Diaporthe strain EB4, originally isolated from roots of a Festuca rubra subsp. pruinosa plant (Pereira et al., 2019), was used in this study. This strain was selected in a previous screening where plants of Lolium perenne were inoculated with several Diaporthe strains and their growth evaluated in a greenhouse experiment (Supplementary Table S1).

The adscription of this strain within the Diaporthe genus was based on a multilocus phylogenetic analysis of the sequences of five genes (Gomes et al., 2013). The primers ITS4 and ITS5 were used to amplify the internal transcribed spacer (ITS) regions of the nuclear ribosomal genes (White et al., 1990). The primers EF1-728F/EF1-986R and CAL-228F/CAL-737R (Carbone and Kohn, 1999) were used to amplify part of the translation elongation factor 1α (TEF1) and the calmodulin (CAL) genes, respectively. The primers CYLH3F (Crous et al., 2004) and H3-1b (Glass and Donaldson, 1995) were used to amplify part of the histone H3 (HIS) gene, and the primers T1 (O’Donnell and Cigelnik, 1997) and Bt-2b (Glass and Donaldson, 1995) to amplify part of the β-tubulin gene (TUB). The amplification conditions were a start step of 2 min at 94°C, followed by 40 cycles of 30 s at 94°C, 1 min at 58°C for CAL, ITS and HIS, or 55°C for TEF-1 and TUB, and 1 min at 72°C, followed by a finishing step of 3 min at 72°C.

PCR amplicons were sequenced at the DNA sequencing service of the University of Salamanca (Spain). A concatenated sequence of the five loci (ITS, TUB, CAL, TEF1, HIS) was aligned with a set of 243 similar sequences belonging to 95 Diaporthe species obtained from Gomes et al. (2013). To determine the taxa closest to strain EB4, a maximum likelihood phylogenetic tree was made using MEGA v.7. After the clades close to strain EB4 were identified, new phylogenetic trees were made using the sequences of type strains of the species included in these clades. Diaporthella corylina was used as an outgroup, and the robustness of the tree topology was evaluated by 1,000 bootstrap replications.

To produce Diaporthe EB4 inoculum, 30 g of sugar beet pulp pellet mixed with 9.0 g CaCO₃, 4.5 g CaSO_4, and 60 ml of water were autoclaved in wide-mouth glass bottles for 30 min at 121°C (Vazquez de Aldana et al., 2020). Each bottle of beet pulp substrate was inoculated with four plugs of mycelium from a potato dextrose agar (PDA) culture and incubated at room temperature (20°C–22°C) for 4 weeks.

To determine the effect of inoculation with Diaporthe EB4 on tritordeum and perennial ryegrass under salinity, a greenhouse experiment was designed for each plant species with two factors: inoculation with Diaporthe (inoculated or non-inoculated) and salinity treatment (0 or 200 mM NaCl). The use of 200 mM NaCl was based on preliminary assays about the response of tritordeum to salinity.

Seeds of tritordeum cv. Aucan (Cecosa Semillas, Spain) were sown in 200 ml pots filled with a substrate composed of peat and perlite (1,1 v/v) previously treated at 80°C for 24 h in a forced air oven. For the inoculation treatments, tritordeum seeds were sown in the same substrate containing fungal inoculum in a proportion inoculum:substrate 1:7 (v/v). Several seeds were sown in each pot and cleared to one seedling after emergence. Each treatment was replicated in 12 pots. Plants subjected to the salinity treatment were watered with a 50 and 100 mM NaCl solution on the first and third day, respectively, to avoid salt shock, and with a 200 mM NaCl solution from day 5 onwards during 3 weeks. After this time, the plants were harvested, roots carefully cleaned, and lyophilized. Dry biomass was recorded and samples were used for chemical analyses.

A similar experiment was performed with perennial ryegrass cv. Tivoli (DLF, Denmark) with the only difference that four plants per pot were grown. Roots of perennial ryegrass, being very thin and challenging to clean, were not collected for biomass analysis.

