**2. Mechanisms of positive effect of BSM on the environment and health and yield of the legume plant**

#### **2.1. Legume-rhizobia Root-Nodule (RN) symbiosis**

Leguminous plants are able to grow in the soil/substrate without any combined nitrogen due to the fixation of atmospheric nitrogen by symbiotic nodule bacteria (collectively called rhizobia). In collaboration with rhizobia, legumes make a large contribution to the global nitrogen balance in natural and agricultural ecosystems [1]. Nitrogen fixation occurs within special plant organs, root nodules (in some associations stem nodules are also formed). Development of these organs represents a well-organized process based on the tightly coordinated expression of specialized symbiotic plant and bacterial genes. The legume nodules provide an ecological niche for bacteria, as well as structure for metabolic/signal exchange between the partners and for the control of symbionts by the hosts [2].

Family Fabaceae contains 17000-19000 species divided between three sub-families (Caesalpi‐ nioideae, Mimosoideae and Papilionoideae) and more than 700 genera of world-wide distri‐ bution [3]. With a single exception (*Parasponia*: Ulmaceae), the ability for symbioses with rhizobia is restricted to Fabaceae, although in eight related dicotyledonous families (Rosid I clade) an ability to form nodules with the nitrogen-fixing actinomycete *Frankia* is known [4].

By contrast to legumes, their nitrogen-fixing microsymbionts do not constitute a taxonomically coherent group of organisms. The majority of rhizobia belong to the α-proteobacteria assigned into the Rhizobiaceae family solely on the basis of their ability to nodulate the legumes (e.g. *Azorhizobium*, *Bradyrhizobium*, *Mesorhizobium*, *Rhizobium*, *Sinorhizobium*). Recently some βproteobacteria (close to *Burkholderia*, *Cupriavidus*, *Pseudomonas* and *Ralstonia*) and even some γ-proteobacteria have been discovered that can form nitrogen-fixing nodules with the legumes [5]. All these bacteria vary enormously in their overall genome organization, location of "symbiotic" (*sym*) genes and their molecular organization and regulation [6, 7].

Root-nodule symbiosis is well known as highly specific plant-microbe interaction. Ac‐ cording to the early surveys of symbiotic specificity [8], legumes were suggested to com‐ prise a range of taxonomically restricted cross-inoculation groups within which the free cross inoculation occurs, while the species from different groups do not cross-inoculate. The best studied examples of this classification are represented by four cross-inoculation groups: "*Trifolium* – *Rhizobium leguminosarum* bv. *trifolii*", "*Pisum, Vicia, Lathyrus, Lens* – *R. leguminosarum* bv. *viciae*", "*Galega* – *R. galegae*", "*Medicago, Melilotus, Trigonella* – *Sino‐ rhizobium meliloti, S. medicae*". However, it was demonstrated later [9, 10] that such strict‐ ly defined specificity is limited to the herbage papilionoid legumes growing in temperate zones and representing the "Galegoid complex".

last decade, a significant progress has been achieved in revealing the genetic bases of symbioses formation and functioning, so the knowledge of the plant genetic control over symbioses can effectively facilitate breeding new varieties of legumes that are needed for modern sustainable agriculture. In this chapter, we describe the present state of the developmental genetics of legume symbioses and depict the potential to organize the multi-component symbioses to be used for optimizing the broad spectrum of plant adaptive functions and to improve the

**2. Mechanisms of positive effect of BSM on the environment and health**

Leguminous plants are able to grow in the soil/substrate without any combined nitrogen due to the fixation of atmospheric nitrogen by symbiotic nodule bacteria (collectively called rhizobia). In collaboration with rhizobia, legumes make a large contribution to the global nitrogen balance in natural and agricultural ecosystems [1]. Nitrogen fixation occurs within special plant organs, root nodules (in some associations stem nodules are also formed). Development of these organs represents a well-organized process based on the tightly coordinated expression of specialized symbiotic plant and bacterial genes. The legume nodules provide an ecological niche for bacteria, as well as structure for metabolic/signal exchange

Family Fabaceae contains 17000-19000 species divided between three sub-families (Caesalpi‐ nioideae, Mimosoideae and Papilionoideae) and more than 700 genera of world-wide distri‐ bution [3]. With a single exception (*Parasponia*: Ulmaceae), the ability for symbioses with rhizobia is restricted to Fabaceae, although in eight related dicotyledonous families (Rosid I clade) an ability to form nodules with the nitrogen-fixing actinomycete *Frankia* is known [4]. By contrast to legumes, their nitrogen-fixing microsymbionts do not constitute a taxonomically coherent group of organisms. The majority of rhizobia belong to the α-proteobacteria assigned into the Rhizobiaceae family solely on the basis of their ability to nodulate the legumes (e.g. *Azorhizobium*, *Bradyrhizobium*, *Mesorhizobium*, *Rhizobium*, *Sinorhizobium*). Recently some βproteobacteria (close to *Burkholderia*, *Cupriavidus*, *Pseudomonas* and *Ralstonia*) and even some γ-proteobacteria have been discovered that can form nitrogen-fixing nodules with the legumes [5]. All these bacteria vary enormously in their overall genome organization, location of

sustainability of legume crop production.

168 Plant Breeding from Laboratories to Fields

**and yield of the legume plant**

**2.1. Legume-rhizobia Root-Nodule (RN) symbiosis**

between the partners and for the control of symbionts by the hosts [2].

"symbiotic" (*sym*) genes and their molecular organization and regulation [6, 7].

