In vitro interaction studies between Glomus intraradices and Armillaria mellea in vines

An interaction study was performed with mycorrhizal plants of the grapevine rootstock Richter 110 (Vitis berlandieri Planch × Vitis rupestris Scheele) and the root pathogenic fungus Armillaria mellea (Vahl:Fr.) P. Kumm using an autotrophic in vitro culture system. Micropropagated plantlets were transferred to Petri plates with MSR medium lacking sugar and vitamins. Inocula of Glomus intraradices (BEG 72) and of Armillaria mellea obtained from a root organ culture and from a mycelium colony grown in malt agar respectively, were added to the plates according to each treatment: non-inoculated, inoculated with G. intraradices, inoculated with A. mellea, and dual-inoculated plants. There were ten replicates per treatment. Fourteen weeks later, the pathogen’s mycelium occupied most of the surface/volume of the plate and had produced rhizomorphs. In dual inoculated plates, A. mellea’s growth was not affected by the presence of G. intraradices, but the latter produced a lower number of spores and its extraradical phase showed granulation, vacuolation and tip swelling. The pathogen induced necrosis and growth decrease in the root. Glomus intraradices alleviated these symptoms and there were no differences in root biomass between non-inoculated plants and plants inoculated with both fungi. There were no symptoms of the disease in shoots and G. intraradices stimulated shoot growth both, although mycorrhizal colonization was lower when A. mellea was present. No direct antagonism or antibiosis against the pathogen was observed, thus the protective effect exerted by the symbiotic fungus in grapevines must be indirect, mediated through the host plant physiology. Additional key words: control, disease symptoms, grapevine, in vitro culture, root pathogenic fungi, tolerance.


Introduction
Armillaria mellea (Vahl:fr) P. Kumm is a root pathogenic fungus causing white root rot in several crop species such as vines and fruit trees.Control measures are scarce and uneff icient, as A. mellea has a high infection capacity and a long-term survival in the soil (Aguín et al., 2006).Alternative biological and cultural control methods are under study (Baumgartner and Rizzo, 2006;Nogales et al., 2009a) and among them, the use of arbuscular mycorrhizal fungi (AMF) has demonstrated to contribute to increase plant tolerance to A. mellea both in the greenhouse (Nogales et al., 2009b) and in the field (Camprubí et al., 2008;Nogales et al., 2009a).
To our knowledge, in vitro interaction studies between AMF and A. mellea have never been reported.In vitro experiments allow to maintain microorganisms, tissues or cells out of their natural environment, limiting the influence of external, non controllable factors, providing a useful experimental tool.Very often, similar results cannot be expected in field conditions, but in vitro studies are a useful experimental approach in research.
The development of a monoxenic culture system for AMF by Bécard and Fortin (1988) and Déclerck et al. (1998) in carrot roots, opened up the possibility to study in vitro interactions between the AMF Glomus intraradices Schenck and Smith and plant pathogenic fungi (Benhamou et al., 1994).Despite its many advantages, the monoxenic culture system has however a severe limitation due to the lack of the whole host plant system.The lack of photosynthetic tissues generates an abnormal hormonal balance and an abnormal physiological relationship between both symbiotic partners (St-Arnaud and Elsen, 2005) that can distort the results of the interaction studies.Voets et al. (2005) developed an autotrophic culture system for the in vitro mycorrhization of micropropagated plantlets, where the roots of these plantlets were inoculated with AMF spores under in vitro conditions.In this system, the shoots grew in the open air under high light intensity allowing plant photosynthesis (Kozai et al., 1988) while roots developed in the dark.Roots were inoculated with individual AMF spores, and the emerging hyphae were capable to colonize the whole root system, to develop extraradical mycelium and to produce new spores.
The adaptation of the autotrophic culture system to woody species such as vine plants, could be useful for studying the arbuscular mycorrhizal (AM) symbiosis development as well as the short term interactions between the mycorrhizal fungus and root pathogens affecting grapevines.Armillaria mellea is a slow-growing pathogen that takes several years to kill grapevines in the field (Baumgartner and Rizzo, 2002), thus, long term studies are needed to assess disease control measures.The development of an in vitro mycorrhizal inoculation system for woody species would enable to observe an early response of mycorrhizal vines to A. mellea.
The aim of this study was to evaluate the interaction between G. intraradices and A. mellea in an autotrophic in vitro culture system using micropropagated Richter 110 grapevines as the host plant.

