LAUSR

In Chile, the planted area of European hazelnut (Corylus avellana L.) reaches around 30,000 hectares, mainly concentrated in the central and southern area of the country where climate and soil provide a natural environment well suited to growing this species. Only a few diseases affect this nut tree in Chile. During the spring seasons in 2018 and 2020, European hazelnut plants (6 to 20% of incidence) exhibited wood necrosis and vascular discoloration of branches, with reduced growth, cankers and wilt branches, in orchards located in San Clemente and Curicó, Maule Region, Bulnes and El Carmen, Ñuble Region, Chile (36°45'-36°54' S; 71°03'-72°26' W). Symptomatic tissues were surface disinfected using a ~1% commercial sodium hypochlorite solution. Disinfected tissues were cut longitudinally, placed onto potato dextrose agar (PDA, Difco) plates, and incubated at 25 °C in the dark for 48 hours. Fungal hyphal tips were taken and placed on PDA medium. A fungal species was consistently isolated from these lignified tissues. The mycelium was initially translucent (turning white in appearance), while the mature mycelium was aerial, varying in color from pale to dark gray (Munsell color code: colony edge mycelium 6Y-6 4 / 5G and colony center mycelium B6-PB 7 / 5PB). The production of pycnidia and conidia was induced using pine needles in water agar medium and incubated in the dark for 10 days. Hyaline unicellular conidia of 25 ± 1.1 µm (range 23.9 to 26.1 µm) long and 11 ± 0.5 µm (Range 10.5 to 11.5 µm) wide (n = 50) were obtained from black pycnidia. Based on the cultural and morphological characteristics observed, the pathogen was identified as a possible species of the family Botryosphaeriaceae (20 isolates). Molecular techniques were used to identify the species of pathogen, and three isolates (F154, F199, and F167) were analyzed by using Multilocus sequence typing to confirm the identity of the pathogen. Genes ITS (internal transcribed spacer region), tef-1 (translation elongation factor 1-alpha) and β-tub (β-tubulin) were amplified using endpoint PCR, with primers ITS1/ITS4 (White et al., 1990), EF1-728F/EF1-986R (Carbone & Kohn, 1999) and Bt2a/Bt2b (Glass & Donaldson, 1995), respectively. The segments were sequenced using the same primers, deposited in Gen Bank, and the accession numbers for each isolate were OM993582, OM993583, ON003481 for ITS, ON054936, ON054938, ON054937 for tef1 and ON054939, ON054941, ON054940 for β-tub, respectively. A phylogenetic tree was constructed using the maximum likelihood statistical method with the Tamura-Nei model based on a concatenated dataset of ITS region, tef1 and β-tubulin gene using Mega-X, and the three Chilean isolates (F154, F199, and F167) formed a single clade with the reference isolates of Diplodia mutila (Fr.) Mont. BLAST algorithm analyses indicated 100% identity to D. mutila for ITS (accession NR_144906), for tef-1 (accession MK573559), and for β-tubulin (accession MG952719). The pathogenicity of the three isolates was validated through Koch's postulates. For this purpose, a trial was established in 6-year-old European hazelnut plants cv. Tonda Di Giffoni. Ten healthy branches were individually inoculated using actively growing mycelial discs from each isolate, while a disc of PDA without fungus was used as a control. Holes of 5-mm diameter were inoculated, making sure the mycelium was in contact with the wood. Finally, the wounds were sealed with plastic film to prevent external contamination and improve humidity conditions. After 120 days, each branch was cut longitudinal-sectioned to verify the presence of wood necrosis which arose between 3.0 to 16.2 mm of length around the point of inoculation. No necrosis was observed in the control. To confirm pathogenicity, infected tissues were cut into small pieces with sterile knives and scalpels, and surface disinfected with a 1% sodium hypochlorite solution for 1 min. The disinfected tissues were placed on PDA medium and incubated at 25°C in the dark until fungal growth was observed. Hyphal tips were taken from the mycelia developed from the pieces of wood, and placed on PDA medium in order to obtain pure isolates. The pathogenicity of the D. mutila isolates F154 and F199 was observed in 100% of the inoculated branches, while isolate F167 showed symptoms in 85% of the branches. The reisolated strains showed similar mycelial growth and microscopic fungal structures to those observed in the isolates used for inoculation. This is the first report of D. mutila affecting European hazelnut in Chile. This fungus has been recently reported affecting hazelnut in Oregon, USA (Wiman et al., 2019), causing similar symptoms to those observed in our study. In addition, D. mutila has been reported infecting walnut in Chile (Diaz et al. 2018) and native forest trees, specifically Araucaria araucana in Chile (Besoain et al., 2017). The presence of D. mutila in commercial hazelnut orchards in Chile highlights the need for epidemiological studies in order to understand the characteristics and impact of this pathogen and, based on this, develop adequate phytosanitary programs for its control.