Strain differentiation in ‘flavescence dorée’ phytoplasmas on SecY and Tuf genes.

Introduction A wide world severe spreading of ‘bois noir’ (BN) disease is recently reported in the majority of grapevine growing areas. Recent findings indicate that molecular variability inside the BN-associated stolbur phytoplasmas is present and for some genes such as elongation factor Tu (tuf gene) it could be related with epidemic features (Langer and Maixner, 2004). Variability in five genes of BN phytoplasmas present in grapevine epidemics was therefore investigated on samples collected in infected grapevine growing areas in Italy and Serbia where BN has been molecularly identified and reported since several years. Materials and Methods During summer 2009-2011, routine surveys carried out to verify identity of phytoplasmas associated with yellows symptoms in grapevine growing areas in Italy and Serbia allow to identify and partially characterize BN phytoplasmas by RFLP analyses with Tru1I and MboII on R16F2n/R2 amplicons (Contaldo et al., 2009). In Serbia 23 plants were sampled from different geographic locations i.e. Bela Crkva, Smederevo, Krčedin, Aleksandrovac and Radmilovac (Table 1), representing some of the major viticultural areas of the Country. In Italy 23 samples were collected in Emilia-Romagna from 6 out of the 9 provinces of the region (Table 1). Total nucleic acids were extracted from midribs and phloem scrapes, and amplicon produced on 16Sr, tuf (Langer and Maixner, 2004), amp (Fabre et al., 2011), secY (Lee et al., 2010) and groEL (J. Mitrović and B. Duduk, unpublished) genes were subjected to RFLP analyses with selected restriction enzymes according with amplicon (Table 1). Reference strains employed maintained in periwinkle were STOL (from Serbia), ASLO (from Slovenia) STOLC, STOL-PO, STOL-CH, MOL (from France). Results and Discussion The ‘bois noir’ infected samples tested from Italy and from Serbia showed two different profiles (a and a+b) on the R16F2n/R2 amplicons digested with MboII; profile (A+B) was possibly indication of mixed strains/double operon presence as previously described (Contaldo et al., 2011) (Table 1). All the samples amplified on 16Sr gene were also successfully amplified on tuf and stamp genes. The comparison of RFLP profiles obtained on tuf gene with HpaII and on stamp gene with Tru1I indicated that tuf-type from Italy was always associated with identical profile on stamp gene: tuf-type a/stamp a and tuf-type b/stamp a+b. This correlation was not clearly defined in the samples from Serbia where only tuf-type b was present but it was associated with three different stamp profiles (stamp a, stamp b and stamp e) (Table 1). It is possible to hypotize a correlation between epidemiological features and tuf/stamp polymorphisms considering that common recognized vector of BN to grapevine is Hyalesthes obsoletus. This cixiid is very likely composed by insect populations having different characteristics not yet defined and not easy to define (J.Y. Rasplus, personal communication) therefore interaction of different insect population could have influenced the polymorphism of these genes in the Serbian samples. Amplification of secY and groEL genes allow to amplify a number of strains but not all those tested. Polymorphism of secY gene on Italian samples with Tru1I and Tsp509I indicates that grapevine profile is distinguishable from those of all the reference strains employed. However three RFLP groups were obtained in reference strains: secY a (STOL), secY b (STOL-C, STOL-PO and STOL-CH) and secY c (ASLO and MOL). Polymorphisms on groEL gene with Hpy188I allow to distinguish two RFLP groups in the Serbian samples, while three different profiles were identified in the samples from Italy.