Serendipita Species Trigger Cultivar-Specific Responses to Fusarium Wilt in Tomato
Serendipita Species Trigger Cultivar-Specific Responses to Fusarium Wilt in Tomato
Ghezel Sefloo, Negar;Wieczorek, Krzysztof;Steinkellner, Siegrid;Hage-Ahmed, Karin
2019-09-28 00:00:00
agronomy Article Serendipita Species Trigger Cultivar-Specific Responses to Fusarium Wilt in Tomato Negar Ghezel Sefloo, Krzysztof Wieczorek, Siegrid Steinkellner and Karin Hage-Ahmed * Institute of Plant Protection, Department of Crop Sciences, University of Natural Resources and Life Sciences, Vienna, 3430 Tulln, Austria; negar.ghezel-sefloo@boku.ac.at (N.G.S.); krzysztof.wieczorek@boku.ac.at (K.W.); siegrid.steinkellner@boku.ac.at (S.S.) * Correspondence: karin.hageahmed@boku.ac.at; Tel.: +43-1-4765-495-305 Received: 21 August 2019; Accepted: 25 September 2019; Published: 28 September 2019 Abstract: The endophytic fungi Serendipita indica and S. vermifera have recently gained increasing attention due to their beneficial eects on plant growth and plant health. Little is known about other species, such as S. williamsii and S. herbamans. To test their biocontrol and growth-promoting potential, susceptible and tolerant tomato cultivars (Kremser Perle and Micro-Tom, respectively) were inoculated with S. williamsii, S. herbamans, S. indica, or S. vermifera and challenged with the soilborne pathogen Fusarium oxysporum f. sp. lycopersici (Fol) in greenhouse experiments. Furthermore, in vitro assays on the direct inhibitory eects of Serendipita spp. against Fol were performed. Negative eects of Fol on phenological growth in the susceptible cultivar were alleviated by all four applied Serendipita spp. Apart from these similar eects on biometric parameters, disease incidence was only reduced by S. herbamans and S. vermifera. In the tolerant cultivar, disease parameters remained unaected although shoot dry mass was negatively aected by S. vermifera. Direct eects of Serendipita spp. against Fol were not evident in the in vitro assays indicating an indirect eect via the host plant. Our results highlight the importance of identifying cultivar-specific eects in pathogen–endophyte–plant interactions to determine the most beneficial combinations. Keywords: Fusarium oxysporum f. sp. lycopersici; tomato; Serendipita herbamans; Serendipita williamsii; Serendipita vermifera; Serendipita indica; biological control; Micro-Tom; fungal endophytes 1. Introduction Fungal endophytes have recently gained increasing attention due to their growth-promoting and bioprotective properties. The testing and exploitation of these fungi might oer new sources for biological control and disease management strategies. One group of endophytic fungi belonging to the family Serendipitaceae (formerly Sebacinales group B) includes such promising candidates and is able to associate with the roots of various plant species [1]. To date, the most well-studied members are Serendipita indica (syn. Piriformospora indica) and S. vermifera (syn. Sebacina vermifera), representing only a small selection of the widespread fungi [1–3]. S. indica was first isolated from a Funneliformis mosseae (syn. Glomus mosseae) spore that originated from the rhizosphere of the woody shrubs Prosopis juliflora and Zizyphus nummularia growing in desert soil in northwestern India [4]. S. indica associates readily with a large number of plant species and can be easily cultured on common artificial media. In previous studies, this fungus was shown to promote plant growth and seed production in a wide range of host plants including wheat and barley [5,6], maize [7], tobacco [8], soybean [9], and Arabidopsis [10]. Studies on biotic stress alleviation as a response to S. indica root colonization have been mostly conducted with barley, wheat, and Arabidopsis as host plants. In these studies, disease control against Blumeria graminis f. sp. hordei [11], Fusarium graminearum [12] and F. culmorum in barley [5,13], and B. graminis f. sp. tritici, F. culmorum Agronomy 2019, 9, 595; doi:10.3390/agronomy9100595 www.mdpi.com/journal/agronomy Agronomy 2019, 9, 595 2 of 17 and Pseudocercosporella herpotrichoides in wheat [6] was shown. Disease control in tomato was reported against F. oxysporum [14], Verticillium dahliae, and Pepino mosaic virus (PepMV), depending on the environmental conditions [15]. S. vermifera was isolated from the roots of the orchid Cyrtostylis reniformis in Australia [16]. Several reports have indicated that S. vermifera is capable of promoting plant growth in a variety of plants [11,17–20]. Some studies on S. vermifera were conducted with respect to its activity as a biocontrol agent against soilborne pathogens including Gaeumannomyces graminis var. tritici causing take-all disease in wheat [21], F. oxysporum f. sp. lentis causing Fusarium wilt in lentil [22], and the leaf pathogen B. graminis f. sp. hordei in barley [11]. Only recently, it was shown that S. vermifera acted against Bipolaris sorokiniana in the rhizosphere of barley by reducing the virulence potential of the pathogen prior to host plant infection [23]. In spite of their potential, using exotic microorganisms such as S. indica and S. vermifera in European agricultural systems might be critical with regard to local fungal endophytic populations and unknown host ranges [24]. Furthermore, there is evidence from molecular data that the prevalence of taxa from Sebacinales is increased in wheat roots from organically managed compared to conventionally managed fields in Switzerland [2], indicating a still hidden source of species in European soils. Consequently, there is a strong need for testing and analyzing the potential of local species and isolates. Recently, S. herbamans was isolated from the roots of Bistorta vivipara in Germany [25] and proved to be a promising candidate for possible application in Europe. Another candidate is S. williamsii (syn. Piriformospora williamsii, ex multinucleate rhizoctonia), which was isolated from a pot culture of Rhizophagus fasciculatus (syn. Glomus fasciculatum originated from soil from Rothamsted Research, UK) associated with clover roots [26,27]. S. williamsii did not show any growth promotion of Arabidopsis in vitro [28] but increased plant biomass when combined with arbuscular mycorrhizal fungi (AMF) in a greenhouse experiment [26]. However, to date, no information is available on the bioprotective potential of S. herbamans and S. williamsii, which might be new biocontrol sources against soilborne pathogens. Fusarium wilt of tomato is caused by the soilborne pathogen F. oxysporum f. sp. lycopersici (Fol) and is responsible for substantial yield losses in open field as well as in greenhouse tomato production (for a review, see McGovern [29]). Diseased plants show wilting, stunting, and yellowing of leaves, which is caused by occlusion of discrete sectors of the vascular tissue and is accompanied by vascular discoloration. Chlamydospores remain viable in the soil for several years and remain an important inoculum source for root infection via the soil. To date, three races of Fol have been identified as well as the corresponding resistance genes [30]. However, not all cultivars carry these resistance genes, and landraces or heirloom cultivars might be of specific interest for organic production or for other traits, such as abiotic stress tolerance [31,32]. The management of Fusarium wilt includes several strategies, such as cultivar selection and chemical, physical, and biological control measures. Biological control based on the use of beneficial microorganisms can oer alternatives for chemical treatments, but detailed knowledge of the underlying mechanisms is necessary for optimum results. This also includes studies at the cultivar level. To date, little is known about specific interactions between cultivars, endophytic fungi, and pathogens, and there is a special focus on induced systemic resistance and increased disease tolerance in tomato. For instance, Steinkellner, et al. [33] showed that tomato cultivars diered in their response to Fol depending on AMF inoculation. Furthermore, cultivar-specific reactions of tomato were reported after resistance induction through rhizobacteria or silicon application, expressed in modifications of the plant cell wall and expression of defense-related enzymes [34,35]. Moreover, there is also evidence of cultivar-specific dierences in growth promotion triggered by S. indica in barley [5], indicating putative cultivar-specific eects of Serendipita spp. The role of cultivar-specificity in induced systemic resistance (ISR) and disease tolerance is a research gap that needs to be addressed in future breeding strategies as proposed for AMF [36]. Hence, identification of the most eective endophytic fungi for certain tomato cultivars could provide further evidence for such complex interactions and tools in a toolbox for integrated management strategies against Fusarium wilt. Agronomy 2019, 9, 595 3 of 17 Therefore, we aimed to investigate (i) the biocontrol potential of the two European isolates, S. herbamans and S. williamsii, against Fol in comparison to