Root-nodule symbiosis is well known as highly specific plant-microbe interaction. Ac‐ cording to the early surveys of symbiotic specificity [8], legumes were suggested to com‐ prise a range of taxonomically restricted cross-inoculation groups within which the free cross inoculation occurs, while the species from different groups do not cross-inoculate. The best studied examples of this classification are represented by four cross-inoculation groups: "*Trifolium* – *Rhizobium leguminosarum* bv. *trifolii*", "*Pisum, Vicia, Lathyrus, Lens* –

The specificity of legume-rhizobia interactions is expressed just during the pre-infection stage when rhizobia recognize the roots of appropriate host plants and colonize their surfaces. The interaction starts when the root-excreted signals, in particular, flavonoids, activate the rhizobial nodulation genes (*nod*/*nol*/*noe*) [11]. These genes control the synthesis of lipochitoo‐ ligosaccharidic (LCO) nodulation factors (Nod factors, NFs) which induce the early stages of RN symbiosis development [12-14]. NFs represent the unique group of bacterial signal molecules not known outside legume-rhizobia symbiosis. They are among the most potent developmental regulators: their effect is expressed at concentrations of 10–8 – 10–12 M only. The core structure of these molecules, common for all rhizobia species, consists of 3-6 residues of N-acetylglucosamine and of a fatty acid (acyl) chain. The type of symbiotic specificity is dependent mainly on the chemical modifications in NF structures [11]. However, a sufficient role in the host specificity of RN symbiosis may also be implemented by the interactions between bacterial surface molecules (some polysaccharides and proteins) [15, 16] and the lectins located on the root hair surfaces [17].

The main enzyme of nitrogen fixation in nodules is a nitrogenase that has a complex structure [18, 19]. Synthesis of nitrogenase (the enzyme catalysing reduction of N2 into NH4 + ) and other proteins involved in nitrogen fixation is induced in bacterial cells after they differentiate into a specific form called bacteroids. Bacteroids are embedded into a membrane structure named symbiosome, which formation as well as bacteroid differentiation is induced by plant [20]. These symbiosomes are organelle-like units of plant cell responsible for nitrogen fixation [21, 22]. Peri-bacteroid membrane (PBM) that surrounds bacteroids is an active interface of RN symbiosis where exchange of metabolites between symbionts occurs [23].

A pronounced differentiation is typical for rhizobia-infected plant cells, such as an increase in internal membrane structures participating in the PBM formation and biosynthetic processes. Polyploidization and chromatin decondensation are typical for these cells correlating with an elevated transcription activity [24]. Biochemically plant cell differentiation is expressed as a *de novo* synthesis of many proteins including leghaemoglobin and nodule-specific isozymes of carbon and nitrogen metabolism [25]. Leghaemoglobin binds oxygen actively ensuring its transport towards symbiosomes (which are characterised by the intensive respiration necessa‐ ry to support energy consuming nitrogen fixation) and microaerobic conditions inside the nod‐ ules (required for the nitrogenase activity). The carbon and nitrogen metabolic enzymes responsible for the energy supply to nitrogenase and for the assimilation of fixed nitrogen are nodule specific [26]. Organic nitrogenous compounds formed from N2 fixation are transported to the upper parts of the plant either as amides (mainly asparagine (Asn), but also glutamine (Gln)) or as ureides (allantoin and allantoate), so that legumes can be classified as amide or ure‐ ide exporters according to the compounds they use for the mobilization of fixed nitrogen [27].

Rhizobial cells also undergo differentiation, but its level varies in different legume species. The terminal bacteroid differentiation (when bacteroids increase their size and DNA content and lack the capacity to divide) is specific for legumes belonging to the inverted repeat–lacking clade (IRLC) such as *Medicago*, *Pisum*, or *Trifolium*, whereas bacteroids in the non-IRLC legumes, such as *Lotus*, show no sign of terminal differentiation as they maintain their normal bacterial size, genome content, and reproductive capacity [28]. The same rhizobia strains that form symbiosis with both IRLC and non-IRLC legumes have different bacteroid differentiation fates in the two legume types. It was demonstrated that in *Medicago* and probably in other IRLC legumes, the nodule-specific NCR peptides act as symbiotic plant effectors to direct the bacteroids into a terminally differentiated state [29]. Possibly, IRLC legumes use nodulespecific NCR peptides to dominate the endosymbionts: NCR peptides interfere with many aspects of the bacteroid metabolism to allow the efficiency of the nitrogen fixation process to be optimized, for example, by stimulation enlargement and polyploidization of bacteroids [30]. Also, the peptides could be part of a mechanism to avoid the "cheating" of rhizobia that could use host resources to accumulate carbon storage compounds instead of fixing nitrogen [29], which is often observed in the non-IRLC legumes but not in the IRLC [30].

It was also found that nodules where terminal bacteroid differentiation takes place are more efficient in terms of energy use. Oono and Denison [31] reported that legume spe‐ cies with terminal bacteroid differentiation (such as peas (*Pisum sativum* L.) and peanuts (*Arachis hypogaea* L.) invest less in nodule construction but have greater fixation efficiency when compared to species with reversible bacteroid differentiation (such as beans (*Pha‐ seolus vulgaris* L.) and cow peas (*Vigna unguiculata* (L.) Walp.). This effect is probably due to genomic endoreduplication of the bacteroids and full contact of single undivided bac‐ teroid with peribacteroid membrane (some reproductive bacteroids can lose contact with PBM after they divide). Still, this is not known if these useful features of terminal bacte‐ roids differentiation in some legumes could be transferred into other legume species. In work of van de Velde et al. [29], expression of NCR genes in nodules of *Lotus japonicus* (Regel.) K. Larsen (with normally reversible bacteroid differentiation) was sufficient to in‐ duce bacteroid morphologies reminiscent of terminally differentiated bacteroids of *Medi‐ cago truncatula* Gaertn. But, no positive effect on nitrogen fixation efficiency was reported, probably because there are much more regulatory genes needed to make bacte‐ roids work propertly in such a heterologous system.

