Beneficial soil-borne bacteria and fungi: a promising way to improve plant nitrogen acquisition

An updated review of how plant-associated bacterial diazotrophs and fungal endophytes may improve plant N nutrition, with an emphasis on the underlying mechanisms and their exploitation for reducing the use of chemically synthesized N fertilizers.


Introduction
After the mid-20th century, world agriculture benefited from unprecedented changes in agronomic practices known as the "Green Revolution". Consequently, over a 50-year period, there was a 100% increase in the yield of main crops per capita, particularly in some regions of the world such as Asia and South America (Lassaletta et al., 2016;Pretty, 2018). This increase was mainly due to the technological and scientific advances over this period: new crop varieties were bred, inorganic fertilizers and chemically synthesized pesticides and herbicides were developed and extensively used, and their application was greatly facilitated by the modernization of agricultural machinery (Lassaletta et al., 2016;Pretty, 2018). In particular, the use of synthetic inorganic nitrogen (N) fertilizers has increased several fold during the past 50 years (Lassaletta et al., 2016;Pretty, 2018). N fertilizers chemically produced via the industrial Haber-Bosch process for agricultural purposes use 1-2% of the world's fossil fuel energy output (Chen et al., 2018a). However, because crops do not take up more than 30-50% of the N available in the soil , the extensive use of N fertilizers has caused major detriments to microbial, animal, and plant biodiversity and to the corresponding nonagricultural ecosystems (Schlesinger, 2009;Withers et al., 2014).
It is now possible to conduct research on the nature and composition of plant microbiota with next-generation sequencing (NGS) techniques. NGS has led to the discovery of the huge diversity of microbe communities present inside (endophytic communities) and on the surface of (epiphytic communities) plant organs, as well as in the rhizosphere (Bulgarelli et al., 2013;Reinhold-Hurek et al., 2015;Cordovez et al., 2019). These microbes play several roles during plant development and are precious allies for crop production and adaptation to various biotic and abiotic stresses (Hardoim et al., 2015;Tkacz and Poole, 2015;Liu et al., 2017). In particular, interactions with rhizospheric and endophytic microorganisms can enhance plant mineral nutrition (Jacoby et al., 2017). For instance, phosphate can be solubilized by organic acids or phytases secreted by soil-borne bacteria or fungi, thus favoring its uptake by plants (Sharma et al., 2013;Alori et al., 2017). Siderophores are low-molecular-weight molecules of microbial origin with a high affinity for iron. They are primarily released to solubilize iron for microbial needs, but consequently make this element more accessible to plants (Lemanceau et al., 2009;Ahmed and Holmström, 2014). Regarding N, microbes play major roles in plant mineral nutrition; these roles are diverse, complex and not fully characterized, particularly during the recycling of inorganic N to mineral or gaseous forms (Coskun et al., 2017). Plants can directly take up the different forms of N, such as nitrate (NO 3 -), ammonium (NH 4 + ), and amino acids, by means of specific transporters (Krapp et al., 2014;Fan et al., 2017;Tegeder and Masclaux-Daubresse, 2018;Wang et al., 2018;Hirel and Krapp, 2020). Indirect uptake of N can also occur following the establishment of beneficial associations with microbes that facilitate plant N acquisition (Santi et al., 2013;Udvardi and Poole, 2013;Courty et al., 2015;Chen et al., 2018b).
In this review, we provide an overview of the major steps involved in N acquisition by plants, with a particular emphasis on the contribution of beneficial microorganisms that extract N from the soil or fix dinitrogen (N 2 ) from the atmosphere before its transfer to the plant. Possible synergistic interactions between these microorganisms is also discussed, as it may offer an avenue for the improvement of plant productivity through the enhancement of N acquisition (Giovannini et al., 2020).

