Effect of putative mitoviruses on in vitro growth of Gremmeniella abietina isolates under different laboratory conditions

Mitoviruses have been found in several forest pathogens (i.e. Cryphonectria parasitica, Gremmeniella abietina), and because they have been shown to reduce the virulence of host fungi there is a growing interest in studying their use as a biocontrol. This study was carried out to test the effect of temperature (5°C, 15°C, 25°C and 35°C), pH (4, 5, 7 and 9) and osmotic potential (–0.6, –1.2, –1.8 and –2.4 MPa) on the mycelial growth of seven G. abietina isolates under controlled laboratory conditions. Four of the isolates hosted mitoviruses and three of them did not. During the experiment, mycelial growth was recorded every week for a period of 8 weeks. Results showed no differences in growth behavior between mitovirus infected and non-infected isolates when placed under different pH modifications. However, the mitovirus-infected isolates presented larger mycelial growth than the mitovirus-free ones when at the fungi’s optimal growing temperature of 15°C. When growing at certain osmotic potentials (–0.6 and –1.8 MPa) a reduction in growth of the mitovirus-infected isolates was observed. The results of this experiment suggest that mycelial growth among non-infected isolates and isolates naturally infected by mitovirus vary under different culture conditions, thus providing further insight into the effects of mitovirus on Gremmeniella abietina isolates.

host physiology and may lead to attenuation (hypovirulence) or enhancement of fungal virulence (hypervirulence) (Ghabrial and Suzuki, 2009).Because some viruses are capable of reducing virulence of fungal pathogens they can potentially be used for control of fungal diseases (McCabe et al., 1999;Boland 2004;Zhang et al., 2010).However, they must fulfill two requirements in order to be suitable for biological control: firstly, have the ability to decrease the fitness of the pathogenic fungus and secondly, transmit the dsRNA efficiently enough to be maintained in a large proportion of the pathogen population (McCabe et al., 1999).Members of genus Mitovirus are only found in fungi and belong to the family Narnaviridae (Ghabrial and Suzuki, 2009).They lack true virions, and have a (+) ssRNA genome of approximately 2.5 kb (Boland, 2004).Mitoviruses have been recorded in several phytopathogenic fungi such as Cryphonectria parasitica (Polashock and Hillman, 1994;Polashock et al., 1997), Ophiostoma novo-ulmi (Brasier, 1983;Rogers et al., 1987), Sclerotina homoeocarpa (Deng et al., 2003;Deng and Boland, 2004), Helicobasidium mompa (Osaki et al., 2005), Chalara elegans (Park et al., 2006) and Botrytis cinerea (Castro et al., 2003;Wu et al., 2007).In most cases, the presence of mitoviruses is associated with reduction of fungal pathogenicity (Ghabrial and Suzuki, 2009;Wu et al., 2010).Members of the genus Mitovirus have also been isolated in G. abietina (Tuomivirta and Hantula, 2003) which, in the Spanish population, has recently been discovered to host (Botella et al., 2012a).
Reduction of virulence could be related, among other reasons, to anomalous mycelial growth in the fungal pathogen caused by mitoviruses (Ghabrial and Suzuki, 2009;Pearson et al. 2009).However mycelial growth is also influenced by environmental and cellular conditions such as temperature, pH and osmotic potential.Temperature limits mycelial growth and production of fruiting bodies in most fungi while the pH determines availability of elements such as nitrogen, calcium and magnesium among others (Carlile et al., 2001).Osmotic potential has also been identified as an important parameter in the ecology and growth of phytopathogenic fungi (Davis et al., 2000).For example, a de-

