From vineyard to genome: optimized enrichment and sequencing of Flavescence dorée phytoplasma from grapevine samples

Phytoplasmas are non-culturable obligate intracellular bacteria that cause considerable economic losses in agriculture. Genome sequencing provides crucial insights into their biology and vector dependence. However, genome studies on phytoplasmas are often hampered by their low abundance in naturally infected plants. Propagation in test plants is usually necessary but time-consuming and resource-intensive, especially for quarantine phytoplasmas such as the phytoplasma causing Flavescence dorée (FD), a serious threat to European viticulture. To overcome these challenges, we aimed to develop a protocol for efficient enrichment of phytoplasma DNA directly from field-collected samples, enabling genome sequencing using both Illumina and Oxford Nanopore Technologies platforms. We evaluated six sample preparation protocols that included stepwise enrichment steps to improve phytoplasma genome coverage and assembly quality. The most effective approach combined differential centrifugation, CTAB extraction and removal of CpG-methylated host DNA and resulted in a notable increase in the relative abundance of phytoplasma reads compared to other protocols. Rarefaction analysis of the dataset generated using this protocol demonstrated that the entire phytoplasma genome was covered by reads in a dataset comprising 3 billion nucleotides. We also evaluated and compared de novo phytoplasma genome assemblies generated from short Illumina reads and long nanopore sequencing reads. While Illumina sequencing yielded more accurate assemblies with longer total lengths, the assemblies derived from nanopore sequencing data contained longer individual contigs. This advantage was reflected in hybrid assemblies that combined both technologies, yielding longer phytoplasma contigs than assemblies from Illumina datasets and lower mismatch rates compared to assemblies from nanopore sequencing datasets. A hybrid de novo assembled genome of the Slovenian FD phytoplasma isolate achieved 96% reference genome coverage, with high contiguity and low error rates. This streamlined and accessible protocol enables high-quality genome sequencing of phytoplasma-infected grapevines without the need for propagation in test plants. This facilitates broader phytoplasma research and can potentially be extended to other naturally infected phytoplasma hosts or organisms infected with other non-culturable microbes.