Nitrogen uptake and assimilation in plants
N is present in the soil in the form of NO 3 -, NH 4 + , or amino acids, with the availability of the various forms depending upon physical factors such as pH and temperature. Plants such as rice, which are adapted to acidic pH and anaerobic conditions, preferably take up NH 4 + (Tabuchi et al., 2007). In contrast, plants that are adapted to more alkaline pH in aerobic soils, as occurs in most arable lands, use mostly NO 3 as their N source (Hirel et al., 2007;Masclaux-Daubresse et al., 2010;O'Brien et al., 2016;Xu, 2018). NO 3 is taken up by the roots and then transported in the plant by means of low-and high-affinity transporters that are, for the most part, located in the plasma membrane of cells (Krapp et al., 2014;Léran et al., 2014;O'Brien et al., 2016;Kant, 2018;Wang et al., 2018;Zhang et al., 2018). Members of the nitrate transporter 1/ peptide transporter family (NRT1/NPF) act as low-affinity NO 3 transporters (Fan et al., 2017;Wang et al., 2018). There are two exceptions in the NPF family that have been identified so far: the Arabidopsis NRTNPF6.3/NRT1.1 and the rice NRT1.1B, which display both low and high affinity for NO 3 - (Ho et al., 2009;Bouguyon et al., 2015). High-affinity NO 3 transport systems involve transporters belonging to the NRT2 family, which belongs to the major facilitator superfamily (MFS) (Krapp et al., 2014;Léran et al., 2014;O'Brien et al., 2016;Kant, 2018;Wang et al., 2018;Zhang et al., 2018). Two other families of transporters are involved in intracellular NO 3 transport: the chloride channel (CLC) family and the slow-type anion channels (SLAC) (Zifarelli and Pusch, 2010;Krapp et al., 2014;Hedrich and Geiger, 2017;Wang et al., 2018).
Following uptake, NO 3 is reduced to nitrite (NO 2 -) by the enzyme nitrate reductase (NR) located in the cytosol. Then, NO 2 is further reduced to NH 4 + by a plastidic nitrite reductase. The resulting NH 4 + is used to synthesize glutamine using glutamate as a substrate (Masclaux-Daubresse et al., 2010;Wang et al., 2018). NH 4 + can also be taken up by plants directly from the soil. However, its availability mainly depends on the soil characteristics (Nieder et al., 2011). Ammonium can be present in its neutral form (ammonia; NH 3 ) or its cationic form (NH 4 + ) depending on the pH of the soil or growth medium; the latter is predominant under most environmental or growth conditions (Ludewig et al., 2007). When used as the sole N source, NH 4 + generally induces physiological and growth perturbations in most plant species (Marino and Moran, 2019). Nevertheless, NH 4 + can be directly taken up by means of high-affinity NH 4 + transporters (AMT) and nonsaturable low-affinity transporters (aquaporin or cation channels) (Tegeder and Masclaux-Daubresse, 2018). NH 4 + transport is mediated by proteins belonging to the AMMONIUM TRANSPORTER/METHYLAMMONIUM PERMEASE RHESUS (AMT/MEP/Rh) family, located in the root cell plasma membrane. These proteins allow the transport of NH 4 + to the different parts of the plant (Ludewig et al., 2007;Yuan et al., 2007;Tegeder and Masclaux-Daubresse, 2018). NH 4 + derived from NO 3 reduction or from direct uptake is finally incorporated into organic molecules via the combined activity of two enzymes, glutamine synthase (GS) and glutamate synthase (GOGAT) (Masclaux-Daubresse et al., 2010;Krapp et al., 2014;Wang et al., 2018;Hirel and Krapp, 2020).
The processes associated with the metabolization and recycling of N are referred to as assimilation and remobilization, respectively. N transport, assimilation, and remobilization contribute to plant NUE (Hirel et al., 2007;Lea and Miflin, 2011), of which the two main components are N uptake efficiency (NupE) and N utilization efficiency (NutE) (Han et al., 2015;Li et al., 2017). This implies that both NupE and NutE need to be considered individually or together to improve NUE, which largely depends on the species, the genotype, and the environmental conditions. Intensive research has thus been conducted over the past three decades to find ways to improve NUE, mostly using transgenic plants or mutants altered for the expression of genes involved in N transport and assimilation (Han et al., 2015;Li et al., 2017;Wang et al., 2018). However, these approaches generally have not led to marked and reproducible positive effects on plant growth and development. Moreover, the positive effects that have been achieved have rarely been confirmed, particularly when plants are grown under agronomic conditions (Plett et al., 2017, Hirel andKrapp, 2020;Cañas et al., 2020). Due to the limited success of such approaches in terms of agronomic applications, as well as the restrictions regarding the use of genetically modified plants in many countries and the growing importance of sustainable agriculture practices, research has been reoriented to evaluate the impact of beneficial microbes on plant N acquisition (Han et al., 2015;Tao et al., 2019). In line with these changes in research strategies, reports suggest that plant NUE can be a key determinant favoring the establishment of beneficial microbes in the rhizosphere of rice (Zhang et al., 2019). Using an elegant genetic and metagenomics approach, Zhang et al. (2019) showed that the rice NRT1.1b nitrate transceptor is associated with the recruitment of bacterial taxa that display enzymatic functions with high N-cycling potential. Thus, the use of beneficial microbes appears to be an attractive alternative to genetic engineering for the implementation of new breeding strategies, although they require several years of development before they can be validated and marketed.