S. indica and S. vermifera in susceptible and tolerant tomato cultivars; (ii) the eects of Serendipita spp. on plant growth and development of the two cultivars; and (iii) direct inhibitory eects of Serendipita spp. on Fol. 2. Materials and Methods 2.1. Fungal Cultivation and Inoculum Production Serendipita indica (DSM 11827), S. williamsii (DAR 29830), S. herbamans (DSM 27534), and S. vermifera (MAFF 305830) were cultivated on modified Käfer medium [37] or malt yeast peptone (MYP) medium at 24 C in darkness for 3–4 weeks. For inoculum production, five plugs of 5 mm diameter from 4-week-old cultures were added to 500 mL of liquid Käfer medium for S. indica and S. williamsii and to liquid MYP medium for S. herbamans and S. vermifera. Subsequently, flasks were incubated on an orbital shaker at 24 C in darkness for 2–3 weeks. Before inoculation, mycelium from liquid culture was homogenized for one minute in a high-speed blender. The suspension was centrifuged for 5 min at 10,000 rpm at 4 C, and the pellet was washed with sterile dH O to remove the media. The chlamydospores and/or mycelial fragments were counted using a Fuchs-Rosenthal counting 5 1 chamber, and the concentration was adjusted to 3 10 cfu mL [15]. F. oxysporum f. sp. lycopersici (Sacc.) W.C. Snyder & H.N. Hansen (race 2, isolate 007), kindly provided by B.J. Cornelissen, (University of Amsterdam, The Netherlands), was cultivated on Czapek Dox (CzD) agar (Duchefa Biochemie, Haarlem, The Netherlands) and incubated at 24 C in darkness for two weeks. For Fol inoculation, chlamydospores were prepared using a slight modification of the method described by Goyal, et al. [38] and Bennett and Davis [39]. Microconidia were harvested by flooding the two-week-old culture plates with dH O and rubbing the surface with a Drigalski spatula. The suspension of conidia and hyphae was filtered through three layers of cheese cloth filter (20–150 m pore diameter; Laporte Ges.m.b.H., Wels, Austria) to separate the conidia from hyphae. The conidial 7 1 concentration was adjusted to 1 10 microconidia mL using a Thoma counting chamber. To prepare soil broth, potting soil (Aussaaterde, Gramoflor GmbH & Co. KG, Vechta, Germany) was autoclaved and mixed with autoclaved dH O at a ratio of 1:4 (w/v); for this purpose, 250 g of soil was added to a volume of 1000 mL of autoclaved dH O and agitated on an orbital shaker for 60 min at 90 rpm. The mixture was sieved through a 1.5 or 1 mm mesh sieve to remove coarse materials and then filtered through 8 layers of fleece cloth filter (150 m). Afterwards, 0.025 g of glucose was added to each 50 mL of filtered soil broth and autoclaved at 121 C for 20 min. The following day, the autoclaved soil broth was transferred to another bottle to reduce insoluble impurities or sediment at the bottom of the bottle and autoclaved for a second time. All remaining insoluble impurities were allowed to settle for a few hours. After sedimentation, 50 mL of the clear soil broth was decanted into individual 125 mL Erlenmeyer flasks. Each Erlenmeyer flask containing 50 mL of soil broth was inoculated with 300 L 7 1 of conidial suspension (1 10 microconidia mL ) and incubated near a window under natural light conditions at room temperature (24–27 C) for 10 to 14 days to produce chlamydospores. Thereafter, the Erlenmeyer flask contents were homogenized in a high-speed blender, and chlamydospores were quantified using a Fuchs-Rosenthal counting chamber. 2.2. Plant Cultivation and Inoculation Tomato seeds (Solanum lycopersicum L. cv. Kremser Perle and cv. Micro-Tom) were surface-sterilized by soaking in 50% household bleach (2.8% NaOCl w/w) for 10 min. The household bleach was washed away by replacing the solution with autoclaved ddH O three times. The surface-sterilized tomato seeds were germinated in pots filled with autoclaved perlite (Granuperl S 3–6, Knauf Perlite GmbH, Vienna, Austria) and incubated in a growth chamber (Rumed, Rubarth Apparate GmbH, Germany) at 2 1 24 C with a 14/10 h light/dark photoperiod (light intensity 296 mol m s ) for three to four weeks. After reaching the first true leaf stage, tomato seedlings were removed from the perlite. Root tips were Agronomy 2019, 9, 595 4 of 17 clipped to facilitate the entry of inoculum. In the first step, roots were inoculated with suspensions of chlamydospores and/or mycelial fragments of selected Serendipita spp. at a concentration of 5 1 3 10 cfu mL [15] by submerging the roots for 24 h. Control plants were treated in a similar way as inoculated plants and were dipped in autoclaved dH O. For Fol inoculation, prepared chlamydospore suspensions were mixed with soil at a final concentration of 5000 chlamydospores g of substrate before transplanting the tomato seedlings. Pots were filled with a mixture of soil (Aussaaterde, Gramoflor GmbH & Co. KG, Vechta, Germany), sand (Quarzsand 0–3 mm, Quarzwerke Österreich GmbH, Melk, Austria) and clay (Liapor fit 1–4 mm, Lias Österreich GmbH, Fehring, Austria) (1:1:1, v/v/v), and the tomato seedlings were transplanted into the prepared pots. Plants were grown in a completely randomized design in a greenhouse and were watered regularly with tap water. Twice a week, plants received 50 mL of a nutrient solution consisting of the following components L : 3.47 mM Ca(NO ) , 1.50 mM K SO , 0.0898 mM KH PO , 3.0715 mM 3 2 2 4 2 4 MgSO , 0.100 mM NH NO , 0,0988 mM Fe H O 3 H O, 0.0048 mM Na Bo O 4 H O, 0.0021 mM 4 4 3 6 5 7 2 2 4 7 2 ZnSO 7 H O, 0.0018 mM CuSO 5 H O, 0.0038 mM MnCl 4 H O, and 0.0005 mM MoO [40]. 4 2 4 2 2 2 3 The experimental setup consisted of the following 10 treatments: (i) control, (ii) S. indica (iii) S. williamsii, (iv) S. herbamans, and (v) S. vermifera with (+Fol) and without Fol ( Fol). Each treatment consisted of 12 replicate pots per time-independent trial. All experiments were run in two time-independent repetitions (trials) in a greenhouse at a day/night temperature of 24 C/18 C and relative humidity of 60%. Additional light was provided under a photoperiod of 16 h when outside photosynthetically 2 1 active radiation (PAR) was below 367.43 mol m s . Experiments were conducted from early August to early October 2016 (trial I) and from the middle of August to the middle of October 2016 (trial II) with the cv. Micro-Tom as well as from September to November 2016 (trial I) and from February to April 2017 (trial II) with the cv. Kremser Perle. Temperature within the greenhouse cabin was recorded every 10 min and daily mean, maximum, and minimum temperatures are presented in Figure S1. After 8 weeks, plants were harvested, and the roots were gently washed under tap water. Afterwards, plants were assessed for the growth parameters, and the phenological development was recorded according to the extended BBCH-scale by using the codes for solaneous fruits [41]. Disease progression was visually evaluated using the scale of Wellman [42] for three sets of five randomly chosen plants per treatment and independent repetition (n = 6). In brief, the stems of tomato plants were split open with a scalpel and the progress of vessel discoloration within the stem together with leaf wilt symptoms were used for rating the plants from 0 (healthy) to 15 (dead). To allow weighting the impact on the host plant, ratings were assigned to five groups (g1 = 0.5, 1; g2 = 2; g3 = 3, 4; g4 = 5, 6; g5 = 7), and disease severity was calculated by using the following formula [43]: 5 n + 2 n + 5 n + 10 n + 20 n g1 g2 g3 g4 g5 Disease severity = (1) n diseased plants The percentage of disease incidence was calculated by counting the number of infected plants relative to the total number of plants for three groups of four randomly chosen plants per treatment and independent repetition (n = 6). Additionally, Fol infection was confirmed by incubating a surface-sterilized piece of the hypocotyl (0.5 cm) on potato dextrose agar plates amended with streptomycin (10 mg L ) at 24 C in darkness followed by microscopic analysis of reisolated Fol morphological structures [33,44]. 2.3. Detection of Serendipita spp. in Tomato Roots To detect different Serendipita spp. in roots of both tomato cultivars, a polymerase chain reaction (PCR) was conducted using a primer pair specific for the Serendipita spp. translation elongation factor EF-1 gene (PiTef ). Using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany), the total DNA was extracted according to the manufacturer ’s instructions from 100 mg tomato roots sampled from the following variants 56 days after inoculation: Noninoculated control plants, plants inoculated with Fol, plants Agronomy 2019, 9, 595 5 of 17 inoculated with different Serendipita spp. and plants inoculated with both Fol and different Serendipita spp. As a positive control, the DNA isolated from the mycelium of different Serendipita spp. was used. PCRs were performed with three randomly selected samples from each treatment. Each 15 L reaction contained 1 L of template DNA, 5 Lof ddH O, 7.5 L of 2 Green GoTaq Reaction Buffer (Promega, Madison, WI, USA), and 0.75 L of forward and reverse primer (PiTeff: ATCGTCGCTGTCAACAAGAT, PiTefr: ACCGTCTTGGGGTTGTATCC modified after Deshmukh et al. [11]). The reactions were performed with an initial 3 min denaturation step at 94 C, followed by 34 cycles at 94 C for 40 s, 54 C for 30 s, and 72 C for 30 s, followed by a final extension at 72 C for 2 min. The presence of amplified PCR products was confirmed by electrophoresis using a 2% agarose gel in 1 TAE buffer. 