Screening, identification and evaluation of potential biocontrol fungal endophytes against Rhizoctonia solani AG3 on potato plants

Abstract

Rhizoctonia solani is an important soilborne pathogen of potato plants whose control typically depends on chemicals. Here, we screened six fungal endophytes for the suppression of R. solani growth both in vitro and in a greenhouse. These isolates were identified using morphology and internal transcribed spacer regions of rDNA as Alternaria longipes, Epicoccum nigrum, Phomopsis sp., and Trichoderma atroviride. Both T. atroviride and E. nigrum showed significant in vitro inhibition of mycelial growth of R. solani, with the greatest inhibition zone observed for E. nigrum species in dual cultures. The highest inhibition was observed for T. atroviride. The inhibition rate was also significantly correlated with the culture filtrates of these isolates. Confocal microscopy showed that T. atroviride acts as a mycoparasite and competitor. However, E. nigrum and A. longipes produce secondary metabolites, while Phomospsis sp. competes for nutrients and space. Greenhouse experiments confirmed that T. atroviride and E. nigrum improved potato yield significantly and decreased the stem disease severity index of sensitive potato.

Introduction

Rhizoctonia solani is one of the most important soilborne pathogens in cultured soils. This pathogen causes disease worldwide, has a wide host range (Woodhall et al., 2007), and is especially prevalent in all potato-growing areas. Stem canker and tuber blemishes are two major diseases associated with R. solani in potato, and both can cause quantitative and qualitative damage to the potato crop. The predominance of the anastomosis group AG-3 in causing potato disease has been reported (Virgen-Calleros et al., 2000).

Biological control is now increasingly considered as an alternative treatment to sustain agriculture. Biological control measures rely on the use of such organisms that are antagonistic to the target pathogens. Mechanisms by which antagonistic organisms act include mycoparasitism that may result from physical interhyphal interference or by the production of volatile and nonvolatile metabolites (Benitez et al., 2004). Several microorganisms, including the obligate mycoparasite fungus Verticillium biguttatum, have been reported as effective biological control agents (BCAs) against R. solani in potato (Van Den Boogert & Jager, 1984). To date, the genus Trichoderma remains an economically efficient BCA that is commercially produced at a large scale and is applied against several fungal pathogens (Samuels, 1996).

Most of the knowledge on BCAs and their functions has been gained by studying endophytic bacteria (Handelsman & Stabb, 1996). An endophyte is often a bacterium or a fungus that colonizes plant tissues for at least part of its life without causing apparent disease symptoms. It has been demonstrated that bacterial endophytes may have beneficial effects on host plants, such as promoting growth and biological control of pathogens (Adhikari et al., 2001). In contrast, fungal endophytes are less well studied to control R. solani on potato, and only fungal genera Ampelomyces, Coniothyrium, and Trichoderma have been tested (Berg, 2009). The author suggests that there is a strong growing market for microbial inoculants worldwide, with an annual growth rate of approximately 10%. Thus, it is important to investigate other fungal genera that may sustain potato crop production. Our objectives were to assess the ability of different fungal endophytes, Trichoderma atroviride, Epicoccum nigrum, Alternaria longipes, and Phomopsis sp. to control R. solani in potato. None of these fungi pose any risk to human or animal health, and are known as potential BCAs. However, their behaviour in the soil cannot be predicted, and could have either synergistic or detrimental effects on beneficial and symbiotic rhizosphere microorganisms. These microorganisms were isolated and identified as fungal endophytes and tested for their performance to compete against R. solani using in vitro dual culture assays. We tested the ability of antagonistic fungal isolates to excrete volatile substances and evaluated the effect of filtrates of liquid cultures of all fungal isolates on the mycelial growth of R. solani. Finally, we evaluated the antagonism under greenhouse conditions.