The presence of Diaporthe in inoculated plants was diagnosed by direct isolation of the fungus from roots, and by microscopy. For this purpose, root samples of each inoculated tritordeum and perennial ryegrass plant were collected at harvest time. For isolation, a root sample was surface-disinfected and root pieces were placed in PDA plates following the method described by Pereira et al. (2019). For microscopy of fungal structures in roots, preparations of cleared (KOH) and non-cleared roots were stained with trypan blue, acid fuchsin, or aniline blue stain (Vierheilig et al., 2005).

The data sets were evaluated for the statistical assumptions of the ANOVA with the Shapiro–Wilk normality test and Brown–Forsythe equal variance test. Then, the effects of salinity and inoculation with Diaporthe on tritordeum and perennial ryegrass parameters were analyzed by two-way ANOVA. Differences between means were evaluated using Tukey’s test. Phytohormone data were analyzed with the Kruskal–Wallis test on ranks followed by Tukey’s test for multiple comparisons.

A preliminary phylogenetic analysis using 242 reference sequences from 95 Diaporthe species helped to identify the clades where strain EB4 was included. This allowed selecting a set of 13 Diaporthe species closely related to EB4, and a more restricted phylogenetic analysis was made using the sequences of type strains from these taxa (Supplementary Figure S1). In the maximum likelihood tree obtained, Diaporthe EB4 is most closely related to Diaporthe scleotioides, but according to the bootstrap values is not included within this taxon. Therefore, this strain appears to belong to a yet undescribed Diaporthe species.

The in vitro evaluation of enzymatic and metabolic activities of Diaporthe EB4 cultures revealed some characteristics that could be associated with plant growth promotion (Table 1). Diaporthe EB4 produced ammonium in peptone water, and IAA in potato dextrose broth. On agar-supplemented media, Diaporthe produced siderophores and extracellular enzymes, such as amylase, cellulase, and protease, but was unable to solubilize phosphate.

The isolation of Diaporthe from roots of inoculated plants was rare, occurring in less than one-fourth of the inoculated plants of each species, and in those cases only one or two root fragments produced Diaporthe isolates. No fungal structures were consistently observed by microscopy in roots of inoculated plants. Therefore, seems that the association of Diaporthe with tritordeum might be rhizospheric, and not endophytic.

No visual symptoms of disease were observed in roots or leaves of plants inoculated with Diaporthe and/or subjected to salinity. This indicates that the strain EB4 is not pathogenic to tritordeum or perennial ryegrass.

Both inoculation with Diaporthe and salinity significantly affected the growth of tritordeum and perennial ryegrass, however, the interaction of both factors was not significant (Figure 1; Supplementary Tables S2, S3). The leaf biomass of both species increased in inoculated plants regardless of salinity. Compared to non-inoculated plants, Diaporthe increased leaf biomass of tritordeum by 84% at 0 mM NaCl, and by 131% at 200 mM NaCl (Figure 1). Similarly, Diaporthe inoculation increased leaf biomass of perennial ryegrass by 29% and 39% at 0 mM and 200 mM NaCl, respectively. The root biomass of tritordeum also increased in inoculated plants compared to non-inoculated at both salinity treatments. Compared to the 0 mM NaCl treatment, 200 mM NaCl significantly reduced the leaf biomass of tritordeum and perennial ryegrass by 20% and 42%, respectively, and the root biomass of tritordeum by 39% (Figure 1).

The comparison of both plant species showed that the relative increase in leaf growth due to Diaporthe was greater in tritordeum than in perennial ryegrass, and the decrease in growth due to salinity was greater in perennial ryegrass than in tritordeum.

Our results showed that Diaporthe EB4 increased growth of tritordeum and perennial ryegrass in terms of leaf and root biomass, modifies the nutrient balance, and alter the level of some phytohormones, regardless of salinity treatment. Increased plant growth mediated by fungal symbiosis could occur because of improved plant nutrition, the production of phytohormones involved in growth regulation, or to a combination of both effects (Rai et al., 2014; Chanclud and Morel, 2016; Begum et al., 2019; Sarkar et al., 2021).

