The effects of fungal root endophytes on plant growth are stable along gradients of abiotic habitat conditions

The effects of fungal root endophytes on plant growth are stable along gradients of abiotic... Abstract Plant symbioses with fungal root endophytes span a continuum from mutualistic to parasitic outcomes, and are highly variable depending on the genotype of each symbiont. The abiotic context in which interactions occur also seems to influence the outcome of plant–endophyte symbioses, but we lack understanding of its relative importance. We aimed to assess if changes in abiotic variables determine the effects of fungal root endophytes on plant growth. We used in vitro co-cultivation assays to test the impact of a selection of endophytic strains from diverse lineages on the growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare along gradients of nutrient availability, light intensity or substrate pH. Most fungi showed a negative but weak effect on plant growth, whereas only a few had persistent detrimental effects across plants and conditions. Changes in abiotic factors affected plant growth but had little influence on their response to fungal inoculation. Of the factors tested, variation in nutrient availability resulted in the most variable plant–endophyte interactions, although changes were feeble and strain-specific. Our findings suggest that the effects of root endophytes on plant growth are robust to changes in the abiotic environment when these encompass the tolerance range of either symbiont. context dependency, environmental gradients, plant–fungus interactions, root, endophytes, symbiosis INTRODUCTION The interactions between species result in diverse effects on the fitness of each organism. Depending on whether the net effects are negative or positive, the interactions commonly are positioned along a continuum between parasitism and mutualism (Ewald 1987). However, the outcomes of interspecies interactions are not static over space and time, as they are affected by the ecological context in which they occur, a process that is frequently termed context dependency of the interactions (Bronstein 1994; Chamberlain, Bronstein and Rudgers 2014). The variation in the outcome of interactions is largely affected by abiotic factors such as temperature or illumination (Davitt, Stansberry and Rudgers 2010; Daskin and Alford 2012), or by biotic factors such as the presence of other species in the community (Agler et al.2016; Laitinen, Hellström and Wäli 2016), to the extent that the net result can change in direction—from mutualism to parasitism, or vice versa—for at least one of the interacting partners. For example, the associations between plants and mycorrhizal fungi are often beneficial for both symbionts, in that the fungus assists the host in the uptake of nutrients and receives organic carbon in return. But these associations have been shown to shift from mutualism to parasitism when soil nutrients are not limiting and the trade-offs between the costs and benefits are reversed (Smith and Read 2010; Andreo-Jimenez et al.2015). Likewise, legume–rhizobacteria symbioses that have been historically considered as mutualistic display differential outcomes depending on the availability of soil nitrogen, as well as on the genotypes of the interacting partners (West et al.2002; Heath and Tiffin 2007). Another widespread and diverse interaction in nature is that occurring between non-mycorrhizal fungal endophytes and plants. Endophytes have been frequently deemed to benefit their hosts through enhancing their resistance and tolerance toward environmental stresses (Clay 1991; Kannadan and Rudgers 2008; Maciá-Vicente et al.2008; Rodriguez and Redman 2008). Most experimental evidence suggests that the outcomes of these associations are very variable across the symbiotic continuum depending on the biotic/abiotic context (Saikkonen et al.1998; Mandyam and Jumpponen 2015; Hiruma et al.2016). However, only a few comprehensive studies have described the range of outcomes and context dependency of endophytic symbioses, and these have largely focused on interactions above-ground (Davitt, Stansberry and Rudgers 2010; Davitt, Chen and Rudgers 2011; Laitinen, Hellström and Wäli 2016). In plant roots, studies on context dependency have mostly dealt with mycorrhizal symbioses (Hoeksema et al.2010), but comparably little is known on the variability of plant associations with non-mycorrhizal root endophytes across environmental gradients. In a recent study, we assessed how the interactions between a diverse array of fungal root endophytes and the three plant species Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare depend on the traits and the phylogenetic affiliations of the fungal partners (Kia et al.2017). Under the assayed conditions, most fungal strains behaved as weak parasites, but their effects on plant growth were strain-dependent and could be partly explained by their morphological, physiological and ecological traits. The study, however, did not afford clues as to how the particular interactions respond to changes in the environmental conditions to which they are subjected. Understanding not only how groups of endophytes differently impact plant fitness, but also how these effects vary across environmental conditions is a necessity to unravel the as yet cryptic role of this fungal guild in ecosystems. For example, there are multiple instances of endophytes phylogenetically related to plant pathogens that provide benefits to their hosts only under particular environmental circumstances (Redman et al.2002; Rodriguez et al.2008; Hiruma et al.2016). Whether these conditional interactions suppose general phenomena with an essential role in the functioning of certain natural systems is not yet clear. Here, we use a selection of the fungal endophytic strains tested in Kia et al. (2017) to assess how their interactions with plants are affected by the abiotic environment. The selection of strains is based on their phylogenetic affiliations, their ecological origins and their observed differential traits and effects on plant growth. We reproduce the interactions of the strains with A. thaliana, M. erraticum and H. vulgare as in Kia et al. (2017), but this time we subjected the symbiotic system to gradients of nutrient availability, light intensity and substrate pH. The selected factors are easy to modify under laboratory conditions, and they are also likely to impact plant–microbe interactions. It is well known that in mycorrhizal associations, fitness of both symbionts depends on abiotic soil conditions (e.g. Piculell, Hoeksema and Thompson 2008), in which levels of macronutrients are important predictors of the plant response to fungal colonization. Nutrient availability may also alter the plant associations with non-mycorrhizal endophytes, with a suspected tendency toward mutualistic interactions in nutritionally limited environments, and to parasitic associations in those that are nutritionally rich (Thrall et al.2007; Newsham 2011; Hiruma et al.2016). Another important factor affecting endophyte symbioses is light availability, which determines plant productivity and hence the availability of photosynthates for plant-associated microbes. There are previous studies showing an interaction of available light and the outcome of plant–endophyte symbioses that show a trend toward parasitism under high light intensities (Bereau et al.2000; Davitt, Stansberry and Rudgers 2010; Álvarez-Loayza et al.2011). Finally, soil pH has been also found to be an important determinant of soil microbial communities and of the performance of plant-associated microbes (Marx and Zak 1965; Wang et al.1993; Belesky and Fedders 1995; Rousk et al.2010). In this study we aimed to assess the relative importance of the selected environmental variables as predictors of the plant response to endophytic inoculation. As indicative measure of the interaction outcomes, we measured changes in the sign and the strength of the endophyte’s effect on plant growth. Specifically, we aimed at answering the following questions: (i) are the outcome of interactions between plants and fungal root endophytes stable across abiotic contexts? (ii) Is the outcome of context-dependent interactions strain-dependent? (iii) Is the outcome of context-dependent interactions stable across host plant species? MATERIALS AND METHODS Fungal strains and plant material A set of 23 strains of endophytic fungi isolated from roots of different plants and geographical locations were selected for this study (Table 1). Most strains originate from a study on the root endophytic diversity associated with Microthlaspi spp. (Glynou et al.2016). In addition, one strain was isolated from Salicornia sp. roots, and Serendipita indica (syn. Piriformospora indica) strain CBS 125645 was obtained from the KNAW-CBS Fungal Biodiversity Centre. Most strains belong to orders Pleosporales, Hypocreales and Helotiales (the most frequent orders found by Glynou et al.2016), and their selection for this study was based on their observed differential combination of morphological and physiological traits, their effects on plant growth (Kia et al.2017), as well as on the association of their natural occurrence with particular ecological factors (Glynou et al.2016). The fungal strains were maintained on corn meal agar medium (CMA, Sigma-Aldrich, St Louis, MO, USA) at approximately 25°C. Table 1. Details of the fungal endophytes included in this study, and of their use in experiments involving different plant hosts and abiotic factors.   Identification  Origin  Experimentb          Strain  Proposed classification  OTUa  Order  ITS accession  Country  Host plant/source  A. thaliana  M. erraticum  H. vulgare  P1188  Thanatephorus sp.  OTU020  Cantharellales  KT268504  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1176  Cadophora sp.  OTU006  Helotiales  KT268493  Croatia  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1312  Cadophora sp.  OTU006  Helotiales  KT268607  Spain  Microthlaspi perfoliatum  p  —  —  P1331  Cadophora sp.  OTU006  Helotiales  KT268626  Spain  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1686  Cadophora sp.  OTU006  Helotiales  KT268959  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1866  Cadophora sp.  OTU006  Helotiales  KT269135  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1940  Cadophora sp.  OTU006  Helotiales  KT269207  Germany  Microthlaspi perfoliatum  p  —  —  P2800  Cadophora sp.  OTU006  Helotiales  KT269998  Germany  Microthlaspi perfoliatum  p  —  —  P1190  Dactylonectria aff. macrodidyma  OTU005  Hypocreales  KT268506  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1076  Fusarium incarnatum-equiseti species complex  OTU010  Hypocreales  KT268395  Turkey  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1141  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268459  France  Microthlaspi erraticum  n  n  n  P1185  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268501  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1304  Fusarium tricinctum species complex  OTU001  Hypocreales  KT268599  Spain  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1020  Alternaria aff. alternata  OTU008  Pleosporales  KT268339  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1603  Alternaria aff. alternata  OTU008  Pleosporales  KU933996  Spain  Salicornia patula  n  n  n  P1191  Alternaria tellustris  OTU002  Pleosporales  KT268507  Croatia  Microthlaspi erraticum  n, l, pc  n, l, p  n, l, p  P1008  unidentified Pleosporales  OTU014  Pleosporales  KT268327  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  —  P1177  unidentified Pleosporales  OTU021  Pleosporales  KT268494  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1004  Leptosphaeria sp.  OTU024  Pleosporales  KT268323  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1134  Paraphoma sp.  OTU007  Pleosporales  KT268452  France  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P2188  Pyrenochaeta sp.  OTU040  Pleosporales  KT269451  Greece  Microthlaspi perfoliatum  n, l, p  n, l, p  n  P2093  Roussoella sp.  OTU043  Pleosporales  KT269356  France  Microthlaspi perfoliatum  p  —  —  CBS 125645  Serendipita indica  OTU033  Sebacinales  DQ411527  India  Rhizospheric soil  n, l, p  n, l, p  n    Identification  Origin  Experimentb          Strain  Proposed classification  OTUa  Order  ITS accession  Country  Host plant/source  A. thaliana  M. erraticum  H. vulgare  P1188  Thanatephorus sp.  OTU020  Cantharellales  KT268504  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1176  Cadophora sp.  OTU006  Helotiales  KT268493  Croatia  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1312  Cadophora sp.  OTU006  Helotiales  KT268607  Spain  Microthlaspi perfoliatum  p  —  —  P1331  Cadophora sp.  OTU006  Helotiales  KT268626  Spain  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1686  Cadophora sp.  OTU006  Helotiales  KT268959  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1866  Cadophora sp.  OTU006  Helotiales  KT269135  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1940  Cadophora sp.  OTU006  Helotiales  KT269207  Germany  Microthlaspi perfoliatum  p  —  —  P2800  Cadophora sp.  OTU006  Helotiales  KT269998  Germany  Microthlaspi perfoliatum  p  —  —  P1190  Dactylonectria aff. macrodidyma  OTU005  Hypocreales  KT268506  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1076  Fusarium incarnatum-equiseti species complex  OTU010  Hypocreales  KT268395  Turkey  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1141  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268459  France  Microthlaspi erraticum  n  n  n  P1185  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268501  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1304  Fusarium tricinctum species complex  OTU001  Hypocreales  KT268599  Spain  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1020  Alternaria aff. alternata  OTU008  Pleosporales  KT268339  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1603  Alternaria aff. alternata  OTU008  Pleosporales  KU933996  Spain  Salicornia patula  n  n  n  P1191  Alternaria tellustris  OTU002  Pleosporales  KT268507  Croatia  Microthlaspi erraticum  n, l, pc  n, l, p  n, l, p  P1008  unidentified Pleosporales  OTU014  Pleosporales  KT268327  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  —  P1177  unidentified Pleosporales  OTU021  Pleosporales  KT268494  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1004  Leptosphaeria sp.  OTU024  Pleosporales  KT268323  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1134  Paraphoma sp.  OTU007  Pleosporales  KT268452  France  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P2188  Pyrenochaeta sp.  OTU040  Pleosporales  KT269451  Greece  Microthlaspi perfoliatum  n, l, p  n, l, p  n  P2093  Roussoella sp.  