#### **2.2. Arbuscular Mycorrhiza (AM)**

Arbuscular mycorrhiza (AM) is formed by at least 80% of contemporary terrestrial plants with fungi of phylum Glomeromycota. The Glomeromycota are unique as the only monophyletic mycorrhizal fungus lineage that has co-evolved with land plants throughout their history. They are obligate biotrophs that colonize plant roots obtaining photosynthates, such as carbohydrates (hexoses), and niches for both their growth and reproduction. The AM is evolved more than 400 million years ago and was considered to play a decisive role in plants achieving a terrestrial existence [32-34]. The AM is supposed to be "the mother of plant root endosymbioses" [35]. Since legumes originated long after AM, about 60 million years ago [36], it may be assumed that all of them have the potential to produce this type of symbioses. *Lupinus* is the only known genus where this ability had apparently been lost [37-39].

Specificity of AM symbiosis is relatively low [34]. Symbiosis establishment starts with molecular dialogue between the partners. Plant roots release sesquiterpenes (also known as inducers of parasitic plant seed germination) as well as different phenolic compounds, including flavonoids, which induce fungal hypha growth and branching [40, 41]. Similar to rhizobia, AM-fungi produce signal molecules termed Myc factors (mycorrhization factors) [42], which can be recognized by the plant. They are a mixture of several lipochitooligosac‐ carides, the structure of which is close to that of rhizobial Nod factors, but is presumably more universal for different plant-fungus combinations [43]. Both sesquiterpenes and Myc factors are released constitutively and in the absence of physical contact with symbiotic partner [44].

lack the capacity to divide) is specific for legumes belonging to the inverted repeat–lacking clade (IRLC) such as *Medicago*, *Pisum*, or *Trifolium*, whereas bacteroids in the non-IRLC legumes, such as *Lotus*, show no sign of terminal differentiation as they maintain their normal bacterial size, genome content, and reproductive capacity [28]. The same rhizobia strains that form symbiosis with both IRLC and non-IRLC legumes have different bacteroid differentiation fates in the two legume types. It was demonstrated that in *Medicago* and probably in other IRLC legumes, the nodule-specific NCR peptides act as symbiotic plant effectors to direct the bacteroids into a terminally differentiated state [29]. Possibly, IRLC legumes use nodulespecific NCR peptides to dominate the endosymbionts: NCR peptides interfere with many aspects of the bacteroid metabolism to allow the efficiency of the nitrogen fixation process to be optimized, for example, by stimulation enlargement and polyploidization of bacteroids [30]. Also, the peptides could be part of a mechanism to avoid the "cheating" of rhizobia that could use host resources to accumulate carbon storage compounds instead of fixing nitrogen [29],

It was also found that nodules where terminal bacteroid differentiation takes place are more efficient in terms of energy use. Oono and Denison [31] reported that legume spe‐ cies with terminal bacteroid differentiation (such as peas (*Pisum sativum* L.) and peanuts (*Arachis hypogaea* L.) invest less in nodule construction but have greater fixation efficiency when compared to species with reversible bacteroid differentiation (such as beans (*Pha‐ seolus vulgaris* L.) and cow peas (*Vigna unguiculata* (L.) Walp.). This effect is probably due to genomic endoreduplication of the bacteroids and full contact of single undivided bac‐ teroid with peribacteroid membrane (some reproductive bacteroids can lose contact with PBM after they divide). Still, this is not known if these useful features of terminal bacte‐ roids differentiation in some legumes could be transferred into other legume species. In work of van de Velde et al. [29], expression of NCR genes in nodules of *Lotus japonicus* (Regel.) K. Larsen (with normally reversible bacteroid differentiation) was sufficient to in‐ duce bacteroid morphologies reminiscent of terminally differentiated bacteroids of *Medi‐ cago truncatula* Gaertn. But, no positive effect on nitrogen fixation efficiency was reported, probably because there are much more regulatory genes needed to make bacte‐

Arbuscular mycorrhiza (AM) is formed by at least 80% of contemporary terrestrial plants with fungi of phylum Glomeromycota. The Glomeromycota are unique as the only monophyletic mycorrhizal fungus lineage that has co-evolved with land plants throughout their history. They are obligate biotrophs that colonize plant roots obtaining photosynthates, such as carbohydrates (hexoses), and niches for both their growth and reproduction. The AM is evolved more than 400 million years ago and was considered to play a decisive role in plants achieving a terrestrial existence [32-34]. The AM is supposed to be "the mother of plant root endosymbioses" [35]. Since legumes originated long after AM, about 60 million years ago [36], it may be assumed that all of them have the potential to produce this type of symbioses. *Lupinus*

is the only known genus where this ability had apparently been lost [37-39].

which is often observed in the non-IRLC legumes but not in the IRLC [30].

roids work propertly in such a heterologous system.

**2.2. Arbuscular Mycorrhiza (AM)**

170 Plant Breeding from Laboratories to Fields

The AM-fungi penetrate the root to colonize inner cortical cells. Plant plays an active role in fungus hosting inside the root tissues using cellular mechanisms similar to those used during rhizobial invasion, such as nucleus reposition, cytoplasm aggregation, special cytoskeletal tunnel assembly and symbiotic membrane formation (reviewed in: [45]). A special intracellular compartment of AM providing tight metabolic exchange between the partners is arbuscule, which is highly branched fungal hypha surrounded by membrane of plant origin [34] similar to symbiosome of RN symbiosis [35, 46].

Inner-root and outer parts of mycelium remain bound with arbuscules and are a single continuum *via* which the fungus is able to translocate mineral nutrient and water from the soil into the root system [47]. Thus, well developed AM-symbiosis allows plant growing well in nutrient-poor and drought-affected soils, increases its resistance against pathogens and pests and heavy metals, and improves soil structure (see below).