Plant material
In vitro micropropagated plantlets of the grapevine rootstock Richter 110 (Vitis berlandieri Planch × Vitis rupestris Scheele) were first propagated from woody cuttings, and the newly elongated shoots were cut and surface-sterilized.Nodal segments of the disinfected shoots were cultured on a modified Murashigue and Skoog (MMS) medium (Murashigue and Skoog, 1962) adapted to vine by Torregrosa and Bouquet (1996) and kept in a growth chamber set at 25 ± 2°C with 50 µmol m -2 s -1 photon flux density (PFD), provided by fluorescent lights (Sylvania cool-white) for a 16 hours-photoperiod.New shoots were formed from each segment and grapevine plantlets were sub-cultured by nodal cuttings every five weeks on MMS medium (Torregrosa et al., 2001).Shoots of 4.5-5.5 cm long with uniform root systems were used in the experiments.

Fungal material
The inoculum of G. intraradices BEG 72 was obtained from mycorrhizal root organ cultures using transformed carrot roots (Daucus carota L.).
The A. mellea strain used in the experiment was isolated from a replant vineyard in Vimbodí (Tarragona) in 2004, and subcultured on malt agar medium.

Experiment set up
Four treatments were considered, with ten replicates each: non-inoculated plants, inoculation with G. intraradices, inoculation with A. mellea and inoculation with both the AMF and the pathogenic fungus.
Petri plates 15 cm in diameter were filled with MSR medium lacking sugar and vitamins (Voets et al., 2005) and two plugs of inoculum isolated from an AM fungal root organ culture of G. intraradices containing approximately 50 spores were placed in the centre of the plate.Simultaneously three plugs of 0.5 cm diameter from an A. mellea mycelium colony grown on malt agar were placed in the edge of the plates to ensure an homogeneous growth of the fungal mycelium in the plate.
This inoculation method was aimed at achieving a simultaneous exposure to both, the AMF and the pathogen, as it occurs in the field.
Approximately 5 cm long in vitro micropropagated plantlets of Richter 110, were then transferred to the Petri plates and an autotrophic culture system was established.The roots remained in the Petri plate on the culture medium, devoid of sucrose and vitamins, while the shoots grew in open air conditions as described for potato plants (Voets et al., 2005).Petri dishes were covered with opaque plastic strips in order to keep the root system in the dark, and plants were kept inside a plastic box at 100% of relative humidity in a growth chamber set at 25°C with 16h photoperiod and a PFD of 200 µmol m -2 s -1 .Ten days later their acclimatization was induced by progressively opening the box.Sterilized (121°C for 15 min) MSR medium lacking sucrose and vitamins was periodically added to the Petri plates to maintain an adequate level of nutrients and liquid in the plates.
After 13 weeks, Petri plates were observed under the microscope (Zeiss, West Germany) for detecting possible interactions between two fungi at hyphal level.The observations were done placing the plates directly under the microscope.
At harvest, 13 weeks after the set up of the experiment, root and shoot biomass, the number of G. intraradices spores and the number of rhizomorphs produced by A. mellea were recorded.Mycelial growth of A. mellea was estimated as the percentage of the total medium surface occupied by the pathogen, and the percentage of mycorrhizal intraradical colonization was estimated by the grid-line intersect method (Giovannetti and Mosse, 1980) on a 1 g portion of each root system which was cleared and stained following the method of Phillips and Hayman (1970), modified by Koske and Gemma (1989).

Statistical analysis
Biomass data were analyzed by a two way ANOVA followed by a Duncan's test (P ≤ 0.05).The number of spores and rhizomorphs, and the percentage of the surface covered by A. mellea's mycelium and the intraradical mycorrhizal colonization percentage were analyzed by a student t test.