OTU043  Pleosporales  KT269356  France  Microthlaspi perfoliatum  p  —  —  CBS 125645  Serendipita indica  OTU033  Sebacinales  DQ411527  India  Rhizospheric soil  n, l, p  n, l, p  n  aClassification into operational taxonomic units as defined by >97% ITS sequence similarity, as described by Kia et al. (2017). bExperiments in which fungal strain were used, involving either inoculations in A. thaliana, M. erraticum, or H. vulgare, under different regimes of nutrient availability (n), light intensity (l), or substrate pH (p). cTreatments removed due to contamination of the batch controls. View Large The Brassicaceae Arabidopsis thaliana ecotype Columbia (Col-0) and Microthlaspi erraticum (Mp_K11), and the Poaceae Hordeum vulgare cv. Barke (barley) were used as host plants in co-cultivation experiments with fungal endophytes. Seeds of A. thaliana were provided by the Laboratory of Plant Physiology of Wageningen University. Seeds of M. erraticum were collected from a field population in Germany (Ali et al.2016). Barley seeds were provided by the company Saatzucht Josef Breun GmbH & Co. KG (Herzogenaurach, Germany). Experimental design Nine independent experiments were set up to test the effects of fungal strains on plant growth under different abiotic conditions. Each experiment was performed using one of the three plant species and one of the three abiotic factors: nutrient availability (four levels), light intensity (three levels), and substrate pH (four levels). A core number of fungal strains representing all orders and most genera were used in all cases, but others were only included in particular experiments (Table 1). In particular, several isolates identified as Cadophora sp. OTU006 were included in experiments with A. thaliana and pH, because their natural occurrence was found to be associated with soil pH (Glynou et al.2016), and this association has also been frequently observed for other phylogenetically close endophytic fungi (Sieber and Grünig 2013; Taylor et al.2014). Due to the number of combinations of fungal strains, plant hosts and abiotic factor levels, most experiments were divided into different experimental batches tested at different times, as described in Kia et al. (2017). In each batch, several fungal treatments were tested simultaneously and compared with a single control treatment consisting of non-inoculated plants. A summary of the experimental designs, including the factor levels and the number of strains tested in each batch, is provided in Supplementary Table S1. Plant inoculation assays and growth conditions Plant–endophyte co-cultivation bioassays were performed in a Binder KBW400 growth chamber (Binder Gmbh, Tuttlingen, Germany) as described in Kia et al. (2017), with modifications to include gradients in the abiotic variables (details of the experimental set-up for each variable are described in the following subsections). In brief, A. thaliana and M. erraticum were grown on half-strength (except in nutrient availability assays) Murashige–Skoog basal salt solid medium (MS; Sigma-Aldrich; Murashige and Skoog 1962) at 23°C under continuous illumination (80 μmol m−2 s−1 [photosynthetic photon flux; PPF], except in light intensity assays), on individual wells of 24-well plates, and they were inoculated with the selected fungal strains from CMA colonies as previously described (Kia et al.2017). Treatments for each fungal strain or uninoculated control plants per abiotic factor level consisted of five replicates, each performed in a separate plate placed in random well positions. Ten days after fungal inoculation, root colonization was assessed via direct observation under a stereomicroscope, and the fresh weight of the plant shoots was measured using a precision scale. Experiments with barley were carried out in glass tubes filled with water-saturated sterilized vermiculite (except in nutrient availability assays), as in Kia et al. (2017). Fungal inoculations were performed by adding four 5-mm plugs taken from the margin of growing fungal colonies on CMA to the substrate. Two-day-old barley seedlings were planted on the vermiculite and grown under half-day conditions (12 h:12 h, light–dark, 80 μmol m−2 s−1, except in light intensity assays) at 23°C. Treatments consisted of 10 replicates each implemented in several batches, as described above. After 10 days of growth, the fresh weight of the plants’ shoots and roots were measured and used to calculate total plant weight. In this case, endophytic root colonization in most treatments was assessed in a subset of five randomly selected plants, except for treatments in which the fungus had a strong detrimental effect on roots that prevented the sampling of enough replicates. For every plant, a 10-cm-long root section was surface-sterilized for 1 min in 0.5% (v/v) sodium hypochlorite and washed thrice with sterilized water. Sterilized roots were then ground in 0.5 ml of 0.1% (w/v) water agar using a Retsch MM200 bead beater (Retsch, Haan, Germany), and 200 μl of the resulting suspension was plated on CMA supplemented with antibiotics (25 mg ml−1 chloramphenicol and 50 mg ml−1 streptomycin) and 0.1% (v/v) Triton X-100. Three to seven days later, development of colony forming units of the respective fungi was assessed to confirm fungal colonization of roots. Nutrient availability assays Experiments of fungal inoculation in A. thaliana and M. erraticum were subjected to gradients of nutrient availability by modifying the strength of the MS medium, using the following levels: full MS, 1/2 MS, 1/4 MS and 1/10 MS. In barley, a similar procedure was used by saturating the vermiculate with 20 ml of water, or with full, 1/50, or 1/100 dilutions of Hoagland's plant nutrients solution (Sigma-Aldrich). Light intensity assays Gradients of light intensity for experiments with all three plant species were applied by modifying the number of active fluorescent daylight tubes in different shelves of the growth chamber. These changes resulted in three levels of light intensity at 80.8, 49.2 and 26.5 μmol m−2 s−1, corresponding to five, three or one active tubes out of a maximum of five, respectively. The light tube cassettes were equipped with a reflector material to maximize light diffusion on the shelves. pH assays In assays with A. thaliana and M. erraticum, gradients of pH of 5.7, 6.5, 7 and 7.5 were achieved by modifying the pH of the MS medium before planting the 7-day-old seedlings. These pH levels encompass the natural range of soil pH covered by the samplings described in Glynou et al. (2016) from where most strains originate, and correspond to soil categories from moderately acidic to slightly alkaline (Ditzler, Scheffe and Monger 2017). The same range of pH values was used for barley assays, in this case by saturating the vermiculite with different solutions of 0.1 M sodium phosphate buffer obtained by mixing solutions of monosodium phosphate and disodium phosphate at different ratios, so that the net number of phosphorus atoms remained constant. Statistical analyses All statistical analyses were performed using R v3.0.2 (R Core Team 2016). The data files and the script with the R command lines for the data analysis have been deposited in Figshare (https://figshare.com/s/6009e8e26a5aff0c55f7, http://dx.doi.org/10.6084/m9.figshare.5240572). We first investigated the changes in plant biomass upon fungal inoculation across abiotic conditions by calculating the effect sizes of each fungal treatment with respect to its respective uninoculated control. Effect size is useful to easily detect changes in the sign and strength of the interactions. Before calculations, measurements from control plants showing fungal contamination were removed from the data. This caused the removal of data from three strains in the experiment of A. thaliana and pH (Table 1), which belonged to a batch where all control plants were removed. Effect sizes with 95% confidence intervals were calculated according to the Cohend's d statistic (Cohen 1988) using the function cohen.d in the R package effsize v0.5.4 (Torchiano 2014), which measures the difference in means and standardizes it by their pooled standard deviation. In order to investigate general patterns of variation in the fungal impact on plant growth across abiotic conditions, we compared the effect sizes at different abiotic factor levels using the Kruskal–Wallis rank sum test. We further investigated the variation of the plant interactions with endophytic strains across conditions using linear fixed-effects models. First, to assess the overall effects of endophytes and abiotic conditions on plant growth for each experiment, we built linear models with plant biomass as a response variable, and the abiotic factor and fungal treatment as explanatory fixed-effect variables including an interaction term. In those experiments performed over the course of different batches, we included the factor experimental batch as an additional explanatory variable. Statistical significance in the effects of explanatory variables and in the interaction term was assessed by means of analysis of variance (ANOVA), after checking that the model's residuals did not strongly deviate from normal distributions. The statistical power of these tests was evaluated using the R package pwr v1.2-1 (Champely 2017), represented as the minimum effect size likely to be detected at P < 0.05 with a power of 95%. We carried out a second set of analyses to summarize the individual variation across conditions of individual plant–strain combinations, by repeating the above linear models independently for each experimental batch, so as to represent variation due to each fungus as compared exclusively with its respective control treatment. The model coefficients with confidence intervals for each fungal treatment were extracted from these models using the function sjp.lmer of the package sjPlot v2.3.1 (Lüdecke 2015). P-values from fitted model objects were calculated with the same function, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. RESULTS Inoculation of plants with endophytic strains resulted in a consistent fungal colonization across experiments, as assessed by direct observation under a dissecting microscope for experiments with A. thaliana and M. erraticum (99–100% of plants colonized), and by cultivation of root samples in H. vulgare (60–93% of plants colonized). A few uninoculated control plants were colonized by fungal contaminants, in which case they were excluded from further analyses. The effect sizes of plant growth in response to fungal inoculation were most often negative (Fig. 1), with overall median values ranging between −2 ± 2 (SD) and −0.05 ± 0.5, indicating a consistent reduction of plant biomass in fungus-inoculated versus uninoculated plants. In all cases, effect sizes showed little overall variation across the abiotic factors tested, namely, nutrient availability, light intensity and substrate pH (Fig. 1). Of all experiments, only those involving A. thaliana and light intensity, and M. erraticum and substrate pH showed a significant variation in effect size across levels of each condition (χ2 = 8.9, df = 2, P = 0.006 and χ2 = 39.7, df = 3, P < 0.001, respectively), although the magnitude of the changes was negligible and neither were clear trends in the direction of the variation. Changes in the magnitude of effect sizes were evident in individual treatments, and mainly ranged between approximately neutral and highly negative values (Fig. 1). Figure 1. View largeDownload slide Effect sizes measured as Cohen's d showing the influence of nutrient availability, light intensity and substrate pH on the interactions between fungal root endophytes and the plants Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare. For each plot, gray lines with points represent the variation in effect sizes for individual fungal strains over the abiotic conditions tested, and the black line with points denote the median value for all strains within the experiment. Note that x-axes are represented as factors, and are not proportional to the variable values. Figure 1. View largeDownload slide Effect sizes measured as Cohen's d showing the influence of nutrient availability, light intensity and substrate pH on the interactions between fungal root endophytes and the plants Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare. For each plot, gray lines with points represent the variation in effect sizes for individual fungal strains over the abiotic conditions tested, and the black line with points denote the median value for all strains within the experiment. Note that x-axes are represented as factors, and are not proportional to the variable values. We used linear regression analyses to assess the influence of the abiotic factors on the interactions between plants and endophytes. An additional explanatory variable was included in analyses for experiments performed in different batches, to account for the data variation among independent assays. Power analyses revealed the capability of tests to detect effect sizes of 10.5–17% in assays with A. thaliana and M. erraticum and of 4.9–9.5% in those with H. vulgare, with a power of 95% (Supplementary Table S2). All experiments showed a significant variation across fungal treatments, and the same was true for all abiotic factors in all plant species with the exception of A. thaliana and pH (Table 2). Overall, there was a positive effect of nutrient availability on plant growth, with model estimates ranging from 0.02 ± 0.005 g (SE) to 0.3 ± 0.02 g. In comparison, the overall effects of light intensity and pH were close to neutrality. The interaction between the factors abiotic condition and fungal inoculation were only consistently significant across plant species in the nutrition experiment, indicating that nutrient availability can systematically affect plant–endophyte interactions (Table 2). In addition to these, growth of H. vulgare in response to endophytes also varied significantly with light availability and substrate pH (Table 1). It is possible, however, that model estimates are somewhat biased in experiments involving A. thaliana in pH and H. vulgare in nutrient and light availability conditions, where different experimental batches were included that had a significant impact in data variation (Table 2). Results obtained in experiments with H. vulgare were similar when using as response variables either total plant weight (Table 2) or shoot and root weights separately (Supplementary Table S3), although in the latter cases light intensity alone had little effect on both variables, and the effect of pH appeared to be strong in roots but not in shoots (Supplementary Table S3). Table 2. Summary statistics for the ANOVA of the effects of abiotic factors and fungal inoculation on plant fresh weight.     