Introduction
Gremmeniella abietina (Lagerberg) Morelet (anamorph Brunchorstia pinea (P.Karsten) Höhnel) is a pathogenic fungus which has caused destruction in plantations and natural conifer forests in Northern and Central Europe, North America, and Japan (Yokota, 1975;Dorworth, 1979;Kaitera and Jalkanen, 1992) producing symptoms such as stem cankers and shoot dieback (Donaubauer, 1972).This fungus has been divided into three races: European, North American and Asian.Within the European race three biotypes have been determined based on the length of spores, number of septa, disease symptoms, and molecular markers.There is biotype A (LTT, large tree type), biotype B (STT, small tree type) and alpine biotype (Uotila, 1983;Hamelin et al., 1993;Hellgren and Hogberg, 1995;Kaitera and Jalkanen, 1996;Hantula and Muller, 1997).In Europe, the fungus mostly affects genera Picea spp.and Pinus spp.although it has also been found on genera Abies and Larix.In Spain, it presence on Pinus pinaster was first reported in 1929 (Martínez, 1933) and later on Pinus halepensis in 1999 (Santamaria et al., 2003).Notwithstanding, it has only been isolated from symptomatic Pinus halepensis trees.The symptoms observed generally consist of dry needles, branches with some distortion of terminal twigs and eventual dieback or death of the trees (Santamaria et al., 2003).Spanish G. abietina is currently recognized as part of the European race (Santamaria et al., 2005) and has recently been related to biotype A, although it has a unique genotype (Botella et al., 2010).
Mycoviruses, which are obligate parasites of fungi, are widespread in all major taxonomic groups of plant pathogenic fungi (Ghabrial and Suzuki, 2009;Pearson et al., 2009).They are transmitted through hyphal anastomosis and/or fungal sporulation (Zhang et al., 2010).Fungal viruses differ in their genomes, which can contain DNA, double-stranded (ds) RNA or single-stranded (ss) RNA genomes (Pearson et al., 2009).Eight families and one genus are currently described in the International Committee on Taxonomy of Viruses (ICTV) (2011): Chrysoviridae, Endornaviridae, Hypoviridae, Narnaviridae, Barnaviridae, Partitiviridae, Reoviridae and Totiviridae and genus Rhizidiovirus (Hausner et al., 2000;Zhang et al., 2010).Mycoviruses usually produce latent infections in nature, affecting sometimes the host's phenotype and/or its growth (Aoki et al., 2009).Symptoms produced by the presence of mycoviruses may vary from zero to severe effects on Effect of mitovirus on mycelial growth of Gremmeniella abietina Although some strains of G. abietina have been shown to host dsRNA mycoviruses, the effect these agents have on the virulence of this problematic phytopathogenic fungus has not yet been investigated.Accordingly, the main objective of the present study has been to evaluate the effect of the occurrence of viral dsRNA molecules (the replicative form of Mitovirus) on the in vitro mycelial growth of G. abietina isolates under different temperature, pH and osmotic potential conditions.

Fungal material
To develop this study seven Spanish isolates of G. abietina were chosen: four isolates were naturally infected by putative mitoviral molecules (P3-12, 00P-07, Hon 3-3 and P1-12) and three were not (Hon 9-2; P1-8 and VAI-13) (Botella et al., 2010).All isolates were selected based on previous studies developed in our laboratory in which RT-PCR and sequencing techniques confirmed the presence or absence of mitoviruses (Botella et al., 2012a(Botella et al., , 2012b)).The isolates were previously stored in 15% glycerol at -80º C and were reactivated on modified orange serum agar medium (MOS-agar;Müller et al., 1994) before performing the experiment.Thus, four weeks before the experiment fungi isolates were sub-cultured in MOS medium and kept in the dark at 15°C in order to obtain sufficient amounts of mycelium.

Mycelial growth
At the bottom of every Petri dish containing 20 ml MOS medium two perpendicular lines were drawn, and a 1mm squared piece of mycelium from each isolate was placed over the intersection of both lines.Mycelial growth was measured weekly for a period of 8 weeks.The response variable was the growth area calculated by the following formula: Area = π/4 (d 1 × d 2 ) where d 1 and d 2 were the two diameters measured along the lines.

Culture conditions for monitoring mycelium growth
The effect of mitovirus infection on mycelial growth under different laboratory conditions was the main focus of this study.Three experiments were conducted, each taking into account a separate factor: changes in temperature, pH or osmotic potential.Within each experiment four variations were tested: four temperatures (5°C, 15°C, 25°C and 35°C), four pH values (4, 5, 7 and 9) and four osmotic potentials (-0.6, -1.2, -1.8  and -2.4 MPa).The effect of temperature on mycelial growth was investigated by placing Petri dishes in several stoves at 5°C, 15°C, 25°C and 35°C.To examine the effect of pH, HCl or KOH 1N was added to MOS medium until the pH required was reached.All these Petri dishes were placed in the dark at 15°C since it is the optimal temperature for fungal development (Santamaria et al., 2004).Finally, in order to evaluate the effect of different osmotic potential on mycelial growth, different concentrations of KCl (250, 500, 750 and 1000mM) were added to MOS medium in order to reach the osmotic potential (ψ π ) values of -0.6 MPa, -1.2 MPa, -1.8 MPa and -2.4 MPa (Lira-Méndez and Mayek-Pérez, 2006).Petri dishes were incubated at 15°C ± 1°C in continuous darkness.Four repetitions of each combination "isolate × treatment" were completed.

