. Red nodes represent hub proteins with multiple interactions

Analysis of linkage disequilibrium and allelic variation effects

To validate the reliability of the identified significant loci and elucidate their genetic background, we performed linkage disequilibrium analysis using genotype data extracted from the 200-kb flanking regions of these loci. The results demonstrated that four of the screened loci exhibited strong linkage disequilibrium: Affx-90,520,409, Affx-91,134,889, Affx-90,790,712, and Affx-90,640,949 (Fig. 7). Furthermore, by evaluating the effects of favorable alleles within the linked region, we found that the alleles of the germination rate-associated loci Affx-90,520,409 and Affx-91,134,889 exhibited significant differences. Specifically, the T/T allele significantly enhanced SER by 9.6–14.4 h/cm compared to the G/G and C/C alleles. For the significant loci associated with SL, the A/A allele at Affx-90,790,712 reduced seedling length by 0.32–13.75 cm compared to the A/G, G/G, A/C and C/C alleles. Therefore, both the T/T and A/A alleles could be identified as favorable alleles for their respective traits.

Fig. 7.

Linkage disequilibrium heatmaps and SNP allele effect analysis. a LD heatmaps for SER and SL associated SNPs, respectively. Triangular boxes highlight high-LD blocks, with blue dots indicating SNP positions. b Allele effects of SNPs associated with SER and SL, respectively; * p < 0.05, ** p < 0.01, *** p < 0.001

Based on the favorable alleles (A/A and T/T) associated with SER and SL in super-sweet corn, this study further analyzed the key genetic loci influencing emergence-related traits (Fig. 8a). PAGE results showed that at the target locus, the hybrid 3176×YN11 simultaneously presented the parental-specific bands of both parents (Fig. 8b). The phenotypic statistics presented in Table 3 indicate that within the inbred line group, the SER of lines F2023 and 90, which carry the unfavorable alleles, was significantly higher than that of lines 3176 and YN11 carrying the favorable alleles. Within the F₁ hybrid group, however, the hybrid 3176×YN11, which combines the favorable alleles, demonstrated significantly superior performance in SER, SL, RL, and SFW traits compared to F2023 × 90. Meanwhile, the hybrid 3176×YN11 showed significant differences compared to its parents in all traits except for RFW. The results above demonstrate that the polymerization of favorable alleles through hybridization can effectively improve the emergence-related traits in super-sweet corn.

Fig. 8.

Favorable allele analysis of significant loci associated with emergence-related traits and genotype validation of F₁ hybrids. a Analysis of favorable alleles at significant loci associated with emergence-related traits. * Red bars represent favorable alleles, and blue bars represent unfavorable alleles. b Genotype validation of F₁ hybrids. * Lanes from left to right are: Marker, ♀3176 (female parent), ♂YN11 (male parent), F₁ (3176×YN11)

Table 3.

Phenotypic statistical analysis of seedling traits in different materials

Materials	SER	SL	RL	SFW	RFW
3176	18.62 ± 0.73	21.68 ± 0.59	19.28 ± 0.39	1.46 ± 0.11	1.115 ± 0.64
YN11	21.21 ± 0.42	18.64 ± 0.44	15.01 ± 0.33	0.87 ± 0.18	0.409 ± 0.14
3176×YN11	13.87 ± 1.16***	24.91 ± 1.17**	22.57 ± 0.35***	1.86 ± 0.06***	1.421 ± 0.13
F2023	11.46 ± 0.39	21.95 ± 0.91	21.41 ± 0.38	1.64 ± 0.13	1.474 ± 0.55
90	19.16 ± 0.35	13.65 ± 1.46	21.46 ± 1.21	0.31 ± 0.04	0.376 ± 0.02
F2023 × 90	26.47 ± 0.26***	21.15 ± 0.61	21.45 ± 0.13	1.48 ± 0.07***	1.497 ± 0.21*

*SER Seedling emergence rate (h/cm), SL Seedling length (/cm), RL Root length (/cm), SFW Shoot fresh weight (g/plant), RFW Root fresh weight (g/plant)

* p < 0.05, ** p < 0.01, *** p < 0.001

Discussion

Analysis of SNP mapping results

Rapid and uniform seedling emergence is a crucial agronomic trait affecting crop yield, and its genetic improvement is of great significance for achieving high and stable production. Seedling emergence rate directly affects population establishment and final yield formation by regulating early plant growth and development processes. SER directly affects population establishment and final yield formation by regulating early plant growth and development processes. Investigating seedling emergence traits in maize germplasm resources enables the selection of superior breeding materials and provides valuable genetic resources for germplasm innovation and marker-assisted breeding.