Nutrients  Light  pH  Plant species  Effect  df  F  P  df  F  P  df  F  P  A. thaliana  Abiotic factor  1,342  124.1  <0.001  1,203  5.9  0.016  1,358  0.7  0.35    Fungus  17,342  21.1  <0.001  15,203  10.9  <0.001  18,358  65.9  <0.001    Experimental batch  1,342  0.5  0.5  n.d.  n.d.  n.d.  1,358  1.3  <0.001    Abiotic factor × fungus  17,342  5.5  <0.001  15,203  0.7  0.8  18,358  1.3  0.21  M. erraticum  Abiotic factor  1,338  68.3  <0.001  1,208  27.5  <0.001  1,234  48.8  <0.001    Fungus  17,338  13.2  <0.001  15,208  5.6  <0.001  15,234  3.2  <0.001    Experimental batch  1,338  3.2  0.07  n.d.  n.d.  n.d.  n.d.  n.d.  n.d.    Abiotic factor × fungus  17,338  2.5  0.001  15,208  0.9  0.6  15,234  0.4  0.97  H. vulgare  Abiotic factor  1,777  89.5  <0.001  1,329  5.4  0.021  1,488  12.4  <0.001    Fungus  16,777  7.8  <0.001  10,329  10.8  <0.001  10,488  6.2  <0.001    Experimental batch  4,777  19.1  <0.001  1,329  22.2  <0.001  2,488  0.6  0.5    Abiotic factor × fungus  16,777  3.8  <0.001  10,329  2.3  0.013  10,488  8.1  <0.001      Nutrients  Light  pH  Plant species  Effect  df  F  P  df  F  P  df  F  P  A. thaliana  Abiotic factor  1,342  124.1  <0.001  1,203  5.9  0.016  1,358  0.7  0.35    Fungus  17,342  21.1  <0.001  15,203  10.9  <0.001  18,358  65.9  <0.001    Experimental batch  1,342  0.5  0.5  n.d.  n.d.  n.d.  1,358  1.3  <0.001    Abiotic factor × fungus  17,342  5.5  <0.001  15,203  0.7  0.8  18,358  1.3  0.21  M. erraticum  Abiotic factor  1,338  68.3  <0.001  1,208  27.5  <0.001  1,234  48.8  <0.001    Fungus  17,338  13.2  <0.001  15,208  5.6  <0.001  15,234  3.2  <0.001    Experimental batch  1,338  3.2  0.07  n.d.  n.d.  n.d.  n.d.  n.d.  n.d.    Abiotic factor × fungus  17,338  2.5  0.001  15,208  0.9  0.6  15,234  0.4  0.97  H. vulgare  Abiotic factor  1,777  89.5  <0.001  1,329  5.4  0.021  1,488  12.4  <0.001    Fungus  16,777  7.8  <0.001  10,329  10.8  <0.001  10,488  6.2  <0.001    Experimental batch  4,777  19.1  <0.001  1,329  22.2  <0.001  2,488  0.6  0.5    Abiotic factor × fungus  16,777  3.8  <0.001  10,329  2.3  0.013  10,488  8.1  <0.001  Significant values (P < 0.05) are shown in bold. n.d., not determined. View Large An assessment of the model coefficients for estimates of each variable supported observations based on effect sizes that most fungal strains negatively impact plant growth (Fig. 2, Supplementary Fig. S1). In this case, the effect of experimental batch was excluded by obtaining coefficients from models independently performed for treatments in each batch, and hence values solely represent the effect of strains as compared with uninoculated plant treatments. These analyses also confirmed the strong interaction between the variables nutrient availability and fungal strain, since they showed a wide variability across strains (Fig. 2, Supplementary Fig. S1). Comparatively, the significant interactions found between light availability or pH and fungal inoculation in H. vulgare showed little variation and hence we deemed them to be marginal (Fig. 2). A visual inspection of the interactions between the three plant species and the fungal strains with effects on plant growth that significantly varied with nutrient availability shows that treatments changing positively with this variable had a rather trivial magnitude and/or did not follow a steady pattern (Fig. 3, Supplementary Fig. S1). On the other hand, interactions varying negatively with nutrient availability seemed to be due to the increase in the gap between the fungus-inoculated and the uninoculated treatments due to fungal parasitism or pathogeny (Fig. 3). In these cases, the increase in plant growth with increasing nutrient availability does not occur in plants hosting fungi that severely compromise their development. We could not detect a tendency of strains within particular fungal lineages to trigger the same responses on plant growth across treatments, but there were individual strains that had a similar impact on plant growth irrespective of the host species or abiotic condition (e.g. Thanatephorus/Rhizoctonia sp. P1188 or Fusarium tricinctum P1304; see Supplementary Fig. S2 for results of all assays). We did not detect a correlation between the overall effect of strains on plant growth and the interaction of these effects with nutrient availability (Spearman's ρ = −0.14 to 0.22, P > 0.4). Figure 2. View largeDownload slide Effects of fungal inoculation on growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, light intensity and substrate pH, as extracted from linear fixed effects models. For each plant species and abiotic condition, data points on the left represent the effects of individual fungal strains respect to uninoculated control plants, whereas points on the right represent the interaction effects of the fungal strain and the abiotic variable. Solid points represent values significant at P < 0.05, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. Horizontal lines represent the median values for all data points in each condition. Additional information for this figure, including the correspondence between data points and fungal strains as well as 95% confidence intervals, is provided in Supplementary Fig. S1. Figure 2. View largeDownload slide Effects of fungal inoculation on growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, light intensity and substrate pH, as extracted from linear fixed effects models. For each plant species and abiotic condition, data points on the left represent the effects of individual fungal strains respect to uninoculated control plants, whereas points on the right represent the interaction effects of the fungal strain and the abiotic variable. Solid points represent values significant at P < 0.05, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. Horizontal lines represent the median values for all data points in each condition. Additional information for this figure, including the correspondence between data points and fungal strains as well as 95% confidence intervals, is provided in Supplementary Fig. S1. Figure 3. View largeDownload slide Variation in the effect of fungal inoculation on total plant biomass of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, as compared with uninoculated controls. Interactions shown correspond to fungal strains with an effect on plant growth that was significantly affected by nutrient availability (see Fig. 2). Solid points with continuous lines represent mean weight values for uninoculated control plants, and open points with dashed lines represent values for fungus-inoculated plants. Error bars represent standard errors. Positive or negative symbols next to the strain names indicate the direction of the variation in the effect of fungi on plant growth with increasing nutrient availability, as obtained by linear models regression analysis. Figure 3. View largeDownload slide Variation in the effect of fungal inoculation on total plant biomass of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, as compared with uninoculated controls. Interactions shown correspond to fungal strains with an effect on plant growth that was significantly affected by nutrient availability (see Fig. 2). Solid points with continuous lines represent mean weight values for uninoculated control plants, and open points with dashed lines represent values for fungus-inoculated plants. Error bars represent standard errors. Positive or negative symbols next to the strain names indicate the direction of the variation in the effect of fungi on plant growth with increasing nutrient availability, as obtained by linear models regression analysis. DISCUSSION We have tested the effect of abiotic conditions—including nutrient availability, light intensity and substrate pH—on the interactions between three different plant species and a diversity of root–endophytic fungi, comprising a variety of taxonomic lineages and geographical origins. Our results show that the effects of fungal root endophytes on the growth of their hosts mainly consist of reductions in plant biomass that are robust to changes in abiotic environmental conditions. No general trends were found in the variation of the magnitude or the direction of the fungal effects across plant species, ruling out a common response to specific abiotic factors by diverse root-colonizing fungi. When changes in the outcome of the symbioses were observed, these occurred in particular combinations of hosts and fungal strains, similarly to what has been found for the interaction between plants and pathogenic fungi (Laine 2007). The differential response of individual plant–fungus combinations is concordant with the frequently reported high variability in the outcome of fungal endophytic symbioses, even when phylogenetically related fungi are compared (Tellenbach, Grünig and Sieber 2011; Mayerhofer, Kernaghan and Harper 2012; Reininger and Sieber 2013; Kia et al.2017). Altogether, our findings suggest that the interplay between the genotypes of the plant host and the root-colonizing fungi is an important determinant of variability in plant growth and likely to affect the host's fitness, which could entail the selection of particular combinations of symbionts in locations under different environmental conditions, as proposed by the geographic mosaic theory of coevolution (Thompson 2005). As expected, both the different abiotic factors and the inoculation with fungal strains had independent effects on plant growth, as estimated by measurements of above-ground or total fresh biomass. In the case of fungal inoculation, the effects of most strains were negative, indicating a parasitism toward the host that is in agreement with previous studies based on in vitro assays of plant–endophyte interactions (Tellenbach, Grünig and Sieber 2011; Keim et al.2014; Mandyam and Jumpponen 2015; Kia et al.2017). The negative impact on plant growth is unsurprising given the trophic dependence of endophytic fungi on the host's resources and the simplicity of the co-cultivation system used, which lacked alternative sources of organic carbon to sustain fungal growth. It is noteworthy, however, that the effects in such conditions mainly consisted of reductions of biomass not accompanied by strong disease symptoms like wilting or chlorosis, and that apparently did not compromise plant survival, with few exceptions in strains related to well-known pathogens such as Thanatephorus/Rhizoctonia sp. or Fusarium sp. The weak plant-parasitic behavior of most root endophytes has already been described for the strains used in this work and others, in bioassays with the same three host plant species also used here (Kia et al.2017). Kia et al. (2017) speculated that the relatively small difference in growth between the endophyte-inoculated and the uninoculated plants in vitro may be readily overcome in natural conditions by changes in the fungal impact on plant growth in response to the environmental context. This is supported by empirical evidence that mutualistic interactions with endophytic fungi can develop in the presence of abiotic or biotic environmental stress (Faeth 2002; Maciá-Vicente et al.2008; Rodriguez et al.2008; Redman et al.2011; Hiruma et al.2016; Almario et al.2017), or that mycorrhizal symbioses tend to be more positive when the accompanying soil microbial communities are more complex (Hoeksema et al.2010). A similar rationale was adopted by Chamberlain, Bronstein and Rudgers (2014) to hypothesize that interactions with weak effect sizes, such as mutualism, are more prone to be context-dependent than other types of symbioses because their outcomes are likely to swing around a neutral effect. Nevertheless, Chamberlain, Bronstein and Rudgers (2014) did not find a strong difference in the context dependency of mutualistic interactions with respect to other types of interactions like competition. Likewise, the low degree of variability that we found in this study is surprising, and suggests that the physiological changes triggered by the tested factors on both the fungus and the plant did not affect the interaction between the two organisms. It must be noted that the ranges applied for each condition were not extreme and encompass magnitudes within the tolerance limit of either symbiont for the short duration of the experiment. It can be expected that use of more severe conditions, either by limitation or excess, would have resulted in a stronger impact in the symbiosis by surpassing the tolerance growth breadth of at least one symbiont. Of the three abiotic gradients tested, the change in nutrient availability had the strongest impact on plant–endophyte interactions. The content of nutrients in the substrate where host plants grow, especially of nitrogen and phosphorus, is well known to impact plant–microbe symbioses. For example, in arbuscular mycorrhizas, mutualistic interactions with a mycorrhizal fungus are preferentially established under phosphorus starvation conditions (Andreo-Jimenez et al.2015), although nitrogen content has also been shown to be relevant for a conducive mutualism (Hoeksema et al.2010). A similar dependency on nutrients has been reported in a few cases for symbioses with non-mycorrhizal fungal endophytes (Behie, Zelisko and Bidochka 2012; Hiruma et al.2016; Almario et al.2017). Newsham (2011) suggested that plant relationships with root endophytic fungi can become beneficial when organic nutrients are present in soil, owing to the ability of fungi to saprotrophically break down complex organic molecules and mobilize sequestered nutrients. In our experiments, no organic nutrients were present, nor did we observe a tendency of endophytes to enhance plant growth under limiting nutrients, which excludes the conditional translocation of particular compounds to the plant under starvation. It seems more likely that the differences in effect sizes triggered by some fungi in response to different nutrient concentrations are related to other kind of physiological changes in either symbiotic partner, such as modifications in the susceptibility to fungal infection in the plant or in the virulence of the fungus, as suggested by Laine (2007). In comparison with nutrient availability, light intensity and pH had little impact on the variability of the interactions. Several studies have reported light intensity as an important factor determining the sign of the interaction between plants and fungal endophytes, in which low light intensities seem to be conducive for more beneficial associations (Davitt, Stansberry and Rudgers 2010; Álvarez-Loayza et al.2011). However, these interactions have been described in leaves, where a strong exposure to light exists, which could drive physiological changes in the fungus important in determining pathogenicity, such as the build-up of reactive oxygen species (Egan et al.2007; Álvarez-Loayza et al.2011). In the case of pH, the differential effects on growth between shoots and roots of H. vulgare indicates