Results
The autotrophic culture system established for vine plants was a useful tool to study the interaction between G. intraradices and A. mellea.The system allowed to observe and quantify the mycelial development of both fungi and the growth of the vine plants.It was also possible to observe non-destructively under the microscope the morphology of AMF spores and the fungal mycelia.
At harvest, 14 weeks after the establishment of the systems, A. mellea had grown around the roots of Richter 110 plantlets and had infected them.The infection was set directly from the A. mellea mycelium and also from the newly formed rhizomorphs.Necrosis and rot symptoms were observed in the roots (Fig. 1) but there were no disease symptoms in the shoots.
In root biomass, a significant effect of the pathogen was detected.Plants inoculated with A. mellea had a lower root weight than non-inoculated plants.The decrease in root growth was alleviated by the mycorrhizal symbiosis, and plants inoculated with both G. intraradices and A. mellea did not differ significantly in root dry weight from healthy plants (Table 1).
The development of A. mellea was not significantly different in plants inoculated only with the pathogen and in plants inoculated with both A. mellea and G. intraradices (Table 2).Moreover, A. mellea produced dicotomically branched rhizomorphs in both treatments (Fig. 2).Concerning the development of the mycorrhizal fungus, both the number of newly produced spores and the percentage of mycorrhizal colonization were lower in mycorrhizal plants inoculated with A. mellea (Table 3).However, the shoot dry weight was not significantly different from that recorded in mycorrhizal plants non-inoculated with A. mellea, and it was higher than the shoot dry weight of non-mycorrhizal plants (Table 1).
In the extraradical phase of dual inoculated plates, mycelia of both fungi grew overlapped.At microscopic level, whereas A. mellea development was not affected by the presence of G. intraradices mycelium, the AM fungus growth was affected by the pathogen.Hyphal vacuolation (Fig. 3a), tip swelling (Fig. 3b) and granulation inside the G. intraradices hyphae (Fig. 3c) were clearly observed.There was also a decrease in spore production, and the spore shape was occasionally abnormal (Fig. 3d).

