The accumulation of arsenic (As) in rice poses a significant threat to food safety and human health. Breeding rice varieties with low As accumulation is an effective strategy for mitigating the health risks associated with arsenic-contaminated rice. However, the genetic mechanisms underlying As accumulation in rice grains remain incompletely understood. We evaluated the As accumulation capacity of 313 diverse rice accessions grown in As-contaminated soils with varying As concentrations. Six rice lines with low As accumulation were identified. Additionally, a genome-wide association studies (GWAS) analysis identified 5 QTLs significantly associated with As accumulation, with qAs4 being detected in both of the experimental years. Expression analysis demonstrated that the expression of LOC_Os04g50680, which encodes an MYB transcription factor, was up-regulated in the low-As-accumulation accessions compared to the high-As-accumulation accessions after As treatment. Therefore, LOC_Os04g50680 was selected as a candidate gene for qAs4. These findings provide insights for exploiting new functional genes associated with As accumulation and facilitating the development of low-As-accumulation rice varieties through marker-assisted breeding.

Arsenic (As) contamination in rice poses a serious risk to food safety and human health. Genetic dissection of As-related quantitative trait loci (QTLs) provides a sustainable strategy for breeding low-As cultivars. In this study, we aimed to improve the detection of minor-effect QTLs for total As accumulation by optimizing both environmental and genetic factors. A recombinant inbred line (RIL) population derived from the cross between Yuzhenxiang (YZX, indica) and YBK (Javanica) was used for initial QTL mapping, and a single locus, qAs1, was identified on chromosome1. To enhance As uptake and phenotypic differentiation, we conducted QTL validation and fine mapping under high-As and continuously flooded conditions using near-isogenic lines (NILs) to minimize background genetic interference. The effect of qAs1 was consistently validated across generations, and the locus was refined to a 159.5 kb genomic interval. Transcriptome analysis revealed three differentially expressed genes (LOC_Os01g52110, LOC_Os01g52214, and LOC_Os01g52260) involved in redox regulation and detoxification. These findings demonstrate the effectiveness of NIL-based fine mapping under optimized environmental conditions and provide promising targets for the genetic improvement of low-As rice cultivars.