that it affects more evidently below-ground tissues, in direct contact with the substratum. In fungi, whereas pH can influence mycelial growth, most fungi can sustain similar growth rates across broad pH ranges (Wheeler, Hurdman and Pitt 1991; Grum-Grzhimaylo et al.2015). The in vitro systems used in this study are artificial and fall short in representing the complex context in which natural plant–endophyte interactions occur. However, simplified systems such as the ones used here have proven adequate to reproduce both beneficial and detrimental interactions between root-colonizing fungi and host plants (e.g. Sesma and Osbourn 2004; Hiruma et al.2016; Almario et al.2017; Venneman et al.2017). Moreover, such systems are necessary to isolate the effects of the study factors from the many biotic and abiotic variables that may confound results in less managed set-ups (Jessup et al.2004). Surprisingly, in spite of such tight control over the non-target sources of variation—and possibly due to it as well—laboratory studies have been shown to yield the highest variability in the interaction outcomes in context-dependency studies, as compared with greenhouse or field studies (Chamberlain, Bronstein and Rudgers 2014). In natural conditions, potential variations in the effects of particular endophytes on the host due to changes in the abiotic environment may become diluted by the combined effects of co-occurring microorganisms and other abiotic and biotic variables. Given the relatively weak variation in the effects of endophytes observed here, it seems unlikely that context-dependent interactions between plants and root endophytes may result in significant changes in the host's fitness when both symbionts grow under conditions within their tolerance ranges. In conclusion, our results do not show an overall strong effect of nutrient availability, light intensity or substrate pH on the interactions between fungal root endophytes and plants, indicating that the outcome of these plant–fungus relationships may be robust to changes in the abiotic environment in the field. Variations in plant growth were observed in interactions between particular fungi and plant species, especially in response to nutrient availability, but these did not seem consistent across fungal lineages or plant species. These different outcomes may ultimately depend on the interplay between the genotypic characteristics of each symbiont, adding to the general variability observed in the interactions between different endophytes and plants (Tellenbach, Grünig and Sieber 2011; Kia et al.2017). Whereas the abiotic conditions studied here seem unlikely determinants of the outcome of plant–endophyte interactions in nature, further research is necessary to assess their importance by assaying more extreme condition and longer term interactions, as well as the implication of biotic factors such as microbe–microbe interactions within the root microbiome. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. Acknowledgements We are grateful to Saatzucht Josef Breun GmbH & Co. KG for kindly providing seeds of barley cv. Barke. We also thank Xiaojuan Xia and Prof. Marco Thines for providing seeds of Microthlaspi erraticum. FUNDING This work was supported by LOEWE (Landes-Offensive zur Entwicklung Wissenschaftlich-ökonomischer Exzellenz) of the state of Hesse and was conducted within the framework of the Cluster for Integrative Fungal Research (IPF). Conflict of Interest. None declared. REFERENCES Agler MT, Ruhe J, Kroll S et al.   Microbial hub taxa link host and abiotic factors to plant microbiome variation. PLoS Biol  2016; 14: e1002352. Google Scholar CrossRef Search ADS PubMed  Ali T, Schmuker A, Runge F et al.   Morphology, phylogeny, and taxonomy of Microthlaspi (Brassicaceae: Coluteocarpeae) and related genera. Taxon  2016; 65: 79– 98. Google Scholar CrossRef Search ADS   Almario J, Jeena G, Wunder J et al.   Root-associated fungal microbiota of nonmycorrhizal Arabis alpina and its contribution to plant phosphorus nutrition. Proc Natl Acad Sci U S A  2017; 114: E9403– 12. Google Scholar CrossRef Search ADS PubMed  Álvarez-Loayza P, White JF Jr, Torres MS et al.   Light converts endosymbiotic fungus to pathogen, influencing seedling survival and niche-space filling of a common tropical tree, Iriartea deltoidea. PLoS One  2011; 6: e16386. Google Scholar CrossRef Search ADS PubMed  Andreo-Jimenez B, Ruyter-Spira C, Bouwmeester HJ et al.   Ecological relevance of strigolactones in nutrient uptake and other abiotic stresses, and in plant-microbe interactions below-ground. Plant Soil  2015; 394: 1– 19. Google Scholar CrossRef Search ADS   Behie SW, Zelisko PM, Bidochka MJ. Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science  2012; 336: 1576– 7. Google Scholar CrossRef Search ADS PubMed  Belesky DP, Fedders JM. Tall fescue development in response to Acremonium coenophialum and soil acidity. Crop Sci  1995; 35: 529– 33. Google Scholar CrossRef Search ADS   Bereau M, Barigah TS, Louisanna E et al.   Effects of endomycorrhizal development and light regimes on the growth of Dicorynia guianensis Amshoff seedlings. Ann For Sci  2000; 57: 725– 33. Google Scholar CrossRef Search ADS   Bronstein JL. Conditional outcomes in mutualistic interactions. Trends Ecol Evol  1994; 9: 214– 7. Google Scholar CrossRef Search ADS PubMed  Chamberlain SA, Bronstein JL, Rudgers JA. How context dependent are species interactions? Ecol Lett  2014; 17: 881– 90. Google Scholar CrossRef Search ADS PubMed  Champely S. pwr: Basic functions for power analysis. R Package Version 1.2-1. 2017, https://cran.r-project.org/web/packages/pwr/index.html (15 May 2017, date last accessed). Clay K. Endophytes as antagonists of plant pests. Microbial Ecology of Leaves . New York: Springer, 1991, 331– 57. Google Scholar CrossRef Search ADS   Cohen J. Statistical Power Analysis for the Behavioral Sciences , 2nd edn. Hillsdale, NJ: Lawrence Earlbaum Associates, 1988, 20– 26. Daskin JH, Alford RA. Context-dependent symbioses and their potential roles in wildlife diseases. Proc R Soc B  2012; 279: 1457– 65. Google Scholar CrossRef Search ADS PubMed  Davitt AJ, Chen C, Rudgers JA. Understanding context-dependency in plant–microbe symbiosis: The influence of abiotic and biotic contexts on host fitness and the rate of symbiont transmission. Environ Exp Bot  2011; 71: 137– 45. Google Scholar CrossRef Search ADS   Davitt AJ, Stansberry M, Rudgers JA. Do the costs and benefits of fungal endophyte symbiosis vary with light availability? New Phytol  2010; 188: 824– 34. Google Scholar CrossRef Search ADS PubMed  Ditzler C, Scheffe K, Monger H. Soil Survey Manual . Washington, DC: Government Printing Office, 2017. Egan MJ, Wang Z-Y, Jones MA et al.   Generation of reactive oxygen species by fungal NADPH oxidases is required for rice blast disease. Proc Natl Acad Sci U S A  2007; 104: 11772– 7. Google Scholar CrossRef Search ADS PubMed  Ewald PW. Transmission modes and evolution of the parasitism-mutualism continuum. Ann N Y Acad Sci  1987; 503: 295– 306. Google Scholar CrossRef Search ADS PubMed  Faeth SH. Are endophytic fungi defensive plant mutualists? Oikos  2002; 98: 25– 36. Google Scholar CrossRef Search ADS   Glynou K, Ali T, Buch A-K et al.   The local environment determines the assembly of root endophytic fungi at a continental scale. Environ Microbiol  2016; 18: 2418– 34. Google Scholar CrossRef Search ADS PubMed  Grum-Grzhimaylo AA, Georgieva ML, Bondarenko SA et al.   On the diversity of fungi from soda soils. Fungal Divers  2015: 1– 48. Heath KD, Tiffin P. Context dependence in the coevolution of plant and rhizobial mutualists. Proc R Soc Lond B Biol Sci  2007; 274: 1905– 12. Google Scholar CrossRef Search ADS   Hiruma K, Gerlach N, Sacristán S et al.   Root endophyte Colletotrichum tofieldiae confers plant fitness benefits that are phosphate status dependent. Cell  2016; 165: 464– 74. Google Scholar CrossRef Search ADS PubMed  Hoeksema JD, Chaudhary VB, Gehring CA et al.   A meta-analysis of context-dependency in plant response to inoculation with mycorrhizal fungi. Ecol Lett  2010; 13: 394– 407. Google Scholar CrossRef Search ADS PubMed  Jessup CM, Kassen R, Forde SE et al.   Big questions, small worlds: microbial model systems in ecology. Trends Ecol Evol  2004; 19: 189– 97. Google Scholar CrossRef Search ADS PubMed  Kannadan S, Rudgers JA. Endophyte symbiosis benefits a rare grass under low water availability. Funct Ecol  2008; 22: 706– 13. Google Scholar CrossRef Search ADS   Keim J, Mishra B, Sharma R et al.   Root-associated fungi of Arabidopsis thaliana and Microthlaspi perfoliatum. Fungal Divers  2014; 66: 99– 111. Google Scholar CrossRef Search ADS   Kia SH, Glynou K, Nau T et al.   Influence of phylogenetic conservatism and trait convergence on the interactions between fungal root endophytes and plants. ISME J  2017; 11: 777– 90. Google Scholar CrossRef Search ADS PubMed  Laine A-L. Pathogen fitness components and genotypes differ in their sensitivity to nutrient and temperature variation in a wild plant–pathogen association. J Evol Biol  2007; 20: 2371– 8. Google Scholar CrossRef Search ADS PubMed  Laitinen RK, Hellström KO, Wäli PR. Context-dependent outcomes of subarctic grass-endophyte symbiosis. Fungal Ecol  2016; 23: 66– 74. Google Scholar CrossRef Search ADS   Lüdecke D. sjPlot: Data visualization for statistics in social science. R Package Version 1. 2015, https://cran.r-project.org/web/packages/sjPlot/index.html (19 June 2017, date last accessed). Maciá-Vicente JG, Jansson H-B, Mendgen K et al.   Colonization of barley roots by endophytic fungi and their reduction of take-all caused by Gaeumannomyces graminis var. tritici. Can J Microbiol  2008; 54: 600– 9. Google Scholar CrossRef Search ADS PubMed  Mandyam KG, Jumpponen A. Mutualism–parasitism paradigm synthesized from results of root-endophyte models. Front Microbiol  2015; 5: 776. Google Scholar CrossRef Search ADS PubMed  Marx DH, Zak B. Effect of pH on mycorrhizal formation of slash pine in aseptic culture. For Sci  1965; 11: 66– 75. Mayerhofer MS, Kernaghan G, Harper KA. The effects of fungal root endophytes on plant growth: a meta-analysis. Mycorrhiza  2012; 23: 119– 28. Google Scholar CrossRef Search ADS PubMed  Murashige T, Skoog F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant  1962; 15: 473– 97. Google Scholar CrossRef Search ADS   Newsham KK. A meta-analysis of plant responses to dark septate root endophytes. New Phytol  2011; 190: 783– 93. Google Scholar CrossRef Search ADS PubMed  Piculell BJ, Hoeksema JD, Thompson JN. Interactions of biotic and abiotic environmental factors in an ectomycorrhizal symbiosis, and the potential for selection mosaics. BMC Biol  2008; 6: 23. Google Scholar CrossRef Search ADS PubMed  R Core Team. R: A Language and Environment for Statistical Computing . Vienna: R Foundation for Statistical Computing, 2016. Redman RS, Kim YO, Woodward CJDA et al.   Increased fitness of rice plants to abiotic stress via habitat adapted symbiosis: a strategy for mitigating impacts of climate change. PLoS One  2011; 6: e14823. Google Scholar CrossRef Search ADS PubMed  Redman RS, Sheehan KB, Stout RG et al.   Thermotolerance generated by plant/fungal symbiosis. Science  2002; 298: 1581. Google Scholar CrossRef Search ADS PubMed  Reininger V, Sieber TN. Mitigation of antagonistic effects on plant growth due to root co-colonization by dark septate endophytes and ectomycorrhiza. Environ Microbiol Rep  2013; 5: 892– 8. Google Scholar CrossRef Search ADS PubMed  Rodriguez R, Redman R. More than 400 million years of evolution and some plants still can’t make it on their own: plant stress tolerance via fungal symbiosis. J Exp Bot  2008; 59: 1109– 14. Google Scholar CrossRef Search ADS PubMed  Rodriguez RJ, Henson J, Van Volkenburgh E et al.   Stress tolerance in plants via habitat-adapted symbiosis. ISME J  2008; 2: 404– 16. Google Scholar CrossRef Search ADS PubMed  Rousk J, Bååth E, Brookes PC et al.   Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME J  2010; 4: 1340– 51. Google Scholar CrossRef Search ADS PubMed  Saikkonen K, Faeth SH, Helander M et al.   Fungal endophytes: A continuum of interactions with host plants. Annu Rev Ecol Syst  1998; 29: 319– 43. Google Scholar CrossRef Search ADS   Sesma A, Osbourn AE. The rice leaf blast pathogen undergoes developmental processes typical of root-infecting fungi. Nature  2004; 431: 582– 6. Google Scholar CrossRef Search ADS PubMed  Sieber TN, Grünig CR. Fungal root endophytes. In: Eshel, A, Beeckman, T (eds). Plant Roots: The Hidden Half , 4th edn. Boca Raton, FL: CRC Press, 2013, 38– 49. Smith SE, Read DJ. Mycorrhizal Symbiosis . Academic Press, 2010. Taylor DL, Hollingsworth TN, McFarland JW et al.   A first comprehensive census of fungi in soil reveals both hyperdiversity and fine-scale niche partitioning. Ecol Monogr  2014; 84: 3– 20. Google Scholar CrossRef Search ADS   Tellenbach C, Grünig CR, Sieber TN. Negative effects on survival and performance of Norway spruce seedlings colonized by dark septate root endophytes are primarily isolate-dependent. Environ Microbiol  2011; 13: 2508– 17. Google Scholar CrossRef Search ADS PubMed  Thompson JN. The Geographic Mosaic of Coevolution . Chicago: University of Chicago Press, 2005. Thrall PH, Hochberg ME, Burdon JJ et al.   Coevolution of symbiotic mutualists and parasites in a community context. Trends Ecol Evol  2007; 22: 120– 6. Google Scholar CrossRef Search ADS PubMed  Torchiano M. effsize: Efficient effect size computation (R package). 2014, https://cran.r-project.org/web/packages/effsize/index.html (12 November 2015, date last accessed). Venneman J, Audenaert K, Verwaeren J et al.   Congolese rhizospheric soils as a rich source of new plant growth-promoting endophytic Piriformospora isolates. Front Microbiol  2017; 8: 212. Google Scholar CrossRef Search ADS PubMed  Wang GM, Stribley DP, Tinker PB et al.   Effects of pH on arbuscular mycorrhiza I. Field observations on the long-term liming experiments at Rothamsted and Woburn. New Phytol  1993; 124: 465– 72. Google Scholar CrossRef Search ADS   West SA, Kiers ET, Pen I et al.   Sanctions and mutualism stability: when should less beneficial mutualists be tolerated? J Evol Biol  2002; 15: 830– 7. Google Scholar CrossRef Search ADS   Wheeler KA, Hurdman BF, Pitt JI. Influence of pH on the growth of some toxigenic species of Aspergillus, Penicillium and Fusarium. Int J Food Microbiol  1991; 12: 141– 9. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Ecology Oxford University Press

The effects of fungal root endophytes on plant growth are stable along gradients of abiotic habitat conditions

Loading next page...
 
/lp/ou_press/the-effects-of-fungal-root-endophytes-on-plant-growth-are-stable-along-986ueVj4Nx
Publisher
Blackwell
Copyright
© FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0168-6496
eISSN
1574-6941
D.O.I.