2.4. Antagonistic Activity Assay The interaction between antagonist fungi and Fol was evaluated by the method described by Ghahfarokhi and Goltapeh [21] with slight modifications. For this purpose, to evaluate the inhibition of Fol mycelium growth, one mycelial disc (5 mm) of each selected endophytic fungus from a 2-week-old culture was placed on one side of a PDA plate (9 cm), 2 cm away from the periphery and incubated at 24 C in darkness for 7 days. After 7 days of incubation, the mycelial discs of Fol from a 2-week-old culture were also placed in a similar way but on the opposite side of the plate. As a control treatment, fresh PDA plates were inoculated with mycelial discs of Fol in a similar way. Both inoculated and control plates were incubated at 24 C in darkness, and the percentage inhibition of radial growth (PIRG) was assessed by measuring the radial growth of Fol every day in both plates in the direction of the antagonist (R ) as well as the control plates (R ), 2 1 until Fol covered the entire plates. Additionally, the growth rates of the endophytes were assessed. The data were calculated using the following formula [45]: (R R ) 100 PIRG = (2) For each treatment, 12 replicates were considered, and the entire experiment was repeated 3 times. 2.5. Statistical Analysis In the first step, biometric data (BBCH, shoot dry mass and root dry mass) of each tomato cultivar were analyzed for dierences between the two independent repetitions by one-way ANOVA. Because dierences between repetitions were significant for some parameters, biometric data were analyzed considering the factors ‘Serendipita’ and ‘trial’ for –Fol and +Fol treatments, separately. According to Levene’s test (p < 0.05), some biometric parameters did not meet the homogeneity of variance assumption and, to avoid data transformation, robust two-way ANOVA was used [46]. Analyses were conducted using the software RStudio 1.1.453 [47] and the ‘WRS2‘ package ver. 1.0-0 [48]. Two-way analyses were conducted considering the factors ‘Serendipita’ and ‘trial’ and their interactions using the function pbad2way(), which calculates M-estimators for location based on medians (p < 0.05). The number of bootstrapping samples was set to 599. When the factors ‘Serendipita’, ‘trial’ or the ‘Serendipita trial’ interactions were significant, post hoc comparisons based on M-estimators for location and bootstrapping were performed using the function mcp2a(). Contrasts were considered to be significant when confidence intervals for Huber ’s Y did not cross zero [49]. Disease incidence and disease severity in cv. Kremser Perle was analysed by one-way ANOVA using the software IBM SPSS Statistics 24 (IBM, Armonk, NY USA). Mean separation was performed using Tukey’s test (p < 0.05). Disease ratings for Micro-Tom were analyzed by the nonparametric Kruskal Wallis test (p < 0.05) due to many ratings with 0. Figures were prepared with the software Sigma Plot 14.0 (Systat Software, Inc., San Jose, CA, USA). The obtained data for the antagonistic activity of Serendipita species against Fol were analyzed based on one-way ANOVA and Tukey’s post hoc test (p < 0.05). Agronomy 2019, 9, 595 6 of 17 3. Results 3.1. Eects of Serendipita spp. and Fol Inoculation on the Phenological Growth Stages of Tomato Plants To reveal the eects of Serendipita spp. on tomato plant development, phenological growth stages were determined. The phenological growth stages of the cv. Kremser Perle without Fol application were significantly aected by the factor ‘trial’ (Table 1). Average (21.2 0.1 C and 20.9 0.1 C) as well as maximum daily temperatures (24.9 0.3 C and 24.1 0.1 C) diered between trial I and trial II of the cv. Kremser Perle, especially during early development (Figure S1A). In the first trial, the median for the BBCH rating for all treatments occurred during the flowering stage (BBCH 61 and 62, respectively) (Figure 1A). In the second trial, plants developed more slowly. Plants in the control, S. williamsii and S. herbamans treatments remained in the leaf development stage (BBCH 19), whereas plants in the S. indica and S. vermifera treatments reached the flowering stage (BBCH 61). However, this trend was not statistically significant (Table 1, Figure 1A). Under Fol disease stress, the factor ‘Serendipita’ significantly aected plant growth performance (Table 1). Plants in the +Fol treatment in both trials did not develop further until the leaf development stage (BBCH 18.5) (Figure 1B). The four Serendipita spp. in combination with Fol alleviated this eect and led to a significant increase in growth development until the flowering stage (BBCH 61 and 62, respectively). Micro-Tom plants were not aected by the factor ‘Serendipita’ or by the factor ‘trial’ in the –Fol and +Fol treatments, respectively (Table 1). Ratings ranged between flowering stages (BBCH 62) and fruit development stages (BBCH 73) in the –Fol treatments (Figure 1C). In the +Fol treatments, plants reached the flowering stages (BBCH 62 and 63) (Figure 1D). Micro-Tom trials were characterized by a high fluctuation in mean and maximum daily temperatures (Figure S1B). The mean maximum daily temperature was 28.7 0.3 C and 27.8 0.3 C in trial I and trial II, respectively. The average daily temperature was 23.3 0.2 C and 22.7 0.2 C, respectively. Table 1. Results from Wilcox robust ANOVA (p = 0.05) for the factors ‘Serendipita‘, ‘trial ‘ and ‘Serendipita trial’ interactions. Treatment Factor BBCH Code Shoot Dry Mass [g] Root Dry Mass [g] F. oxysporum f. sp. lycopersici cv. Kremser Perle Serendipita 0.0801 <0.0001 <0.01 trial <0.0001 <0.0001 <0.0001 Serendipita trial 0.0684 <0.0001 0.0601 cv. Micro-Tom Serendipita 0.8331 <0.05 <0.05 trial 0.0501 0.1235 <0.05 Serendipita trial 0.4007 <0.0001 0.2588 +F. oxysporum f. sp. lycopersici cv. Kremser Perle Serendipita <0.01 <0.0001 <0.0001 trial 0.5643 <0.0001 0.4574 Serendipita trial 0.8464 <0.05 0.1252 cv. Micro-Tom Serendipita 0.6745 <0.05 0.0551 trial 0.6260 <0.05 0.0918 Serendipita trial 0.6795 0.3556 0.7780 1 2 BBCH code according to Feller et al. [41]. Significant values (p < 0.05) are highlighted in bold. Agronomy 2018, 8, x FOR PEER REVIEW 8 of 18 Agronomy 2019, 9, 595 7 of 17 Figure 1. Median BBCH values for the cv. Kremser Perle and the cv. Micro-Tom treated with dierent Figure 1. Median BBCH values for the cv. Kremser Perle and the cv. Micro-Tom treated with different Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without (A,C) and Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without (A,C) and with (B,D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II, 56 days after inoculation. with (B,D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II, 56 days after Contrasts for the main factor ‘Serendipita spp.’ are indicated by upper case letters. Contrasts for the inoculation. Contrasts for the main factor ‘Serendipita spp.’ are indicated by upper case letters. main factor ‘trial’ are indicated by lower case letters. Columns followed by the same letters are not Contrasts for the main factor ‘trial’ are indicated by lower case letters. Columns followed by the same significantly dierent (Wilcox robust post hoc test, p < 0.05, n = 12). letters are not significantly different (Wilcox robust post hoc test, p < 0.05, n = 12). 3.2. Eects of Serendipita spp. and Fol Inoculation on Tomato Plant Biomass 3.2. Effects of Serendipita spp. and Fol Inoculation on Tomato Plant Biomass The shoot dry mass of the cv. Kremser Perle was significantly aected by the ‘Serendipita trial’ The shoot dry mass of the cv. Kremser Perle was significantly affected by the ‘Serendipita × trial’ interaction in –Fol (p < 0.0001) and +Fol (p < 0.05) treatments (Table 1). In the –Fol treatments, the first interaction in –Fol (p < 0.0001) and +Fol (p < 0.05) treatments (Table 1). In the –Fol treatments, the first trial was characterized by a higher shoot dry mass development, i.e., 59%, compared to the second trial was characterized by a higher shoot dry mass development, i.e., 59%, compared to the second trial (Figure 2A). ‘Serendipita trial’ interaction eects were observed between the control and S. indica trial (Figure 2A). ‘Serendipita × trial’ interaction effects were observed between the control and S. indica ˆ ˆ (Y = 1570, p < 0.0001) and S. williamsii (Y = 1.050, p < 0.01) treatments, showing an increase of shoot ̂ ̂ (𝛹 = −1,570, p < 0.0001) and S. williamsii (𝛹 = −1.050, p < 0.01) treatments, showing an increase of shoot dry mass of 42% and 38%, respectively, in the first trial. Another source of interaction was observed dry mass of 42% and 38%, respectively, in the first trial. Another source of interaction was observed ˆ ˆ between S. vermifera and S. indica (Y = 4.185, p < 0.0001), S. williamsii (Y = 2.890, p < 0.0001) and ̂ ̂ between