DsRNA extractions
Fungal mycelium of mitovirus-infected isolates from significative treatments was incubated in MOS medium covered with cellophane for two weeks.Mycelia were first freeze-dried and then ground for 20 minutes into a fine powder.DsRNA was extracted following a modified version of the protocol described by Morris and Dodds (1979).The dsRNA presence in every isolate was verified by electrophoresis.Samples were loaded in a 1% agarose gel, which contained 1x TAE buffer and GelRed TM 10,000X.The test was run in a 1x TAE buffer during 60 min at 90V/30 cm, and immediately afterwards observed under UV light and photographed.The marker used to estimate the lengths of the dsRNA molecules was λ-DNA Hind III -ΦX174Hae III (DyNAzyme TM ).

Statistical analysis
All statistical analyses were done with SAS program (SAS Institute Inc., 2004).The response variable in all models was growth area (mm 2 ).A repeated-measures ANOVA for every treatment was calculated by means of Repeated Procedure by SAS to test the effect of the time on the mycelial growth of the isolates.In this case, the growth areas of every week were used as responses variables.Furthermore, for every experiment (temperature, pH and osmotic potential) a model was calculated to evaluate the effect of the putative presence of mitovirus (yes/no), the treatments (4) and their interactions by a two-way analysis of variance.A significance of 95% was taken in all of the analyses.A Tukey HSD test was used on means of factors when significant differences were found in the ANOVA model.Before the analyses were performed, normality, linearity and homocedasticity for the residuals were probed with Shapiro-Wilk test and graphical procedures.

Effect of temperature
A significant effect of time on the colony growth area (p < 0.001) was observed.Although the interaction between time and mitovirus presence was significant (p = 0.017) as well as the interaction between time and temperature (p < 0.001), only the effect of temperature and mitovirus presence on mycelial growth at the end of the experiment (eight weeks after plating) is shown in Table 1.The average growth from mitovirus-infected and mitovirus-free isolates is shown in Figure 1 at 5°C (A), 15°C (B), 25°C (C) and 35°C (D) throughout the eight weeks.Growth at 25°C was minimal and there was no growth at 35°C.Mean growth area was significantly different among mitovirus-infected and mitovirus-free isolates (p = 0.0030), temperatures (p < 0.001) and their interactions (p < 0.001).According to the Tukey test, the largest colony areas were found at 15°C whereas the smallest were found at 35°C.The overall mean colony size of mitoviruses-infected isolates was significantly bigger than that of the mitoviruses-free ones.When temperatures were considered separately, significant differences among mitoviruses-infected and mitovirusesfree isolates were found only at 15°C (p = 0.0043), the temperature that produced the most growth.

Effect of pH
A significant effect of time was observed on the growth area in the pH experiment (p < 0.001) and in its interaction with the pH treatments (p < 0.001) but not in the mitovirus presence (p = 0.7265).Average growth from mitovirus-infected and mitovirus-free isolates is shown in Figure 2 at pH 4 (A), pH 5 (B), pH 7 (C) and pH 9 (D) throughout the eight weeks.Only the data from the effect of pH values on mycelial growth at the end of the experiment (week 8) is shown in Table 2.The growth area was affected by the pH value (p < 0.001), but it was neither affected by the mitovirus presence (p = 0.9459) nor their interaction (p = 0.2753).The largest mycelial growth for all samples was observed at pH 4 while the smallest was shown at pH 9.No differences were shown between mitovirus-infected and mitovirus-free isolates in any pH treatment.

Effect of osmotic potential (ψ π )
In this experiment, time also affected the growth (p < 0.001) and interacted as well with the osmotic potential (p < 0.001) and the mitovirus presence (p < 0.0447).Data taken in the eight week showed the greatest differences (Table 3).The average growth of mitovirus-infected and mitovirus-free isolates is shown in Figure 3 at -0.6MPa (A), -1.2MPa (B), -1.8MPa  (C) and -2.4MPa (D) throughout the eight weeks.The model was significative (p = 0.027) although it was not the mitovirus presence (p = 0.1378) nor osmotic potential (p = 0.0805), but the interaction was significative (p = 0.0034), that is, the effect of mitovirus presence was different among the different osmotic potentials.