Plant nitrogen nutrition mediated by beneficial microbes
Microbes play essential roles in the conversion of different forms of N. These different forms of N can be used by plants following chemical reactions carried out by living organisms, and contribute to the N cycle that occurs between the soil and the atmosphere. These reactions involve bacteria, archaea, and fungi. First, atmospheric N 2 enters the N cycle following its reduction to NH 4 + , a reaction catalyzed by the enzyme nitrogenase (Nase). Nase is present in N 2 -fixing symbiotic and associative symbiotic microorganisms called diazotrophs, which are mostly represented by bacteria and archaea (Coskun et al., 2017;Lehnert et al., 2018a). An estimated 50% of land N originates from biological N 2 fixation (agricultural and natural) that occurs mostly in the Rhizobium-legume symbiosis and generates inorganic N mainly in the form of NH 4 + (Fowler et al., 2013;Lehnert et al., 2018a). Then, NH 4 + is incorporated into carbon-containing molecules for the synthesis of amino acids or oxidized by microorganisms such as nitrifying bacteria that produce nitric oxides and nitrates (Coskun et al., 2017;Lehnert et al., 2018a). These oxidized forms of N can be reduced by denitrifying bacteria, archaea, or fungi that convert NO 3 to NO 2 -, then to nitric oxide (NO), nitrous oxide (N 2 O), and finally back to N 2 (Fowler et al., 2013;Coskun et al., 2017). Although most microbes involved in the N cycle do not directly interact with plants, part of the microbial community present in the soil is associated with plant roots. These microbes can have several beneficial effects, in particular on plant growth and for the acquisition of nutrients such as N, a process that is tightly linked to their biological characteristics and their ability to fix N 2 .

Improved plant nitrogen nutrition with diazotrophs
Biological N 2 fixation occurs only in prokaryotes. Converting N 2 to NH 4 + is an energy-consuming process catalyzed by the multimeric enzyme Nase, which is known to be inhibited by oxygen (Fig. 1) (Hoffman et al., 2014). There are two main types of symbiotic associations with N 2 -fixing bacteria ( Table 1). The first involves a symbiotic association between plants and diazotrophic bacteria, leading to highly efficient atmospheric N 2 fixation and the transfer of the fixed N to the host plant. The second type leads to much less specific and less efficient endophytic associative symbiotic interactions.