S. vermifera and S. indica (𝛹 = 4.185, p < 0.0001), S. williamsii (𝛹 = −2.890, p < 0.0001) and S. S. herbamans (Y = 2.865, p < 0.0001), which was mainly characterized by a stable development of the herbamans (𝛹 = 2.865, p < 0.0001), which was mainly characterized by a stable development of the shoot dry mass in S. vermifera treatments with 4.7 and 4.2 g, respectively, over the two trials compared shoot dry mass in S. vermifera treatments with 4.7 and 4.2 g, respectively, over the two trials compared to S. indica, S. williamsii, and S. herbamans treatments. to S. indica, S. williamsii, and S. herbamans treatments. The shoot dry mass of the cv. Kremser Perle was significantly aected by the ‘Serendipita trial’ The shoot dry mass of the cv. Kremser Perle was significantly affected by the ‘Serendipita × trial’ interaction in +Fol treatments (p < 0.05) (Table 1, Figure 2B). Fol reduced the shoot dry mass to 2.79 and interaction in +Fol treatments (p < 0.05) (Table 1, Figure 2B). Fol reduced the shoot dry mass to 2.79 1.73 g in trials I and II, respectively (Figure 2B). All Serendipita spp. treatments increased the shoot and 1.73 g in trials I and II, respectively (Figure 2B). All Serendipita spp. treatments increased the shoot dry mass by between 57% and 170% compared to the +Fol treatment. Significant ‘Serendipita trial’ dry mass by between 57% and 170% compared to the +Fol treatment. Significant ‘Serendipita × trial’ Agronomy 2019, 9, 595 8 of 17 Agronomy 2018, 8, x FOR PEER REVIEW 9 of 18 interactions occurred between S. indica and S. vermifera (𝛹 = 2.609, p < 0.05), mainly characterized by interactions occurred between S. indica and S. vermifera (Y = 2.609, p < 0.05), mainly characterized by a stable development of S. vermifera over the two trials (4.37 and 4.67 g, respectively). a stable development of S. vermifera over the two trials (4.37 and 4.67 g, respectively). Figure 2. Shoot dry mass (mean S.E.) in g for the cv. Kremser Perle and cv. Micro-Tom treated with Figure 2. Shoot dry mass (mean ± S.E.) in g for the cv. Kremser Perle and cv. Micro-Tom treated with dierent Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without different Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without (A,C) and with (B,D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II (56 days after (A, C) and with (B, D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II (56 days inoculation). Contrasts for the main factor ‘Serendipita’ are indicated by upper case letters. Contrasts after inoculation). Contrasts for the main factor ‘Serendipita’ are indicated by upper case letters. for the main factor ‘trial’ are indicated by lower case letters. Columns followed by the same letters are Contrasts for the main factor ‘trial’ are indicated by lower case letters. Columns followed by the same not significantly dierent (Wilcox robust post hoc test, p < 0.05, n = 12). When main eects could not letters are not significantly different (Wilcox robust post hoc test, p < 0.05, n = 12). When main effects be interpreted due to interaction eects, contrasts for significant ‘Serendipita trial’ interactions are could not be interpreted due to interaction effects, contrasts for significant ‘Serendipita × trial’ presented within the text of the results section. interactions are presented within the text of the results section. For the cv. Micro-Tom, shoot dry mass was significantly aected by a ‘Serendipita trial’ interaction For the cv. Micro-Tom, shoot dry mass was significantly affected by a ‘Serendipita × trial’ in the –Fol treatments (Table 1, Figure 2C). In the second trial, the shoot dry mass of the control treatment interaction in the –Fol treatments (Table 1, Figure 2C). In the second trial, the shoot dry mass of the was increased by 29% compared to the first trial (Figure 2C). A similar pattern was observed for the control treatment was increased by 29% compared to the first trial (Figure 2C). A similar pattern was S. herbamans treatment. This is in contrast to the S. indica treatment (Y = 0.235, p < 0.0001), where observed for the S. herbamans treatment. This is in contrast to the S. indica treatment (𝛹 = −0.235, p < the shoot dry mass with 1.01 and 1.06 g, respectively, was similar over the two trials. Furthermore, 0.0001), where the shoot dry mass with 1.01 and 1.06 g, respectively, was similar over the two trials. the S. vermifera treatment showed an increase in shoot dry mass in trial I whereas there was a reduction Furthermore, the S. vermifera treatment showed an increase in shoot dry mass in trial I whereas there in shoot dry mass in trial II compared to the control (Y = 0.645, p < 0.0001) and the S. herbamans was a reduction in shoot dry mass in trial II compared to the control (𝛹 = −0.645, p < 0.0001) and the treatment (Y = 0.690, p < 0.0001), respectively. S. herbamans treatment (𝛹 = −0.690, p < 0.0001), respectively. The shoot dry mass of the cv. Micro-Tom was significantly aected by the factors ‘Serendipita’ The shoot dry mass of the cv. Micro-Tom was significantly affected by the factors ‘Serendipita’ and ‘trial’ under Fol disease stress (Table 1, Figure 2D). Shoot dry weights in the Fol treatments were and ‘trial’ under Fol disease stress (Table 1, Figure 2D). Shoot dry weights in the Fol treatments were significantly lower in trial I than in trial II (Y = 0.545, p < 0.05) (Figure 2D). Furthermore, the application significantly lower in trial I than in trial II ( 𝛹 = 0.545, p < 0.05) (Figure 2D). Furthermore, the application of S. vermifera reduced the shoot dry mass by 32% compared to the +Fol control treatment over the two trials (𝛹 = −0.410, p < 0.05). Agronomy 2018, 8, x FOR PEER REVIEW 10 of 18 Agronomy 2019, 9, 595 9 of 17 The root dry mass of the cv. Kremser Perle in the –Fol treatments was significantly affected by the main factors ‘Serendipita’ (p < 0.01) and ‘trial’ (p < 0.0001) (Table 1, Figure 3A). Trial I was characterized by a higher root dry mass development than trial II (𝛹 = 0.528, p < 0.0001). Furthermore, of S. vermifera reduced the shoot dry mass by 32% compared to the +Fol control treatment over the two a signi ˆ ficant increase in the root dry mass of the S. indica treatment (0.61 g) compared to the control trials (Y = 0.410, p < 0.05). ̂ ̂ (0.42 g) (𝛹 = −0.37500, p < 0.01) and S. vermifera treatment (0.43 g) (𝛹 = 0.440, p < 0.0001) was evident. The root dry mass of the cv. Kremser Perle in the –Fol treatments was significantly aected The root dry mass of the cv. Kremser Perle in the +Fol treatments was significantly affected by the by the main factors ‘Serendipita’ (p < 0.01) and ‘trial’ (p < 0.0001) (Table 1, Figure 3A). Trial I was factor ‘Serendipita’ (p < 0.0001) (Table 1, Figure 3B). In the +Fol treatme ˆ nts, the application of Serendipita characterized by a higher root dry mass development than trial II (Y = 0.528, p < 0.0001). Furthermore, spp. increased root dry weights by 105 to 136% compared to +Fol alone. a significant increase in the root dry mass of the S. indica treatment (0.61 g) compared to the control Thˆe root dry mass of the cv. Micro-Tom in the –Fol treatments ˆ differed significantly over the two (0.42 g) (Y = 0.37500, p < 0.01) and S. vermifera treatment (0.43 g) (Y = 0.440, p < 0.0001) was evident. trials (𝛹 = −0.135, p < 0.05), but for the main effect, ‘Serendipita’ post hoc procedures did not reveal The root dry mass of the cv. Kremser Perle in the +Fol treatments was significantly aected by the any significant differences (Table 1, Figure 3C). Fol application did not have an effect, alone or in factor ‘Serendipita’ (p < 0.0001) (Table 1, Figure 3B). In the +Fol treatments, the application of Serendipita combination with Serendipita spp., on the root dry mass of Micro-Tom plants (Table 1, Figure 3D). spp. increased root dry weights by 105 to 136% compared to +Fol alone. Figure 3. Root dry mass (mean S.E.) in g for the cv. Kremser Perle and cv. Micro-Tom treated with Figure 3. Root dry mass (mean ± S.E.) in g for the cv. Kremser Perle and cv. Micro-Tom treated with dierent Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without different Serendipita spp. (Si = S. indica, Sw = S. williamsii, Sh = S. herbamans, Sv = S. vermifera) without (A,C) and with (B,D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II (56 days after (A, C) and with (B, D) F. oxysporum f. sp. lycopersici (Fol) inoculation over trial I and trial II (56 days inoculation). Contrasts for the main factor ‘Serendipita’ are indicated by upper case letters. Contrasts after inoculation). Contrasts for the main factor ‘Serendipita’ are indicated by upper case letters. for the main factor ‘trial’ are indicated by lower case letters. Columns followed by the same letters are Contrasts for the main factor ‘trial’ are indicated by lower case letters. Columns followed by the same not significantly dierent (Wilcox robust post hoc test, p < 0.05, n = 12). letters are not significantly different (Wilcox robust post hoc test, p < 0.05, n = 12). The root dry mass of the cv. Micro-Tom in the –Fol treatments diered significantly over the two 3.3. Fol Disease Incidence and Severity in Planta