Genetic Dissection of Tobacco (Nicotiana tabacum L.) Plant Height Using Single-Locus and Multi-Locus Genome-Wide Association Studies

Tobacco (Nicotiana tabacum L.) plant height (PH) is a biologically important plant architecture trait linked to yield and controlled by polygenes. However, limited information is available on quantitative trait nucleotides (QTNs), alleles, and candidate genes. The plant height of 94 tobacco accessions and their 126,602 SNPs were measured to conduct a genome-wide association study (GWAS) using four multi-locus (ML) and two single-locus (SL) models to better understand its genetic basis. The ML and SL models detected 181 and 29 QTNs, respectively, across four environments/BLUP; LOD scores ranged from 3.01–13.45, and the phenotypic variance explained (PVE) ranged from 0.69–25.37%. Fifty-two novel, stable QTNs were detected across at least two methods and/or two environments/BLUP, with 0.64–24.76% PVE. Among these, 49 QTNs exhibited significant phenotypic differences between two alleles; the distribution of elite and alternative alleles for each accession ranged from 3–42 and 6–46, respectively, in the mapping population. Seven cross combinations in two directions were predicted using alleles of validated QTNs, including Qinggeng × KY14 for taller plants and RG112 × VA115 for shorter plants. We identified 27 candidate genes in the vicinity of 49 stable QTNs based on comparative genomics, gene ontology (GO), and KEGG enrichment analysis, including AP2, Nitab4.5_0000343g0250.1 (ROC1), Nitab4.5_0000197g0010.1 (VFB1), CDF3, AXR6, KUP8, and NPY2. This is the first study to use genotyping-by-sequencing (GBS) of SNPs to determine QTNs, potential candidate genes, and alleles associated with plant height. These findings could provide a new avenue for investigating the QTNs in tobacco by combining SL and ML association mapping and solid foundations for functional genomics, the genetic basis, and molecular breeding for PH in tobacco.