Seven F. circinatum isolates were selected for the study. Two paired monosporic cultures (mycovirus-infected and mycovirus-free) were used per isolate (Table 1). The presence of different mycovirus strains in each isolate was reported in a previous study by our research group ( Vainio et al. , 2015 ). In the present study, the presence or absence of mycoviruses (FcMV1 and/or FcMV2-2) was confirmed according to Martínez-Álvarez et al. (2014b) . For in vitro experiments, an isolate (FcCa06) previously confirmed as mycovirus-free and pathogenic ( Martínez-Álvarez et al. , 2012 , 2014a ) was also used.

Mycelial growth of F. circinatum isolates was evaluated on Petri dishes containing PDAS medium (potato dextrose agar and 0.5 mg/L streptomycin). A 4-mm mycelial plug of the pathogen was placed in the centre of the plate. The plates were sealed with Parafilm ^© and kept in darkness at 25ºC for seven days ( Leslie & Summerell, 2006 ). Seven replicates of each isolate were used in the assays. The growth of the colonies (along 2 perpendicular axes) was evaluated daily, and the total area of growth was calculated by the following equation:

Area = ∏ /4 (D ₁ x D ₂ ),

where D= diameter. A sample of 100 µL of conidial suspension of F. circinatum isolates was grown in Potato Dextrose Broth (PDB) supplemented with 5 g/L streptomycin sulphate (PDBS) ( Martinez-Alvarez et al. , 2012 ). Three flasks were prepared for each isolate. The spores were incubated for 6 hours under stirring at 180 cycles/min at 25° C in darkness and at 60% humidity. Three aliquots (each 10 µL) of the spore suspension were removed from each flask for evaluation of germination (200 conidia in each sample, i.e. a total of 1,800 conidia were scored). The spore suspensions were examined under a light microscope (Nikon E600) with a 40x lens. A hemocytometer was used to count the number of germinated spores as the grid helped to avoid counting the same spore more than once.

Pathogenicity tests were performed with two-year-old seedlings of P. radiata from Galicia, Spain. The plants were raised individually in root trainers (cells of 200 mL) and incubated in a growth chamber at 21º C with a 16-h photoperiod, following a completely randomized design.

Inoculations were carried out by the stem inoculation technique ( Martínez-Álvarez et al. , 2016 ). U-shaped wounds were cut into the bark of the seedlings, at 5-7 cm above ground, with a sterile scalpel ( Correll et al. , 1991 ). The conidia suspension was obtained from F. circinatum isolates cultured for seven days in PDB at 25ºC and darkness. Aliquots (100 µL) of spore suspension (adjusted to 10 ⁶ conidia/mL using a hemocytometer) were inoculated into the wound, which was covered with Parafilm ^© to prevent drying ( Martínez-Álvarez et al. , 2014a ). Twenty-seven seedlings were inoculated for each isolate tested. Control seedlings (n=27) were prepared in the same way with sterilized distilled water instead of spore suspensions.

The total length of the seedlings, the root collar diameter and the length of necrotic damage were measured in all seedlings. In order to obtain a more accurate view of the extent of the necrosis, the stems were split lengthways with a scalpel. To avoid the effect of the size of the seedlings, the relative necrosis was estimated as the ratio between length of the necrotic lesion and total length of the seedling.

Mycelial growth. For the statistical analysis, we used a linear mixed model of analysis of variance with repeated measurements, considering four between-subject factors and one within-subject factor with seven levels. The mathematical formulation of the model was as follows:

In vitro growth rates of the F. circinatum isolates varied significantly depending on the presence/absence of mycoviruses (F=6.03, p =0.002). In particular, F. circinatum isolates infected with micoviruses, both single infections with FcMV1 and FcMV2-2 (t=2.39, p =0.02 and t=3.63, p <0.001, respectively) and co-infection with FcMV1 + FcMV2-2 (t=4.27, p <0.001), yielded lower growth rates than mycovirus-free F. circinatum isolates (Fig. 1A). There were no significant differences between mycovirus-infected F. circinatum isolates in relation to the type of mycovirus (FcMV1, FcMV2-2 or co-infection) (Fig. 1A). A reduction in the mycelial growth of different pathogens infested with mycoviruses has been reported to be related to hypovirulence in fungi in several studies ( Bottacin et al. , 1994 ; Ahn & Lee, 2001 ; Castro et al. , 2003 ; Robin et al. , 2010 ; Rodríguez-García et al. , 2014 ; Zheng et al. , 2014 ; Yu et al. , 2015 ). Nevertheless, the effect on mycelial growth may also be determined by the species ( Vainio et al. , 2012 ; Hyder et al. , 2013 ), isolate ( Hunst et al. , 1986 ; Rodríguez-García et al. , 2014 ; Yu et al. , 2015 ) temperature ( Vainio et al. , 2010 ; Bryner & Rigling, 2011 ; Romeralo et al. , 2012 ; Zamora et al. , 2016) or the type of mycovirus ( Cryphonectria parasitica CHV1, mycovirus hypovirulences). These factors, among others, may explain the inconsistencies in the findings of different studies. The conditions of the experiment appear to be related to the effect of the virus, and use of the same stains as those used by Muñoz-Adalia et al. (2016b) thus failed to yield any effect of mycoviruses on the mycelial growth of F. circinatum .

A large reduction in spore germination was observed in the mycovirus-infected isolates of F. circinatum, relative to mycovirus-free isolates (F=28.21, p <0.001). Again, mycovirus-infected isolates, both single infections with FcMV1 and FcMV2-2 (t=28.76, p <0.001 and t=13.18, p <0.001, respectively) and co-infection with FcMV1 + FcMV2-2 (t=8.81, p <0.001), yielded lower germination rates than mycovirus-free isolates (Fig. 1B). No significant differences were observed in relation to the mycovirus. Some studies have demonstrated that the presence of hypovirulence-causing mycoviruses also reduces sporulation ( Kazmierczak et al. , 1996 , Moleleki et al. , 2003 ; Robin et al. , 2010 ; Zamora et al. , 2016). Thus, Ihrmark et al. (2004) observed that the frequency of germination of basidiospores was reduced by the presence of mycoviruses. However, to our knowledge no specific research has been carried out to study the effect of mycoviruses on spore germination in ascomycetes.

The pattern observed in the in vivo experiment was not consistent with the findings of the in vitro assays (both mycelial growth and spore germination). In fact, no significant differences in the length of lesions caused by F. circinatum were found in relation to the presence/absence of mycovirus (F=0.62, p <0.61) (Table 2). Although necrosis has traditionally been considered an indicator of the disease progress, because F. circinatum kills stem tissue during colonization ( Glazebrook et al. , 2005 ), recent research has demonstrated that F. circinatum can colonize the plants without causing necrosis, at least at the early stages of infection (J. Martín-García, unpubl. data). Further studies should therefore be carried out at different temperatures and with other techniques, such as fluorescence ( Oßwald et al. , 2014 ) and molecular methods ( Bodles et al. , 2006 ), to determine the damage caused beyond the visible symptoms.