Starch serves as a critical storage component, significantly influencing the grain yield and quality of maize (Zea mays L.). Understanding the genetic basis of natural variation in kernel starch content (SC) is essential for maize breeding to meet future demands. A genome-wide association study (GWAS) identified 84 and 96 loci associated with kernel SC across two years, overlapping with 185 candidate genes. The candidate gene ZmMYB71, encoding a MYB-related transcription factor, demonstrated the highest co-expression frequency with starch synthesis genes. Analysis revealed that ZmMYB71 functions as a nuclear located transcription repressor, and mutants exhibited increased kernel SC by over 2.32%, with minimal impact on amylose content or 100-grain weight. Sh1, Sh2, and GBSSI exhibited up-regulation in mutants by 1.56-, 1.45- and 1.32-fold, respectively, aligning with RNA sequencing results; their promoter activities appear directly repressed by ZmMYB71 through the GATATC and TTAGGG motifs. Additionally, the ZmMYB71 elite haplotype Hap1 occurred in over 55% of the high-starch maize sub-populations Iowa Stiff Stalk Synthetic (BSSS) and Partner B (PB), but only in 7.14% of the low-starch sub-population Partner A (PA). Analysis of Hap1 haplotype frequencies across breeding stages revealed a significant increase to 40.28% in inbred groups released after 2010, compared to 28.57 and 27.94% in 1980 and 1990, and 2000, respectively. These findings enhance understanding of natural variation in maize kernel SC and establish ZmMYB71 as a negative regulator with potential applications in SC improvement.

Genomic prediction (GP) in plant breeding has the potential to predict and identify the best-performing hybrids based on the genotypes of their parental lines. In a GP experiment, 34 elite inbred lines were selected to make 285 single-cross hybrids in a partial-diallel cross design. These lines represented a mini-core collection of Chinese maize germplasm and comprised 18 inbred lines from the Stiff Stalk heterotic group and 16 inbred lines from the Non-Stiff Stalk heterotic group. The parents were genotyped by sequencing and the 285 hybrids were phenotyped for nine yield and yield-related traits at two locations in the summer sowing area (SUS) and three locations in the spring sowing area (SPS) in the main maize-producing regions of China. Multiple GP models were employed to assess the accuracy of trait prediction in the hybrids. By ten-fold cross-validation, the prediction accuracies of yield performance of the hybrids estimated by the genomic best linear unbiased prediction (GBLUP) model in SUS and SPS were 0.51 and 0.46, respectively. The prediction accuracies of the remaining yield-related traits estimated with GBLUP ranged from 0.49 to 0.86 and from 0.53 to 0.89 in SUS and SPS, respectively. When additive, dominance, epistasis effects, genotype-by-environment interaction, and multi-trait effects were incorporated into the prediction model, the prediction accuracy of hybrid yield performance was improved. The ratio of training to testing population and size of training population optimal for yield prediction were determined. Multiple prediction models can improve prediction accuracy in hybrid breeding.

Flowering time (FT) is a key maize domestication trait, variation in which allows maize to grow in a wide range of latitudes. Although previous studies have investigated the genetic control of FT-related traits per se, few studies of FT hybrid performance have been published. We characterized the genomic architecture associated with hybrid performance for FT in a hybrid panel by testcrossing Chang 7–2 with 328 Ye478×Qi319 recombinant inbred lines (RILs). We identified 11 quantitative trait loci (QTL) for hybrid performance in FT-related traits, including a major QTL qFH10 that controls hybrid performance and heterosis in a summer maize-growing region. However, this locus acts in regulating FT traits per se only in a spring maize-growing region. We validated ZmCCT10 as a candidate gene for qFH10 and found that differences between hybrids and their parental lines in DNA methylation in the differentially methylated region (DMR, –700 to –1520) of the ZmCCT10 promoter affected gene expression pattern and thereby FT in the summer maize-growing region.