trials (Y = 0.135, p < 0.05), but for the main eect, ‘Serendipita’ post hoc procedures did not reveal any significant dierences (Table 1, Figure 3C). Fol application did not have an eect, alone or in To evaluate the effects of selected Serendipita spp. on plant resistance to Fusarium wilt, disease combination with Serendipita spp., on the root dry mass of Micro-Tom plants (Table 1, Figure 3D). incidence and severity were scored eight weeks after pathogen inoculation for each tomato cultivar. With respect to the cv. Kremser Perle, the application of Serendipita spp. had a significant (F (4, 25) = 7.597, p < 0.0001) effect on the disease incidence (Table 2). The disease incidence was significantly Agronomy 2019, 9, 595 10 of 17 3.3. Fol Disease Incidence and Severity in Planta To evaluate the eects of selected Serendipita spp. on plant resistance to Fusarium wilt, disease incidence and severity were scored eight weeks after pathogen inoculation for each tomato cultivar. With respect to the cv. Kremser Perle, the application of Serendipita spp. had a significant (F 7.597, (4, 25) = p < 0.0001) eect on the disease incidence (Table 2). The disease incidence was significantly reduced from 93 4% in the Fol treatment to 47 7% and 48 7% when coinoculated with S. herbamans and S. vermifera, respectively. Inoculation with S. indica (87 10%) and S. williamsii (64 9%) did not decrease disease incidence significantly. However, disease severity was not aected by the application of Serendipita spp. [F = 1.590, p = 0.208] (Table 2). (4, 25) Table 2. Disease parameters (mean S.E.) 56 days after pathogen inoculation (n = 6). Mean values followed by the same letters are not significantly dierent. Cultivar Treatment Disease Incidence [%] Disease Severity [%] Kremser Perle Fol 93 4 B 27 4 A Si + Fol 87 10 B 31 3 A Sw + Fol 64 9 AB 31 6 A Sh + Fol 47 7 A 17 3 A Sv + Fol 48 7 A 28 6 A F(4,25) = 7.597 F(4,25) = 1.590 p < 0.0001 p = 0.208 Micro-Tom Fol 20 7 A 12 5 A Si + Fol 3 3 A 1 1 A Sw + Fol 7 4 A 1 1 A Sh + Fol 0 0 A 0 0 A Sv + Fol 18 6 A 4 2 A Fol = F. oxysporum f. sp. lycopersici; Si = Serendipita indica; Sw = S. williamsii; Sh = S. herbamans; Sv = S. vermifera, 2 3 ANOVA, Tukey’s test (p < 0.05) Kruskal Wallis test (p < 0.05). Disease incidence in the tolerant cv. Micro-Tom reached 20 7% when Fol was applied alone and 18 7% when S. vermifera was coinoculated (Table 2). The disease incidence in the treatments for S. indica, S. williamsii, and S. herbamans was reduced by 3, 7, and 0%, respectively. However, this reduction was not statistically significant. In all experiments, the Fol treatments did not show any disease symptoms, and reisolation procedures did not show any Fol outgrowth. 3.4. Detection of Endophytic Fungi in Tomato Roots To detect dierent endophytic fungi in the tomato roots, PCR analysis was performed using a primer pair specific for the Serendipita spp. translation elongation factor EF-1 (PiTef ) gene that is widely conserved among species and has been shown to be a useful tool for screening root samples for the presence of dierent Serendipita spp. [11]. PCR with the DNA isolated from roots of plants inoculated with all selected endophytic fungi Fol resulted in amplification of a specific 200 bp product (Figure 4A,B). This PCR product was not detected in control plants, which were inoculated only with Fol or were untreated (Figure 4A,B). All reactions performed with DNA isolated from the mycelia of selected endophytic fungi resulted in the amplification of the specific PCR product (Figure 4C). Agronomy 2019, 9, 595 11 of 17 Agronomy 2018, 8, x FOR PEER REVIEW 12 of 18 Agronomy 2018, 8, x FOR PEER REVIEW 12 of 18 Figure 4. Polymerase chain reaction (PCR) detection of Serendipita spp. in tomato roots. Gel electrophoresis Figure Figure 4. 4. P P ol ol ym ymerase erase chain chain reaction reaction ((PC PCR R)) de dete te ct ct io io n n oo f f Se Se re re ndipita ndipita spp. spp. in in tomato tomato roots roots . G . el G el of PCR products amplified with the primer pair (PiTeff and PiTefr) specific for Serendipita spp. PiTef electrophoresis of PCR products amplified with the primer pair (PiTeff and PiTefr) specific for electrophoresis of PCR products amplified with the primer pair (PiTeff and PiTefr) specific for gene. The PCRs were performed with the genomic DNA extracted from tomato roots inoculated Serendipita spp. PiTef gene. The PCRs were performed with the genomic DNA extracted from tomato Serendipita spp. PiTef gene. The PCRs were performed with the genomic DNA extracted from tomato with selected endophytic fungi (Si, Sw, Sh, Sv) without Fol and with Fol as well as untreated plants roots inoculated with selected endophytic fungi (Si, Sw, Sh, Sv) without Fol and with Fol as well as roots inoculated with selected endophytic fungi (Si, Sw, Sh, Sv) without Fol and with Fol as well as and plants inoculated only with Fol as negative controls (C). A, cv. Kremser Perle; B, cv. Micro-Tom; untreated plants and plants inoculated only with Fol as negative controls (C). A, cv. Kremser Perle; B, untreated plants and plants inoculated only with Fol as negative controls (C). A, cv. Kremser Perle; B, C: Positive controls, PCR products amplified with DNA isolated from mycelia of selected endophytic cv. Micro-Tom; C: positive controls, PCR products amplified with DNA isolated from mycelia of cv. Micro-Tom; C: positive controls, PCR products amplified with DNA isolated from mycelia of fungi. C = control; Si = Serendipita indica; Sw = S. williamsii; Sh = S. herbamans; Sv = S. vermifera; Fol = F. selected endophytic fungi. C = control; Si = Serendipita indica; Sw = S. williamsii; Sh = S. herbamans; Sv = selected endophytic fungi. C = control; Si = Serendipita indica; Sw = S. williamsii; Sh = S. herbamans; Sv = oxysporum f. sp. lycopersici. S. vermifera; Fol = F. oxysporum f. sp. lycopersici. S. vermifera; Fol = F. oxysporum f. sp. lycopersici. 3.5. Antagonistic Activity Assay against Fol 3.5. Antagonistic Activity Assay against Fol 3.5. Antagonistic Activity Assay against Fol In vitro confrontation assays showed that Fol grew rapidly towards the Serendipita spp. colonies. In vitro confrontation assays showed that Fol grew rapidly towards the Serendipita spp. colonies. In vitro confrontation assays showed that Fol grew rapidly towards the Serendipita spp. colonies. In the treatments S. indica and S. williamsii, a contact zone could be observed (Figure 5); however, In the treatments S. indica and S. williamsii, a contact zone could be observed (Figure 5); however, Fol In the treatments S. indica and S. williamsii, a contact zone could be observed (Figure 5); however, Fol Fol passed this zone and was able continue its growth on the respective Serendipita spp. colonies. passed this zone and was able continue its growth on the respective Serendipita spp. colonies. Due to passed this zone and was able continue its growth on the respective Serendipita spp. colonies. Due to Due to the lack of inhibitory eects, PIRG could not be calculated. Phase contrast microscopy of these the lack of inhibitory effects, PIRG could not be calculated. Phase contrast microscopy of these the lack of inhibitory effects, PIRG could not be calculated. Phase contrast microscopy of these interaction zones revealed that Fol continued its growth next to the mycelia of the respective Serendipita interaction zones revealed that Fol continued its growth next to the mycelia of the respective interaction zones revealed that Fol continued its growth next to the mycelia of the respective spp. (Figure S2). Furthermore, neither hyphal coiling on Fol hyphae nor lysis of Fol hyphae was Serendipita spp. (Figure S2). Furthermore, neither hyphal coiling on Fol hyphae nor lysis of Fol hyphae Serendipita spp. (Figure S2). Furthermore, neither hyphal coiling on Fol hyphae nor lysis of Fol hyphae observed at 6 and 7 dpi. The mycelium of Fol growing on S. herbamans and S. vermifera colonies was was observed at 6 and 7 dpi. The mycelium of Fol growing on S. herbamans and S. vermifera colonies was observed at 6 and 7 dpi. The mycelium of Fol growing on S. herbamans and S. vermifera colonies more compact (Figure S3C,D) compared to the mycelium growing on S. indica and S. williamsii colonies was more compact (Figure S3C,D) compared to the mycelium growing on S. indica and S. williamsii was more compact (Figure S3C,D) compared to the mycelium growing on S. indica and S. williamsii (Figure S3A,B). Due to overgrowth of the Serendipita spp. colonies by Fol, PIRG was not calculated. colonies (Figure S3A,B). Due to overgrowth of the Serendipita spp. colonies by Fol, PIRG was not colonies (Figure S3A,B). Due to overgrowth of the Serendipita spp. colonies by Fol, PIRG was not calculated. calculated. Figure 5. In vitro assay on antifungal activity of (A) S. indica, (B) S. williamsii, (C) S. herbamans, and Figure 5. In vitro assay on antifungal activity of (A) S. indica, (B) S. williamsii, (C) S. herbamans, and (D) (D) S. vermifera against F. oxysporum f. sp. lycopersici (Fol) after 9 days of confrontation on PDA plates Figure 5. In vitro assay on antifungal activity of (A) S. indica, (B) S. williamsii, (C) S. herbamans, and S. vermifera against F. oxysporum f. sp. lycopersici (Fol) after 9 days of confrontation on PDA plates and and control plates with S. indica (E), S. williamsii (F), S. herbamans (G), S. vermifera (H), and Fol (I). (D) S. vermifera against F. oxysporum f. sp. lycopersici (Fol) after 9 days of confrontation on PDA plates control plates with S. indica (E), S. williamsii (F), S. herbamans (G), S. vermifera (H), and Fol (I). and control plates with S. indica (E), S. williamsii (F), S. herbamans (G), S. vermifera (H), and Fol (I). 