Abstract
Sophora flavescens is a cornerstone of Traditional Chinese Medicine (TCM), particularly valued for its root, known as Kushen. Despite its pharmaceutical significance, genomic resources for this species have been limited. In this study, we present a high-quality, chromosomal-level de novo assembly of the S. flavescens genome using Nanopore long-read sequencing and Hi-C scaffolding. The resulting 2.08 Gb genome is approximately five times larger than its closest sequenced relative, Lupinus albus, a disparity driven by a recent and rapid proliferation of transposable elements (TEs), which constitute over 80% of the genome. We annotated 60,485 protein-coding genes and characterized the biosynthetic pathways for major bioactive compounds, including alkaloids and flavonoids. This assembly provides a critical genetic blueprint for understanding the evolution of the Fabaceae family and accelerating the molecular breeding of this medicinal powerhouse.
Introduction
Sophora flavescens (Kushen) has a long history of use in TCM formulations, most notably in the "Compound Kushen Injection" (CKI), a state-approved adjunct therapy for cancer treatment in China. Its therapeutic properties are derived from a rich repertoire of bioactive compounds, primarily quinolizidine alkaloids (QAs) like matrine and oxymatrine, as well as various flavonoids and isoflavonoids.
Prior to this study, the lack of a reference genome hindered efforts to map the evolutionary history of S. flavescens or pinpoint the precise genetic mechanisms responsible for its complex chemical profile. The plant’s genome is characterized by high heterozygosity (~1.4%) and a high percentage of repetitive sequences, making traditional sequencing approaches insufficient.
Methods and Materials
To overcome the structural complexity of the S. flavescens genome, a hybrid sequencing and scaffolding strategy was implemented.
Sequencing Strategy
- Nanopore Long-Read Sequencing: Generated 222 Gb of data to provide the structural backbone and span repetitive regions.
- Illumina Short-Read Sequencing: Produced 226 Gb of high-accuracy data used for error correction (polishing) of the long-read assembly.
- Hi-C Scaffolding: Utilized 242 Gb of chromatin interaction data to organize contigs into nine distinct chromosomal scaffolds.
Assembly Pipeline
The team benchmarked 17 different assembly strategies using various bioinformatics tools (e.g., CANU, Flye, SMARTdenovo) to identify the optimal pipeline for handling the plant's high heterozygosity. The final workflow involved draft assembly, multiple rounds of polishing with Racon and Medaka, error correction with Nextpolish, and haplotig purging to ensure a non-redundant representative genome.
Key Results and Genomic Insights
Assembly Quality
The final assembly achieved a scaffold N50 of 233.47 Mb, indicating exceptional contiguity. The completeness of the genome was validated with a BUSCO score of 93.3%, confirming that the vast majority of conserved orthologous genes are present and intact.
The Transposable Element (TE) Explosion
A defining feature of the S. flavescens genome is its size (2.08 Gb). Analysis revealed that ~83% of the genome consists of TEs, specifically LTR retrotransposons such as Gypsy and Copia. This suggests that S. flavescens underwent a massive, recent expansion of these elements after diverging from its relatives, explaining why its genome is significantly larger than other legumes like Lupinus albus.
Biosynthetic Pathways and Gene Expression
By integrating the genome assembly with RNA-Seq data from leaf, stem, and root tissues, the study characterized 21 key genes involved in flavonoid/isoflavonoid biosynthesis. Furthermore, the expression profiles support the hypothesis that quinolizidine alkaloids are synthesized in the aerial "green parts" of the plant and subsequently translocated to the roots for storage.
Conclusion
The chromosomal-level assembly of Sophora flavescens represents a major milestone for medicinal plant research. By identifying the candidate genes for bioactive compound production and revealing the impact of TE expansion on genome evolution, this study provides the foundation for future functional genomics and CRISPR-based metabolic engineering. These resources will ultimately facilitate the development of standardized, high-potency Kushen varieties for clinical use.
Summary updated