Phosphorous (P) is one of the mineral nutrients essential for the plant growth (constitut‐ ing up to 0.2% of the dry weight of the plant cell) and development. It plays the diverse regulatory, structural, and energy transfer roles and consequently is required in signifi‐ cant amounts [48, 49]. The plants can acquire soluble forms of phosphorous directly from soil through the plant specific phosphate transporters (PTs). The dominant availa‐ ble forms in soil (orthophosphate ions, Pi ) are very poorly mobile because of the abun‐ dance of cations such as Ca2+, Fe3+ and Al3+ [50]. In such environments where inorganic phosphorous is the predominant form in soil, a range of root adaptations, most of them primarily involved in mobilization and assimilation of phosphorous, are described in‐ cluding plant dependence on arbuscular mycorrhizas (see for review: [34, 51]). In most cases there is a preferential uptake *via* fungal hyphae (the mycorrhizal uptake pathway) [52]. Studies employing radioactive tracers to track hyphal Pi uptake from soil have shown considerable AM contributions to phosphorous uptake [53-55]. The process in‐ volves several fungal transport systems some of which have an extremely high affinity for Pi [56].

After transporting into hyphae, the major part of Pi is polymerized by polyphosphate kinase into polyphosphates (poly-P), the linear chains of Pi . The granules rich in poly-P together with phosphorous-containing esters are packed into the cylindrical vacuoles which are transported along the hyphae by tubulin fibrils. After reaching the arbuscules, phosphorous compounds are destroyed by phosphatases and the released Pi cross the partners' interface [56-58]. The arbuscule is the site of phosphate transfer from fungus to plant. It is well documented that plants possess many classes of phosphate transport proteins, including those which are expressed only in AM symbiosis [59-61]. It was discovered that five plant and one fungal PT genes are consistently expressed inside the arbusculated cells [60]. A plant phosphate trans‐ porter MtPt4 was shown to be expressed specifically on the peri-arbuscular membrane in *Medicago truncatula* [62].

The mycorrhizal Pi uptake pathway is controlled by the plant host. Many results suggest that the plant phosphorous status is a major regulator controlling induction/repression of plant PT genes at both the soil-root interface and the inner-root symbiotic interface [63-65]. It was shown that high phosphorous concentrations counteract the induction of the mycorrhizal Pi transporter genes by phospholipid extracts from mycorrhizal roots containing the mycorrhiza signal lysophosphatidylcholine [65]. The efflux of Pi probably occurs in coordination with its uptake and the fungus, on its side, might exert the con‐ trol over the amount of Pi delivered to the plant [66].

Although Pi acquisition receives more attention, the important advances in investigations on nitrogen uptake by AM-fungi have been made in recent years. AM-fungi directly uptake ammonium (NH4 + ), nitrate and amino acids [67] with preference to NH4 + [68]. The first step in the nitrogen uptake requires the activity of specific transporters located at the interface between the soil and extraradical mycelium. A fungal transporter gene (*GintAMT1*) involved in the process and having high affinity with NH4 + was characterized [68].

Inorganic nitrogen that was taken up by the extraradical mycelium should then be incorpo‐ rated into the amino acids and translocated to the intraradical mycelium, mainly as arginine (Arg) since this is the predominant free amino acid in the external hyphae [69]. The glutamine synthetase/glutamate synthase (GS/GOGAT) cycle is possibly responsible for a subsequent NH4 <sup>+</sup> assimilation in AM extraradical hyphae [70, 71], although the involvement of glutamate dehydrogenase has not been experimentally excluded. Arg similar to Poly-P is stored and is translocated along hyphae in vacuoles and is later released to the plant apoplast [66].

A mycorrhizal-specific NH4 + transporter *LjAMT2;2* has been revealed recently in transcrip‐ tomic analysis of *Lotus japonicus* roots upon colonization with *Gigaspora margarita*. The gene has been characterized as a high-affinity AMT belonging to the AMT2 subfamily. It is strongly up-regulated and exclusively expressed in the mycorrhizal roots, but not in the nodules, and transcripts have preferentially been located in the arbusculated cells [72].

The plants colonized by AM-fungi have been demonstrated to manifest an increased re‐ sistance to attack of some pathogenic microorganisms, such as fungi, nematodes, bacte‐ ria, phytoplasma, and plant viruses (reviewed in: [73]) as well as to plant feeding insects [74, 75]. However, it is still unknown whether such increased resistance to pathogens is a consequence of improved plant overall fitness or it is due to the specific defense respons‐ es induced by AM-fungi.

Actually, a range of processes occurring as a result of pathogen invasion (plant defense responses) also takes place in mycorrhized root tissues. They include the signal percep‐ tion, signal transduction and defense-related gene activation [76-80]. The elements of hy‐ persensitive responses have been observed to take place at both compatible and noncompatible combinations of plants with AM-fungi; reactions similar to the "oxidative burst" are typical for AM during fungus penetration into the epidermal cell [81]. In AM, as in other compatible biotrophic interactions, the defense-like response appears to be weak and occurs transitorily during the early phases of colonization, suggesting that the suppression of plant defense responses by the fungal signals may contribute to success‐ ful, compatible AM fungal colonization [76, 82]. AM-fungi are known to alter both con‐ stitutive and induced defenses in foliar tissues [83-85].

plants possess many classes of phosphate transport proteins, including those which are expressed only in AM symbiosis [59-61]. It was discovered that five plant and one fungal PT genes are consistently expressed inside the arbusculated cells [60]. A plant phosphate trans‐ porter MtPt4 was shown to be expressed specifically on the peri-arbuscular membrane in