4. Discussion 4. Discussion Agronomy 2019, 9, 595 12 of 17 4. Discussion Here, we report the eects of four dierent Serendipita spp. on two tomato cultivars diering in their susceptibility to Fol. It could be shown that, for the susceptible cv. Kremser Perle, the negative eects of Fol on phenological growth were alleviated by all four applied Serendipita spp. Furthermore, negative eects on shoot and root dry mass were compensated by between 57% and 170% and 105% and 136%, respectively. Apart from these similar eects in biometric parameters, the eects on disease incidence diered. Hence, S. herbamans and S. vermifera reduced the disease incidence from 93 4% in the Fol treatment to 47 7% and 48 7%, respectively. For the treatments S. indica and S. williamsii, the reduction of disease incidence to 84% and 64%, respectively, was not significant. Based on these findings, we speculate that the mode of action diers between the applied Serendipita spp. 4.1. In Vitro Antagonistic Activity Assay against Fol In the biological control of plant pathogens, direct modes of action such as antibiosis, competition, and parasitism and indirect ways of action via activation of host plant defenses can occur. In our in vitro confrontation assays on PDA, we addressed the question of whether the applied Serendipita spp. can directly aect Fol; however, none of the selected Serendipita spp. could suppress the growth of Fol in vitro. On the contrary, all Serendipita species were suppressed in their growth and Fol was able to overgrow S. herbamans and S. vermifera with a dense and S. indica and S. williamsii with a more sparse mycelium in vitro. In heat-treated soil, Fol forms a dense mycelial network on the root surface within two days after inoculation [50], and also in vitro Fol colonizes artificial medium quite rapidly. To date, successful biocontrol of Fol could for instance be achieved by inoculating a nonpathogenic F. oxysporum isolate that competes for nutrients on the rhizoplane [50] and induces resistance responses in the host plant [51,52]. In our experiments, all Serendipita spp. isolates appeared to be poor competitors for nutrients on PDA even when incubated seven days prior to the addition of Fol. S. herbamans, and S. vermifera grew very slowly on PDA and might have needed more incubation time prior to the addition of Fol in order to compete. Additionally, we could not observe evidence of inhibitory diusible or volatile secondary metabolite production on PDA in vitro. This is in line with other studies [5,12] where S. indica did not show any inhibition of F. culmorum and F. graminearum in vitro but reduced the negative eects on the biomass of barley. Furthermore, barley plants inoculated with S. indica had higher ascorbate levels over a three-week post inoculation period [5], which might be responsible for the protective eect [13]. Only recently, it was shown that S. vermifera inhibits B. sorokiniana in vitro and in planta [23]. In vitro, signs of mycoparasitism such as hyphal coiling, penetration, and colonization of hyphae of B. sorokiniana were observed. Furthermore, the authors showed via transcriptomic analysis that S. vermifera reduced the virulence potential of the pathogen prior to host plant infection rather than causing extensive host transcriptional reprogramming [23]. These findings indicate dierent levels of complexity in such multitrophic interactions and might also be very specific with regard to pathogen–endophyte–plant combinations. 4.2. Fol Disease Incidence and Severity in Planta In our experiments, plants were inoculated with Serendipita spp. via root dipping for 24 h and were then potted in substrate amended with Fol chlamydospores. Chlamydospores are the main source of Fol inoculum in the field and were selected to mimic the field situation. Evaluation of Fol disease incidence and severity was performed 56 days. In other studies, plants were cocultivated with S. indica for one to three weeks before plants were additionally challenged with dierent pathogens, showing a bioprotective eect in barley, tomato, and wheat [6,11–13,15]. This might also explain why S. indica did not reduce disease incidence in our study, although the negative eects on plant development were alleviated. However, studies on simultaneous and even delayed inoculation showed a reduction of disease development in maize, lentils, and wheat [6,22,53]. Similar eects could be observed for S. vermifera in lentils [22]. Another dierence can be noted in the observed eects. In the presented Agronomy 2019, 9, 595 13 of 17 studies, eects of the pathogen were alleviated by mitigating the loss of biomass and through the reduction of disease severity, but eects on disease incidence were not observed. Another important factor is the longevity of this bioprotective eect. The bioprotective eect could be transient by delaying disease development, as was also observed against F. graminearum in barley [12]. This could also have been the case for S. indica in our study where a putative protective eect in the beginning might have been transient. In contrast, the eects of S. herbamans and S. vermifera remained stable eight weeks after inoculation, indicating a superior performance to S. indica with the cv. Kremser Perle and Fol. Hence, the European isolate S. herbamans and the Australian isolate S. vermifera proved to be very eective against Fusarium wilt in the greenhouse. However, detailed studies are not available yet on whether host priming is involved and if so, the duration that plants stay primed and which defense pathways are involved. Additionally, transcriptomic studies on the endophyte–pathogen–plant level could give more insight on putative fungal–fungal interactions relevant for disease reduction in planta [23,54,55]. 4.3. Eects of Serendipita spp. and Fol Inoculation on Tomato Plant Biomass An often-mentioned trait of S. indica and S. vermifera is their role in growth promotion in many dierent plant species. Here, the main focus was on investigating the eects against Fol, but Serendipita spp. control treatments revealed that the eects depended on the fungal species, the environmental conditions, and the tomato cultivar. For Kremser Perle in the first trial, which was characterized by higher mean and maximum temperatures, especially during early development (Figure S1A), growth promotion eects of S. indica and S. williamsii were evident. This might be due to the dierent anities of these two fungi with respect to the temperature and temperature-dependent growth rates of tomato plants [56]. Apart from dierent eects on growth promotion, the eects on the cv. Kremser Perle under Fol disease stress remained stable over the trials. For the cv. Micro-Tom, shoot dry mass was significantly aected by a ‘Serendipita trial’ interaction in the –Fol treatments. In addition, the mean and maximum temperatures diered between the trials (Figure S1B). Eects in the –Fol treatments were variable, ranging from neutral to growth impairment (S. indica) and from a growth increase to growth impairment (S. vermifera). There is evidence that S. indica can also reduce tomato plant development depending on the inoculum concentration, tomato growth stage, and substrate composition [15]. However, how dierent temperatures aect the interactions of Serendipita spp. with their host plants is not known yet. Further studies on such temperature dynamics would be very relevant for putative applications in crop production. Apart from factors mentioned above, light intensities have been proven to aect the interaction between PepMV and tomato plants inoculated with S. indica [15]. Plants cultivated under low light intensities showed an increased content of PepMV particles compared to plants without S. indica. Under high light intensities, PepMV particles were even significantly reduced. In our trials, plants received irradiation either from natural light or from high pressure sodium lamps depending on 2 1 the available natural light (PAR < 367.43 mol m s ). Spectra of natural light fluctuate depending on factors such as weather conditions, time of day, and season [57] whereas spectral composition and intensity of artificial light remain constant. The irradiance spectrum has specific eects on plant responses such as secondary metabolite production and photosynthesis [57,58], which also impacts on the carbon content of cells. The amount of PAR provided by natural and artificial light varied between the two trials due to variation in day length, thereby eventually altering the carbon content