10.1093/femsec/fix162
Publisher site
See Article on Publisher Site

Abstract

Abstract Plant symbioses with fungal root endophytes span a continuum from mutualistic to parasitic outcomes, and are highly variable depending on the genotype of each symbiont. The abiotic context in which interactions occur also seems to influence the outcome of plant–endophyte symbioses, but we lack understanding of its relative importance. We aimed to assess if changes in abiotic variables determine the effects of fungal root endophytes on plant growth. We used in vitro co-cultivation assays to test the impact of a selection of endophytic strains from diverse lineages on the growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare along gradients of nutrient availability, light intensity or substrate pH. Most fungi showed a negative but weak effect on plant growth, whereas only a few had persistent detrimental effects across plants and conditions. Changes in abiotic factors affected plant growth but had little influence on their response to fungal inoculation. Of the factors tested, variation in nutrient availability resulted in the most variable plant–endophyte interactions, although changes were feeble and strain-specific. Our findings suggest that the effects of root endophytes on plant growth are robust to changes in the abiotic environment when these encompass the tolerance range of either symbiont. context dependency, environmental gradients, plant–fungus interactions, root, endophytes, symbiosis INTRODUCTION The interactions between species result in diverse effects on the fitness of each organism. Depending on whether the net effects are negative or positive, the interactions commonly are positioned along a continuum between parasitism and mutualism (Ewald 1987). However, the outcomes of interspecies interactions are not static over space and time, as they are affected by the ecological context in which they occur, a process that is frequently termed context dependency of the interactions (Bronstein 1994; Chamberlain, Bronstein and Rudgers 2014). The variation in the outcome of interactions is largely affected by abiotic factors such as temperature or illumination (Davitt, Stansberry and Rudgers 2010; Daskin and Alford 2012), or by biotic factors such as the presence of other species in the community (Agler et al.2016; Laitinen, Hellström and Wäli 2016), to the extent that the net result can change in direction—from mutualism to parasitism, or vice versa—for at least one of the interacting partners. For example, the associations between plants and mycorrhizal fungi are often beneficial for both symbionts, in that the fungus assists the host in the uptake of nutrients and receives organic carbon in return. But these associations have been shown to shift from mutualism to parasitism when soil nutrients are not limiting and the trade-offs between the costs and benefits are reversed (Smith and Read 2010; Andreo-Jimenez et al.2015). Likewise, legume–rhizobacteria symbioses that have been historically considered as mutualistic display differential outcomes depending on the availability of soil nitrogen, as well as on the genotypes of the interacting partners (West et al.2002; Heath and Tiffin 2007). Another widespread and diverse interaction in nature is that occurring between non-mycorrhizal fungal endophytes and plants. Endophytes have been frequently deemed to benefit their hosts through enhancing their resistance and tolerance toward environmental stresses (Clay 1991; Kannadan and Rudgers 2008; Maciá-Vicente et al.2008; Rodriguez and Redman 2008). Most experimental evidence suggests that the outcomes of these associations are very variable across the symbiotic continuum depending on the biotic/abiotic context (Saikkonen et al.1998; Mandyam and Jumpponen 2015; Hiruma et al.2016). However, only a few comprehensive studies have described the range of outcomes and context dependency of endophytic symbioses, and these have largely focused on interactions above-ground (Davitt, Stansberry and Rudgers 2010; Davitt, Chen and Rudgers 2011; Laitinen, Hellström and Wäli 2016). In plant roots, studies on context dependency have mostly dealt with mycorrhizal symbioses (Hoeksema et al.2010), but comparably little is known on the variability of plant associations with non-mycorrhizal root endophytes across environmental gradients. In a recent study, we assessed how the interactions between a diverse array of fungal root endophytes and the three plant species Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare depend on the traits and the phylogenetic affiliations of the fungal partners (Kia et al.2017). Under the assayed conditions, most fungal strains behaved as weak parasites, but their effects on plant growth were strain-dependent and could be partly explained by their morphological, physiological and ecological traits. The study, however, did not afford clues as to how the particular interactions respond to changes in the environmental conditions to which they are subjected. Understanding not only how groups of endophytes differently impact plant fitness, but also how these effects vary across environmental conditions is a necessity to unravel the as yet cryptic role of this fungal guild in ecosystems. For example, there are multiple instances of endophytes phylogenetically related to plant pathogens that provide benefits to their hosts only under particular environmental circumstances (Redman et al.2002; Rodriguez et al.2008; Hiruma et al.2016). Whether these conditional interactions suppose general phenomena with an essential role in the functioning of certain natural systems is not yet clear. Here, we use a selection of the fungal endophytic strains tested in Kia et al. (2017) to assess how their interactions with plants are affected by the abiotic environment. The selection of strains is based on their phylogenetic affiliations, their ecological origins and their observed differential traits and effects on plant growth. We reproduce the interactions of the strains with A. thaliana, M. erraticum and H. vulgare as in Kia et al. (2017), but this time we subjected the symbiotic system to gradients of nutrient availability, light intensity and substrate pH. The selected factors are easy to modify under laboratory conditions, and they are also likely to impact plant–microbe interactions. It is well known that in mycorrhizal associations, fitness of both symbionts depends on abiotic soil conditions (e.g. Piculell, Hoeksema and Thompson 2008), in which levels of macronutrients are important predictors of the plant response to fungal colonization. Nutrient availability may also alter the plant associations with non-mycorrhizal endophytes, with a suspected tendency toward mutualistic interactions in nutritionally limited environments, and to parasitic associations in those that are nutritionally rich (Thrall et al.2007; Newsham 2011; Hiruma et al.2016). Another important factor affecting endophyte symbioses is light availability, which determines plant productivity and hence the availability of photosynthates for plant-associated microbes. There are previous studies showing an interaction of available light and the outcome of plant–endophyte symbioses that show a trend toward parasitism under high light intensities (Bereau et al.2000; Davitt, Stansberry and Rudgers 2010; Álvarez-Loayza et al.2011). Finally, soil pH has been also found to be an important determinant of soil microbial communities and of the performance of plant-associated microbes (Marx and Zak 1965; Wang et al.1993; Belesky and Fedders 1995; Rousk et al.2010). In this study we aimed to assess the relative importance of the selected environmental variables as predictors of the plant response to endophytic inoculation. As indicative measure of the interaction outcomes, we measured changes in the sign and the strength of the endophyte’s effect on plant growth. Specifically, we aimed at answering the following questions: (i) are the outcome of interactions between plants and fungal root endophytes stable across abiotic contexts? (ii) Is the outcome of context-dependent interactions strain-dependent? (iii) Is the outcome of context-dependent interactions stable across host plant species? MATERIALS AND METHODS Fungal strains and plant material A set of 23 strains of endophytic fungi isolated from roots of different plants and geographical locations were selected for this study (Table 1). Most strains originate from a study on the root endophytic diversity associated with Microthlaspi spp. (Glynou et al.2016). In addition, one strain was isolated from Salicornia sp. roots, and Serendipita indica (syn. Piriformospora indica) strain CBS 125645 was obtained from the KNAW-CBS Fungal Biodiversity Centre. Most strains belong to orders Pleosporales, Hypocreales and Helotiales (the most frequent orders found by Glynou et al.2016), and their selection for this study was based on their observed differential combination of morphological and physiological traits, their effects on plant growth (Kia et al.2017), as well as on the association of their natural occurrence with particular ecological factors (Glynou et al.2016). The fungal strains were maintained on corn meal agar medium (CMA, Sigma-Aldrich, St Louis, MO, USA) at approximately 25°C. Table 1. Details of the fungal endophytes included in this study, and of their use in experiments involving different plant hosts and abiotic factors.   Identification  Origin  Experimentb          Strain  Proposed classification  OTUa  Order  ITS accession  Country  Host plant/source  A. thaliana  M. erraticum  H. vulgare  P1188  Thanatephorus sp.  OTU020  Cantharellales  KT268504  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1176  Cadophora sp.  OTU006  Helotiales  KT268493  Croatia  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1312  Cadophora sp.  OTU006  Helotiales  KT268607  Spain  Microthlaspi perfoliatum  p  —  —  P1331  Cadophora sp.  OTU006  Helotiales  KT268626  Spain  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1686  Cadophora sp.  OTU006  Helotiales  KT268959  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1866  Cadophora sp.  OTU006  Helotiales  KT269135  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1940  Cadophora sp.  OTU006  Helotiales  KT269207  Germany  Microthlaspi perfoliatum  p  —  —  P2800  Cadophora sp.  OTU006  Helotiales  KT269998  Germany  Microthlaspi perfoliatum  p  —  —  P1190  Dactylonectria aff. macrodidyma  OTU005  Hypocreales  KT268506  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1076  Fusarium incarnatum-equiseti species complex  OTU010  Hypocreales  KT268395  Turkey  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1141  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268459  France  Microthlaspi erraticum  n  n  n  P1185  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268501  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1304  Fusarium tricinctum species complex  OTU001  Hypocreales  KT268599  Spain  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1020  Alternaria aff. alternata  OTU008  Pleosporales  KT268339  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1603  Alternaria aff. alternata  OTU008  Pleosporales  KU933996  Spain  Salicornia patula  n  n  n  P1191  Alternaria tellustris  OTU002  Pleosporales  KT268507  Croatia  Microthlaspi erraticum  n, l, pc  n, l, p  n, l, p  P1008  unidentified Pleosporales  OTU014  Pleosporales  KT268327  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  —  P1177  unidentified Pleosporales  OTU021  Pleosporales  KT268494  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1004  Leptosphaeria sp.  OTU024  Pleosporales  KT268323  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1134  Paraphoma sp.  OTU007  Pleosporales  KT268452  France  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P2188  Pyrenochaeta sp.  OTU040  Pleosporales  KT269451  Greece  Microthlaspi perfoliatum  n, l, p  n, l, p  n  P2093  Roussoella sp.  OTU043  Pleosporales  KT269356  France  Microthlaspi perfoliatum  p  —  —  CBS 125645  Serendipita indica  OTU033  Sebacinales  DQ411527  India  Rhizospheric soil  n, l, p  n, l, p  n    Identification  Origin  Experimentb          Strain  Proposed classification  OTUa  Order  ITS accession  Country  Host plant/source  A. thaliana  M. erraticum  H. vulgare  P1188  Thanatephorus sp.  OTU020  Cantharellales  KT268504  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1176  Cadophora sp.  OTU006  Helotiales  KT268493  Croatia  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1312  Cadophora sp.  OTU006  Helotiales  KT268607  Spain  Microthlaspi perfoliatum  p  —  —  P1331  Cadophora sp.  OTU006  Helotiales  KT268626  Spain  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1686  Cadophora sp.  OTU006  Helotiales  KT268959  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1866  Cadophora sp.  OTU006  Helotiales  KT269135  Bulgaria  Microthlaspi perfoliatum  p  —  —  P1940  Cadophora sp.  OTU006  Helotiales  KT269207  Germany  Microthlaspi perfoliatum  p  —  —  P2800  Cadophora sp.  OTU006  Helotiales  KT269998  Germany  Microthlaspi perfoliatum  p  —  —  P1190  Dactylonectria aff. macrodidyma  OTU005  Hypocreales  KT268506  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n, l, p  P1076  Fusarium incarnatum-equiseti species complex  OTU010  Hypocreales  KT268395  Turkey  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1141  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268459  France  Microthlaspi erraticum  n  n  n  P1185  Fusarium oxysporum species complex  OTU003  Hypocreales  KT268501  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1304  Fusarium tricinctum species complex  OTU001  Hypocreales  KT268599  Spain  Microthlaspi perfoliatum  n, l, pc  n, l, p  n, l, p  P1020  Alternaria aff. alternata  OTU008  Pleosporales  KT268339  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1603  Alternaria aff. alternata  OTU008  Pleosporales  KU933996  Spain  Salicornia patula  n  n  n  P1191  Alternaria tellustris  OTU002  Pleosporales  KT268507  Croatia  Microthlaspi erraticum  n, l, pc  n, l, p  n, l, p  P1008  unidentified Pleosporales  OTU014  Pleosporales  KT268327  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  —  P1177  unidentified Pleosporales  OTU021  Pleosporales  KT268494  Croatia  Microthlaspi erraticum  n, l, p  n, l, p  n  P1004  Leptosphaeria sp.  OTU024  Pleosporales  KT268323  Turkey  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P1134  Paraphoma sp.  OTU007  Pleosporales  KT268452  France  Microthlaspi perfoliatum  n, l, p  n, l, p  n, l, p  P2188  Pyrenochaeta sp.  OTU040  Pleosporales  KT269451  Greece  Microthlaspi perfoliatum  n, l, p  n, l, p  n  P2093  Roussoella sp.  