Discussion
In vitro experiments are research tools that allow the control of many parameters, thus enabling detailed interaction studies.The extrapolation of these results to field conditions should be done with caution, as The effect of the mycorrhizal fungus, the effect of the pathogen and the interaction between both was determined by a two way ANOVA.The asterisk indicates significant differences at a probability level of P ≤ 0.05.environmental factors such as the soil/substrate physicochemical parameters and the rhizospheric microbial populations can have a strong influence in the expected outcome.In our experimental system, A. mellea development was not affected by the extraradical phase of G. intraradices.The number of rhizomorphs and the growth of the mycelium were not different in both A. mellea inoculated treatments.Several studies have demonstrated that AM fungi have an influence on rhizosphere microorganisms by affecting the host plant, due in part to modifications in root exudates (Schwab et al., 1983;McAllister et al., 1995).It has been suggested that exudates from mycorrhizal roots may be implicated in an altered susceptibility of mycorrhizal plants against soil borne pathogens (Vierheilig and Piché, 2002;Jones et al., 2004;Vierheilig, 2004a).
Hyphal interference is a form of antagonism where growth inhibition and a subsequent vacuolation, granulation and lysis of the cells occur as the hyphae of both species come into close proximity (Ikediugwu and Webster, 1970;Skidmore and Dickinson, 1976;Shankar et al., 1994).These interactions are mediated by nonenzimatic diffusible metabolites (antibiotics) that alter the permeability of cell membranes leading to plasmolysis and hyphal death (Ikediugu and Webster, 1970;Skidmore and Dickinson, 1976;Boddy, 2000).The negative interactions observed in the extraradical mycelium of G. intraradices in the presence of A. mellea may indicate the potential production of antibiotics or toxic compounds by the pathogen.Hepper (1979) observed that the spore germination and the growth of AM fungal mycelium could be stimulated or inhibited by different compounds, and the production of toxic secondary metabolites by Armillaria sp. has been reported by several authors (Peipp and Sonnenbichler, 1992;Sonnenbichler et al., 1994).The release of these compounds, induced by the presence of antagonistic fungi and plant cells, can inhibit the growth of other microorganisms and even induce cell death before any contact occurs (Peipp and Sonnenbichler, 1992).Soluble or volatile compounds produced by the pathogenic fungus could therefore have inhibited the sporulation and the development of the extraradical mycelium of G. intraradices.
The lower root AMF colonization found in plants inoculated with both fungi, G. intraradices and A. mellea, might be a consequence of the reduced development of the extraradical mycelium of G. intraradices observed in these systems.Although the pathogen had a negative effect on plant roots, where early symptoms of the disease such as necrosis and decreased growth were observed, the presence of the AMF in co-inoculated plants reduced the symptoms, and root biomass was not different from that recorded in plants non-inoculated with the pathogen.This could be due to an improvement in the root development caused by the AMF colonization, especially in the absorption zone, which can compensate the loss in root biomass caused by the pathogen (Pozo et al., 2002).
The development of pathogenic infections has been inversely correlated with the intraradical AMF colonization (Caron et al., 1986;Cordier et al., 1998;Vierheilig, 2004b;Scheffknecht et al., 2006), but in our results no direct relationship between the intraradical colonization and the decrease in disease symptoms was determined.
The pathogen did not have a significant effect in shoot growth although G. intraradices had a lower intraradical development in the presence of A. mellea, while the mycorrhizal fungus had a stimulatory effect on shoot development.These results indicate that the AM symbiosis has a protective effect at plant level, reducing disease symptoms.A direct action of AMF against pathogens through antagonism, antibiosis or degradation has never been shown, therefore, the bioprotection effect of the AM symbiosis against diseases must be  indirect, through changes in the host physiology, inducing resistance mechanisms, improving plant nutrition, through competition for infection sites or for space or nutrients with pathogens, or by producing changes in the soil microbiota that can negatively affect the pathogen (Rodríguez-Kábana and Calvet, 1994;St-Arnaud et al., 1995).
The in vitro grapevine culture system provides a method where factors as improved nutrition and changes in the rhizospheric microbiota can be discarded.In the particular case of A. mellea there is neither competition for infection loci, nor for host photosyntates, as both fungi colonise different host tissues, and have different trophic requirements.The increased tolerance observed in the plant roots could also be accounted for by an accumulation of newly formed products in the AMF infection site (Rosendahl and Rosendahl, 1990), including symbiosis related proteins, phenolic compounds, hydrolases like quitinases with antimicrobial potential and structural polymers, such as lignin.Although these changes are still controversial, and many have been shown to be transient, from our results the effect of AMF on pathogen tolerance seems to be exerted through the host plant physiology.
The system allowed the quantification of plant and fungal development and non destructive observations of the interaction of both fungi under the microscope.Although the pathogen negatively influenced the symbiont's extraradical and intraradical development, symbiotic plants had a higher root and shoot biomass than healthy non symbiotic plants, despite the presence of the pathogen.The results show an indirect bioprotection effect of G. intraradices against A. mellea in the early stages of the disease development, under our experimental conditions.

Figure 1 .
Figure 1.Roots of a Richter 110 grapevine rootstock plant infected by Armillaria mellea after 13 weeks growth in an autotrophic culture system.

Figure 2 .
Figure 2. Rhizomorphs of Armillaria mellea after 13 weeks growth in an autotrophic culture system with a Richter 110 plant.

Table 1 .
Response of Richter 110 grapevine rootstock plants after 13 weeks growth in autotrophic culture systems inoculated or not with Glomus intraradices and Armillaria mellea Data are means of ten replicates per treatment.Different letters indicate statistically significant differences according to Duncan's test (P ≤ 0.05).

Table 2 .
Growth development of Armillaria mellea in autotrophic in vitro culture systems of Richter 110 plants inoculated or not with Glomus intraradices aData are means of 10 plants per treatment.Data followed by a different letter differ significantly at P ≤ 0.05 (Student t test).

Table 3 .
Growth development of Glomus intraradices in autotrophic in vitro culture systems of Richter 110 plants inoculated or not with Armillaria mellea bData are means of 10 plants per treatment.Data followed by a different letter differ significantly at P ≤ 0.05 (Student t test).