The development of sustainable technologies for crop management aiming to reduce the use of fertilizers and pesticides is an ongoing global challenge in agriculture (Pretty and Bharucha, 2018). In this scenario, the potential of beneficial symbiotic microbes to improve plant nutrition and stress resistance/tolerance is well established (Berendsen et al., 2012; Pieterse et al., 2014; Pozo et al, 2020). However, their performance under agronomic conditions is not optimal, as the outcomes of plant-microbe interactions are highly context-dependent and may vary according to the plant growing conditions (van der Heyde et al., 2017; Lee Díaz et al., 2021). Among the beneficial organisms used as inoculants in agriculture, fungi, and particularly arbuscular mycorrhizal fungi (AMF), are receiving increasing interest (Pozo et al., 2021). These soil-borne microorganisms, belonging to the phylum Glomeromycota, are widespread in natural and agricultural ecosystems and colonize the roots of more than 80% of terrestrial plant species, including major crops (Smith and Smith, 2011). This mutualistic association, known as mycorrhiza, has major benefits for both partners: AMF receive plant organic carbon in the form of carbohydrates and lipids (Keymer et al., 2017; Salmeron-Santiago et al, 2021). In return, AMF improve water and nutrients uptake by the plant, especially phosphorus (P) (Chen et al., 2018; Ferrol et al, 2019). In addition, they also increase plant phenotypic and metabolic plasticity to cope with biotic and abiotic stressors (Rivero et al., 2018; Sanchez-Bel et al, 2018; Campo et al, 2020; Mitra et al., 2021; Rivero et al., 2021; Pozo de la Hoz et al., 2021).

AM symbiosis usually renders the plant more resistant to certain soil-borne and aboveground pathogens and chewing herbivores, as shown in multiple systems (Jung et al., 2012; Song et al., 2013; Song et al., 2015; Selvaraj and Thangavel, 2021; Dowarah et al., 2022). This Mycorrhiza-Induced Resistance (MIR) is related to an improved ability of mycorrhizal plants to trigger defense responses upon challenge, a cost-efficient strategy known as defense priming (Pozo and Azcón-Aguilar, 2007; Martinez-Medina et al., 2016; Mauch-Mani et al., 2017). Priming is a common mechanism during induced systemic resistance triggered by beneficial microbes, consisting of a stronger and faster activation of plant defense mechanisms, usually dependent on jasmonate signaling (Pieterse et al., 2014; Mora-Romero et al., 2014; Gruden et al., 2020). Despite priming of plant defenses in response to pathogens and herbivores in mycorrhizal plants is well documented (Campos-Soriano et al., 2012; Song et al., 2013; Song et al., 2015; Sanchez-Bel et al., 2016; Fiorilli et al., 2018; Schoenherr et al., 2019; Sanmartín et al., 2020; Rivero et al., 2021; Manresa-Grao et al., 2022) consequences of AM symbiosis for insect herbivores are complex to predict. Indeed, the nutritional benefits of the symbiosis may increase food availability or improve the diet quality for the pest, depending on the fungal and insect identity and growing conditions (Koricheva et al., 2009; Frew and Price, 2019). Thus, the interaction outcome results from the interplay of opposite effects: nutritional improvement, benefiting the herbivore, and defense priming, potentially reducing herbivore performance (Pozo et al., 2020). Indeed, plant-microbe-insect interactions are very complex, with plants coordinating their responses according to multiple external and internal cues through a precise regulation orchestrated by phytohormone networks (Pozo et al., 2015; Gruden et al., 2020).

Many biotic and abiotic factors influence mycorrhizal colonization (Diagne et al., 2020; Mitra et al., 2021). Indeed, the plant-AMF interaction is finely regulated by several factors, including the plant and fungal genotypes and environmental conditions, with nutrients availability as one of the most influential aspects (Pozo et al., 2015; Lee Díaz et al., 2021). P is a central regulator of mycorrhizal establishment, root colonization and fungal growth within the host plant. Thus, the abuse of P fertilizers inhibits the establishment of mycorrhizal symbiosis in different plants (Smith and Smith, 2011; Smith et al., 2011; Higo et al., 2020). P, an essential nutrient for plants, is a non-renewable resource and poorly available in the field, being a limiting factor for plant growth (Wang et al., 2021). This element is taken up from the soil in the form of inorganic P by the roots (Rouached et al., 2010), and its deficiency triggers important changes in plant physiology and biochemistry including changes in plant growth, metabolism, hormonal balance, gene expression and root architecture. P deficiency further promotes the production of the phytohormones strigolactones, accumulation of anthocyanins and other phenolic compounds (Ha and Tran, 2014; Vysotskaya et al., 2016; Marro et al., 2022).