OTU043  Pleosporales  KT269356  France  Microthlaspi perfoliatum  p  —  —  CBS 125645  Serendipita indica  OTU033  Sebacinales  DQ411527  India  Rhizospheric soil  n, l, p  n, l, p  n  aClassification into operational taxonomic units as defined by >97% ITS sequence similarity, as described by Kia et al. (2017). bExperiments in which fungal strain were used, involving either inoculations in A. thaliana, M. erraticum, or H. vulgare, under different regimes of nutrient availability (n), light intensity (l), or substrate pH (p). cTreatments removed due to contamination of the batch controls. View Large The Brassicaceae Arabidopsis thaliana ecotype Columbia (Col-0) and Microthlaspi erraticum (Mp_K11), and the Poaceae Hordeum vulgare cv. Barke (barley) were used as host plants in co-cultivation experiments with fungal endophytes. Seeds of A. thaliana were provided by the Laboratory of Plant Physiology of Wageningen University. Seeds of M. erraticum were collected from a field population in Germany (Ali et al.2016). Barley seeds were provided by the company Saatzucht Josef Breun GmbH & Co. KG (Herzogenaurach, Germany). Experimental design Nine independent experiments were set up to test the effects of fungal strains on plant growth under different abiotic conditions. Each experiment was performed using one of the three plant species and one of the three abiotic factors: nutrient availability (four levels), light intensity (three levels), and substrate pH (four levels). A core number of fungal strains representing all orders and most genera were used in all cases, but others were only included in particular experiments (Table 1). In particular, several isolates identified as Cadophora sp. OTU006 were included in experiments with A. thaliana and pH, because their natural occurrence was found to be associated with soil pH (Glynou et al.2016), and this association has also been frequently observed for other phylogenetically close endophytic fungi (Sieber and Grünig 2013; Taylor et al.2014). Due to the number of combinations of fungal strains, plant hosts and abiotic factor levels, most experiments were divided into different experimental batches tested at different times, as described in Kia et al. (2017). In each batch, several fungal treatments were tested simultaneously and compared with a single control treatment consisting of non-inoculated plants. A summary of the experimental designs, including the factor levels and the number of strains tested in each batch, is provided in Supplementary Table S1. Plant inoculation assays and growth conditions Plant–endophyte co-cultivation bioassays were performed in a Binder KBW400 growth chamber (Binder Gmbh, Tuttlingen, Germany) as described in Kia et al. (2017), with modifications to include gradients in the abiotic variables (details of the experimental set-up for each variable are described in the following subsections). In brief, A. thaliana and M. erraticum were grown on half-strength (except in nutrient availability assays) Murashige–Skoog basal salt solid medium (MS; Sigma-Aldrich; Murashige and Skoog 1962) at 23°C under continuous illumination (80 μmol m−2 s−1 [photosynthetic photon flux; PPF], except in light intensity assays), on individual wells of 24-well plates, and they were inoculated with the selected fungal strains from CMA colonies as previously described (Kia et al.2017). Treatments for each fungal strain or uninoculated control plants per abiotic factor level consisted of five replicates, each performed in a separate plate placed in random well positions. Ten days after fungal inoculation, root colonization was assessed via direct observation under a stereomicroscope, and the fresh weight of the plant shoots was measured using a precision scale. Experiments with barley were carried out in glass tubes filled with water-saturated sterilized vermiculite (except in nutrient availability assays), as in Kia et al. (2017). Fungal inoculations were performed by adding four 5-mm plugs taken from the margin of growing fungal colonies on CMA to the substrate. Two-day-old barley seedlings were planted on the vermiculite and grown under half-day conditions (12 h:12 h, light–dark, 80 μmol m−2 s−1, except in light intensity assays) at 23°C. Treatments consisted of 10 replicates each implemented in several batches, as described above. After 10 days of growth, the fresh weight of the plants’ shoots and roots were measured and used to calculate total plant weight. In this case, endophytic root colonization in most treatments was assessed in a subset of five randomly selected plants, except for treatments in which the fungus had a strong detrimental effect on roots that prevented the sampling of enough replicates. For every plant, a 10-cm-long root section was surface-sterilized for 1 min in 0.5% (v/v) sodium hypochlorite and washed thrice with sterilized water. Sterilized roots were then ground in 0.5 ml of 0.1% (w/v) water agar using a Retsch MM200 bead beater (Retsch, Haan, Germany), and 200 μl of the resulting suspension was plated on CMA supplemented with antibiotics (25 mg ml−1 chloramphenicol and 50 mg ml−1 streptomycin) and 0.1% (v/v) Triton X-100. Three to seven days later, development of colony forming units of the respective fungi was assessed to confirm fungal colonization of roots. Nutrient availability assays Experiments of fungal inoculation in A. thaliana and M. erraticum were subjected to gradients of nutrient availability by modifying the strength of the MS medium, using the following levels: full MS, 1/2 MS, 1/4 MS and 1/10 MS. In barley, a similar procedure was used by saturating the vermiculate with 20 ml of water, or with full, 1/50, or 1/100 dilutions of Hoagland's plant nutrients solution (Sigma-Aldrich). Light intensity assays Gradients of light intensity for experiments with all three plant species were applied by modifying the number of active fluorescent daylight tubes in different shelves of the growth chamber. These changes resulted in three levels of light intensity at 80.8, 49.2 and 26.5 μmol m−2 s−1, corresponding to five, three or one active tubes out of a maximum of five, respectively. The light tube cassettes were equipped with a reflector material to maximize light diffusion on the shelves. pH assays In assays with A. thaliana and M. erraticum, gradients of pH of 5.7, 6.5, 7 and 7.5 were achieved by modifying the pH of the MS medium before planting the 7-day-old seedlings. These pH levels encompass the natural range of soil pH covered by the samplings described in Glynou et al. (2016) from where most strains originate, and correspond to soil categories from moderately acidic to slightly alkaline (Ditzler, Scheffe and Monger 2017). The same range of pH values was used for barley assays, in this case by saturating the vermiculite with different solutions of 0.1 M sodium phosphate buffer obtained by mixing solutions of monosodium phosphate and disodium phosphate at different ratios, so that the net number of phosphorus atoms remained constant. Statistical analyses All statistical analyses were performed using R v3.0.2 (R Core Team 2016). The data files and the script with the R command lines for the data analysis have been deposited in Figshare (https://figshare.com/s/6009e8e26a5aff0c55f7, http://dx.doi.org/10.6084/m9.figshare.5240572). We first investigated the changes in plant biomass upon fungal inoculation across abiotic conditions by calculating the effect sizes of each fungal treatment with respect to its respective uninoculated control. Effect size is useful to easily detect changes in the sign and strength of the interactions. Before calculations, measurements from control plants showing fungal contamination were removed from the data. This caused the removal of data from three strains in the experiment of A. thaliana and pH (Table 1), which belonged to a batch where all control plants were removed. Effect sizes with 95% confidence intervals were calculated according to the Cohend's d statistic (Cohen 1988) using the function cohen.d in the R package effsize v0.5.4 (Torchiano 2014), which measures the difference in means and standardizes it by their pooled standard deviation. In order to investigate general patterns of variation in the fungal impact on plant growth across abiotic conditions, we compared the effect sizes at different abiotic factor levels using the Kruskal–Wallis rank sum test. We further investigated the variation of the plant interactions with endophytic strains across conditions using linear fixed-effects models. First, to assess the overall effects of endophytes and abiotic conditions on plant growth for each experiment, we built linear models with plant biomass as a response variable, and the abiotic factor and fungal treatment as explanatory fixed-effect variables including an interaction term. In those experiments performed over the course of different batches, we included the factor experimental batch as an additional explanatory variable. Statistical significance in the effects of explanatory variables and in the interaction term was assessed by means of analysis of variance (ANOVA), after checking that the model's residuals did not strongly deviate from normal distributions. The statistical power of these tests was evaluated using the R package pwr v1.2-1 (Champely 2017), represented as the minimum effect size likely to be detected at P < 0.05 with a power of 95%. We carried out a second set of analyses to summarize the individual variation across conditions of individual plant–strain combinations, by repeating the above linear models independently for each experimental batch, so as to represent variation due to each fungus as compared exclusively with its respective control treatment. The model coefficients with confidence intervals for each fungal treatment were extracted from these models using the function sjp.lmer of the package sjPlot v2.3.1 (Lüdecke 2015). P-values from fitted model objects were calculated with the same function, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. RESULTS Inoculation of plants with endophytic strains resulted in a consistent fungal colonization across experiments, as assessed by direct observation under a dissecting microscope for experiments with A. thaliana and M. erraticum (99–100% of plants colonized), and by cultivation of root samples in H. vulgare (60–93% of plants colonized). A few uninoculated control plants were colonized by fungal contaminants, in which case they were excluded from further analyses. The effect sizes of plant growth in response to fungal inoculation were most often negative (Fig. 1), with overall median values ranging between −2 ± 2 (SD) and −0.05 ± 0.5, indicating a consistent reduction of plant biomass in fungus-inoculated versus uninoculated plants. In all cases, effect sizes showed little overall variation across the abiotic factors tested, namely, nutrient availability, light intensity and substrate pH (Fig. 1). Of all experiments, only those involving A. thaliana and light intensity, and M. erraticum and substrate pH showed a significant variation in effect size across levels of each condition (χ2 = 8.9, df = 2, P = 0.006 and χ2 = 39.7, df = 3, P < 0.001, respectively), although the magnitude of the changes was negligible and neither were clear trends in the direction of the variation. Changes in the magnitude of effect sizes were evident in individual treatments, and mainly ranged between approximately neutral and highly negative values (Fig. 1). Figure 1. View largeDownload slide Effect sizes measured as Cohen's d showing the influence of nutrient availability, light intensity and substrate pH on the interactions between fungal root endophytes and the plants Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare. For each plot, gray lines with points represent the variation in effect sizes for individual fungal strains over the abiotic conditions tested, and the black line with points denote the median value for all strains within the experiment. Note that x-axes are represented as factors, and are not proportional to the variable values. Figure 1. View largeDownload slide Effect sizes measured as Cohen's d showing the influence of nutrient availability, light intensity and substrate pH on the interactions between fungal root endophytes and the plants Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare. For each plot, gray lines with points represent the variation in effect sizes for individual fungal strains over the abiotic conditions tested, and the black line with points denote the median value for all strains within the experiment. Note that x-axes are represented as factors, and are not proportional to the variable values. We used linear regression analyses to assess the influence of the abiotic factors on the interactions between plants and endophytes. An additional explanatory variable was included in analyses for experiments performed in different batches, to account for the data variation among independent assays. Power analyses revealed the capability of tests to detect effect sizes of 10.5–17% in assays with A. thaliana and M. erraticum and of 4.9–9.5% in those with H. vulgare, with a power of 95% (Supplementary Table S2). All experiments showed a significant variation across fungal treatments, and the same was true for all abiotic factors in all plant species with the exception of A. thaliana and pH (Table 2). Overall, there was a positive effect of nutrient availability on plant growth, with model estimates ranging from 0.02 ± 0.005 g (SE) to 0.3 ± 0.02 g. In comparison, the overall effects of light intensity and pH were close to neutrality. The interaction between the factors abiotic condition and fungal inoculation were only consistently significant across plant species in the nutrition experiment, indicating that nutrient availability can systematically affect plant–endophyte interactions (Table 2). In addition to these, growth of H. vulgare in response to endophytes also varied significantly with light availability and substrate pH (Table 1). It is possible, however, that model estimates are somewhat biased in experiments involving A. thaliana in pH and H. vulgare in nutrient and light availability conditions, where different experimental batches were included that had a significant impact in data variation (Table 2). Results obtained in experiments with H. vulgare were similar when using as response variables either total plant weight (Table 2) or shoot and root weights separately (Supplementary Table S3), although in the latter cases light intensity alone had little effect on both variables, and the effect of pH appeared to be strong in roots but not in shoots (Supplementary Table S3). Table 2. Summary statistics for the ANOVA of the effects of abiotic factors and fungal inoculation on plant fresh weight.     