Bulletin of Insectology 60 (2): 115-116, 2007 ISSN 1721-8861 Purification of phytoplasma DNA for sequencing the “flavescence dorée” and stolbur phytoplasma genomes Patricia CARLE, Sylvie MALEMBIC-MAHER, Xavier FOISSAC UMR1090 Génomique Diversité Pouvoir Pathogène & ISVV, INRA & Université Bordeaux 2, Villenave d'Ornon, France Abstract In order to compare genomes of the “flavescence dorée” and stolbur phytoplasmas, we produced fractions from infected plants enriched in phytoplasma DNA. DNA of phytoplasma strains FD92 and stolbur PO were respectively prepared from broad beans and tomato plants by repeated bis-benzymide-CsCl buoyant density gradient centrifugations. Sequence of cloned DNAs were ho- mologous to known phytoplasma sequences in 27 % to 54 % of the cases. According to these data, the sequencing of the genome of these two organisms can be undertaken. A draft of the FD92 phytoplasma genome using the pyrosequencing technology will be first envisaged. Key words: Mollicutes, “bois noir”, CsCl gradient, genome sequencing. Introduction Vineyards of southern Europe are affected by the “fla- vescence dorée” phytoplasma (FD), inducing an epi- demic and quarantine disease of the grapevine. FD is transmitted by the ampelophagic leafhopper of nearctic origin Scaphoideus titanus Ball, and has an important negative economic impact on grapevine production (Boudon-Padieu, 2002). Another grapevine yellows causing serious damage in Euro-mediterranean vine- yards is the “bois noir” (BN) disease. BN is associated with the endemic stolbur phytoplasma which is mainly transmitted by a polyphagous planthopper, Hyalesthes obsoletus Signoret (Fos et al., 1992). Stolbur phyto- plasmas (Stolp) are reported across Europe, parts of Asia and Africa. They affect a wide range of crops in- cluding tomato, potato pepper, tobacco, lavender and strawberry. Whereas important progress has been made in phyto- plasma classification and ecology, little is known about mechanisms of phytopathogenicity and transmission by insects; these research areas should benefit from the knowledge and comparative analysis of phytoplasma genomes. Recently, the complete genome sequences of 2 lines of Candidatus Phytoplasma asteris’, OY-M (860 kbp) and AY-WB (706 kbp) have been published. Genome char- acterization is underway for ‘Ca. P. pruni’, ‘Ca. P. mali’ and ‘Ca. P. australiense’. A modified suppression sub- tractive hybridization (SSH) method allowed to isolate some nearly pure Stolp DNA (line PO) from infected Catharanthus roseus periwinkle plants, and to charac- terize 15% of stolbur phytoplasma genome (Cimerman et al., 2006). However this subtractive method did not provide gene libraries representative enough for se- quencing the Stolp genome. For FDp no SSH gene li- braries could be prepared. As a consequence, the clas- sical CsCl density gradient method was preferred for the preparation of enriched sources of phytoplasma DNA. Materials and methods FD92 phytoplasma was transmitted to Vicia faba var. Aqua Dulce using leafhoppers collected in FD-affected vineyards in South-west of France in 1992, and main- tained since that time by continuous serial transmissions from broad bean to broad bean using the alternative leafhopper Euscelidius variegatus (Kirschbaum). Stolbur phytoplasma (PO) was transmitted by grafting Solanum lycopersicum tomato plants var. Elsa Craig. Phytoplasma DNA was enriched from host plant DNA, by 4 repeated bis-benzymide Cesium chloride density gra- dients (Kollar and Seemüller, 1989). The bis-benzymide Hoechst 33258 binds preferentially to adenine + thymine (A+T)-rich sequences thereby reducing their buoyant density. So, the phytoplasma DNA band was character- ized by a lower buoyant density than that of the host DNA. Total nucleic acids were extracted from stems and leaf midribs of infected broad beans, periwinkles or to- matoes by the cetyltrimethylammonium bromide (CTAB) procedure (Murray and Thompson, 1980). Purified nucleic acid pellets were resuspended in 1X TE buffer. One mg of total DNA was used for each CsCl gradient. The same procedure was performed with healthy plants as a control. DNA was mechanically sheared or cut by HindIII. Nebulization of DNA was performed using nebulizers (Invitrogen life technologies) according to the manufac- turer’s instructions to achieve a 1-10 kbp size range. Af- ter reparation, DNA was ligated in dephosphorylated pBS vector, previouly linearized by HincII and SmaI or by HindIII. Plasmids were cloned in E.coli DH10B cells, and purified using SV Miniprep Promega Kit. In- serts of recombinant plasmids were sequenced on ABI Prism sequencer. Results After the first centrifugation, the expected enriched phy- toplasma DNA was visible as a faint band above the main band of the host DNA (figure 1, gradient 1) by