In agriculture, chemical and organic fertilizers are strongly and widely applied to solve P soil deficiency, causing contamination of aquifers and altering plant interactions with beneficial microbes, including AMF (Smith and Smith, 2011; Cordell and White, 2015). P levels also impact plant interactions with other organisms by altering plant tissues’ nutritional value for organisms feeding on them (pathogens and herbivores) and by modulating plant defenses (Breuillin et al, 2010; Castrillo et al, 2017; Chan et al., 2021). The role of P in plant immunity is currently under intense scrutiny, and the data reveal a very complex scenario (Chan et al., 2021; Qu et al., 2021; Val-Torregrosa et al, 2022). The phosphate starvation response in plants has been suggested to increase plant resistance to necrotrophic pathogens and leaf chewing herbivores in several plant species (Arabidopsis, tomato, tobacco) pointing to a positive cross-talk between the JA- and P- starvation signaling pathways that activates plant immunity (Khan et al., 2016). However, opposite results have been also found in different plant species facing different aggressors (Campo et al., 2020; Val-Torregrosa et al, 2022). It should be noted that most basic studies dealing with the molecular mechanisms mediating P effects on plant immunity compare very contrasting P levels, usually far from those used in agricultural settings.

Despite the increasing number of studies addressing P influence on plant defenses, only a few have investigated the relationship between P and MIR, suggesting a complex interaction with contrasting results (Wang et al., 2020; Qu et al., 2021). While evidence of the role of P in regulating mycorrhizal colonization, plant growth and immunity responses exist, the interplay among the different effects and mechanisms behind this interplay remain poorly studied (Wang et al., 2020; Qu et al., 2021). Our study aims to elucidate whether the mycorrhizal effect on tomato resistance to biotic stressors is regulated by P availability. For that, we evaluated the impact of different P fertilization regimes on tomato growth, mycorrhizal colonization and herbivore resistance using the tomato/Funneliformis mosseae system and the necrotrophic pathogen Botrytis cinerea and the generalist chewing herbivore Spodoptera exigua. We found that P availability strongly influences plant growth and resistance to biotic stresses, and that these effects are modulated by the AM symbiosis, showing that MIR is indeed dependent on P availability. By analyzing plant nutrients, defense-related phytohormones and transcriptional regulation of defenses in response to the damage-associated signals oligogalacturonides (OGs), we aimed to uncover the mechanistic basis of the impact of P availability on MIR. This knowledge will contribute to improve AMF application and management for sustainable crop protection.

Evidence of the major role of P in regulating plant immunity is increasing in recent years (Campos-Soriano et al., 2020; Chan et al., 2021; Val-Torregrosa et al., 2022). To identify key molecular regulators and their function, studies usually compare very extreme P levels, addressing null P to dissect P starvation responses (Dixon et al., 2020; Higo et al., 2020). However, even moderate differences in P concentration may have an impact on plant physiology, and severe P deficiencies are unlikely to occur in agricultural settings. In the present study, we assess how moderate differences in P fertilization, which are more likely to occur in agricultural conditions than severe ones, affect mycorrhizal colonization, growth, and resistance to S. exigua in tomato, using varying P levels that still allow a good plant development. Our results illustrate that while P regulates plant fitness and pest resistance, its effects depend on the mycorrhizal status of the plant. Moreover, they confirm that plant protection by mycorrhiza depends on P availability, highlighting that P fertilization is a key element in the context-dependency of MIR.

We show the dose-dependent effects of P on tomato growth and resistance against B. cinerea and S. exigua. Changes in most parameters were not linearly related to the P supply (0.3, 0.7 and 1.0 mM). Instead, P-related differences in most parameters (including plant biomass, anthocyanin content, mycorrhizal colonization, pathogen and herbivore performance on plant) were more pronounced when comparing 0.3 and 0.7 mM P, with no significant differences between 0.7 and 1.0 mM P -except for shoot fresh weight. Therefore, we selected 0.3 and 0.7 mM P as limiting and sufficient P conditions, respectively, to address the mechanisms that may underlie the differences observed in the plant resistance to the herbivore and in MIR under different P scenarios. We observed that the moderate P fertilization (0.7 mM) allowed mycorrhizal colonization while maintaining good plant development.