Nutrients  Light  pH  Plant species  Effect  df  F  P  df  F  P  df  F  P  A. thaliana  Abiotic factor  1,342  124.1  <0.001  1,203  5.9  0.016  1,358  0.7  0.35    Fungus  17,342  21.1  <0.001  15,203  10.9  <0.001  18,358  65.9  <0.001    Experimental batch  1,342  0.5  0.5  n.d.  n.d.  n.d.  1,358  1.3  <0.001    Abiotic factor × fungus  17,342  5.5  <0.001  15,203  0.7  0.8  18,358  1.3  0.21  M. erraticum  Abiotic factor  1,338  68.3  <0.001  1,208  27.5  <0.001  1,234  48.8  <0.001    Fungus  17,338  13.2  <0.001  15,208  5.6  <0.001  15,234  3.2  <0.001    Experimental batch  1,338  3.2  0.07  n.d.  n.d.  n.d.  n.d.  n.d.  n.d.    Abiotic factor × fungus  17,338  2.5  0.001  15,208  0.9  0.6  15,234  0.4  0.97  H. vulgare  Abiotic factor  1,777  89.5  <0.001  1,329  5.4  0.021  1,488  12.4  <0.001    Fungus  16,777  7.8  <0.001  10,329  10.8  <0.001  10,488  6.2  <0.001    Experimental batch  4,777  19.1  <0.001  1,329  22.2  <0.001  2,488  0.6  0.5    Abiotic factor × fungus  16,777  3.8  <0.001  10,329  2.3  0.013  10,488  8.1  <0.001      Nutrients  Light  pH  Plant species  Effect  df  F  P  df  F  P  df  F  P  A. thaliana  Abiotic factor  1,342  124.1  <0.001  1,203  5.9  0.016  1,358  0.7  0.35    Fungus  17,342  21.1  <0.001  15,203  10.9  <0.001  18,358  65.9  <0.001    Experimental batch  1,342  0.5  0.5  n.d.  n.d.  n.d.  1,358  1.3  <0.001    Abiotic factor × fungus  17,342  5.5  <0.001  15,203  0.7  0.8  18,358  1.3  0.21  M. erraticum  Abiotic factor  1,338  68.3  <0.001  1,208  27.5  <0.001  1,234  48.8  <0.001    Fungus  17,338  13.2  <0.001  15,208  5.6  <0.001  15,234  3.2  <0.001    Experimental batch  1,338  3.2  0.07  n.d.  n.d.  n.d.  n.d.  n.d.  n.d.    Abiotic factor × fungus  17,338  2.5  0.001  15,208  0.9  0.6  15,234  0.4  0.97  H. vulgare  Abiotic factor  1,777  89.5  <0.001  1,329  5.4  0.021  1,488  12.4  <0.001    Fungus  16,777  7.8  <0.001  10,329  10.8  <0.001  10,488  6.2  <0.001    Experimental batch  4,777  19.1  <0.001  1,329  22.2  <0.001  2,488  0.6  0.5    Abiotic factor × fungus  16,777  3.8  <0.001  10,329  2.3  0.013  10,488  8.1  <0.001  Significant values (P < 0.05) are shown in bold. n.d., not determined. View Large An assessment of the model coefficients for estimates of each variable supported observations based on effect sizes that most fungal strains negatively impact plant growth (Fig. 2, Supplementary Fig. S1). In this case, the effect of experimental batch was excluded by obtaining coefficients from models independently performed for treatments in each batch, and hence values solely represent the effect of strains as compared with uninoculated plant treatments. These analyses also confirmed the strong interaction between the variables nutrient availability and fungal strain, since they showed a wide variability across strains (Fig. 2, Supplementary Fig. S1). Comparatively, the significant interactions found between light availability or pH and fungal inoculation in H. vulgare showed little variation and hence we deemed them to be marginal (Fig. 2). A visual inspection of the interactions between the three plant species and the fungal strains with effects on plant growth that significantly varied with nutrient availability shows that treatments changing positively with this variable had a rather trivial magnitude and/or did not follow a steady pattern (Fig. 3, Supplementary Fig. S1). On the other hand, interactions varying negatively with nutrient availability seemed to be due to the increase in the gap between the fungus-inoculated and the uninoculated treatments due to fungal parasitism or pathogeny (Fig. 3). In these cases, the increase in plant growth with increasing nutrient availability does not occur in plants hosting fungi that severely compromise their development. We could not detect a tendency of strains within particular fungal lineages to trigger the same responses on plant growth across treatments, but there were individual strains that had a similar impact on plant growth irrespective of the host species or abiotic condition (e.g. Thanatephorus/Rhizoctonia sp. P1188 or Fusarium tricinctum P1304; see Supplementary Fig. S2 for results of all assays). We did not detect a correlation between the overall effect of strains on plant growth and the interaction of these effects with nutrient availability (Spearman's ρ = −0.14 to 0.22, P > 0.4). Figure 2. View largeDownload slide Effects of fungal inoculation on growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, light intensity and substrate pH, as extracted from linear fixed effects models. For each plant species and abiotic condition, data points on the left represent the effects of individual fungal strains respect to uninoculated control plants, whereas points on the right represent the interaction effects of the fungal strain and the abiotic variable. Solid points represent values significant at P < 0.05, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. Horizontal lines represent the median values for all data points in each condition. Additional information for this figure, including the correspondence between data points and fungal strains as well as 95% confidence intervals, is provided in Supplementary Fig. S1. Figure 2. View largeDownload slide Effects of fungal inoculation on growth of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, light intensity and substrate pH, as extracted from linear fixed effects models. For each plant species and abiotic condition, data points on the left represent the effects of individual fungal strains respect to uninoculated control plants, whereas points on the right represent the interaction effects of the fungal strain and the abiotic variable. Solid points represent values significant at P < 0.05, based on conditional F-tests with Kenward–Roger approximation for the degrees of freedom. Horizontal lines represent the median values for all data points in each condition. Additional information for this figure, including the correspondence between data points and fungal strains as well as 95% confidence intervals, is provided in Supplementary Fig. S1. Figure 3. View largeDownload slide Variation in the effect of fungal inoculation on total plant biomass of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, as compared with uninoculated controls. Interactions shown correspond to fungal strains with an effect on plant growth that was significantly affected by nutrient availability (see Fig. 2). Solid points with continuous lines represent mean weight values for uninoculated control plants, and open points with dashed lines represent values for fungus-inoculated plants. Error bars represent standard errors. Positive or negative symbols next to the strain names indicate the direction of the variation in the effect of fungi on plant growth with increasing nutrient availability, as obtained by linear models regression analysis. Figure 3. View largeDownload slide Variation in the effect of fungal inoculation on total plant biomass of Arabidopsis thaliana, Microthlaspi erraticum and Hordeum vulgare under varying conditions of nutrient availability, as compared with uninoculated controls. Interactions shown correspond to fungal strains with an effect on plant growth that was significantly affected by nutrient availability (see Fig. 2). Solid points with continuous lines represent mean weight values for uninoculated control plants, and open points with dashed lines represent values for fungus-inoculated plants. Error bars represent standard errors. Positive or negative symbols next to the strain names indicate the direction of the variation in the effect of fungi on plant growth with increasing nutrient availability, as obtained by linear models regression analysis. DISCUSSION We have tested the effect of abiotic conditions—including nutrient availability, light intensity and substrate pH—on the interactions between three different plant species and a diversity of root–endophytic fungi, comprising a variety of taxonomic lineages and geographical origins. Our results show that the effects of fungal root endophytes on the growth of their hosts mainly consist of reductions in plant biomass that are robust to changes in abiotic environmental conditions. No general trends were found in the variation of the magnitude or the direction of the fungal effects across plant species, ruling out a common response to specific abiotic factors by diverse root-colonizing fungi. When changes in the outcome of the symbioses were observed, these occurred in particular combinations of hosts and fungal strains, similarly to what has been found for the interaction between plants and pathogenic fungi (Laine 2007). The differential response of individual plant–fungus combinations is concordant with the frequently reported high variability in the outcome of fungal endophytic symbioses, even when phylogenetically related fungi are compared (Tellenbach, Grünig and Sieber 2011; Mayerhofer, Kernaghan and Harper 2012; Reininger and Sieber 2013; Kia et al.2017). Altogether, our findings suggest that the interplay between the genotypes of the plant host and the root-colonizing fungi is an important determinant of variability in plant growth and likely to affect the host's fitness, which could entail the selection of particular combinations of symbionts in locations under different environmental conditions, as proposed by the geographic mosaic theory of coevolution (Thompson 2005). As expected, both the different abiotic factors and the inoculation with fungal strains had independent effects on plant growth, as estimated by measurements of above-ground or total fresh biomass. In the case of fungal inoculation, the effects of most strains were negative, indicating a parasitism toward the host that is in agreement with previous studies based on in vitro assays of plant–endophyte interactions (Tellenbach, Grünig and Sieber 2011; Keim et al.2014; Mandyam and Jumpponen 2015; Kia et al.2017). The negative impact on plant growth is unsurprising given the trophic dependence of endophytic fungi on the host's resources and the simplicity of the co-cultivation system used, which lacked alternative sources of organic carbon to sustain fungal growth. It is noteworthy, however, that the effects in such conditions mainly consisted of reductions of biomass not accompanied by strong disease symptoms like wilting or chlorosis, and that apparently did not compromise plant survival, with few exceptions in strains related to well-known pathogens such as Thanatephorus/Rhizoctonia sp. or Fusarium sp. The weak plant-parasitic behavior of most root endophytes has already been described for the strains used in this work and others, in bioassays with the same three host plant species also used here (Kia et al.2017). Kia et al. (2017) speculated that the relatively small difference in growth between the endophyte-inoculated and the uninoculated plants in vitro may be readily overcome in natural conditions by changes in the fungal impact on plant growth in response to the environmental context. This is supported by empirical evidence that mutualistic interactions with endophytic fungi can develop in the presence of abiotic or biotic environmental stress (Faeth 2002; Maciá-Vicente et al.2008; Rodriguez et al.2008; Redman et al.2011; Hiruma et al.2016; Almario et al.2017), or that mycorrhizal symbioses tend to be more positive when the accompanying soil microbial communities are more complex (Hoeksema et al.2010). A similar rationale was adopted by Chamberlain, Bronstein and Rudgers (2014) to hypothesize that interactions with weak effect sizes, such as mutualism, are more prone to be context-dependent than other types of symbioses because their outcomes are likely to swing around a neutral effect. Nevertheless, Chamberlain, Bronstein and Rudgers (2014) did not find a strong difference in the context dependency of mutualistic interactions with respect to other types of interactions like competition. Likewise, the low degree of variability that we found in this study is surprising, and suggests that the physiological changes triggered by the tested factors on both the fungus and the plant did not affect the interaction between the two organisms. It must be noted that the ranges applied for each condition were not extreme and encompass magnitudes within the tolerance limit of either symbiont for the short duration of the experiment. It can be expected that use of more severe conditions, either by limitation or excess, would have resulted in a stronger impact in the symbiosis by surpassing the tolerance growth breadth of at least one symbiont. Of the three abiotic gradients tested, the change in nutrient availability had the strongest impact on plant–endophyte interactions. The content of nutrients in the substrate where host plants grow, especially of nitrogen and phosphorus, is well known to impact plant–microbe symbioses. For example, in arbuscular mycorrhizas, mutualistic interactions with a mycorrhizal fungus are preferentially established under phosphorus starvation conditions (Andreo-Jimenez et al.2015), although nitrogen content has also been shown to be relevant for a conducive mutualism (Hoeksema et al.2010). A similar dependency on nutrients has been reported in a few cases for symbioses with non-mycorrhizal fungal endophytes (Behie, Zelisko and Bidochka 2012; Hiruma et al.2016; Almario et al.2017). Newsham (2011) suggested that plant relationships with root endophytic fungi can become beneficial when organic nutrients are present in soil, owing to the ability of fungi to saprotrophically break down complex organic molecules and mobilize sequestered nutrients. In our experiments, no organic nutrients were present, nor did we observe a tendency of endophytes to enhance plant growth under limiting nutrients, which excludes the conditional translocation of particular compounds to the plant under starvation. It seems more likely that the differences in effect sizes triggered by some fungi in response to different nutrient concentrations are related to other kind of physiological changes in either symbiotic partner, such as modifications in the susceptibility to fungal infection in the plant or in the virulence of the fungus, as suggested by Laine (2007). In comparison with nutrient availability, light intensity and pH had little impact on the variability of the interactions. Several studies have reported light intensity as an important factor determining the sign of the interaction between plants and fungal endophytes, in which low light intensities seem to be conducive for more beneficial associations (Davitt, Stansberry and Rudgers 2010; Álvarez-Loayza et al.2011). However, these interactions have been described in leaves, where a strong exposure to light exists, which could drive physiological changes in the fungus important