All the pupae of L. botrana used for the in vitro and field trials were obtained from the breeding project of the Entomology Laboratory of INIA La Cruz. The strain B. pseudobassiana RGM 1747 was isolated from Polistes gallicus (Hymenoptera: Vespidae) and stored in the Bank of the Chilean Collection of Microbial Genetic Resources of INIA Quilamapu.
DNA extraction, PCR conditions and sequencing
The conidial DNA of strain RGM 1747 was extracted using the Quick DNA Fungal / Bacterial Kit (Zymo Research, CA, USA) following the manufacturer's instructions. The partial sequences amplified by PCR were for the rpb1, rpb2, tef and Bloc loci , which were sequenced at Macrogen (Seoul, South Korea). The DNA sequences determined for the four loci were submitted to the GENBANK Nucleotide Sequence Database under accession numbers MH048640, MH048641, MH048642 and MH048643.
Multilocus sequence analysis (MLSA) of Beauveria sp
Concatenated alignments were performed with MUSCLE for the Bloc, tef, rpb1 and rpb2 loci in 34 strains . Poorly aligned and dissimilar regions were eliminated from the analysis. The evolutionary history was inferred using neighbor joining method . The evolutionary distances were computed using the Tamura 3-parameter method . The strength of the internal branches of the resulting trees was statistically evaluated by bootstrap analysis with 1000 bootstrap replications . The rate of the variation of the nucleotide substitutions was modeled using a gamma distribution (shape parameter=0.68). The evolutionary analyses were conducted in MEGA7 .
The entomopathogenic fungi were spread over the surface of PDA (Difco Ô, NJ, USA) in a plate. The fungi were grown for one week at 25°C, and the spores and hyphae produced were collected through the addition of 10 mL of PBS (pH 7.4) containing 0.05% Tween 20 (Merck, Darmstadt, Germany) to a plate. The conidial concentration was determined by counting in a Neubauer chamber and adjusted to 106 conidia/mL. The formulation was generated to maintain the fungal biomass in the winter using 2.5% rice flour, 2.5% skim milk, 20% glycerol (Merck), 0.2% Silwet L-77 Ag (Arysta LifeScience, Santiago, Chile), 73.8% distilled water and 1% EPF at 106 conidia/mL. All the components were mixed in a beaker at 300 rpm for 30 min.
In vitro EPF formulation evaluation against L. botrana
The assessment of the formulation was carried out in in vitro assays at 25°C. One milliliter of the formulation mix was placed in the center of a paper disk that covered the bottom of a plate, and ten moth pupae were deposited on the paper. As a control, water was used instead of the formulation. The infection of pupae with EPF was monitored every day for eight days. The experiment included three replicates.
The assay was developed in the V. vinifera ´Red Globe´ field at La Platina research station in the Metropolitan region of Santiago, Chile, in natural environmental conditions in July, during the winter. The field trial was arranged in a randomized block design with four treatments, each with four replications with three vines each (4x4x3). To infest the vines, ten pupae without cocoons per plant were deposited under the rhytidome, covered with a tulle mesh and fixed with adhesive tape to avoid dispersion of the pupae and the entry of other arthropods. The first treatment was performed using one application of the formulation. In the second treatment, we used two applications on days one and seven. In the third treatment, we performed applications during days one, seven and fourteen. The fourth treatment was the control using water. The EPF concentration for treatments one, two and three was 109 CFU/L. The applications were performed with a hand pump, and wetting was performed with a volume of 500 mL per vine. Each treatment was inspected seven days after each application. For the inspection of the infected pupae, we removed the pupae from under the rhytidome and deposited them in a plate containing a humid paper disk when the EPF had completed colonization. The evaluation of efficacy was performed over 48 h at 25°C.
The assay was developed in V. vinifera plants located in urban residences in the Metropolitan region of Santiago, Chile, under natural environmental conditions in August, during the winter diapause of pupae with cocoons. The urban trial was arranged in a randomized block design with four treatments, each with three replications with three vines each (4x3x3). The distribution of the blocks in the urban area was set up according the captures of the moths in pheromone (E7, Z9-dodecadienyl acetate) traps during the spring and summer by SAG. The treatments and the methodology for the application of the formulation and inspection of the vines were the same as those described for the field trial. Finally, the pupae grown in cocoons with EPF were dissected to determine infection by the fungus. The pupae were dyed with lactophenol blue (Merck) to observe the penetration of the fungus into their structure.
Temperature and humidity data during the field and urban trials were collected through the weather station network (http://agromet.inia.cl/).
The efficacy of the formulation in vitro and in field trials was determined for uniform populations , and the efficacy in the urban trials was determined for nonuniform populations . The percentages of efficacy in the field test and urban test were compared by Tukey´s HSD test (a=0,05). All the experiments were analyzed using the software Statgraphics Centurion XVII (Statpoint Technologies Inc., VA, USA).