The highly efficient N 2 -fixing symbiotic associations involve bacteria belonging to the family Rhizobiaceae and Actinorhizae such as Frankia sp. Rhizobia are alphaproteobacteria that establish an N 2 -fixing symbiosis mostly with legumes (Fabaceae). This is the best characterized process of endosymbiosis in plants involving N 2 -fixing bacteria. Upon molecular recognition of bacterial nodulation factors, root epidermal cells begin to differentiate into an organ called a nodule that hosts modified bacterial cells called bacteroids. Bacteroids perform N 2 fixation in the root nodules, allowing NH 4 + to be transferred to the plant via its vascular system. Biological N 2 fixation occurring in the Rhizobium-legume symbiosis has been extensively reviewed (Oldroyd et al., 2011;Udvardi and Poole, 2013) and will not be covered here. Some rhizobial species are able to induce the formation of N 2 -fixing root nodules in the non-legume plant Parasponia sp. (Behm et al., 2014;van Velzen et al., 2018). In this type of symbiotic association, the bacteroids are not fully differentiated as they are in legumes. In turn, both Nase activity and the number of bacterial cells per nodule are much lower compared with that in legumes (Behm et al., 2014;van Velzen et al., 2018). Filamentous actinobacteria of the genus Frankia establish symbioses with species from eight plant families belonging to the orders Fagales, Rosales, and Cucurbitales. Compared with rhizobia, the specificity of Frankia to their host plant is less strict (Froussart et al., 2016). Interestingly, Frankia cells are able to fix N 2 regardless of the symbiotic or non-symbiotic conditions, whereas rhizobia fix N 2 only under symbiotic conditions (Santi et al., 2013). Frankia host species are shrubs and trees growing under a large variety of environmental and climatic conditions. Some N 2 -fixing cyanobacteria can associate with plants and provide NH 4 + to their hosts without forming specialized organs such as nodules. Generally, these symbiotic N 2 -fixing cyanobacteria belong to the genus Nostoc. They have the capacity to differentiate specialized cells called heterocysts that carry out N 2 fixation. These cyanobacteria can associate with a wide range of bryophytes (liverworts and hornworts), pteridophytes (Azolla sp.) gymnosperms (Cycadaceae), and angiosperms (Gunneraceae) (Santi et al., 2013;Mus et al., 2016). Cyanobacteria colonize specific structures in their hosts, which can be glands located on the dorsal lobe of leaves (Azolla) or cavities in stems (Gunnera) (Santi et al., 2013). Most interactions between plants and cyanobacteria are facultative, except for Azolla sp., which has been used as an N biofertilizer for thousands of years in rice paddy fields (Kollah et al., 2016). Many plant species can associate with associative symbiotic free-living N 2 -fixing bacteria (AS-NFB). AS-NFB are present in a number of taxa, including alpha-proteobacteria, beta-proteobacteria, gamma-proteobacteria, and firmicutes. They include Azoarcus, Azospirillum, Herbaspirillum, Azotobacter, Burkholderia, Klebsiella, Gluconobacter, Acetobacter, and Pseudomonas spp. (Santi et al., 2013;Reinhold-Hurek et al., 2015;Mus et al., 2016). A number of these microorganisms are present in the rhizosphere and remain at the root surface, whereas others can colonize the plant endosphere; they are classified as plant-interacting or endophytic bacteria, respectively. Diazotrophic bacteria are Polyphosphate is hydrolyzed and then inorganic phosphate is transported to the peri-arbuscular interface. Fungal cells can also take up N, ammonium (NH 4 + ), amino acids, and small peptides. In the fungal cells, nitrate (NO 3 -) is reduced to NH 4 + by two successive reactions catalyzed by the enzymes nitrate reductase (NR) and nitrite reductase (NiR). NH 4 + is then incorporated into organic molecules via the combined action of the enzymes glutamine synthase (GS) and glutamate synthase (GOGAT), leading to the production of glutamine (Gln) and glutamate (Glu), two amino acids that foster fungal cell growth. N is finally stored in the form of arginine (Arg), which is bound to polyphosphate in the vacuolar fungal cells. Arginine is broken down into urea and then into NH 4 + , which is finally transferred to the peri-arbuscular interface. NH 4 + is considered to be the major form of N transferred from the fungus to the plant, although it is likely that NO 3 can also be transported. Symbiotic diazotrophs such as rhizobia or Frankia sp. can establish highly efficient N 2 -fixing symbiosis via nitrogenase (Nase) activity and transfer N to the plant, particularly legumes, in specialized organs called nodules. A number of associative symbiotic N 2 -fixing bacteria (NFB) interact with root cells or colonize plant roots and shoots. Some of these bacteria can also secrete auxins (AUX) that promote plant growth. Diazotrophic bacteria associated with a symbiotic fungus can also contribute AMF and thus plant N acquisition. AAT, Amino acid transporter; AMT, ammonium transporter; NRT, nitrate transporter. present in the rhizosphere of many wild and cultivated plant species, such as cereals. Due to their ability to reduce N 2 to NH 4 + , a form of N readily utilizable by the host plant, there is growing interest in determining the physiological and molecular mechanisms underlying this plant-bacterial association (Rosenblueth et al., 2018). Therefore, several studies have been undertaken to isolate and characterize these diazotrophic bacteria, particularly when the associated or host plants are grown under low N fertilization conditions. The characterization of these bacteria is primarily based on the detection of nif gene sequences encoding Nase.
Studies have demonstrated that N 2 fixation occurs in bacteria harboring nif genes and that there is a subsequent transfer of NH 4 + to the plant. Nase activity can be measured using the acetylene reduction assay (ARA) and by means of 15 N dilution experiments. The ARA is based on the capacity of Nase to reduce acetylene to ethylene, a molecule that is easily quantified by gas chromatography (Kifle and Laing, 2015;Brusamarello-Santos et al., 2017). 15 N dilution experiments can estimate the contribution of diazotrophic 14 N 2 fixation to total plant N acquisition following 15 N labelling (Chalk, 2016). A few studies have shown that AS-NFB can fix N 2 whether they are associated with or colonize the plant, and that the resulting inorganic N is transferred to the plant (Sevilla et al., 2001;Hurek et al., 2002;Iniguez et al., 2004;Pankievicz et al., 2015). To increase the efficiency of AS-NFB in terms of their ability to transfer N to plants, genetic manipulations (mutations or introduction of constitutive promoters) can be conducted to deregulate NH 4 + production and excretion for direct incorporation by shoots and roots (Ambrosio et al., 2017;Santos et al., 2017). Wheat inoculation with a spontaneous mutant excreting large amounts of NH 4 + resulted in an increase in plant yield .
In addition to providing N to the plant, several AS-NFB have the ability to promote root and shoot growth of wheat, maize, rice, and sorghum through the production of hormones, such as auxins, known to promote organ growth (Dobbelaere et al., 1999;Santi et al., 2013;Vacheron et al., 2013). These bacteria are referred to as plant-growth-promoting rhizobacteria (PGPR) (Cassán and Diaz-Zorita, 2016;Kudoyarova et al., 2019). Because the effect of PGPR can be multifactorial, it has been described by the "Multiple Mechanisms Theory" (Bashan and de-Bashan, 2010).
In a symbiotic Rhizobium-legume association, the plant benefits from the increased availability of reduced N 2 , while the bacteria utilize carbohydrates provided by the host plant (Udvardi and Poole, 2013;Hoffman et al., 2014). When there is an interaction between a plant and diazotrophic bacteria, the vicinity of the roots is enriched with carbohydrates present in the root exudates. It is thus likely that these carbohydrates are used to provide the energy required for N 2 fixation. In line with this hypothesis, reports indicate that, in a tropical maize genotype, aerial roots that produce a mucilage constitute a chemoattractive niche for diazotrophic bacteria (Van Deynze  , 2018). This mucilage provides substantial amounts of carbohydrates to the bacteria, leading to a remarkably active N 2 fixation process that is greatly favored in this sugarenriched environment (Knee et al., 2001;Van Deynze et al., 2018). Future agronomic developments utilizing the ability of plant genotypes to release carbohydrates in root exudates not only to attract AS-NFB but also to boost beneficial bacterial colonization and N 2 fixation thus requires further research (Coskun et al., 2017;Jacoby et al., 2017).

Improved plant nitrogen nutrition in plantfungus associations
Plant N acquisition can also be enhanced when plants develop symbiotic mycorrhizal associations with fungi of the order Glomeromyta by means of mycelium that colonizes root cells (endomycorrhiza) or is attached extracellularly to them (ectomycorrhiza). This type of symbiotic association makes it possible to capture nutrients in the surrounding rhizosphere.