The mycorrhizal Pi uptake pathway is controlled by the plant host. Many results suggest that the plant phosphorous status is a major regulator controlling induction/repression of plant PT genes at both the soil-root interface and the inner-root symbiotic interface [63-65]. It was shown that high phosphorous concentrations counteract the induction of the mycorrhizal Pi transporter genes by phospholipid extracts from mycorrhizal roots containing the mycorrhiza signal lysophosphatidylcholine [65]. The efflux of Pi probably occurs in coordination with its uptake and the fungus, on its side, might exert the con‐

Although Pi acquisition receives more attention, the important advances in investigations on nitrogen uptake by AM-fungi have been made in recent years. AM-fungi directly uptake

the nitrogen uptake requires the activity of specific transporters located at the interface between the soil and extraradical mycelium. A fungal transporter gene (*GintAMT1*) involved

Inorganic nitrogen that was taken up by the extraradical mycelium should then be incorpo‐ rated into the amino acids and translocated to the intraradical mycelium, mainly as arginine (Arg) since this is the predominant free amino acid in the external hyphae [69]. The glutamine synthetase/glutamate synthase (GS/GOGAT) cycle is possibly responsible for a subsequent

+

<sup>+</sup> assimilation in AM extraradical hyphae [70, 71], although the involvement of glutamate dehydrogenase has not been experimentally excluded. Arg similar to Poly-P is stored and is

tomic analysis of *Lotus japonicus* roots upon colonization with *Gigaspora margarita*. The gene has been characterized as a high-affinity AMT belonging to the AMT2 subfamily. It is strongly up-regulated and exclusively expressed in the mycorrhizal roots, but not in the nodules, and

The plants colonized by AM-fungi have been demonstrated to manifest an increased re‐ sistance to attack of some pathogenic microorganisms, such as fungi, nematodes, bacte‐ ria, phytoplasma, and plant viruses (reviewed in: [73]) as well as to plant feeding insects [74, 75]. However, it is still unknown whether such increased resistance to pathogens is a consequence of improved plant overall fitness or it is due to the specific defense respons‐

Actually, a range of processes occurring as a result of pathogen invasion (plant defense responses) also takes place in mycorrhized root tissues. They include the signal percep‐ tion, signal transduction and defense-related gene activation [76-80]. The elements of hy‐ persensitive responses have been observed to take place at both compatible and non-

translocated along hyphae in vacuoles and is later released to the plant apoplast [66].

transcripts have preferentially been located in the arbusculated cells [72].

was characterized [68].

transporter *LjAMT2;2* has been revealed recently in transcrip‐

+

[68]. The first step in

), nitrate and amino acids [67] with preference to NH4

*Medicago truncatula* [62].

172 Plant Breeding from Laboratories to Fields

ammonium (NH4

NH4

+

A mycorrhizal-specific NH4

es induced by AM-fungi.

trol over the amount of Pi delivered to the plant [66].

in the process and having high affinity with NH4

+

Drought stress is a major agricultural constraint in the semi-arid tropics. In most cases symbiosis with AM-fungi has been shown to increase host plant growth rates during drought stress and plant resistance to drought. Several mechanisms explaining this phenomenon have been proposed: an influence of AM on plant hormone profiles, increasing intensity of gaseous exchange and photosynthesis in leaves, direct water transport via fungal hyphae from soil into the host plant, enhanced water uptake through improved hydraulic conductivity and increas‐ ing leaf conductance and photosynthetic activity, nitrate assimilation by fungal hyphae, enhanced activity of plant enzymes involved in defence against oxidative stress, plant osmosis regulation, and changes in cell-wall elasticity (reviewed in: [86-89]).

The AM fungal hyphae grow into the soil matrix and create conditions conducive to the formation of microaggregates and then their packing into macroaggregates due to production copious amounts of the glycoprotein glomalin [90, 91]. Through AM-fungi-mediated effects on soil structure, it seems logical to suggest that AM colonization of a soil might affect its moisture retention properties and, in turn, the behaviour of plants growing in the soil, particularly when it is relatively dry [88].

AM-fungi were found to play an important role in heavy metal detoxification and the estab‐ lishment of vegetation in strongly polluted areas (see for review: [92]). Fungal strains isolated from old zinc wastes also decrease heavy metal uptake by plants growing on metal rich substrates, limiting the risk of increasing the levels of these elements in the food chain [93]. Phytoremediation of metal contaminated areas attracts the increasing interest as a cheaper alternative to chemical methods, more friendly for environment and nondestructive to soil biota. The effectiveness of the bioremediation techniques depends on the appropriate selection of both the plant and the fungal partners. Plants conventionally introduced in contaminated areas disappear relatively soon, while those appearing during natural succession are better adapted to harsh conditions. Much more stable are plants that appear on the wastes sponta‐ neously, but, it takes a long time till they establish and form stable communities. Symbiotic partners selected on the basis of such research are often the best choice for future phytoreme‐ diation technologies [93-96]. Introduction of plants from xerothermic grasslands into the soils contaminated with industrial metal rich wastes is supposed to be a new solution for waste revegetation [97]. Further improvements can be obtained by optimization of diverse micro‐ biota including various groups of rhizospheric bacteria and shoot endophytes [92].

#### **2.3. Associations of roots with Plant Growth-Promoting Rhizobacteria (PGPR)**

Plant Growth Promoting Rhizobacteria (PGPR) are the taxonomically diverse group including different bacteria (*Arthrobacter*, *Azospirillum, Bacillus, Enterobacter, Pseudomonas, Paenibacillus,* *Streptomyces*) and even some archaea [98]. The PGPR are inhabitants of soil in the vicinity of plant roots and are dependent on consuming root exudates. Many PGPR are able to attach to root surfaces and to AM and other fungal hyphae. The PGPR provide several benefits affecting the host plant either directly (due to mineral nutrient improvement and stimulation of root development) or indirectly (due to defence of plants from soil-borne pathogens and improving host tolerance to abiotic stresses).