Material and methods

Fungal isolates and molecular identification

Rhizoctonia solani R14 and Phomopsis sp. R24 strains were isolated from infected potato plants from a field in August 2007 in Montreal region (Canada). Fungal endophytes (E1, E2 E8, E13, and E18) were isolated from the leaves of Norway maples in October 2007 in Montreal based on the methods described by Berg et al. (2005). These endophytes were evaluated for antagonism against R. solani. Fungal strains were identified by PCR and sequencing of internal transcribed spacer (ITS) regions of rDNA. Mycelia, grown in liquid potato dextrose broth at 25 °C, were harvested by filtration and used to extract DNA using the plant DNA extraction kit (Qiagen, Canada). PCR was performed using primers ITS1 and ITS4 to amplify ITS regions of seven isolates (R14, R44, E1, E2, E8, E13, and E18) (Tables 1 and 2). Amplification reactions were carried out in a volume of 50 μL using the Dream Taq kit (Fermentas, Canada) according to the manufacture's recommendations. PCR was performed using a Mastercycler (Eppendorf, Canada) following the programme: 5 min at 94 °C, followed by 29 cycles of 30 s at 94 °C, 30 s at 59 °C and 1 min at 72 °C, and 7 min at 72 °C. PCR amplicons were sequenced at the Genome Quebec Innovation Center (Montreal, Canada). Sequences were blasted using the nucleotide blast search at NCBI. Sequences were deposited in EMBL under accession numbers FN646616–FN646622. Morphological observations such as colony growth, colour, type of mycelia, size, and form arrangement of conidia were used to confirm molecular data (Alexopoulos et al., 1996).

Dual culture assays

Fungal isolates were screened for their ability to suppress the mycelial growth of R. solani strain R14 by in vitro dual culture assays on potato dextrose agar (PDA) (Lahlali et al., 2007). Each combination of pathogen/antagonist was replicated 10 times and plates were randomly placed in the dark and incubated at 25 °C until the PDA medium was completely covered with pathogen mycelia. As negative controls, 10 Petri dishes were inoculated only with an R. solani agar disc and a water agar disc. The radial mycelial growth of R. solani towards the antagonistic fungus (Ri) and that on a control plate (Rc) were measured and the mycelial growth inhibition was calculated according to the formula: (Rc−Ri)/Rc × 100. Statistical analyses were performed with anova using the sas statistical package (SAS Institute, Cary, NC). When the effect was found to be significant, the LSD was performed for mean separation at P ≤0.05.

Confocal microscopy

Fungal strains that showed a significant inhibition of R. solani growth (T. atroviride, A. longipes, Phomopsis sp., and E. nigrum E1, E8, and E18) were prepared for confocal microscopy. Agar plugs containing mycelia of both strains were placed in opposite sides of a plate containing 20 mL of PDA. Microscope coverslips were placed on the top agar between the antagonistic strains. When hyphae were observed on the surface of the coverslips, they were removed and immediately stained with SytoGreen 13 dye (Invitrogen, Canada) for 30 min at room temperature. Coverslips were mounted in an 80% glycerol solution on a microscope slide and visualized using a Zeiss LSM 5 DUO confocal microscope. Images were acquired by excitation at 488 nm and emission with a long pass 506-nm filter. We used three replicates for each combination pathogen/antagonist.

Ability of the antagonists to produce volatile compounds on agar

PDA plates were inoculated in the centre with a 0.5 cm diameter mycelial disc containing both antagonists and pathogen. Fungal isolates including R. solani were separately cultivated per plate. The lids were removed and two plates containing each R. solani and one fungal endophyte, and one plate was inverted and placed on top of the other plate. The two plate bases were then sealed with a double layer of parafilm. All plates were randomized and placed at room temperature. Controls were prepared using the same experimental setup, except that a water agar disc was used instead of the antagonist culture. We used 10 replicates per treatment. The inhibition rate of each antagonist against pathogenic fungus was calculated and statistical analyses were performed as described above.

Growth of R. solani with antagonistic culture filtrates

This experiment was carried out using the protocol described by Campanile et al. (2007). Radial growth was recorded by measuring the mean colony diameter at 1-day intervals for the time required to reach the margin of the dish in controls. Statistical analyses were used as described above.