With continuous updates and improvements in maize reference genome sequencing, GWAS has become a crucial research method for deciphering complex genetic traits in maize. For instance, Zhang et al. conducted a GWAS analysis of 14 phenotypic traits in maize and identified a total of 30 SNPs and 82 candidate genes related to low-temperature tolerance during the growth and development stages of maize [48]. Using the Farm CPU model, Yang et al. conducted GWAS analysis on 126 maize inbred lines under two environments and detected 116 significant loci associated with ear position traits [32]. Through genome-wide association analysis, Li et al. identified 36 candidate genes in-volved in the regulation of maize seed germination [49]. Chen et al. detected 13 SNPs as-sociated with kernel color and identified 95 candidate genes [50].

To elucidate the molecular genetic mechanisms of seedling emergence in super-sweet corn, this study performed comparative mapping between the identified loci and previously reported QTL regions. The significant SNP Affx-91,134,889 on chromosome 3, identified in this study, was located within the mQTL3-3 interval (175,554,472 − 184,720,973 bp) identified by Han [51]. Additionally, the significantly associated SNP Affx-91,247,800 on chromosome 1 identified in this study was physically located 1.81 Mb apart from SYN18499 mapped by Meng et al. [52]. The SNP Affx-91,134,889 on chromosome 3 was physically located 1.69 Mb from S3_186453015 identified by Sun et al. [53], while Affx-90,761,633 was positioned 1.38 Mb away from PZE-101,111,994 reported by Meng et al. [52]. In summary, the four significant SNP loci identified in this study exhibit a high degree of consistency with previously reported loci. These findings demonstrate the reliability of our association mapping, and the significant SNP identified may offer valuable in-sights for studying seedling emergence traits in super-sweet corn.

Candidate gene analysis

Based on the GWAS analysis, this study identified 12 SNP loci, from which a total of 294 candidate genes were screened. Among them, 49 genes have well-defined functional annotations. Further functional analysis identified 15 candidate genes significantly associated with seedling emergence traits in super-sweet corn. Genetic analysis revealed 10 genes regulating SER and 5 genes governing seedling length SL.

Regarding the SER trait, four candidate genes were found to influence seed emergence by regulating root system development in plants. Of particular significance, the candidate gene GRMZM2G148723 encodes WPP domain-interacting protein 1 (WIP1), an essential factor for nuclear envelope anchoring in root tip cells. Functionally, WIP1 interacts with nucleoporin NUP136 as a scaffold protein to coordinate nucleocytoplasmic transport and auxin signal transduction, thereby maintaining the root stem cell niche and regulating root system growth, while its loss leads to impaired primary root growth and abnormal development [54, 55]. GRMZM2G123540 encodes ELD1 protein, a key negative regulator in light signaling that coordinates photomorphogenesis through a dual regulatory mechanism. Under low-light conditions, ELD1 maintains cellulose biosynthesis homeostasis and stimulates cell expansion to ensure normal skotomorphogenesis. Loss-of-function mutations cause photomorphogenic defects including impaired hypocotyl elongation and light sensitivity abnormalities [56]. The candidate gene GRMZM2G131275 encodes RHO guanyl-nucleotide exchange factor 3 (RhoGEF3), which orchestrates auxin signaling to promote lateral root growth and differentiation [57]. GRMZM2G027627 encodes a THDP-binding superfamily protein. Its IAR4 domain-containing variant regulates primary root growth under salt stress by mediating ROS-auxin crosstalk and PIN-dependent auxin transport [58], and concurrently regulates organ development through mitochondrial pyruvate dehydrogenase activity [59].

Among the three candidate genes influencing SER through enzymatic biosynthesis, GRMZM2G003853 annotated (HYD1) C-8,7 sterol isomerase represents a pivotal plant enzyme. This enzyme modulates root hair elongation by regulating sterol distribution, demonstrating significant potential for crop genetic improvement. Targeted regulation of HYD1 expression could optimize root system architecture and enhance stress resistance in cultivated varieties [60]. GRMZM2G337766 within the Affx-91,247,800 confidence interval encodes a leucine-rich repeat receptor kinase (PEPR1), which mediates plant immune responses triggered by PEPs while simultaneously suppressing root growth. Mechanistically, PEPR1 regulates root hair differentiation and elongation by modulating the expression of key regulators CPC and GL2 [61, 62]. GRMZM2G417954 encodes the NCED9 protein, a key rate-limiting enzyme in the abscisic acid (ABA) biosynthetic pathway. It works synergistically with the β-1,3-glucanase activity of BG14 to co-regulate seed longevity and dormancy. Loss-of-function of this gene significantly reduces seed lifespan [63]. These findings suggest that NCED9 in maize seeds precisely regulates seed dormancy and longevity at the molecular level by promoting β-1,3-glucanase-mediated callose degradation in embryonic tissues, thereby modulating ABA biosynthesis and associated signaling transduction pathways. Notably, this study has identified two pivotal candidate genes that significantly influence seedling emergence rate by modulating chloroplast developmental pathways. For example, GRMZM2G087804 is located at the Affx-90,520,409 locus on chromosome 3. The encoded key regulatory proteins GPRI1 and ATGLK1 cell-autonomously regulate chloroplast development in plants and primarily interact with the specific interactor protein BGH2 to suppress GLK-mediated transcriptional activation under dark conditions. Through precise modulation of the chlorophyll biosynthesis pathway, they thereby promote chloroplast differentiation, maintain photomorphogenesis in seedlings, and sustain optimal photosynthetic efficiency in plants [64]. GRMZM2G026614 encodes a chloroplast RNA-binding protein. Critically, CP33A protein plays a global role in processing and stabilizing chloroplast RNAs, and is functionally essential for seedling development [65, 66]; This study further identified GRMZM2G146163, a candidate gene annotated as DNAJ-like 20, that regulates SER through protein degradation pathways. The encoded protein exhibits high expression during early seed germination, where it specifically recognizes misfolded proteins and mediates targeted degradation via the ubiquitin-proteasome system. This mechanism maintains the stability of key chloroplast biosynthetic enzymes, ensuring normal metabolic activity during germination and promoting healthy seedling growth [46, 67–69].

Candidate genes associated with the SL trait included, within the confidence interval of the significantly associated locus Affx-90,790,712, the candidate gene GRMZM2G173124 encodes a C-x8-C-x5-C-x3-H-type zinc finger family protein. The TZF1 domain of this protein may regulate root meristem cell proliferation downstream of PRRs [47]. GRMZM2G080375 encodes PFK3, which is crucial for maintaining sugar homeostasis during leaf growth. During early seedling development, this gene regulates sugar stability by modulating PFK mutant activity, thereby influencing tissue-specific accumulation of glucosinolate defense compounds and ensuring proper leaf growth and development [70]. GRMZM2G080588 encodes a member of the Ser kinase family, whose activity peaks during cell division and serves as a critical regulator of mitosis [71]. After cytoplasmic localization and expression, it transitions into a purely nuclear protein, activating AUR1 activity and promoting seed embryo development [72].

Conclusions

This study utilized 201 super-sweet corn inbred lines as an association panel and employed three analytical models to conduct a GWAS of emergence traits, including SER and SL. A total of 10 loci significantly associated with seed emergence traits were identified, and 49 genes with clear functional annotations were screened within the confidence intervals. Based on a comprehensive evaluation of literature review and functional analysis, 15 key genes associated with seedling emergence traits were ultimately identified. Among them, the A/A and T/T genotypes were identified as superior alleles for enhancing the emergence rate in super-sweet corn. Pyramiding these favorable alleles can effectively improve the seedling emergence rate of super-sweet corn seeds. This study provides important insights into the genetic and molecular mechanisms underlying seedling emergence in super-sweet corn, and lays a theoretical foundation for breeding elite varieties with rapid and uniform emergence.

Abbreviations

GWAS

Genome-wide association study

SNP

Single Nucleotide Polymorphism

SER

Seedling emergence rate (h/cm)

ET

Emergence time

RT

Radicle emergence time

SL

Seedling length (/cm)

SD

Standard deviation

CV

Coefficient of variation

h²

Heritability

Chr

Chromosome

PVE

Phenotypic variation explained

LD

Linkage disequilibrium

RL

Root length (/cm)

SFW

Shoot fresh weight (g/plant)

RFW

Root fresh weight (g/plant)

Sh2

Shrunken-2

Authors’ contributions

NC and XC designed the research. NC, YW and YY analyzed the data. NC and YW drafted the manuscript. NC, YY and AR revised the manuscript. NC conducted the experiments. YW, AR, HY, and YW participated in the spike germination experiments. All the authors read and approved the final version of the manuscript.

Funding

This research was funded by the Emergency Management of the National Natural Science.

Foundation of China (31440066), Key Discipline Construction Funds for Crop Science of Anhui.

Sciences and Technology University (No. XK-XJGF001), the Research Development Fund of Anhui.

Science and Technology University (FZ230126), Joint Research Project for Elite Maize Varieties of Anhui Province (2020AHMS01), the Science and Technology Innovation Groups of Anhui Science and Technology University (2023KJCXTD002), Supported by Bio-breeding Laboratory of Anhui Province (2025SWYZ0200).

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.