Genomic analyses provide insights into spinach domestication and the genetic basis of agronomic traits

Spinach (*Spinacia oleracea*), a globally significant leafy vegetable, presents a rich source of vitamins and minerals. Despite its importance, ongoing efforts are needed to improve spinach for traits like reduced oxalate content (which can cause health issues) and enhanced disease resistance. Previous research identified QTLs, markers, and genes related to leaf morphology, bolting, flowering, and nutritional quality; however, a high-quality, complete genome assembly has been lacking. This study aims to address this gap, utilizing advanced sequencing technologies to generate a chromosome-scale reference genome, conduct population genomic analysis on a large collection of cultivated and wild spinach accessions, and perform GWAS to pinpoint genes associated with key agronomic traits. This comprehensive approach will provide a deeper understanding of spinach evolution, domestication, and the genetic basis of economically important traits, ultimately guiding future breeding efforts.

Existing research on spinach genetics has primarily focused on identifying quantitative trait loci (QTLs) and markers associated with traits such as leaf morphology, bolting time, flowering, and nutritional composition. While some progress has been made in identifying candidate genes for these traits, a high-quality reference genome was lacking, hindering more comprehensive genomic analyses. Several draft genomes have been previously published, but none achieved the chromosome-level completeness necessary for detailed comparative genomics, genetic mapping, and gene cloning studies. This study aimed to overcome these limitations by generating a significantly improved genome assembly and then integrating this with population genomic and GWAS analyses to obtain a more holistic understanding of the genetic basis of spinach traits.

The researchers employed a combination of advanced sequencing technologies to generate a highly accurate and complete spinach genome. A highly homozygous inbred line, Monoe-Viroflay, was selected for genome sequencing. This involved generating a substantial amount of PacBio long reads and Illumina short reads. PacBio reads were self-corrected and assembled into contigs, which were further polished using Illumina paired-end reads. The contigs were then scaffolded into chromosomes using Hi-C data (high-throughput chromosome conformation capture) and Chicago libraries. The final genome assembly was rigorously evaluated for quality, completeness, and continuity using several metrics. Gene prediction was performed integrating various approaches, including homology searches, transcriptome data, and ab initio predictions. For the population genomic analysis, 305 accessions of cultivated and wild spinach were sequenced, resulting in the identification of a large number of SNPs and SVs. GWAS was then performed using a linear mixed model and a large panel of SNPs. The study also included detailed phenotyping of 20 agronomic traits across a wide range of spinach accessions. The traits included measures of plant architecture, leaf morphology, petiole characteristics, bolting time, flowering time, downy mildew resistance, and oxalate content. Finally, the researchers identified domestication sweeps by comparing the cultivated spinach genomes with those of their closest wild relative, *S. turkestanica*, using several different statistical methods.

The researchers successfully assembled a high-quality chromosome-scale reference genome for spinach, significantly improving on previously available assemblies. The genome is characterized by a high level of completeness and accuracy, along with an improved contig N50 size, indicating fewer gaps and a higher level of continuity. Analysis of the genome revealed that spinach has undergone substantial genome rearrangements after its divergence from the ancestral Chenopodiaceae karyotype. This was associated with high repeat content in the genome. Population genomic analysis showed that *S. turkestanica* is the closest wild relative to cultivated spinach, with relatively low genetic differentiation (Fst) between the two, suggesting a weak bottleneck during spinach domestication. The nucleotide diversity (π) was higher in the Asian population of cultivated spinach compared to the European population, supporting an Asian origin of spinach. The GWAS identified numerous regions associated with 20 important agronomic traits. A significant region associated with downy mildew resistance was found on chromosome 3, containing several candidate genes including R genes. Other regions associated with leaf morphology, plant architecture, flowering time, and oxalate content were identified and candidate genes implicated for further study. The analysis identified almost 1000 domestication sweeps in cultivated spinach, some of which overlapped with GWAS signals related to important traits, suggesting a role of human selection in driving the evolution of spinach phenotypes.

The findings of this study provide crucial insights into the evolutionary history and domestication of spinach, as well as the genetic basis of key agronomic traits. The high-quality reference genome assembled represents a significant improvement over previous resources. The observation of substantial genome rearrangements in spinach, along with a high repeat content, provides valuable insights into genome evolution in this species. The results of the population genomic analysis confirm the close relationship between cultivated spinach and *S. turkestanica*, highlighting the potential of wild relatives as valuable resources for future breeding programs. The GWAS results, combined with the domestication sweep analysis, clearly demonstrate the influence of artificial selection on spinach phenotypic evolution. The identification of numerous associated regions and candidate genes for traits of economic importance will accelerate spinach breeding efforts. The discovery of candidate genes for downy mildew resistance opens new avenues for developing resistant cultivars, crucial for sustainable spinach production. Similarly, the identification of candidate genes for oxalate reduction is important for enhancing the nutritional value and safety of spinach for human consumption.

This study provides a highly improved chromosome-scale reference genome for spinach, along with comprehensive population genomic and GWAS analyses. These findings elucidate the evolutionary history and genetic architecture of key agronomic traits in spinach, advancing our understanding of its domestication and providing crucial resources for future breeding efforts. The characterization of candidate genes associated with downy mildew resistance and oxalate content provides promising targets for marker-assisted selection (MAS) strategies. Future research could focus on functional validation of these candidate genes and the development of new molecular markers for use in MAS programs.

The study focused on a specific set of agronomic traits and geographical regions, limiting the overall generalizability of some findings. While the Monoe-Viroflay inbred line provided a high-quality genome assembly, it might not fully capture the genetic diversity present in all spinach populations. The GWAS analysis relied on association mapping; therefore, causal relationships between identified genes and phenotypic variation need further experimental validation.