In particular, the fungi provide minerals to the host plant and improve plant tolerance to biotic and abiotic stresses (Wang et al., 2017;Choi et al., 2018;Ferlian et al., 2018). Arbuscular mycorrhizal fungi (AMF) are the most widespread symbiotic association: they colonize 80-90% of terrestrial plants (Gianinazzi et al., 2010). Following spore germination, AMF colonize intra-and inter-cellular spaces and then develop highly branched structures called arbuscules in the cortical cells of the host plant. These structures are surrounded by a peri-arbuscular interface where nutrient exchanges with the host plant cells take place. The arbuscules are connected to an intraradical mycelium that is present inside the root tissues. The intraradical mycelium is also connected to an extraradical mycelium that spreads in the soil, allowing the uptake of minerals such as phosphate and N ( Fig. 1) (Chen et al., 2018b;Choi et al., 2018;Ferrol et al., 2019). Following colonization by AMF, plant transcripts of genes coding for NO 3 and NH 4 + transporters are up-regulated (Hildebrandt et al., 2002;Courty et al., 2015;Garcia et al., 2016;Koegel et al., 2017). Similarly, fungal NO 3 and NH 4 + transporters are up-regulated during the plant colonization process (Courty et al., 2015;Koegel et al., 2017). So far, most of the studies aiming to understand the mechanisms involved in plant N nutrition in the presence of AMF have been based on gene transcriptional activation (Guether et al., 2009;Pérez-Tienda et al., 2014;Garcia et al., 2016;Chen et al., 2018b). Interestingly, specific up-regulation of some transporters has been observed in roots colonized by AMF, indicating that under these conditions the transport mechanism is modified compared with that found in non-colonized plants . For instance, the AMT3.1 plant NH 4 + transporter transcript is specifically up-regulated in cereals such as sorghum, maize, and rice colonized by AMF (Koegel et al., 2017). The same study proposed that AMT3.1 is the major driver of N and phosphorus (P) transfer to the plant colonized by AMF (Koegel et al., 2017). Thus, the whole N uptake machinery available to the plant in the absence of mycorrhizae is not entirely recruited during these interactions.
Part of the N acquired by the fungal cells is used for their own metabolism. Storage of N as arginine (Arg) is indicated by the large amounts of Arg found in the extraradical mycelium and by the up-regulation of the genes and enzymes involved in the biosynthesis of Arg in the extraradical mycelium and its breakdown into urea in the intraradical mycelium (Cruz et al., 2007;Chen et al., 2018b). Arg is strongly bound to polyphosphate in vacuolar fungal cells, which may reflect a connection between P and N metabolisms (Fig. 1) (Woods and Ferré, 2005;Kikuchi et al., 2014). NH 4 + is considered to be the major form of N transfer from the fungus to the plant, although there may be NO 3 transfer as well because NRT genes from both the plant and the fungus have been found to be up-regulated (for review, see Courty et al., 2015;Garcia et al., 2016;Chen et al., 2018b). Carbohydrates and long-chain fatty acids are excreted by plant cells in the peri-arbuscular interface and then acquired by fungal cells via specific transporters (MacLean et al., 2017, Choi et al., 2018. Further investigations are now needed to determine whether improved plant N acquisition is due to the transfer of N from the AMF or to an increase in N uptake triggered by the fungus, considering at the same time that the fungus and the plant can compete for limited resources (Makarov, 2019). Developing such studies may be a means to further assess whether AMF symbiosis has a direct or an indirect impact on plant N nutrition. In addition, it is necessary to determine whether there is competition for N acquisition between the plant and the fungus, which might imply that, under agronomic conditions, N fertilization needs to be optimized for the plant to benefit from the symbiotic association.

Improvement in plant nitrogen nutrition by combining bacterial and mycorrhizal colonization
A promising strategy to enhance plant nutrition and thus plant performance is through the development of tripartite associations with bacteria and mycorrhizal fungi (Wu et al., 2005, Artursson et al., 2006Anca, 2009, Giovannini et al., 2020). AMF are in some cases associated with other microbes in the rhizosphere. Such associations can be either beneficial or detrimental to the AMF (Jansa et al., 2013). Although these tripartite associations are not very well characterized yet, they seem to strongly depend on nutritional exchanges between AMF and bacteria. These exchanges involve the production of exudates by the fungi that attract the bacteria, and the ability of the bacteria to facilitate the access to nutrients of the fungi (Kaiser et al., 2015;Zhang et al., 2016). For instance, bacteria can more easily solubilize phosphate than fungi, thereby improving both fungal and plant phosphate nutrition (Zhang et al., 2016). Interestingly, bacteria of the genus Paenibacillus isolated from the roots of Sorghum bicolor are able to stimulate mycorrhizal colonization of the plant by Glomus mosseae (Budi et al., 1999). Moreover, Paenibacillus sp. has been detected inside cells of the ectomycorrhizal fungus Laccaria bicolor (Bertaux et al., 2003). Paenibacillus validus is a species that can promote the growth in vitro and the formation of spores in Rhizophagus irregularis (Hildebrandt et al., 2006). This stimulating effect on spore production may be due to the presence of sugars secreted by the bacterium that enhance fungal growth until spores form. Thus, enhanced fungal growth may be beneficial to the development of a mycorrhizal association with a plant (Hildebrandt et al., 2006). In addition, a number of Paenibacillus species are potential N 2 fixers that can solubilize phosphate and iron and secrete phytohormones (Grady et al., 2016). Several plant-associated fungi are colonized by endosymbiotic diazotrophs that can provide N to the fungi (Minerdi et al., 2001;Sharma et al., 2008, Torres-Cortés et al., 2015Paul et al., 2020).