DsRNA banding patterns
The presence of the different putative mitoviruses was confirmed by dsRNA extraction and gel electrophoresis after significant treatments were carried out  Infected Mitovirus free Effect of mitovirus on mycelial growth of Gremmeniella abietina (Figure 4).Isolate P3-12 was found to maintain a 2.5 kb band despite receiving the treatments of ψ π -0.6 MPa and -2.4 MPa.Conversely, the 2.5 kb band was not sustained in isolate Hon3-3 after treatments of ψ π -0.6MPa and -1.8 MPa.These results suggested that putative mitovirus occurrence is not affected equally by similar osmotic potential and therefore KCL concentrations.In addition, dsRNA bands that appeared in P3-12 suggested the occurrence of other putative mycoviruses, which did not seem to be affected by the different treatments either.According to the size of the bands and the previous work developed in the laboratory (Botella et al., 2010) they possibly belonged to genera Totivirus (ca 6kb) and Partitivirus (three bands of ca 1-2 kb).

Discussion
Mycelial growth depends on the temperature of the environment.In our study, all the isolates showed an optimal growth at 15°C, which was in accordance with Santamaría et al. (2004) who demonstrated that Spanish isolates of G. abietina had the best growth at this particular temperature.Furthermore, the presence of mitovirus seemed to have a significant effect on G. abietina isolates at its optimal growing temperature of 15°C because the isolates with mitovirus present had higher mycelial growth than isolates without mitoviruses.This increase in the mycelial growth of our isolates could be related to a higher virulence of the pathogen since, in general terms, a suppression of mycelial growth has been reported to be closely associated with hypovirulence of fungi (Ghabrial et al., 2009;Pearson et al., 2009) although it could also be related to other factors (e.g., poor sporulation).
Heat tolerance was previously observed in several fungi among virus-infected and virus-free isolates (Marquez et al., 2007;Herrero et al., 2011) but in our study neither mitovirus-infected nor mitovirus-free isolates were able to endure the heat (few isolates hardly grew at 25°C and no growth was observed at 35°C).Marquez et al. (2007) observed that plants inoculated with the virus-infected wild type isolate of Curvularia protuberata R.R. Nelson and Hodges, with presence of the virus named CThTV, tolerated soils temperatures as high as 65°C for two weeks whereas plants inoculated with the virus-free isolate of the fungus dried-up and became chlorotic.Light evidence of heat tolerance was also observed in Tolypocladium cylindrosporum W. Gams due to the different behaviors displayed between virus-infected and virus-free isolates at 30°C (Herrero et al., 2011).
The pH value determines the availability of elements such as nitrogen, calcium and magnesium, among others, taken up by the fungus.In other fungi the effects of viruses have been shown to undergo variations when the composition of substrates, and therefore the availability of elements, differ.Nevertheless, Van Diepeningen (2006) observed that abundance of available nutrients in rich medium could mask viral effects on Aspergillus isolates.In our study no statistical differences were shown between mitovirus-infected and mitovirus-free strain growth under any treatment variation.Fungal cellular activity measured by means of growth and metabolism rates tend to decrease if the fungi are grown at different pH values from their optimal (Perez et al., 2000).According to our results, highest mycelial growths of all the isolates were observed at the initial pH 4, which is consistent with the general statement that most fungi will grow properly over a broad pH range on the acidic side of neutrality, i.e., pH from 4 to 7 (Carlile et al., 2001).Nevertheless, it is known that several species of isolates are able to modify the initial pH of the media in order to stabilize the acidity or alkalinity of the substrate (Carlile et al., 2001;Vazquez Garcia et al., 2002).
In our study, there wasn't any clear evidence that a decrease in osmotic potential produced a reduction of mycelial growth as previously observed in other fungal species (Imolehin et al., 1980;Lira-Mendez and Mayek-Perez, 2006;Palmero et al., 2008;Armengol et al., 2011).A reduction of the growth of the mitovirus-infected isolates was observed at -0.6 MPa and -1.8 MPa which can be linked to a decrease in the virulence of the isolates.Changes in behavior were also observed in isolates with and without viral infection when growing at certain osmotic potentials for Monosporascus cannonballus (Armengol et al., 2011).
Plant pathologists have been interested for a long time in mycoviruses (and among them, the mitoviruses) because of their potential use as biological control agents (Pearson et al. 2009).Although many viruses produce no obvious phenotypic changes, it is reasonable to assume that many virus infections will have some effect on growth (McCabe et al. 1999).The results from this experiment suggest that the presence of mitoviruses affects mycelial growth under different culture conditions as previously observed (Vainio et al. 2010).Nevertheless, the differences in growth among isolates may be also having been due to a genetic influence (Zharare et al., 2010) a possibility not tested here due to not working with genetically similar strains.Previous studies have shown that viruses found in many fungi, e.g., Cryphonectria parasitica, Fusarium graminearum or Botrytis cinerea, produce several phenotypic changes such as reduction in growth and sporulation of the fungal strains they infect (Chu et al., 2002;Boland, 2004;Van Diepeningen et al., 2006;Robin et al., 2010;Wu et al., 2010;Zhang et al., 2010).In our study, the isolates growing at osmotic potential medium of -0.6 MPa and -1.8 MPa also showed a reduction of the mycelial growth.However, in the virus-infected Fusarium oxysporum strains when growing on PDA only slight morphological alterations were evident (Lee et al., 2011).Furthermore, it has been observed that Effect of mitovirus on mycelial growth of Gremmeniella abietina several Cryphonectria parasitica virus-infected strains grow as well as virus free isolates on most artificial media although they are incapable of producing grilling cankers on chestnut trees and sporulate poorly (McCabe et al. 1999).In other cases, the presence of dsRNAs did not cause any fungal specific symptoms, such as reduced mycelial growth (Aoki et al., 2009).In some Alternaria spp.species there was no correlation between the radial growth of isolates and the presence of the dsRNAs (Zabalgogeazcoa, 1998).In contrast to hypovirulent interactions, there is evidence that some mycoviruses are beneficial to their hosts.Tan et al., (2007) observed statistically significant differences in in vitro growth rates of virus-infected versus uninfected isolates, with the infected cultures growing more rapidly.In our results, an increase of the mycelial growth was observed at treatment 15°C, the optimal growing conditions of G. abietina.
This study provides additional knowledge on the effects of mitovirus infection on G. abietina isolates.However, further research including other virulenceassociated parameters such as sporulation rates and in vivo virulence are recommended to establish an association between mycovirus infection and fungal virulence in Spanish G. abietina isolates.The development of a biocontrol protocol may create opportunities for biological control of this disease.