Antagonism tests in greenhouse

Greenhouse trials were performed in pots filled with Pro-Mix (Premier Tech, Canada). Seed tubers of the potato cultivar ‘Riba’ were obtained from the market. The inoculum of R. solani and antagonist isolates were prepared by subculturing an infected agar disc on PDA medium. Bags containing 1 kg rye seeds were inoculated with six plates of pathogen or antagonist cultures and stored at room temperature for 30 days. Sterilized Pro-Mix was infected with R. solani at an amount corresponding to 5% of the total weight and was placed in a greenhouse (90% relative humidity and 16 h of light). After 2 weeks, the infested and noninfested Pro-Mix were inoculated separately with each antagonist and then placed in a greenhouse. After 1 week, the disinfected potato seed tubers with sodium hypochlorite were planted at a rate of one tuber seed for each pot culture. The planted pots were left in the greenhouse (22–25 °C day, 18–20 °C night) for 3 months. The following tested treatments are summarized in Table 3. Fifteen pots per treatment arranged in a completely randomized design were used in this experiment. Disease symptoms were measured including stem lesions after 10 weeks of planting. Stem lesions were evaluated using a scale of 1–5 as described previously by Sturz et al. (1995). After 3 months, the yielded tubercles (g), per pot treatment, were recorded. Statistical analyses were used as described above.

Results

Molecular characterization and identification of fungal isolates

All fungal isolates were identified using ITS regions of rDNA and blast search. All isolates showed 100% homology with E. nigrum, A. longipes, R. solani, and T. atroviride (Table 1). One isolate showed 99.6% homology with Phomopsis subordinaria and was therefore named as Phomopsis sp. The blast scores are summarized in Table 1.

Dual culture assays

The confrontation cultures between R. solani and isolates E1, E8, and E18 (identified as E. nigrum) showed clear inhibition zones and different patterns of interactions (Fig. 1). Isolates E2 and R24, identified as T. atroviride and Phomopsis sp., respectively, showed fast growth and covered the plate completely including the mycelium of R. solani. Isolate E13, identified as A. longipes, also showed an inhibition zone against the pathogenic fungus. Antagonistic isolates showed different inhibition rates when confronted with R. solani (Table 1). The highest inhibition rate was observed with T. atroviride, followed by Phomopsis sp., A. longipes, and E. nigrum. Nevertheless, these inhibition rates were statistically significant at P ≤0.05.

Confocal microscopy

Figure 2 shows the different patterns of interactions between antagonistic isolates. The antagonist mycelium was easily distinguished from R. solani mycelium by hyphal morphology (Fig. 2f). Trichoderma atroviride hyphae established close contact with those of R. solani by coiling (Fig. 2e). The coils were usually very dense and appeared to tightly encircle the R. solani hyphae. After 7 days, T. atroviride hyphae penetrated R. solani hyphae and caused a loss of turgor. Phomopsis sp. invaded the R. solani colony and limited its growth (Fig. 2d). The hyphal density of Phomopsis sp. was higher than R. solani. Alternaria longipes also showed a denser hyphae than R. solani, but no evidence of any hyphal penetration was observed. However, these cocultured R. solani hyphae showed an abnormal morphology in comparison with hyphae of R. solani grown alone (Fig. 2f). This may be due to a reduction in cell turgor. Epicoccum nigrum isolates grow alongside of R. solani hyphae and then wind around it, causing lysis of its hyphae (Fig. 2a and b). Epicoccum nigrum did not show any evidence of penetration, although clear inhibition zones were observed where R. solani mycelia were almost dead.

Ability of the antagonists to produce volatile compounds on agar

All antagonistic fungal isolates are capable of producing volatile compounds when grown on PDA media. Table 2 shows a significant difference between various antagonist isolates. The highest inhibition was recorded by T. atrovirde (81.81%), followed by Phomopsis sp. (38.63%), A. longipes (21.02%), and E. nigrum E18 (20.73%), E1 (11.36%), and E8 (10.22%), respectively.

Growth of pathogenic fungus with antagonistic culture filtrates

The influence of antagonistic culture filtrates on the mycelial growth of pathogenic fungus R. solani was evaluated at different concentrations (Fig. 3). The inhibition varied according to the type of antagonistic fungal isolate. This inhibition increased proportionally with the filtrate concentration of the antagonist isolate. The greatest inhibition was observed with T. atroviride culture filtrates.

Greenhouse trials

This experiment revealed that potato seed tubers planted in a substrate inoculated with both R. solani and antagonist germinated within 1 week. The emergence of potato seed tubers was significantly low as compared with those planted in pots containing antagonist fungal isolates (data not shown) or compared with the control treatment (only treated with pathogen). In the case of pots uninfected with pathogen (untreated control), seed tubers planted in the presence of antagonistic fungal isolates started emerging at the same time as the untreated control.

Compared with the inoculated, untreated control, plants receiving antagonist isolates had a significantly reduced index of stem disease (Table 3). The highest disease index of R. solani in stems was observed in the infected control treatment (4.46). The disease index differed significantly among the different treatments, ranging from T. atroviride (0.1), E. nigrum E8 (1.13), E. nigrum E18 (1.80), E. nigrum E1 (1.86), Phomopsis sp. (1.86) to A. longipes (2.86), with the untreated control at unity (1.00). The highest severity of disease was observed in the infected and noninoculated treatment (0.89) followed by A. longipes, Phomopsis sp., E. nigrum E18, E. nigrum E1, E. nigrum E8, and T. atroviride and the untreated control (0.57, 0.37, 0.36, 0.36, 0.22, 0.20, and 0.20, respectively).

All treatments that were inoculated with R. solani and treated with antagonist had a significantly higher yield than the inoculated treatment (R. solani alone). The highest tuber weight (yield) was recorded in T. atroviride (211 g per plant), followed by the untreated treatment (199 g per plant), and then E. nigrum E8, E. nigrum E18, E. nigrum E18, A. longipes, and Phomopsis sp. Results also showed significant differences in fresh weight, plant height, and root weight depending on the treatment. The best results were observed for treatments based using T. atroviride or E. nigrum and the untreated control (results not shown).

Discussion

Our findings show that fungal endophytes have significant antagonistic activity against R. solani when tested by an in vitro dual culture. These fungi were identified as T. atroviride, Phomopsis sp., A. longipes, and E. nigrum (E1, E8, and E18) using ITS regions of rDNA. The inhibition rate varied significantly according to the type of antagonist. The highest inhibition rate against R. solani was recorded using T. atroviride, followed by Phomopsis sp., A. longipes, and three E. nigrum isolates. This experiment also showed that there are two types of mechanisms used by antagonistic fungal isolates, one in which the entire plate on PDA media is invaded such as with T. atroviride and Phomopsis sp., and the other in which R. solani growth is weakly inhibited (A. longipes, E. nigrum). In this study, T. atroviride and Phomopsis sp. were found to be the best antagonists against R. solani.

Confocal microscopy observations of all the fungal BCAs used in this study confirmed that they act differently against R. solani. The active antagonists limit themselves to the pathogens and block their development by winding around the hyphae. However, T. atroviride showed evidence of penetration into pathogen hyphae. This mechanism has been reported (Benhamou & Chet, 1996) using electron microscopy. Whipps (2001) showed that Trichoderma spp. includes several species that produce antibiotics against different plant pathogens and, indeed, many were studied and some have been used as commercial BCAs. Whipps (2001) also mentioned that competition for nutrients and space is another possible mechanism by which BCAs suppress or reduce pathogen infections. For example, T. atroviride can parasitize many soilborne pathogens, such as R. solani, Sclerotium rolfstii, Fusarium sp., Phytophthora sp., and Pythium sp. Trichoderma has been reported to form specialized structures upon contact with its target, in particular, the mycoparasite coils around the host hyphae (Herrera-Estrella & Chet, 1999). There are several studies showing the implication of the genes encoding hydrolytic enzymes and the secretion of these enzymes in the mycoparasitism interactions (Kim et al., 2002). On the other hand, E. nigrum limits pathogen development by growing along R. solani hyphae and inducing their lysis. Epicoccum nigrum, also known in the literature as Epicoccum purpurascens Ehrenb, ex Schlecht., is an anamorphic fungus that produces darkly pigmented (Fig. 1e) muriform conidia on short conidiophores borne on the surface of a sporodochium, a superficial, cushion-like mass of pseudoparenchyma-like hyphal cells. It has been used as a BCA for certain fungal diseases of plants, apple brown rot (Monilia laxa) and damping-off (Hashem & Ali, 2004). However, its efficacy has never been evaluated against Rhizoctonia diseases. Consequently, our work is the first investigation showing the role of this fungus in controlling R. solani diseases on potato.