These various tripartite associations can favor the establishment of more efficient fungal and N 2 -fixing symbioses, potentially leading to better acquisition of N by the plant (Paul et al., 2020). Thus, further investigations to identify bacteria favoring the establishment of a symbiotic association between plants and AMF are worth conducting to determine whether this tripartite association can boost plant N acquisition, particularly under reduced fertilization conditions (Giovannini et al., 2020). Associating plants with more complex microbial consortia is another strategy that can boost plant performance, because AMF inoculation in combination with a microbial consortium isolated from nonfertilized soils increase N uptake in Brachypodium dystachion (Hestrin et al., 2019).

Concluding remarks
To fully exploit the beneficial effect of soil microbiota to improve plant N acquisition, more research is required to study and thus optimize the associations between plants, bacteria, and fungi in order to use them for the implementation of more sustainable agricultural practices (Hirel et al., 2011;Verzeaux et al., 2017a). First, it will be necessary to perform large-scale surveys and experiments to isolate bacterial and fungal species or strains associated with crop species or crop varieties and determine in which area in the world improved plant performance is observed, not only under laboratory conditions but also in the field (Cassán and Diaz-Zorita, 2016). This will allow the identification of microorganisms (or combinations of microorganisms) and crop genotypes (races collected around the world, lines, hybrids, etc.) that develop efficient interactions, particularly when N fertilization is lowered or limiting.
Furthermore, there clearly is genetic diversity among crop varieties for beneficial plant-microorganism associations, which suggests that, both among different species and within the same species, there are favorable alleles involved in the control of such associations (Sawers et al., 2018;Valente et al., 2020, Chandra et al., 2020. More ambitious studies at the genome level could be undertaken to identify which genes or loci are involved in the genetic control of plant performance in response to inoculation with N 2 -fixing bacteria and AMF (Lehnert et al., 2018b;Vidotti et al., 2019). Fig. 2. Strategies to optimize interactions with beneficial microbes to improve plant nitrogen (N) acquisition. Various strategies can be combined to reduce the use of N fertilizers and improve crop N acquisition. They include (i) capitalizing on the beneficial impact of microbes either individually or when they are present in tripartite associations; (ii) utilizing plant genetic deversity to select the most efficient interactions in terms of the acquisition of nutrients, N in particular; and (iii) developing agronomic practices such as no-till and those based on the use of cover crops that favor plant-microbe interactions (Hirel et al., 2011).
Another important issue will be to assess the persistence and competitiveness of beneficial microbes when there is a positive interaction with plants, particularly when microbial inoculants are used to improve plant performance (Romano et al., 2020). In parallel, fundamental research needs to be conducted to characterize the interactions regarding N acquisition between plants and beneficial microorganisms at the physiological and molecular levels. This type of study can help identify markers that can further help breeders to screen or develop new crop varieties for their responsiveness to microorganisms, either already present in the rhizosphere (van der Heijden et al., 2008) or provided as bio-inoculants. The agronomic practices adapted to or developed for these new varieties must be "microbe friendly" so that they favor and do not reduce soil-borne beneficial microbes (Fig. 2) (Hirel et al., 2011;Verzeaux et al., 2017a). Because fungicides are detrimental to soil microbial populations, the reduction in pesticide use can promote the benefits of soil microbiota ecosystemic services (Jacobsen and Hjelmsø, 2014). No-till farming, which is based on sowing seeds without disturbing the soil, is being adopted by growers because it allows the conservation of mycorrhizal hyphal soil networks (Verzeaux et al., 2017a, b). The use of cover crops, grown between the main crop, mown, and then left in place before sowing the next crop, helps conserve soil organic matter, favoring soil microbe proliferation (Verzeaux et al., 2016). A wise combination of all these strategies will help reduce the use of N fertilizers (Verzeaux et al., 2017a).