The inhibitory role of P on mycorrhizal colonization is well documented but varies with the plant and AMF species (Smith and Smith, 2011; Chen et al., 2014; Higo et al., 2020). Mycorrhizal colonization in tomato is indeed highly dependent on P levels. In our study, the medium and high P doses allowed mycorrhizal establishment, but prevented the progression of mycorrhizal colonization in time. However, although mycorrhizal colonization was around 10%, this was enough to significantly affect several parameters. Some growth promotion was observed at the moderate P level, and anthocyanin content, a good indicator of P starvation, were significantly reduced by mycorrhiza under these conditions. The relationship between the extension of mycorrhizal colonization in roots and the benefits of symbiosis has been discussed for years (Treseder and Kathleen, 2013). While higher colonization levels can lead to a stronger P uptake by the mycorrhizal pathway, benefits are not proportional to colonization, and even low colonization levels may have important benefits for the plant.

Regarding the attackers performance, a clear effect of P fertilization was observed. P levels affect both B. cinerea lesion development and S. exigua performance. Caterpillars performed worst on plants growing under low P, displaying very low weight and very high mortality compared to the other fertilization regimes. This is in agreement with P being an essential nutrient also for caterpillars (Perkins et al., 2004), as previously shown in other caterpillar-plant systems as Mamestra brassicae-Plantago lanceolata or Spodoptera littoralis-Nicotiana benthamiana (Khan et al., 2016; Qu et al., 2021). In our system, mycorrhization per se did not affect any of the parameters, but the data analyses revealed a significant interaction between the factors P and mycorrhiza (M). Remarkably, the impact of P levels on larval performance was significant in non-mycorrhizal plants, but not in mycorrhizal plants. A similar pattern was found in the interaction with the pathogen: lesion development was significantly impacted by the P levels in Nm plants, but not in Fm plants. Thus, the mycorrhizal symbiosis buffered the influence of P fertilization on the susceptibility to the attacker. Because of the limitations of performing herbivory tests in detached leaves, we performed more detailed analyses allowing the caterpillars to feed directly in the plant using the low (0.3 mM) and moderate (0.7 mM) P doses, which previously showed contrasting effects on plant and S. exigua performance. Using whole plants, not only local, but also systemic plant responses may contribute to the final outcome, as for example roots and distal leaves play a key role in response to damage and to herbivory (Kundu et al., 2018; Erb and Reymond, 2019; Gamir et al., 2020).

Under the lowest P level, there was no effect of the mycorrhizal symbiosis on larval mortality nor on their development to pupae. Noteworthy, caterpillar weight was higher for those feeding on mycorrhizal plants than for those on non-mycorrhizal ones, pointing to a potential benefit of the AM symbiosis at the nutritional level. In contrast, under 0.7 mM P, functional MIR was observed, as the symbiosis with F. mosseae reduced S. exigua survival, weight and the proportion of individuals reaching the pupa state. The prevalence of MIR at 0.7 but not at 0.3 mM P was also observed in the interaction with B. cinerea. Thus, the results show that mycorrhiza effects on plant resistance to pathogens and herbivores performance are P-dependent, and that MIR was impaired under P deficiency. A modulatory role of nitrogen deficiency on MIR against pathogens was also previously shown in tomato (Sanchez-Bel et al., 2016).

The effects of mycorrhization on plant interaction with herbivores are very complex. Enhanced resistance against chewing herbivores had been described in multiple systems (Schoenherr et al., 2019; Selvaraj et al., 2020; Jiang et al, 2021; Shafiei et al, 2022), but opposite effects have also been reported (Bennett et al., 2009; Hoffmann et al, 2009; Bernaola et al., 2018; Zeng et al., 2022), as the improved nutritional status of AM plants may benefit insects (Koricheva et al, 2009; Pozo et al., 2020). To understand the mechanisms underlying the contrasting effects of P availability on AM impact on insect performance, we investigated several plant nutritional and defensive traits that may contribute to the observed resistance phenotypes. Regarding plant nutrition and growth, mycorrhizal colonization enhanced shoot growth at low P doses and significantly elevated P levels in the plants regardless of the fertilization doses. C and N, recognized as key limiting-nutrients for insect herbivores (Mattson, 1980), were both influenced by P, but mycorrhiza did only change C levels under low P fertilization. The increase by mycorrhiza in C content under low P may have contributed to the better performance of the larvae in mycorrhizal plants under the lowest fertilization conditions. However, the contribution of changes in these nutrients to the worst performance of the herbivore in mycorrhizal plants under the 0.7 mM P is unlikely, since P is increased -and this would be positive for the pest- and C and N did not differ between Nm and Fm plants. Nevertheless, other nutritional aspects, such as the changes in other micronutrients, and potential variations in the content of sugars and amino acids upon herbivory may have contributed to the resistance phenotype and deserve further attention in follow-up studies. In fact, nutrition related changes, including protein and amino acid contents in mycorrhizal plants, have been proposed to contribute to resistance against S. exigua in maize plants (Ramírez-Serrano et al, 2022). In tomato, significant changes in sugar and amino acid contents occurs in response to herbivory in systemic tomato leaves may contribute to defense (Kundu et al., 2018). Whether this systemic regulation is modulated in mycorrhizal plants remain to be determined.

P effects can go beyond nutritional aspects. P starvation has been associated with an increase in JA signaling and expression of JA-related genes in different plant species, potentially leading to increased defense against pathogens and chewing herbivores in Arabidopsis (Khan et al., 2016; Luo et al., 2021). Although we found lower B. cinerea lesion development and higher mortality of S. exigua when feeding in plants under deficient (0.3 mM) P as compared to the medium 0.7 mM dose, no basal changes in the JA content among these plants were observed. Only the JA precursor OPDA differed among those plants, with higher levels at the higher P doses. Therefore, increased mortality at low P is not likely to be explained by a higher basal JA accumulation under such growing conditions, but rather by poor quality of the leaves (e.g. low nutrients and high anthocyanins) or by a stronger activation of defenses in those leaves as indicated by the transcriptional analysis of marker genes discussed below. The levels of IAA and ABA were also evaluated, as they are also regulated during responses to herbivory (Vos et al., 2013; Machado et al., 2016; Erb and Reymond, 2019). P fertilization impacted ABA, IAA and SA levels in the leaves, with ABA and SA concentrations being higher and IAA lower under low P conditions. Mycorrhizal colonization did not alter hormone basal levels except for a slight increase in auxins at P deficiency. Accordingly, basal differences in phytohormone levels among mycorrhizal and non-mycorrhizal plants are not likely to mediate the observed mycorrhizal effect on pathogen and larval performance under moderate P conditions.

We then addressed whether the differences may be related to a differential plant capacity to activate defenses. In order to analyze early defense responses and avoid variability associated with pathogen and caterpillar preference, we chemically elicited damage responses by application of OGs. OGs are released from the plant cell wall by pathogen polygalacturonases (Cervone et al., 1989; Ferrari et al., 2013), but also by endogenous polygalacturonases induced upon wounding, as shown in multiple plant species including tomato (Bergey et al., 1999; Orozco-Cárdenas and Ryan, 1999; De Lorenzo et al., 2011). As chewing herbivores elicit wound responses in the plant, they are also thought to trigger OGs release (Aljbory and Chen, 2018; Erb and Reymond, 2019) although this assumption needs further experimental evidences. Endogenous and exogenously applied OG elicitors activate plant defense responses and enhance resistance to different aggressors (De Lorenzo et al., 2011; Benedetti et al., 2015; Gamir et al., 2020). In the present study, OG application promoted JA, OPDA and IAA accumulation in leaves from plants under both, low and moderate, fertilization levels. Furthermore, OPDA and IAA levels were higher under the higher P fertilization, while JA was not affected by P availability. Although JA, OPDA and IAA increased their content under elicitation, no differences in their levels between Fm and Nm plants were found. We further analyzed well characterized JA regulated, anti-herbivory defense marker genes as those coding for LapA, Pin II, MC and TD (Uppalapati et al, 2005; Yan et al, 2013). All these marker genes showed a priming profile in mycorrhizal plants under 0.7 mM P, with a significantly higher expression upon OG treatment than in Nm plants. Conversely, under the most limiting P level, most of those genes showed much higher expression in non-elicited Nm plants, suggesting that Nm plants activate JA-related defenses upon P starvation in the absence of damage elicitation. It should be noted that although no differences were found between mycorrhizal and non-mycorrhizal plants elicited with OGs under the lowest P level (0.3 mM), there were differences when considering the fold change between non-elicited and elicited plants within each type of plants. While Nm plants showed a repression of defense-related genes upon OG treatment under P-deficient conditions, mycorrhizal Fm plants showed an increase in their expression. Thus, the main difference is the basal, non-elicited status of these plants: Nm plants are expressing JA-dependent defenses at the basal level, but fail to elicit defenses upon elicitation, in contrast to the behavior of mycorrhizal plants. The enhanced growth promotion of mycorrhizal plants under the lowest P fertilization, higher P content in leaves and lower anthocyanin levels suggest that mycorrhizal plants prioritize the improvement in nutrition under P deficiency, but prioritize defense activation under non-deficient conditions. Priming of plant defenses against herbivores has been shown in different systems (Song et al, 2013; Rivero et al, 2021; Jiang et al., 2021), but a failure of defense priming has been also reported (Qu et al., 2021; Ramírez-Serrano et al, 2022; Zeng et al., 2022). Analysis of chemical defenses, as the accumulation of polyphenols, lignins and other defensive metabolites in future studies will contribute to better characterize the differences in the defense responses under the different nutrition scenarios. Moreover, the potential contribution of nutrition to the different outcomes in those experiments is difficult to address, as the experimental systems were very different, but it is tempting to speculate that nutrient availability is a key player in defining the final resistance phenotype. The results of the present study support this hypothesis by evidencing contrasting defense-responses in mycorrhizal and non-mycorrhizal plants to elicitation depending on the P nutritional status.

It is well established that when subjected to multiple stresses, plants prioritize their responses to promote growth or defense to optimize their resources according to the specific needs in a given context (Smakowska et al., 2016). JA signaling has been described to be a key player in such prioritization, with negative regulators of JA signaling acting when defenses are not needed (Major et al., 2017; Guo et al., 2018). We explored the regulation of two JAZ transcriptional repressors characterized as JA-inducible in tomato leaves (Chini et al., 2017). We found that while no significant differences related to the mycorrhizal status of the plant were observed at the higher P level, JAZ expression was higher in mycorrhizal plants under P limitation, in agreement with the lower expression levels of the defense-related genes LapA, PinII, MC and TD. In contrast, upon OG elicitation the expression of JAZ genes was suppressed in mycorrhizal plants, likely releasing the repression of the JA signaling pathway to activate plant defenses. Finally, to test whether this role in mediating the defense-growth trade-off reported for JAZ may be leading to differences in the regulation of the primary and secondary metabolism in mycorrhizal plants, we analyzed the expression of carbohydrate-related genes. The genes encoding the sucrose synthase SUS1, the invertase LIN6 and the sucrose transporter SUT4 (Sanmartin et al, 2020) were all showing higher expression in mycorrhizal plants under P deficiency in the absence of challenge, matching the expression profile found for the JAZ genes. These higher expression levels were not maintained upon OG elicitation. Remarkably, no differences were found in these carbohydrate-related genes at moderate P levels. Therefore, mycorrhizal plants showed reduced expression of defense-related genes in the absence of challenge under P-limiting conditions, which correlated with higher expression levels of the genes coding for the JA negative regulators JAZ1 and JAZ2 and of genes related to C metabolism. Thus, under P deficiency, mycorrhizal symbiosis drives the plant response towards the activation of primary metabolism, that is, growth-related responses but this effect disappears upon a challenge or pest attack. On the contrary, when P levels are higher, no differences between mycorrhizal and non-mycorrhizal plants are found in the absence of challenge. However, upon damage signaling elicitation defense responses are primed in mycorrhizal plants, thus promoting defenses. Differential regulation of the growth-defense balance in mycorrhizal plants was also shown under N depleted environments (Sanchez-Bel et al., 2018). Thus, experimental evidences support a fine-tuned regulation of the growth-defense balance in mycorrhizal plants according to the nutrients availability.

In summary, our results show that P availability plays a regulatory role in JA-dependent plant defense responses, and the effect is modulated by the mycorrhizal status of the plant. We also show that P availability can be a key element in the reported context dependency of MIR, as mycorrhization effects ranged from moderately positive (increase in larval biomass under the lowest P level) to negative effects on disease and herbivore development (enhanced mortality and reduced biomass and individuals reaching pupa stage) under moderate P levels, thus displaying MIR. Our hormonal and transcriptional analyses point to a differential regulation of the growth-defense balance in mycorrhizal plants, where the symbiosis tailors the plant to prioritize primary metabolism and nutritional-growth related responses under P deficiency, but promotes priming of plant defenses when this key element is not limiting. This fine-tuned regulation of defenses may contribute to the enhanced resilience of mycorrhizal plants under varying conditions and multi-stress contexts. Understanding the underlying molecular mechanisms will contribute to improve the efficacy of mycorrhizal inoculants in crop protection.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

LD, JAL-R, and MP conceptualized the study. LD performed plant and insect bioassays, physiological, biochemical, and gene expression analyses. JR collaborated in the performance of the experiments and JG was responsible of the hormone analyses. LD and BR-S performed statistical analyses. All authors contributed to data interpretation and paper redaction. LD and MP wrote the manuscript. All authors contributed to the final version of the article and approved the submission.