of cells, which is also a crucial factor in the mutualism–parasitism continuum of AM symbiosis [59]. Whether and how nutrient dynamics in plants mediated by Serendipita spp. are responsible for these interactions remains to be elucidated but would be very relevant for putative applications in crop production. Changes in nutrient dynamics might also be responsible for the reduced shoot dry mass of Micro-Tom plants inoculated with S. vermifera and challenged with Fol. The cultivar Micro-Tom does not harbor the I2 resistance gene but shows tolerance to Fol race 2 [60]. Tolerance can be associated with negative eects such as reduced growth and yield and is still quite unexplored [61]. In the case of the cv. Micro-Tom, plants inoculated with Fol grew slower in the beginning than the respective Agronomy 2019, 9, 595 14 of 17 control plants; however, this eect was transient and could not be observed after 3 to 4 weeks (data not shown). Apart from the reduced shoot dry mass, no eects on the disease ratings occurred, which might indicate that plants prioritize the response to Fusarium wilt. Conversely, in Nicotiana attenuata, S. vermifera improved plant growth without increasing N or P contents, but this eect was accompanied by increased development of Manduca sixta larvae and reduced production of trypsin proteinase inhibitors [17]. Thus, how nutrient dynamics, defense responses, and the less explored mechanisms of disease tolerance interact with endophytic fungi such as S. vermifera are worth investigating in future studies, especially within the context of cultivar-specific growth response. 5. Conclusions In conclusion, we show that all tested Serendipita species alleviated negative eects on plant development in the susceptible cultivar Kremser Perle, whereas Fol disease incidence was reduced only by S. herbamans and S. vermifera. Moreover, the shoot dry mass of the Fol-tolerant cv. Micro-Tom was negatively aected by S. vermifera under Fol disease stress. Observed dierences between Serendipita spp. in their protective eects could be based on the involvement of dierent mechanisms such as induced resistance or growth compensation or variation in the longevity of the eect. Hence, it is pivotal to investigate cultivar-specific eects in pathogen–endophyte–plant interactions to prevent such adverse eects and to determine the most beneficial combinations. Furthermore, we are confident that it is worth investigating other Serendipita species and strains for their potential in plant disease control. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4395/9/10/595/s1. Figure S1: Average, maximum and minimum daily air temperature [ C] inside of the greenhouse cabin for the cv. Kremser Perle (A) and Micro-Tom (B) over trial I and trial II, respectively. Figure S2. Interaction zone of F. oxysporum f. sp. lycopersici (Fol) with S. indica (A, 6 dpi), S. williamsii (B, 6 dpi), S. herbamans (D, 7 dpi) and S. vermifera (E, F, 7 dpi) on PDA. Hyphae of S. herbamans (C) on PDA (7dpi). Phase contrast. Arrows indicate hyphae of Fol. Scale bar: 50 m (A, B, D, E, F), 20 m (C). Figure S3. In vitro antifungal activity assay of (A) S. indica, (B) S. williamsii, (C) S. herbamans and (D) S. vermifera against F. oxysporum f. sp. lycopersici (Fol) 14 days after inoculation on PDA plates. Author Contributions: Conceptualization, N.G.S., S.S. and K.H.-A.; methodology, K.W., S.S. and K.H.-A.; formal analysis, N.G.S., K.H.-A.; investigation, N.G.S., K.W.; writing—original draft preparation, N.G.S. and K.H.-A.; writing—review and editing, N.G.S., K.W., S.S. and K.H.-A.; visualization, N.G.S., KW and K.H.-A.; supervision, S.S. and K.H.-A.; project administration, K.H.-A.; funding acquisition, K.H.-A. Funding: Part of this work was supported by the Hochschuljubiläumsstiftung of the Municipiality of Vienna (H-268043/2016) and by the Austrian Science Fund (FWF, P30051-B32). Acknowledgments: We thank Michael Weiß (University of Tübingen, Germany) for kindly providing S. williamsii (DAR 29830), S. herbamans (DSM 27534) and S. vermifera (MAFF 305830). We thank Karin Baumgartner, Sabine Daxböck-Horvath and Tina Austerlitz for their excellent technical assistance. Conflicts of Interest: The authors declare no conflict of interest. References 1. Weiß, M.; Sýkorová, Z.; Garnica, S.; Riess, K.; Martos, F.; Krause, C.; Oberwinkler, F.; Bauer, R.; Redecker, D. Sebacinales everywhere: Previously overlooked ubiquitous fungal endophytes. PLoS ONE 2011, 6. [CrossRef] [PubMed] 2. Verbruggen, E.; Rillig, M.C.; Wehner, J.; Hegglin, D.; Wittwer, R.; van der Heijden, M.G.A. Sebacinales, but not total root associated fungal communities, are aected by land-use intensity. New Phytol. 2014, 203, 1036–1040. [CrossRef] [PubMed] 3. Venneman, J.; Audenaert, K.; Verwaeren, J.; Baert, G.; Boeckx, P.; Moango, A.M.; Dhed’a, B.D.; Vereecke, D.; Haesaert, G. Congolese rhizospheric soils as a rich source of new plant growth-promoting endophytic Piriformospora isolates. Front. Microbiol. 2017, 8. [CrossRef] [PubMed] 4. Verma, S.; Varma, A.; Rexer, K.-H.; Hassel, A.; Kost, G.; Sarbhoy, A.; Bisen, P.; Bütehorn, B.; Franken, P. Piriformospora indica, gen. et sp. nov., a new root-colonizing fungus. Mycologia 1998, 90, 896–903. [CrossRef] Agronomy 2019, 9, 595 15 of 17 5. Waller, F.; Achatz, B.; Baltruschat, H.; Fodor, J.; Becker, K.; Fischer, M.; Heier, T.; Hückelhoven, R.; Neumann, C.; von Wettstein, D.; et al. The endophytic fungus Piriformospora indica reprograms barley to salt-stress tolerance, disease resistance, and higher yield. Proc. Natl. Acad. Sci. USA 2005, 102, 13386–13391. [CrossRef] [PubMed] 6. Serfling, A.; Wirsel, S.G.R.; Lind, V.; Deising, H.B. Performance of the biocontrol fungus Piriformospora indica on wheat under greenhouse and field conditions. Phytopathology 2007, 97, 523–531. [CrossRef] 7. Varma, A.; Savita, V.; Sudha; Sahay, N.; Bütehorn, B.; Franken, P. Piriformospora indica, a cultivable plant-growth-promoting root endophyte. Appl. Environ. Microbiol. 1999, 65, 2741–2744. [PubMed] 8. Sherameti, I.; Shahollari, B.; Venus, Y.; Altschmied, L.; Varma, A.; Oelmuller, R. The endophytic fungus Piriformospora indica stimulates the expression of nitrate reductase and the starch-degrading enzyme glucan-water dikinase in tobacco and Arabidopsis roots through a homeodomain transcription factor that binds to a conserved motif in their promoters. J. Biol. Chem. 2005, 280, 26241–26247. [CrossRef] 9. Kharkwal, A.C.; Prasad, R.; Kharkwal, H.; Das, A.; Bhatnagar, K.; Varma, A.; Sherameti, I.; Oelmüller, R. Co-Cultivation with Sebacinales. In Advanced Techniques in Soil Microbiology; Varma, A., Oelmüller, R., Eds.; Springer: Berlin/Heidelberg, Germany, 2007; pp. 247–270. [CrossRef] 10. Peškan-Berghöfer, T.; Shahollari, B.; Giong, P.; Hehl, S.; Markert, C.; Blanke, V.; Varma, A.K.; Oelmüller, R. Association of Piriformospora indica with Arabidopsis thaliana roots represents a novel system to study beneficial plant-microbe interactions and involves early plant protein modifications in the endoplasmatic reticulum and at the plasma membrane. Physiol. Plant. 2004, 122, 465–477. [CrossRef] 11. Deshmukh, S.; Hueckelhoven, R.; Schaefer, P.; Imani, J.; Sharma, M.; Weiss, M.; Waller, F.; Kogel, K.H. The root endophytic fungus Piriformospora indica requires host cell death for proliferation during mutualistic symbiosis with barley. Proc. Natl. Acad. Sci. USA 2006, 103, 18450–18457. [CrossRef] 12. Deshmukh, S.; Kogel, K.H. Piriformospora indica protects barley from root rot caused by Fusarium graminearum. J. Plant Dis. Protect. 2007, 114, 263–268. [CrossRef] 13. Harrach, B.D.; Baltruschat, H.; Barna, B.; Fodor, J.; Kogel, K.-H. The mutualistic fungus Piriformospora indica protects barley roots from a loss of antioxidant capacity caused by the necrotrophic pathogen Fusarium culmorum. Mol. Plant-Microbe Interact. 2013, 26, 599–605. [CrossRef] [PubMed] 14. Qiang, X.; Weiss, M.; Kogel, K.H.; Schäfer, P. Piriformospora indica-a mutualistic basidiomycete with an exceptionally large plant host range. Mol. Plant Pathol. 2012, 13, 508–518. [CrossRef] [PubMed] 15. Fakhro, A.; Andrade-Linares, D.R.; von Bargen, S.; Bandte, M.; Büttner, C.; Grosch, R.; Schwarz, D.; Franken, P. Impact of Piriformospora indica on tomato growth and on interaction with fungal and viral pathogens. Mycorrhiza 2010, 20, 191–200. [CrossRef] [PubMed] 16. Warcup, J.H. Mycorrhizal associations of isolates of Sebacina vermifera. New Phytol. 1988, 110, 227–231. [CrossRef] 17. Barazani, O.; Benderoth, M.; Groten, K.; Kuhlemeier, C.; Baldwin, I.T. Piriformospora indica and Sebacina vermifera increase growth performance at the expense of herbivore resistance in Nicotiana attenuata. Oecologia 2005, 146, 234–243. [CrossRef] 18. Ghimire, S.R.; Craven, K.D. Enhancement of switchgrass (Panicum virgatum L.) biomass production under drought conditions by the ectomycorrhizal fungus Sebacina vermifera. Appl. Environ. Microbiol. 2011, 77, 7063–7067. [CrossRef] [PubMed] 19. Ray, P.; Craven, K.D. Sebacina vermifera: A unique root symbiont with vast agronomic potential. World J. Microbiol. Biotechnol. 