Some fungal root endophytes are reported to play an important role in the recycling and mineralization of organic nutrient pools in poor soils, like those found in Antarctica (Upson et al., 2009; Oses-Pedraza et al., 2020). Similarly, Festuca rubra subsp. pruinosa grows in a nutrient-poor rock substrate in sea cliffs. In such habitat, the nutrients found in dead roots and other plant debris could be recycled for further plant use by components of the root microbiome. Characteristics of Diaporthe EB4 cultures like the production of ammonium, siderophores, and extracellular enzymatic activities like cellulase, amylase, and protease, could support a nutrient cycling role similar to that of Antarctic endophytes (Hill et al., 2019; Oses-Pedraza et al., 2020). Ammonium is a preferential nitrogen source for plants, and Diaporthe EB4 cultures produced this compound in an assay where peptone, composed of peptides and amino acids, was used as a nitrogen source. In addition, a gelatin hydrolysis test confirmed the extracellular protease activity of this fungal strain. Therefore, Diaporthe EB4, like some Antarctic endophytes, can degrade protein to amino acids and ammonium, readily available plant nutrients, thus favoring plant growth (Upson et al., 2009; Hill et al., 2019). The increase in total nitrogen detected in leaves and roots of inoculated plants would support this hypothesis. Other hydrolytic enzymes produced by Diaporthe, like cellulase and amylase may also improve nutrient availability via the degradation of the plant cell walls from dead plant tissues and other organic constituents of the soil (van den Brink and de Vries, 2011). The enzymatic degradation of cellulose and other cell wall polysaccharides can produce glucose and other monosaccharides, which can be absorbed by plant roots, as well as by rhizospheric microorganisms (Yamada et al., 2011). The saprobic capability of Diaporthe EB4 further supports its possible function in nutrient cycling. This strain grew profusely in the beet pulp medium used to prepare its inoculum, and another Diaporthe strain showed the greatest saprobic capability among several root endophytes grown in this polysaccharide-rich medium (Vazquez de Aldana et al., 2020). Therefore, degradation of organic compounds as a result of the saprobic growth of Diaporthe on plant remains could improve the availability of plant nutrients in the rhizosphere.

Another beneficial effect for plants could be derived from the production of fungal siderophores, chelator agents with specificity for binding insoluble Fe³⁺ (Ahmed and Holmström, 2014). Diaporthe EB4 cultures produced siderophores, and this might be related to the significantly greater accumulation of Fe observed in roots of inoculated plants, regardless of salinity treatment. Fe is essential for chlorophyll biosynthesis and photosynthesis (Ahmed and Holmström, 2014). The greater Fe accumulation in roots than in leaves could occur because plants acquire Fe from the rhizosphere through the root system, but the translocation to leaves is regulated by the response to a Fe deficiency signal, as reported in barley (Tsukamoto et al., 2009). Other essential nutrients like Mg, Ca, and Zn also increased in response to Diaporthe inoculation.

There is no strong evidence for Diaporthe EB4 being an endophyte in tritordeum or ryegrass. Neither the fungus was consistently isolated, nor fungal structures were observed in roots. Therefore, it is possible that the beneficial activity observed in inoculated plants is derived from an epiphytic association, where the fungus is limited to the rhizospheric space. In contrast, Diaporthe strains were frequently isolated from roots of Festuca rubra subsp. pruinosa, the original host of the EB4 strain (Pereira et al., 2019). Many Diaporthe species are host generalists (Gomes et al., 2013), and the capability of EB4 to infect root tissues might be greater in its original Festuca host than in our agricultural grasses. Nevertheless, in a way similar to what we have observed here, some dark septate endophytes enhance plant growth without clear evidence of root colonization (Newsham, 2011; Yakti et al., 2018).

Coupled with increased nutrient availability, the production of IAA by Diaporthe itself, as well as the increased content of this hormone observed in inoculated plants, could explain the increased biomass observed. Similarly, it has been suggested that the growth promotion in rice stimulated by Phomopsis (=Diaporthe) may be due to its high IAA production capacity (Chithra et al., 2017). A recent investigation reveal that the inoculation of common bean plants with endophytic fungi surpassed the effect of the exogenously applied hormones (IAA), which resulted in the improvement of plant biomass, photosynthetic pigments, carbohydrate and protein contents (Ismail et al., 2021). IAA is involved in vascular development and cell elongation, enhancing plant growth and inducing lateral root formation (Chanclud and Morel, 2016; Egamberdieva et al., 2017). The stimulation of lateral roots improves nutrient uptake, and consequently increases plant growth. Auxins are also involved in plant–fungal symbiotic interactions, for instance, they are required for the colonization of mycorrhizal fungi (Chanclud and Morel, 2016).

Soil salinity inhibits plant growth for two reasons: first, a high concentration of salt reduces the water uptake capability of plants (osmotic stress), and second, an excessive salt concentration within the plant can be toxic (ionic stress; Munns and Tester, 2008). Plants have a rapid response to an increase in external osmotic pressure, and a slower response to the internal Na⁺ accumulation. In our experiments, salinity reduced tritordeum and perennial ryegrass growth. However, Diaporthe helped plants to overcome the stress, improving biomass accumulation and altering related physiological parameters, such as proline and gibberellins.

As expected, proline increased in both plant species in response to salinity, but the concentration of this osmolyte further increased due to Diaporthe. Therefore, although inoculated plants showed a greater Na⁺ leaf uptake than their non-inoculated counterparts, increased proline synthesis might compensate the deleterious effect of this ion. Due to the Na⁺ sequestration in cell vacuoles, an osmotic adjustment of the cytoplasm is needed to maintain cell turgor, and this is supposedly achieved by the active accumulation of organic solutes (e.g., proline and sugars) and inorganic ions (mainly K⁺ and Cl⁻; Munns and Tester, 2008; Zhao et al., 2020). In addition to osmotic adjustment, proline has multiple functions stabilizing the structure of proteins and scavenging ROS (Acosta-Motos et al., 2017). In accordance with our results, other endophytes have been shown to enhance proline synthesis in plants, improving salt stress tolerance (Zhang et al., 2016; Bouzouina et al., 2021; Mahgoub et al., 2021). In contrast, in Festuca rubra subsp. pruinosa, the original host of Diaporthe EB4, proline synthesis in response to salinity was independent of the symbiosis with three different fungal endophytes, Epichloë festucae, Periconia macrospinosa, and Fusarium oxysporum (Pereira et al., 2021). The fact that this is a halophytic grass might be related to this.

A high Na⁺ content in the soil can induce an iron deficiency or imbalance due to competition for the uptake of nutrients, such as K⁺ and Ca²⁺, thereby affecting plant physiological parameters (Hu and Schmidhalter, 2005; Evelin et al., 2019). Na⁺ accumulation in the plant due to salinity did not decrease the K⁺ content, which increased in tritordeum and did not change significantly in perennial ryegrass. However, Diaporthe had a negative effect on K⁺ content in both plant species, and thus the Na⁺/K⁺ ratio increased in inoculated plants. Fungi, such as Aspergillus aculeatus, Piriformospora indica, or Fusarium, are able to increase K⁺ in plants under salinity improving the Na⁺/K⁺ homoeostasis maintaining a low ratio (Xie et al., 2017; Ghorbani et al., 2019). Other Festuca rubra endophytes also have a positive effect on the adaptation to salinity of their host plants. For instance, Periconia macrospinosa has similar effects in its host plant to the ones here observed with Diaporthe in tritordeum and ryegrass under salinity: Festuca rubra inoculated with Periconia showed increased biomass and Na⁺ content, and decreased K⁺ content respect to uninoculated plants (Pereira et al., 2021). Fusarium oxysporum, the most abundant component of the Festuca root microbiome, promoted the growth of its host plant under salinity, but in contrast to Periconia and Diaporthe, the Na⁺ content was significantly reduced in inoculated plants, and K⁺ remained unchanged. This suggests that Na⁺, K⁺, and their ratio, can be altered in plants experiencing symbiotic growth promotion.

Ca²⁺ availability could be seriously reduced under salinity because Na⁺ readily displaces Ca²⁺ from its extracellular binding sites. This could be detrimental for plant growth because Ca²⁺ plays an essential role in the structure and function of plant membranes and in proline metabolism (Hu and Schmidhalter, 2005; Iqbal et al., 2014). However, we found that under salinity Ca²⁺ increased slightly in tritordeum leaves, but its accumulation in roots greatly increased in inoculated plants (under saline and non-saline conditions). Thus, a Diaporthe mediated improvement in Ca²⁺ uptake that accumulates in roots, might improve the ionic homeostasis, and be related to cellular stability and the increased proline accumulation and salinity tolerance observed. Exogenous application of CaCl₂ has been found to increase proline content, thus mitigating salt stress (Khan et al., 2010). In addition to Ca²⁺, Diaporthe also improved Mg²⁺ in roots (regardless of salinity) and Zn in roots of plants under salinity. These nutrients are important in plants, regulating various metabolic pathways involved in abiotic stress tolerance (Khan et al., 2018). The fact that those nutrients accumulate in roots and are not mobilized to shoots could be due to an adequate concentration in shoots.

Phytohormones play an important role in plant tolerance to stress by modulating the physiological responses and defense systems of the plant. ABA synthesis is one of the fastest responses to abiotic stress factors, causing stomatal closure, thereby reducing water loss via transpiration (Chaves et al., 2009; Chanclud and Morel, 2016). As expected, ABA increased in tritordeum exposed to salinity, and Diaporthe enhanced this response. Similarly, GAs increased due to Diaporthe under salinity, and this can help plant growth maintenance during stress conditions. In fact, exogenous application of GA to plants mitigated saline stress, and the same effect was observed when fungal endophytes producing GA were inoculated (Tuna et al., 2008; Khan et al., 2012; Hamayun et al., 2017). Among other effects, GAs can influence ion uptake and proline metabolism (Iqbal et al., 2014; Al-Harthi et al., 2021). We did not analyze if Diaporthe EB4 can produce GAs in vitro, but this occurs in fungi, such as Aspergillus, Fusarium, Porostereum spadiceum, or Paecilomyces formosus (Khan et al., 2012; Waqas et al., 2012; Hamayun et al., 2017; Bilal et al., 2018). SA is involved in defense signaling against biotrophic pathogens (Pieterse et al., 2014), but also improves plant stress tolerance through modulation of antioxidative enzymes and increasing the proline metabolism (Misra and Saxena, 2009; Khan et al., 2015). We found that in response to salinity SA decreased in non-inoculated plants but it was maintained at non-stress levels in Diaporthe-inoculated plants. Although SA is involved in the modulation of antioxidative enzymes, the increase in SA mediated by Diaporthe under salinity was not reflected in the total antioxidant capacity of tritordeum plants, which decreased under salinity. However, this could be related to the increase in proline, which also has a role in scavenging reactive oxygen species (Iqbal et al., 2014). Association with fungal endophytes has been reported to enhance the effectiveness of the plant host antioxidative system by manipulating the production or activity of antioxidants under stress (White and Torres, 2010).

In conclusion, our study shows that a Diaporthe strain isolated from roots of an halophytic grass promotes the growth of two agricultural grasses, tritordeum and perennial ryegrass, by increasing nutrient uptake and the production of indole 3-acetic acid. Improved growth and nutrient uptake might be due to the fact that Diaporthe produces several extracellular enzymes capable of degrading organic nutrient pools, and siderophores that improve iron uptake. In connection with improved nutrient availability, Diaporthe increased the IAA content in the plant, which is related to root growth promotion. Under salinity, Diaporthe EB4 ameliorated stress through accumulation of GA and IAA in plants, increased proline synthesis, and nutrient uptake in roots, which subsequently led to improved tritordeum and perennial ryegrass growth. These findings might provide a foundation for further research about this possibly new Diaporthe species from a halophytic grass as a natural plant growth-promoting agent in agricultural grasses.