in determining pathogenicity, such as the build-up of reactive oxygen species (Egan et al.2007; Álvarez-Loayza et al.2011). In the case of pH, the differential effects on growth between shoots and roots of H. vulgare indicates that it affects more evidently below-ground tissues, in direct contact with the substratum. In fungi, whereas pH can influence mycelial growth, most fungi can sustain similar growth rates across broad pH ranges (Wheeler, Hurdman and Pitt 1991; Grum-Grzhimaylo et al.2015). The in vitro systems used in this study are artificial and fall short in representing the complex context in which natural plant–endophyte interactions occur. However, simplified systems such as the ones used here have proven adequate to reproduce both beneficial and detrimental interactions between root-colonizing fungi and host plants (e.g. Sesma and Osbourn 2004; Hiruma et al.2016; Almario et al.2017; Venneman et al.2017). Moreover, such systems are necessary to isolate the effects of the study factors from the many biotic and abiotic variables that may confound results in less managed set-ups (Jessup et al.2004). Surprisingly, in spite of such tight control over the non-target sources of variation—and possibly due to it as well—laboratory studies have been shown to yield the highest variability in the interaction outcomes in context-dependency studies, as compared with greenhouse or field studies (Chamberlain, Bronstein and Rudgers 2014). In natural conditions, potential variations in the effects of particular endophytes on the host due to changes in the abiotic environment may become diluted by the combined effects of co-occurring microorganisms and other abiotic and biotic variables. Given the relatively weak variation in the effects of endophytes observed here, it seems unlikely that context-dependent interactions between plants and root endophytes may result in significant changes in the host's fitness when both symbionts grow under conditions within their tolerance ranges. In conclusion, our results do not show an overall strong effect of nutrient availability, light intensity or substrate pH on the interactions between fungal root endophytes and plants, indicating that the outcome of these plant–fungus relationships may be robust to changes in the abiotic environment in the field. Variations in plant growth were observed in interactions between particular fungi and plant species, especially in response to nutrient availability, but these did not seem consistent across fungal lineages or plant species. These different outcomes may ultimately depend on the interplay between the genotypic characteristics of each symbiont, adding to the general variability observed in the interactions between different endophytes and plants (Tellenbach, Grünig and Sieber 2011; Kia et al.2017). Whereas the abiotic conditions studied here seem unlikely determinants of the outcome of plant–endophyte interactions in nature, further research is necessary to assess their importance by assaying more extreme condition and longer term interactions, as well as the implication of biotic factors such as microbe–microbe interactions within the root microbiome. SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. Acknowledgements We are grateful to Saatzucht Josef Breun GmbH & Co. KG for kindly providing seeds of barley cv. Barke. We also thank Xiaojuan Xia and Prof. Marco Thines for providing seeds of Microthlaspi erraticum. FUNDING This work was supported by LOEWE (Landes-Offensive zur Entwicklung Wissenschaftlich-ökonomischer Exzellenz) of the state of Hesse and was conducted within the framework of the Cluster for Integrative Fungal Research (IPF). Conflict of Interest. None declared. REFERENCES Agler MT, Ruhe J, Kroll S et al.   Microbial hub taxa link host and abiotic factors to plant microbiome variation. PLoS Biol  2016; 14: e1002352. Google Scholar CrossRef Search ADS PubMed  Ali T, Schmuker A, Runge F et al.   Morphology, phylogeny, and taxonomy of Microthlaspi (Brassicaceae: Coluteocarpeae) and related genera. Taxon  2016; 65: 79– 98. Google Scholar CrossRef Search ADS   Almario J, Jeena G, Wunder J et al.   Root-associated fungal microbiota of nonmycorrhizal Arabis alpina and its contribution to plant phosphorus nutrition. Proc Natl Acad Sci U S A  2017; 114: E9403– 12. Google Scholar CrossRef Search ADS PubMed  Álvarez-Loayza P, White JF Jr, Torres MS et al.   Light converts endosymbiotic fungus to pathogen, influencing seedling survival and niche-space filling of a common tropical tree, Iriartea deltoidea. PLoS One  2011; 6: e16386. Google Scholar CrossRef Search ADS PubMed  Andreo-Jimenez B, Ruyter-Spira C, Bouwmeester HJ et al.   Ecological relevance of strigolactones in nutrient uptake and other abiotic stresses, and in plant-microbe interactions below-ground. Plant Soil  2015; 394: 1– 19. Google Scholar CrossRef Search ADS   Behie SW, Zelisko PM, Bidochka MJ. Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science  2012; 336: 1576– 7. Google Scholar CrossRef Search ADS PubMed  Belesky DP, Fedders JM. Tall fescue development in response to Acremonium coenophialum and soil acidity. Crop Sci  1995; 35: 529– 33. Google Scholar CrossRef Search ADS   Bereau M, Barigah TS, Louisanna E et al.   Effects of endomycorrhizal development and light regimes on the growth of Dicorynia guianensis Amshoff seedlings. Ann For Sci  2000; 57: 725– 33. Google Scholar CrossRef Search ADS   Bronstein JL. Conditional outcomes in mutualistic interactions. Trends Ecol Evol  1994; 9: 214– 7. Google Scholar CrossRef Search ADS PubMed  Chamberlain SA, Bronstein JL, Rudgers JA. How context dependent are species interactions? Ecol Lett  2014; 17: 881– 90. Google Scholar CrossRef Search ADS PubMed  Champely S. pwr: Basic functions for power analysis. R Package Version 1.2-1. 2017, https://cran.r-project.org/web/packages/pwr/index.html (15 May 2017, date last accessed). Clay K. Endophytes as antagonists of plant pests. Microbial Ecology of Leaves . New York: Springer, 1991, 331– 57. Google Scholar CrossRef Search ADS   Cohen J. Statistical Power Analysis for the Behavioral Sciences , 2nd edn. Hillsdale, NJ: Lawrence Earlbaum Associates, 1988, 20– 26. Daskin JH, Alford RA. Context-dependent symbioses and their potential roles in wildlife diseases. Proc R Soc B  2012; 279: 1457– 65. Google Scholar CrossRef Search ADS PubMed  Davitt AJ, Chen C, Rudgers JA. Understanding context-dependency in plant–microbe symbiosis: The influence of abiotic and biotic contexts on host fitness and the rate of symbiont transmission. Environ Exp Bot  2011; 71: 137– 45. Google Scholar CrossRef Search ADS   Davitt AJ, Stansberry M, Rudgers JA. Do the costs and benefits of fungal endophyte symbiosis vary with light availability? New Phytol  2010; 188: 824– 34. Google Scholar CrossRef Search ADS PubMed  Ditzler C, Scheffe K, Monger H. Soil Survey Manual . Washington, DC: Government Printing Office, 2017. Egan MJ, Wang Z-Y, Jones MA et al.   Generation of reactive oxygen species by fungal NADPH oxidases is required for rice blast disease. Proc Natl Acad Sci U S A  2007; 104: 11772– 7. Google Scholar CrossRef Search ADS PubMed  Ewald PW. Transmission modes and evolution of the parasitism-mutualism continuum. Ann N Y Acad Sci  1987; 503: 295– 306. Google Scholar CrossRef Search ADS PubMed  Faeth SH. Are endophytic fungi defensive plant mutualists? Oikos  2002; 98: 25– 36. Google Scholar CrossRef Search ADS   Glynou K, Ali T, Buch A-K et al.   The local environment determines the assembly of root endophytic fungi at a continental scale. Environ Microbiol  2016; 18: 2418– 34. Google Scholar CrossRef Search ADS PubMed  Grum-Grzhimaylo AA, Georgieva ML, Bondarenko SA et al.   On the diversity of fungi from soda soils. Fungal Divers  2015: 1– 48. Heath KD, Tiffin P. Context dependence in the coevolution of plant and rhizobial mutualists. Proc R Soc Lond B Biol Sci  2007; 274: 1905– 12. Google Scholar CrossRef Search ADS   Hiruma K, Gerlach N, Sacristán S et al.   Root endophyte Colletotrichum tofieldiae confers plant fitness benefits that are phosphate status dependent. Cell  2016; 165: 464– 74. Google Scholar CrossRef Search ADS PubMed  Hoeksema JD, Chaudhary VB, Gehring CA et al.   A meta-analysis of context-dependency in plant response to inoculation with mycorrhizal fungi. Ecol Lett  2010; 13: 394– 407. Google Scholar CrossRef Search ADS PubMed  Jessup CM, Kassen R, Forde SE et al.   Big questions, small worlds: microbial model systems in ecology. Trends Ecol Evol  2004; 19: 189– 97. Google Scholar CrossRef Search ADS PubMed  Kannadan S, Rudgers JA. Endophyte symbiosis benefits a rare grass under low water availability. Funct Ecol  2008; 22: 706– 13. Google Scholar CrossRef Search ADS   Keim J, Mishra B, Sharma R et al.   Root-associated fungi of Arabidopsis thaliana and Microthlaspi perfoliatum. Fungal Divers  2014; 66: 99– 111. Google Scholar CrossRef Search ADS   Kia SH, Glynou K, Nau T et al.   Influence of phylogenetic conservatism and trait convergence on the interactions between fungal root endophytes and plants. ISME J  2017; 11: 777– 90. Google Scholar CrossRef Search ADS PubMed  Laine A-L. Pathogen fitness components and genotypes differ in their sensitivity to nutrient and temperature variation in a wild plant–pathogen association. J Evol Biol  2007; 20: 2371– 8. Google Scholar CrossRef Search ADS PubMed  Laitinen RK, Hellström KO, Wäli PR. Context-dependent outcomes of subarctic grass-endophyte symbiosis. Fungal Ecol  2016; 23: 66– 74. Google Scholar CrossRef Search ADS   Lüdecke D. sjPlot: Data visualization for statistics in social science. R Package Version 1. 2015, https://cran.r-project.org/web/packages/sjPlot/index.html (19 June 2017, date last accessed). Maciá-Vicente JG, Jansson H-B, Mendgen K et al.   Colonization of barley roots by endophytic fungi and their reduction of take-all caused by Gaeumannomyces graminis var. tritici. Can J Microbiol  2008; 54: 600– 9. Google Scholar CrossRef Search ADS PubMed  Mandyam KG, Jumpponen A. Mutualism–parasitism paradigm synthesized from results of root-endophyte models. Front Microbiol  2015; 5: 776. Google Scholar CrossRef Search ADS PubMed  Marx DH, Zak B. Effect of pH on mycorrhizal formation of slash pine in aseptic culture. For Sci  1965; 11: 66– 75. Mayerhofer MS, Kernaghan G, Harper KA. The effects of fungal root endophytes on plant growth: a meta-analysis. Mycorrhiza  2012; 23: 119– 28. Google Scholar CrossRef Search ADS PubMed  Murashige T, Skoog F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant  1962; 15: 473– 97. Google Scholar CrossRef Search ADS   Newsham KK. A meta-analysis of plant responses to dark septate root endophytes. New Phytol  2011; 190: 783– 93. Google Scholar CrossRef Search ADS PubMed  Piculell BJ, Hoeksema JD, Thompson JN. Interactions of biotic and abiotic environmental factors in an ectomycorrhizal symbiosis, and the potential for selection mosaics. BMC Biol  2008; 6: 23. Google Scholar CrossRef Search ADS PubMed  R Core Team. R: A Language and Environment for Statistical Computing . Vienna: R Foundation for Statistical Computing, 2016. Redman RS, Kim YO, Woodward CJDA et al.   Increased fitness of rice plants to abiotic stress via habitat adapted symbiosis: a strategy for mitigating impacts of climate change. PLoS One  2011; 6: e14823. Google Scholar CrossRef Search ADS PubMed  Redman RS, Sheehan KB, Stout RG et al.   Thermotolerance generated by plant/fungal symbiosis. Science  2002; 298: 1581. Google Scholar CrossRef Search ADS PubMed  Reininger V, Sieber TN. Mitigation of antagonistic effects on plant growth due to root co-colonization by dark septate endophytes and ectomycorrhiza. Environ Microbiol Rep  2013; 5: 892– 8. Google Scholar CrossRef Search ADS PubMed  Rodriguez R, Redman R. More than 400 million years of evolution and some plants still can’t make it on their own: plant stress tolerance via fungal symbiosis. J Exp Bot  2008; 59: 1109– 14. Google Scholar CrossRef Search ADS PubMed  Rodriguez RJ, Henson J, Van Volkenburgh E et al.   Stress tolerance in plants via habitat-adapted symbiosis. ISME J  2008; 2: 404– 16. Google Scholar CrossRef Search ADS PubMed  Rousk J, Bååth E, Brookes PC et al.   Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME J  2010; 4: 1340– 51. Google Scholar CrossRef Search ADS PubMed  Saikkonen K, Faeth SH, Helander M et al.   Fungal endophytes: A continuum of interactions with host plants. Annu Rev Ecol Syst  1998; 29: 319– 43. Google Scholar CrossRef Search ADS   Sesma A, Osbourn AE. The rice leaf blast pathogen undergoes developmental processes typical of root-infecting fungi. Nature  2004; 431: 582– 6. Google Scholar CrossRef Search ADS PubMed  Sieber TN, Grünig CR. Fungal root endophytes. In: Eshel, A, Beeckman, T (eds). Plant Roots: The Hidden Half , 4th edn. Boca Raton, FL: CRC Press, 2013, 38– 49. Smith SE, Read DJ. Mycorrhizal Symbiosis . Academic Press, 2010. Taylor DL, Hollingsworth TN, McFarland JW et al.   A first comprehensive census of fungi in soil reveals both hyperdiversity and fine-scale niche partitioning. Ecol Monogr  2014; 84: 3– 20. Google Scholar CrossRef Search ADS   Tellenbach C, Grünig CR, Sieber TN. Negative effects on survival and performance of Norway spruce seedlings colonized by dark septate root endophytes are primarily isolate-dependent. Environ Microbiol  2011; 13: 2508– 17. Google Scholar CrossRef Search ADS PubMed  Thompson JN. The Geographic Mosaic of Coevolution . Chicago: University of Chicago Press, 2005. Thrall PH, Hochberg ME, Burdon JJ et al.   Coevolution of symbiotic mutualists and parasites in a community context. Trends Ecol Evol  2007; 22: 120– 6. Google Scholar CrossRef Search ADS PubMed  Torchiano M. effsize: Efficient effect size computation (R package). 2014, https://cran.r-project.org/web/packages/effsize/index.html (12 November 2015, date last accessed). Venneman J, Audenaert K, Verwaeren J et al.   Congolese rhizospheric soils as a rich source of new plant growth-promoting endophytic Piriformospora isolates. Front Microbiol  2017; 8: 212. Google Scholar CrossRef Search ADS PubMed  Wang GM, Stribley DP, Tinker PB et al.   Effects of pH on arbuscular mycorrhiza I. Field observations on the long-term liming experiments at Rothamsted and Woburn. New Phytol  1993; 124: 465– 72. Google Scholar CrossRef Search ADS   West SA, Kiers ET, Pen I et al.   Sanctions and mutualism stability: when should less beneficial mutualists be tolerated? J Evol Biol  2002; 15: 830– 7. Google Scholar CrossRef Search ADS   Wheeler KA, Hurdman BF, Pitt JI. Influence of pH on the growth of some toxigenic species of Aspergillus, Penicillium and Fusarium. Int J Food Microbiol  1991; 12: 141– 9. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Microbiology EcologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial