The intensification of agriculture in recent decades has led to significant increases in yields, but has also contributed to biodiversity loss, land degradation, soil pollution, greenhouse gas emissions and water eutrophication. There is an urgent need for more sustainable methods of food production that use fewer agrochemicals (mineral fertilisers and pesticides). One promising solution is to promote native beneficial soil organisms that have the potential to act as “bio-fertilisers” and “bio-pesticides”.

Mycorrhizal fungi play a crucial role in this context. The word “Mycorrhiza” means “fungal root” and refers to one of the most ancient and widespread symbiosis between fungi and plants. Mycorrhizal fungi establish symbiotic relationships with 80% of plant species, including major crops like maize and wheat. Arbuscular mycorrhizal fungi (AMF) form extensive networks of hyphae (fungal filaments) in the soil through which they obtain essential nutrients. They colonise plant roots and form tree-like structures ("arbuscules") (Fig. 1.). The fungi provide nutrients (including phoshorus), and in exchange they obtain carbohydrates and fatty acids generated by the plant through photosynthesis. This vital function effectively positions them as "bio-fertilisers”.

In addition to improving plant nutrient uptake, AM fungi also help play pivotal roles in improving soil structure, retaining nutrients within the soil, mitigating greenhouse gas emissions, providing drought tolerance and conferring resistance to diseases resistance. This last role is referred as their "bio-pesticides" function. They are therefore very useful members of the natural soil microbiome and make an important contribution to soil health, plant nutrition and productivity.

The beneficial properties of AMF can be harnessed in two ways. First, native AMF communities can be supported through the adoption of favourable agricultural practices such as reducing tillage intensity, diversifying crops and organic farming. Second, AMF can be deliberately introduced into the soil, a strategy particularly valuable for restoring depleted soils characterized by low levels of native AMF. While greenhouse trials commonly yield positive effects, the outcome from field applications of these beneficial AM fungi is highly variable, ranging from beneficial to detrimental depending on the field. Consequently, the use of AM fungi to improve crop performance in agriculture is unpredictable.

Nevertheless, their large-scale implementation in agriculture hinges on one crucial factor: a dependable, consistent increase in yield. Only when the use of AM fungi can reliably and predictably boost crop yield will they truly become a significant contributor to more sustainable agriculture.

In order to determine the context-dependency of inoculation success, we conducted large-scale trials with AM fungi on 54 Swiss arable fields (comprising 864 individual plots) and investigated their effect on maize growth (Fig. 2). Maize plants were inoculated either with the AM fungus Rhizoglomus irregulare SAF22 (Swiss arbuscular fungi collection isolate #22) or with a control substrate without fungus (Fig. 2c) The success of the inoculation was determined by the mycorrhizal growth response (MGR), a measure of the effect of AMF inoculation on crop yield. This entailed comparing the biomass of plants inoculated with AM fungi to the biomass of control plants (mock-inoculated) for each field.

Remarkably, the mycorrhizal growth response was highly variable across the 54 fields, ranging from a decrease of 12% to an increase of 40% (Fig. 3). In one quarter of the fields, we observed significant growth improvements, with yields up to 40% higher. However, in the remaining fields, the growth response was either neutral or, in two fields, even significantly negative (-12%). This disparity underscores the potential concern for the farmers, as the benefits of AM fungi inoculation are only evident in one out of four fields.

To identify the factors that explain the underlying factors responsible for this wide-ranging variability in MGR, we carried out a detailed analysis of soil samples taken at the beginning of the growing season, just before planting the crops. Our primary objective was to find out the precise soil conditions conducive to the successful AM inoculation in the field. We measured 52 chemical, physical and biological soil properties and used state-of-the-art molecular genetic techniques to determine the composition of the native fungal community in the soil. In addition, we analyzed the composition of the fungal community in the roots at the time of harvest. This comprehensive investigation allowed us to predict inoculation success.

We employed various modelling approaches to identify the key soil properties that could best explain the variation in MGR. Whitin the soil properties, carbon and phosphorus levels and were negatively correlated with inoculation success while mineralised nitrogen and magnesium levels were positively correlated. Within the fungal microbiome, the abundance of fungal plant pathogens (e.g. Fusarium, Olpidium, Myrothecium) emerged as strong predictors of high yield gains from inoculation with AM fungi (33% of variation in MGR explained). Surprisingly, however, the composition of the native soil fungal microbiome had almost twice the influence on inoculation success (53%) than soil properties (29%). Overall, using a linear regression model, we were able to predict 86% of the variation in MGR: this means that we can successfully predict MGR, and therefore inoculation success, in 5 out of 6 fields.

We also investigated the root microbiome at the end of the season and found that in fields with high MGR, the inoculated AM fungus significantly suppressed pathogenic fungi in the roots (Fig. 4). The ability of AM fungi to protect plant roots from attack by soil-borne pathogens can be explained by several mechanisms, including enhanced uptake of plant nutrients and thus improved plant health, induced systemic resistance (plant immune response analogous to inoculation in humans), and alteration of the root microbiome. Our data on the root microbiome suggest, in part, direct competition for root colonisation.

In this project, we developed soil microbiome diagnostics as a valuable tool to improve the efficacy of mycorrhizal field inoculation. We identified the most important factors for predicting inoculation success. This, in turn, will empower farmers with reliable recommendations regarding the profitability of mycorrhizal inoculation.

Although our work was focused on a single maize variety, it represents the initial step in our journey. Further studies will include a wider range of crop varieties, soil types and climatic regions to broaden the scope of the work. Drawing an analogy from "personalised medicine", our objective is to establish the foundation for soil diagnostics to play a pivotal role in what is known as "precision farming". This concept involves tailoring agronomic practices to specific soil properties. Ultimately, the targeted use of AM fungi can become a reliable and cost-effective alternative to the conventional use of agrochemicals and thus contribute to more sustainable agriculture.