Similar to rhizobia, *Azospirillum* possess nitrogenase and therefore is able to fix atmos‐ pheric nitrogen. In the early papers, plant growth promoting activity was attributed mainly to associative nitrogen fixation. A broad distribution of cereal-*Azospirillum* associ‐ ations was identified (reviewed in: [99]). It was demonstrated later, however, that a par‐ tial role in these plant–PGPR associations was due to phytohormone auxin (indole-3 acetic acid, IAA) synthesis [100] which improves the root growth and assimilatory capabilities and hence aids nitrogen uptake by plants. In spite of absence of the visible anatomic differentiation in root-*Azospirillum* associations, its development involves a range of molecular interactions some of which may be common to endosymbiotic associ‐ ations with rhizobia. Moreover, there is a visible taxonomic relatedness between *Azospir‐ illum* and *Bradyrhizobium* genera. Thus, azospirilla and those slow-growing rhizobia might originate from a common *Azospirillum*-like ancestor (see for review: [101, 102]).

Many PGPR are able to solubilize sparingly soluble phosphates, usually by releasing che‐ lating organic acids. Phosphate solubilizing bacteria (PSB) have been identified, but their effectiveness in the soil-plant system is unclear. The ability of an inoculated PSB to sup‐ ply phosphopous to plant may be limited, either because the compounds released by PSB to solubilize phosphate are rapidly degraded or because the solubilized phosphate is re-fixed before it reaches the root surface [103].

The best studied examples of bacteria providing efficient defense from phytopathogens are: *Pseudomonas* (*P. fluorescens, P. chlororaphis, P. putida*), *Bacillus* (*B. cereus, B. subtilis*) and some *Serratia* (e.g., *S. marcescens*) species. Many of these bacteria are capable of preventing attacks by pathogenic fungi, nematodes and bacteria [98, 104, 105]. Diverse mechanisms may be involved in host protection offered by PGPR.

The best studied mechanism is the competitive exclusion of pathogens often related to their direct suppression by the bacterial antibiotic substances: phenazine-1-carboxamide, 2,4 diacetylphloroglucinol, kanosamine, oligomycin A, oomycin A, pyoluterin, pyrrolnitrin, xanthobaccin, zwittermycin A, volatile dyes (HCN) and cyclic lipopeptides [98, 104, 105].

An important mechanism for the suppression of pathogens by biocontrol microbes may result from competition for iron or other metals, that involves bacterial siderophores which may possess much greater affinities for ferric ions than those for fungal siderophores [104, 105]. The value of siderophores in biocontrol effects under natural conditions is predominantly associ‐ ated with their ability to induce forms of systemic resistance in plant [98, 106].

Competitive exclusion of pathogens by PGPR is best achieved when the bacteria exhibit high root-colonizing activity. Application of the technique of genetic labeling with Green Fluores‐ cent Protein (GFP) suggested that these bacteria do not regularly colonize the root interiors, and only rarely they can be observed inside the outer root tissues [98]. Most PGPR cells are concentrated on the root surface where the micro-colonies [98, 104] or bio-films are formed [107]. Since the interactions of plants with root-associated bacteria are not specific, the bacteria will colonize the roots of a broad spectrum of hosts. Specificity of the defensive association may be expressed however, at the point when antimicrobial compounds are being synthesized and this does not always correlate with bacteria taxonomy; many strains of *Bacillus* and *Pseudomonas* which possess plant-protective properties have close relatives amongst phyto‐ pathogenic types [106, 108, 109].

*Streptomyces*) and even some archaea [98]. The PGPR are inhabitants of soil in the vicinity of plant roots and are dependent on consuming root exudates. Many PGPR are able to attach to root surfaces and to AM and other fungal hyphae. The PGPR provide several benefits affecting the host plant either directly (due to mineral nutrient improvement and stimulation of root development) or indirectly (due to defence of plants from soil-borne pathogens and improving

Similar to rhizobia, *Azospirillum* possess nitrogenase and therefore is able to fix atmos‐ pheric nitrogen. In the early papers, plant growth promoting activity was attributed mainly to associative nitrogen fixation. A broad distribution of cereal-*Azospirillum* associ‐ ations was identified (reviewed in: [99]). It was demonstrated later, however, that a par‐ tial role in these plant–PGPR associations was due to phytohormone auxin (indole-3 acetic acid, IAA) synthesis [100] which improves the root growth and assimilatory capabilities and hence aids nitrogen uptake by plants. In spite of absence of the visible anatomic differentiation in root-*Azospirillum* associations, its development involves a range of molecular interactions some of which may be common to endosymbiotic associ‐ ations with rhizobia. Moreover, there is a visible taxonomic relatedness between *Azospir‐ illum* and *Bradyrhizobium* genera. Thus, azospirilla and those slow-growing rhizobia might originate from a common *Azospirillum*-like ancestor (see for review: [101, 102]).

Many PGPR are able to solubilize sparingly soluble phosphates, usually by releasing che‐ lating organic acids. Phosphate solubilizing bacteria (PSB) have been identified, but their effectiveness in the soil-plant system is unclear. The ability of an inoculated PSB to sup‐ ply phosphopous to plant may be limited, either because the compounds released by PSB to solubilize phosphate are rapidly degraded or because the solubilized phosphate is

The best studied examples of bacteria providing efficient defense from phytopathogens are: *Pseudomonas* (*P. fluorescens, P. chlororaphis, P. putida*), *Bacillus* (*B. cereus, B. subtilis*) and some *Serratia* (e.g., *S. marcescens*) species. Many of these bacteria are capable of preventing attacks by pathogenic fungi, nematodes and bacteria [98, 104, 105]. Diverse mechanisms may be

The best studied mechanism is the competitive exclusion of pathogens often related to their direct suppression by the bacterial antibiotic substances: phenazine-1-carboxamide, 2,4 diacetylphloroglucinol, kanosamine, oligomycin A, oomycin A, pyoluterin, pyrrolnitrin, xanthobaccin, zwittermycin A, volatile dyes (HCN) and cyclic lipopeptides [98, 104, 105].