The results obtained for the production of volatile substances showed that all antagonist isolates produce volatile substances acting against this pathogenic fungus. However, the inhibition of radial pathogenic fungus growth remains inferior to that observed in the dual culture assay. It has been shown that Trichoderma species are highly effective BCAs of soilborne plant pathogens and can produce volatile and nonvolatile antibiotics that inhibit the growth of other pathogens (R. solani, Heterobasidium annosum, and Fusarium oxysporum) (Haran et al., 1996). Our work is the first investigation to test both fungal genera Phomopsis and Alternaria for a role in controlling R. solani diseases. These fungi have been widely reported as potential BCAs against weeds Cirsium arvense, Amaranthus spp., and water hyacinth (Leth et al., 2008).

We found that the in vitro mycelial growth of R. solani declined significantly with increasing amount of culture filtrates of all the antagonistic fungal isolates tested. Whatever the amount of filtrate cultures used, the highest inhibition was obtained with T. atroviride, followed, respectively, by E. nigrum E8, E. nigrum E1, A. longipes, E. nigrum E18, and Phomopsis sp. The slight inhibition obtained with Epicoccum isolate E18 in comparison with both other species of this genus may be due to its poor growth under the in vitro conditions used in this study. Using the same conditions, Campanile et al. (2007) reported that culture filtrates from Epicoccum species had a greater inhibition than those of T. viride against Diplodia corticola, the causal agent of cankers on oaks. This contradiction may be due to the different pathogen tested in the two studies. In our view, the secondary metabolites synthesized by E. nigrum act negatively on R. solani and render them very sensitive. The inhibition zone observed in Petri dish cultures during direct confrontation analysis could be explained by the synthesis of these substances. It has been reported that the production of secondary metabolites was influenced by compounds in the growth medium of the fungal pathogen or antagonist, as well as by temperature and pH. Several reports demonstrated the ability of Trichoderma species to produce volatile and nonvolatile antibiotics that inhibit the growth of plant pathogenic fungi (Haran et al., 1996).

The greenhouse trials showed a consistent and significant antagonistic activity of all fungi against R. solani. Furthermore, a significant positive correlation was observed between the in vitro and the in planta assays. Trichoderma atroviride significantly increased the potato yield and significantly reduced the stem diseases (disease index and severity) compared with the infected and noninoculated control. This result confirms previous reports on Trichoderma species (Whipps, 2001; Campanile et al., 2007). Epicoccum species are in second place with an efficacy similar to untreated and noninoculated treatment, followed by A. longipes and Phomopsis sp. These results confirmed those obtained by in vitro assays and showed that the microorganisms producing the secondary metabolites, in particular, T. atroviride and E. nigrum are the best effective microorganisms against this pathogenic fungus. The low efficacy of Phomopsis sp. and A. longipes in situ could be explained by its use in the literature as the BCAs against weeds that may act directly in plant rather than pathogen. However, application of these microorganisms under field conditions warrants more investigations about their mass of production, their formulation, and their delivery methods. Most fungal BCAs for soilborne pathogens were formulated as powder and applied in the field by drenching the soil surface or in furrows as well as by treating potato seeds directly. Tsror et al. (2001) reported that the application of Trichoderma harzianum in furrows reduced the incidence of black scurf significantly as compared with its application to the soil surface, which showed a relatively small effect.

In summary, our results demonstrated that all fungi tested are effective for controlling R. solani diseases on potato. In our view, some constraints that could limit their effectiveness are rhizosphere complexity and soil environment. In this context, their adaptability to field conditions, their toxicity for humans and animals as formulated products, and their time of application should be studied.

Acknowledgement

This work was supported by the NSERC discovery grant to M.H. We thank the Canada Foundation for Innovation (CFI) for confocal microscopy facility support. We also thank Amandine Honore for technical assistance and Dr David Morse for comments and English editing.

References

Author notes