2016, 32, 1–10. [CrossRef] 20. Ray, P.; Ishiga, T.; Decker, S.R.; Turner, G.B.; Craven, K.D. A novel delivery system for the root symbiotic fungus, Sebacina vermifera, and consequent biomass enhancement of low lignin COMT switchgrass lines. Bioenergy Res. 2015, 8, 922–933. [CrossRef] 21. Ghahfarokhi, R.; Goltapeh, M. Potential of the root endophytic fungus Piriformospora indica; Sebacina vermifera and Trichoderma species in biocontrol of take-all disease of wheat Gaeumannomyces graminis var. tritici in vitro. J. Agric. Technol. 2010, 6, 11–18. 22. Dolatabadi, H.; Goltapeh, E.; Mohammadi, N.; Rabiey, M.; Rohani, N.; Varma, A. Biocontrol potential of root endophytic fungi and Trichoderma species against Fusarium wilt of lentil under in vitro and greehouse conditions. J. Agric. Sci. Technol. 2012, 14, 407–420. 23. Sarkar, D.; Rovenich, H.; Jeena, G.; Nizam, S.; Tissier, A.; Balcke, G.U.; Mahdi, L.; Bonkowski, M.; Langen, G.; Zuccaro, A. The inconspicuous gatekeeper: Endophytic Serendipita vermifera acts as extended plant protection barrier in the rhizosphere. New Phytol. 2019. [CrossRef] [PubMed] Agronomy 2019, 9, 595 16 of 17 24. Ray, P.; Guo, Y.; Kolape, J.; Craven, K.D. Non-targeted colonization by the endomycorrhizal fungus, Serendipita vermifera, in three weeds typically co-occurring with switchgrass. Front. Plant Sci. 2018, 8. [CrossRef] [PubMed] 25. Riess, K.; Oberwinkler, F.; Bauer, R.; Garnica, S. Communities of endophytic sebacinales associated with roots of herbaceous plants in agricultural and grassland ecosystems are dominated by Serendipita herbamans sp. nov. PLoS ONE 2014, 9, e94676. [CrossRef] [PubMed] 26. Williams, P. Orchidaceous rhizoctonias in pot cultures of vesicular-arbuscular mycorrhizal fungi. Can. J. Bot. 1985, 63, 1329–1333. [CrossRef] 27. Basiewicz, M.; Weiss, M.; Kogel, K.H.; Langen, G.; Zorn, H.; Zuccaro, A. Molecular and phenotypic characterization of Sebacina vermifera strains associated with orchids, and the description of Piriformospora williamsii sp. nov. Fungal Biol. 2012, 116, 204–213. [CrossRef] [PubMed] 28. Banhara, A.; Ding, Y.; Kühner, R.; Zuccaro, A.; Parniske, M. Colonization of root cells and plant growth promotion by Piriformospora indica occurs independently of plant common symbiosis genes. Front. Plant Sci. 2015, 6. [CrossRef] [PubMed] 29. McGovern, R.J. Management of tomato diseases caused by Fusarium oxysporum. Crop Protect. 2015, 73, 78–92. [CrossRef] 30. Takken, F.; Rep, M. The arms race between tomato and Fusarium oxysporum. Mol. Plant Pathol. 2010, 11, 309–314. [CrossRef] 31. Corrado, G.; Caramante, M.; Pianelli, P.; Rao, R. Genetic diversity in Italian tomato landraces: Implications for the development of a core collection. Sci. Hortic. 2014, 168, 138–144. [CrossRef] 32. Rivard, C.L.; Louws, F.J. Grafting to manage soilborne diseases in heirloom tomato production. HortScience 2008, 43, 2104–2111. [CrossRef] 33. Steinkellner, S.; Hage-Ahmed, K.; Garcia-Garrido, J.M.; Illana, A.; Ocampo, J.A.; Vierheilig, H. A comparison of wild-type, old and modern tomato cultivars in the interaction with the arbuscular mycorrhizal fungus Glomus mosseae and the tomato pathogen Fusarium oxysporum f. sp. lycopersici. Mycorrhiza 2012, 22, 189–194. [CrossRef] [PubMed] 34. Diogo, R.V.C.; Wydra, K. Silicon-induced basal resistance in tomato against Ralstonia solanacearum is related to modification of pectic cell wall polysaccharide structure. Physiol. Mol. Plant Pathol. 2007, 70, 120–129. [CrossRef] 35. Kurabachew, H.; Wydra, K. Induction of systemic resistance and defense-related enzymes after elicitation of resistance by rhizobacteria and silicon application against Ralstonia solanacearum in tomato (Solanum lycopersicum). Crop Protect. 2014, 57, 1–7. [CrossRef] 36. Hohmann, P.; Messmer, M.M. Breeding for mycorrhizal symbiosis: Focus on disease resistance. Euphytica 2017, 213, 113. [CrossRef] 37. Hill, T.; Käfer, E. Improved protocols for Aspergillus minimal medium: Trace element and minimal medium salt stock solutions. Fungal Genet. Newsl. 2001, 48, 20–21. [CrossRef] 38. Goyal, J.P.; Maraite, H.; Meyer, J.A. Abundant production of chlamydospores by Fusarium oxysporum f. sp. melonis in soil extract with glucose. Neth. J. Plant Pathol. 1973, 79, 162–164. [CrossRef] 39. Bennett, R.S.; Davis, R.M. Method for rapid production of Fusarium oxysporum f. sp. vasinfectum chlamydospores. J. Cotton Sci. 2013, 17, 52–59. 40. Steineck, O. Nährlösungen der Pflanzenkultur. Die Bodenkultur 1951, 5, 313–324. 41. Feller, V.C.; Bleiholder, H.; Buhr, L.; Hack, H.; Heß, M.; Klose, R.; Meier, U.; Strauß, R.; van den Boom, T.; Weber, E. Phenological growth stages of vegetable crops—II. Fruit vegetables and pulses. Nachrbl Dtsch Pflanzenschutzd 1995, 47, 217–232. 42. Wellman, F.L. A technique for studying host resistance and pathogenicity in tomato Fusarium wilt. Phytopathology 1939, 29, 945–956. 43. Hage-Ahmed, K.; Moyses, A.; Voglgruber, A.; Hadacek, F.; Steinkellner, S. Alterations in root exudation of intercropped tomato mediated by the arbuscular mycorrhizal fungus Glomus mosseae and the soilborne pathogen Fusarium oxysporum f. sp. lycopersici. J. Phytopathol. 2013, 161, 763–773. [CrossRef] 44. Nelson, P.; Toussoun, T.; Marasas, W. Fusarium Species. An Illustrated Manual for Identification; Pennsylvania State University Press: University Park, PA, USA, 1983; p. 193. Agronomy 2019, 9, 595 17 of 17 45. Rahman, M.A.; Begum, M.F.; Alam, M.F. Screening of Trichoderma isolates as a biological control agent against Ceratocystis paradoxa causing pineapple disease of sugarcane. Mycobiology 2009, 37, 277–285. [CrossRef] [PubMed] 46. Wilcox, R. Chapter 7-One-way and higher designs for independent groups. In Introduction to Robust Estimation and Hypothesis Testing, 4th ed.; Wilcox, R., Ed.; Academic Press: Cambridge, MA, USA, 2017; pp. 319–415. [CrossRef] 47. R Core Team. RStudio: Integrated development for R; RStudio Inc.: Boston, MA, USA, 2016. 48. Mair, P.; Wilcox, R. Robust statistical methods in R using the WRS2 package. Behav. Res. Methods 2019. [CrossRef] [PubMed] 49. Field, A.P.; Miles, J.; Field, Z. Factorial ANOVA (GLM 3). In Discovering Statistics Using R; Sage: London, UK, 50. Olivain, C.; Humbert, C.; Nahalkova, J.; Fatehi, J.; L’Haridon, F.; Alabouvette, C. Colonization of tomato root by pathogenic and nonpathogenic Fusarium oxysporum strain inoculated together and separately into the soil. Appl. Environ. Microbiol. 2006, 72, 1523–1531. [CrossRef] [PubMed] 51. Larkin, R.P.; Fravel, D.R. Mechanisms of action and dose-reponse relationships governing biological control of Fusarium wilt of tomato by nonpathogenic Fusarium spp. Phytopathology 1999, 89, 1152–1161. [CrossRef] [PubMed] 52. De Lamo, F.J.; Constantin, M.E.; Fresno, D.H.; Boeren, S.; Rep, M.; Takken, F.L.W. Xylem sap proteomics reveals distinct dierences between R gene- and endophyte-mediated resistance against Fusarium wilt disease in tomato. Front. Microbiol. 2018, 9. [CrossRef] [PubMed] 53. Kumar, M.; Yadav, V.; Tuteja, N.; Johri, A.K. Antioxidant enzyme activities in maize plants colonized with Piriformospora indica. Microbiology 2009, 155, 780–790. [CrossRef] [PubMed] 54. Laur, J.; Ramakrishnan, G.B.; Labbé, C.; Lefebvre, F.; Spanu, P.D.; Bélanger, R.R. Eectors involved in fungal–fungal interaction lead to a rare phenomenon of hyperbiotrophy in the tritrophic system biocontrol agent–powdery mildew–plant. New Phytol. 2018, 217, 713–725. [CrossRef] 55. Atanasova, L.; Crom, S.L.; Gruber, S.; Coulpier, F.; Seidl-Seiboth, V.; Kubicek, C.P.; Druzhinina, I.S. Comparative transcriptomics reveals dierent strategies of Trichoderma mycoparasitism. BMC Genomics 2013, 14, 121. [CrossRef] 56. Adams, S.R.; Cockshull, K.E.; Cave, C.R.J. Eect of temperature on the growth and development of tomato fruits. Ann. Bot. 2001, 88, 869–877. [CrossRef] 57. Hogewoning, S.W.; Douwstra, P.; Trouwborst, G.; van Ieperen, W.; Harbinson, J. An artificial solar spectrum substantially alters plant development compared with usual climate room irradiance spectra. J. Exp. Bot. 2010, 61, 1267–1276. [CrossRef] [PubMed] 58. Hernández, R.; Eguchi, T.; Deveci, M.; Kubota, C. Tomato seedling physiological responses under dierent percentages of blue and red photon flux ratios using LEDs and cool white fluorescent lamps. Sci. Hortic. 2016, 213, 270–280. [CrossRef] 59. Johnson, N.C.; Graham, J.H.; Smith, F.A. Functioning of mycorrhizal associations along the mutualism–parasitism continuum. New Phytol. 1997, 135, 575–585. [CrossRef] 60. Takahashi, H.; Shimizu, A.; Arie, T.; Rosmalawati, S.; Fukushima, S.; Kikuchi, M.; Hikichi, Y.; Kanda, A.; Takahashi, A.; Kiba, A.; et al. Catalog of Micro-Tom tomato responses to common fungal, bacterial, and viral pathogens. J. Gen. Plant Pathol. 2005, 71, 8–22. [CrossRef] 61. Pagán, I.; García-Arenal, F. Tolerance to plant pathogens: Theory and experimental evidence. Int. J. Mol. Sci. 2018, 19, 810. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png
Agronomy
Multidisciplinary Digital Publishing Institute
http://www.deepdyve.com/lp/multidisciplinary-digital-publishing-institute/serendipita-species-trigger-cultivar-specific-responses-to-fusarium-vh9TLUbeAU