An important mechanism for the suppression of pathogens by biocontrol microbes may result from competition for iron or other metals, that involves bacterial siderophores which may possess much greater affinities for ferric ions than those for fungal siderophores [104, 105]. The value of siderophores in biocontrol effects under natural conditions is predominantly associ‐

Competitive exclusion of pathogens by PGPR is best achieved when the bacteria exhibit high root-colonizing activity. Application of the technique of genetic labeling with Green Fluores‐ cent Protein (GFP) suggested that these bacteria do not regularly colonize the root interiors,

ated with their ability to induce forms of systemic resistance in plant [98, 106].

host tolerance to abiotic stresses).

174 Plant Breeding from Laboratories to Fields

re-fixed before it reaches the root surface [103].

involved in host protection offered by PGPR.

Microscopic observations demonstrated that suppression of fungi may be correlated with bacterial attachment to pathogen hyphae. As a result of this attachment, some PGPR strains commence their biocontrol functions by behaving as hyper-parasites of pathogenic fungi. This suppression may be related to the production of bacterial enzymes which destroy the pathogen cell walls [104, 105, 110, 111].

Sometimes the biocontrol activities of PGPR do not correlate with intensive colonization of host roots and plant protection results from only a small number of bacteria cells. This occurs when PGPR inoculation induces the systemic resistance mechanisms that make the root nonaccessible by pathogens. Initially this effect of PGPR was called ISR (Induced Systemic Resistance) and was attributed exclusively to nonpathogenic systems [112]; SAR (Systemic Acquired Resistance) reactions, by contrast, were considered to be typical for the interactions with plant pathogens. Nevertheless, it was later found that the reactions of both types occur in either pathogen or nonpathogen systems and are distinguished by the nature of their endogenous elicitors (reviewed in: [106, 113]). The conventional SAR reaction is characterized by an accumulation of salicylic acid as signal molecules and pathogenesis-related proteins (PRproteins), whereas ISR reaction is based on signal transduction pathways regulated by jasmonates and ethylene. The systemic defence responses of both types may be elicited exogenously by PGPR cells attached to the roots or penetrating their outer tissues. Some molecules produced by PGPR (cell wall and cyclic lipopolysaccharides, flagella components, exoenzymes, phytohormones, type III secretion system (TTSS) effectors, siderophores, salicylic acid, and toxins) may be perceived by the plant and elicit a defensive response [106].

It has been reported that PGPR which produce an enzyme which is involved in the catabolism of 1-aminocyclopropane-1-carboxylate (ACC) – the ACC deaminase, can lower ethylene concentration in a developing or stressed plant, protecting it against the deleterious effects of ethylene induced stress and facilitating the formation of longer roots [114, 115]. This demon‐ strates that ethylene is negative regulator of plant interaction with PGPR.

Despite relatively low specificity of plant associations with PGPR, plant genotype has been shown to influence their effectiveness (i.e. genetic integration exists between the partners), and a series of genome loci (QTL) was identified controlling its quantitative variation [106, 116, 117]. The most pronounced plant species-specificity has been observed in the manifestation of ISR reactions caused by PGPR [118, 119].

Both highly effective direct promotion of plant growth and biocontrol may be due to an ability of the host to regulate PGPR functions by modulating the composition of root exudates. Rootexcreted organic acids, but not sugars, are optimal for support of different types of PGPR [120-122]. Additionally, some plants (including the legumes, pea and alfalfa) regulate their PGPR functions by exudating specialised signals from the roots which mimic the bacterial "quorum sensing" regulators required for root colonization and antifungal activities [123]. These observations suggest that improvement of biocontrol functions in root-PGPR associa‐ tions may be achieved *via* manipulations with the bacterial and plant host genotypes.

#### **2.4. Mutually beneficial associations of plants with endophytic bacteria**

Healthy naturally propagated plants grown in the field or in pot cultures are colonized by populations of endophytic bacteria. The spectrum of endophytic bacteria isolated from the roots of various plants covers a wide range of species; representatives of the genera *Pseudo‐ monas*, *Bacillus* or *Streptomyces* are most frequently encountered as endophytes (reviewed in: [124]). Newly developed molecular methods enable complete analyses of the diversity of culturable and non-culturable bacteria [125]. Most of the known genera include some phyto‐ pathogenic endophytes. Endophytes and pathogens both possess many similar virulence factors (reviewed in: [124]).

Some endophytes are seed-borne, but others have mechanisms for colonizing plants that have yet to be elucidated [126]. Although there are occasional poorly substantiated reports of intracellular colonization of bacteria providing a consistent and effective increase in the productivity of crops, it is still considered that the intercellular apoplastic space is the most suitable niche for endophytes [127]. It is suggested that many bacterial 'endophytes' may not colonize the living tissues, but occupy protective niches in dead surface tissues or closely adhering soil of rhizosheaths. Consistent entry of endophytes into living root tissues in the field is supposed to require a bacterial capability to hydrolyse the hydrophobic incrustations of the walls of epidermal, hypodermal, endodermal, and other cortical cells [128].

Plant associations with endophytic bacteria can increase plant growth and promote gen‐ eral development or improve plant resistance to pathogens and other environmental stresses enhancing the host's ability to acquire nutrients, or by production of plant growth-regulating, allelopathic or antibiotic compounds [127, 129]. Sometimes improved plant resistance can be linked to induced systemic resistance caused by bacterial elicitors coming from the endophyte [130].

It is necessary to study the natural associations between bacterial endophytes and their hosts for the purposes of employing such systems most productively in sustainable agriculture [127]. Delivery of endophytes to the environment or agricultural fields should be carefully evaluated to avoid introducing plant, animal and human pathogens [131].

#### **2.5. Synergistic effect of microbes in rhizosphere**

Microorganisms in the rhizosphere are under the influence of root exudates and plant as a whole as well as of interspecies interactions with each other. Many fungi including AM-fungi can interact with different bacterial species which frequently attach to fungal mycelium (reviewed in: [103, 132, 133]). For those bacteria known to stimulate mycelial growth of mycorrhizal fungi and/or enhance root mycorrhization the term 'mycorrhiza-helper-bacteria' has been proposed [133, 134]. Particularly, the bacteria may encourage growth of AM-fungi at the perisymbiotic stage of development, which precedes the establishment of a direct contact of the microsymbiont with the plant root [103, 135]. On the other hand, AM-fungi directly modify the environment due to mycelial exudation [136], forming the so-called 'mycorrhizo‐ sphere' [137]. In addition, the stimulation of root exudation as a result of interactions with AMfungi leads to qualitative and quantitative changes of the bacterial community in the rhizosphere (reviewed in: [103, 132]).

excreted organic acids, but not sugars, are optimal for support of different types of PGPR [120-122]. Additionally, some plants (including the legumes, pea and alfalfa) regulate their PGPR functions by exudating specialised signals from the roots which mimic the bacterial "quorum sensing" regulators required for root colonization and antifungal activities [123]. These observations suggest that improvement of biocontrol functions in root-PGPR associa‐

Healthy naturally propagated plants grown in the field or in pot cultures are colonized by populations of endophytic bacteria. The spectrum of endophytic bacteria isolated from the roots of various plants covers a wide range of species; representatives of the genera *Pseudo‐ monas*, *Bacillus* or *Streptomyces* are most frequently encountered as endophytes (reviewed in: [124]). Newly developed molecular methods enable complete analyses of the diversity of culturable and non-culturable bacteria [125]. Most of the known genera include some phyto‐ pathogenic endophytes. Endophytes and pathogens both possess many similar virulence

Some endophytes are seed-borne, but others have mechanisms for colonizing plants that have yet to be elucidated [126]. Although there are occasional poorly substantiated reports of intracellular colonization of bacteria providing a consistent and effective increase in the productivity of crops, it is still considered that the intercellular apoplastic space is the most suitable niche for endophytes [127]. It is suggested that many bacterial 'endophytes' may not colonize the living tissues, but occupy protective niches in dead surface tissues or closely adhering soil of rhizosheaths. Consistent entry of endophytes into living root tissues in the field is supposed to require a bacterial capability to hydrolyse the hydrophobic incrustations

Plant associations with endophytic bacteria can increase plant growth and promote gen‐ eral development or improve plant resistance to pathogens and other environmental stresses enhancing the host's ability to acquire nutrients, or by production of plant growth-regulating, allelopathic or antibiotic compounds [127, 129]. Sometimes improved plant resistance can be linked to induced systemic resistance caused by bacterial elicitors

It is necessary to study the natural associations between bacterial endophytes and their hosts for the purposes of employing such systems most productively in sustainable agriculture [127]. Delivery of endophytes to the environment or agricultural fields should be carefully evaluated

Microorganisms in the rhizosphere are under the influence of root exudates and plant as a whole as well as of interspecies interactions with each other. Many fungi including AM-fungi can interact with different bacterial species which frequently attach to fungal mycelium (reviewed in: [103, 132, 133]). For those bacteria known to stimulate mycelial growth of

of the walls of epidermal, hypodermal, endodermal, and other cortical cells [128].

to avoid introducing plant, animal and human pathogens [131].

**2.5. Synergistic effect of microbes in rhizosphere**

tions may be achieved *via* manipulations with the bacterial and plant host genotypes.

**2.4. Mutually beneficial associations of plants with endophytic bacteria**

factors (reviewed in: [124]).

176 Plant Breeding from Laboratories to Fields

coming from the endophyte [130].

Synergistic effect between RN and AM symbioses of legumes was described by many authors [95, 138-142]. AM formation is known to promote nodule development and nitrogen fixation by rhizobia, in particular, by means of improvement of mineral (predominantly phosphorous) host plant nutrition (see for review: [103]). AM-fungi also manifest synergism during interac‐ tions with PGPR (both indigenous and introduced), which perform biocontrol, nitrogen fixation, and phosphate mobilization during double and complex inoculation [103, 132]. The synergetic effect of plant inoculation by rhizobia and PGPR (*Azospirillum*, *Bacillus*, and *Pseudomonas*) is well known. In particular, it is associated with PGPR production of indole-3 acetic acid, which encourages nodule formation [103, 143]. Triple inoculation of a model legume *Anthyllis cytisoides* with PGPR, AM-fungi and rhizobia was shown to be the most effective approach for revegetation in mediterranean semi-arid ecosystems [94].

Thus, the potential of microbial synergism allows us to speak about high prospects of bio‐ technologies focused on creation multicomponent symbioses (MCS) that increase the fertility and quality of agricultural legume and nonlegume crops. At the same time, the results of experiments with plant symbioses with AM-fungi, rhizobia and PGPR, including multimi‐ crobial systems, show the important role of physiological and genetic adaptation of microor‐ ganisms to local environmental conditions [92, 94, 95, 144]. Hence, during the development of such biotechnologies, it is recommended to use a complex of local microbial isolates adapted to particular environmental conditions.