Conclusions
In our study mycelial growth depended on the treatment and the presence of mitoviruses.The presence of mitoviruses did not reduce mycelial growth of Gremmeniella abietina at its optimal growing temperature of 15°C.No effects of the occurrence of mitoviruses were shown among the mitovirus-infected and the mitovirus-free ones at any pH value.When growing at certain osmotic potentials (-0.6 and -1.8 MPa) a reduction in the growth of the mitovirus-infected isolates compared to the mitovirus-free ones was observed.Further research including other virulence-associated parameters is recommended.

Table 1 .
Mycelial growth (mm²) after 8 weeks at different temperatures.Mean value ± standard error (SE).Treatments tagged with * presented significant differences among isolates ¹ If the isolate was naturally-infected with mitovirus.²Averagegrowthwhencombinig all the temperatures together.³Differentletters in the same column show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test).4Differentletters in the same row show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test).5Averagegrowth when combinig all the isolates together.Effect of mitovirus on mycelial growth of Gremmeniella abietina

Table 2 .
Mycelial growth (mm²) after 8 weeks at different pHs.Mean value ± standard error (SE) ¹ If the isolate was naturally-infected with mitovirus.²Averagegrowth when combinig all the pH values together.³Differentletters in the same column show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test).4Differentletters in the same row show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test). 5Average growth when combinig all the isolates together.

Table 3 .
Mycelial growth (mm²) after 8 weeks at different osmotic potentials.Mean value ± standard error (SE).Treatments tagged with * presented significant differences among isolates ¹ If the isolate was naturally-infected with mitovirus.²Averagegrowthwhencombinig all the osmotic potentials together.³Differentletters in the same column show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test).4Differentletters in the same row show values significantly different from p < 0.05 (ANOVA Tukey's HSD Test).5Averagegrowth when combinig all the isolates together.