Flowering time is an indicator of adaptation in maize and a key trait for selection in breeding. The genetic basis of flowering time in maize, especially in response to plant density, remains unclear. The objective of this study was to identify maize quantitative trait loci (QTL) associated with flowering time-related traits that are stably expressed under several plant densities and show additive effects that vary with plant density. Three hundred recombinant inbred lines (RIL) derived from a cross between Ye 478 and Qi 319, together with their parents, were planted at three plant densities (90,000, 120,000, and 150,000 plants ha⁻¹) in four environments. The five traits investigated were days to tasseling (DTT), days to silking (DTS), days to pollen shed (DTP), interval between anthesis and silking (ASI), and interval between tasseling and anthesis (TAI). A high-resolution bin map was used for QTL mapping. In the RIL population, the DTT, DTS, and DTP values increased with plant density, whereas the ASI and TAI values showed negligible response to plant density. A total of 72 QTL were identified for flowering time-related traits, including 15 stably expressed across environments. Maize flowering time under different densities seems to be regulated by complex pathways rather than by several major genes or an independent pathway. The effects of some stable QTL, especially qDTT8-1 and qDTT10-4, varied with plant density. Fine mapping and cloning of these QTL will shed light on the mechanism of flowering time and assist in breeding early-maturing maize inbred lines and hybrids.

Gray leaf spot (GLS), caused by Cercospora zeae-maydis, is an important foliar disease of maize (Zea mays L.) worldwide, resistance to which is controlled by multiple quantitative trait loci (QTL). To gain insights into the genetic architecture underlying the resistance to this disease, an association mapping population consisting of 161 inbred lines was evaluated for resistance to GLS in a plant pathology nursery at Shenyang in 2010 and 2011. Subsequently, a genome-wide association study, using 41,101 single-nucleotide polymorphisms (SNPs), identified 51 SNPs significantly (P < 0.001) associated with GLS resistance, which could be converted into 31 QTL. In addition, three candidate genes related to plant defense were identified, including nucleotide-binding-site/leucine-rich repeat, receptor-like kinase genes similar to those involved in basal defense. Two genic SNPs, PZE-103142893 and PZE-109119001, associated with GLS resistance in chromosome bins 3.07 and 9.07, can be used for marker-assisted selection (MAS) of GLS resistance. These results provide an important resource for developing molecular markers closely linked with the target trait, enhancing breeding efficiency.

Artificial selection during domestication and post-domestication improvement results in loss of genetic diversity near target loci. However, the genetic locus associated with cob glume color and the nature of the genomic pattern surrounding it was elusive and the selection effect in that region was not clear. An association mapping panel consisting of 283 diverse modern temperate maize elite lines was genotyped by a chip containing over 55,000 evenly distributed SNPs. Ten-fold resequencing at the target region on 40 of the panel lines and 47 tropical lines was also undertaken. A genome-wide association study (GWAS) for cob glume color confirmed the P1 locus, which is located on the short arm of chromosome 1, with a −log₁₀ P value for surrounding SNPs higher than the Bonferroni threshold (α/n, α < 0.001) when a mixed linear model (MLM) was implemented. A total of 26 markers were identified in a 0.78 Mb region surrounding the P1 locus, including 0.73 Mb and 0.05 Mb upstream and downstream of the P1 gene, respectively. A clear linkage disequilibrium (LD) block was found and LD decayed very rapidly with increasing physical distance surrounding the P1 locus. The estimates of π and Tajima's D were significantly (P < 0.001) lower at both ends compared to the locus. Upon comparison of temperate and tropical lines at much finer resolution by resequencing (180-fold finer than chip SNPs), a more structured LD block pattern was found among the 40 resequenced temperate lines. All evidence indicates that the P1 locus in temperate maize has not undergone neutral evolution but has been subjected to artificial selection during post-domestication selection or improvement. The information and analytical results generated in this study provide insights as to how breeding efforts have affected genome evolution in crop plants.