The Bacillus genus

Classification

The Bacillus genus was first reported by ^{Cohn (1872)}, who described it as heat-resistant, endospore-producing bacteria. The species of Bacillus belong to the bacteria kingdom; Phylus Firmicutes; Class Bacilli; Order Bacillales and Family Bacillaceae (^{Maughan and van der Auwera, 2011}). Currently, the genus includes over 336 species, which, because of their genetic similarity, can be classified into different groups, the most important of which are a) the group of B. cereus, related to pathogenicity, which includes B. cereus-anthracis-thuringiensis; b) the environmental bacilli, which are characteristically present in different habitats, such as the group of Bacillus subtilis, composed of B. subtilis-licheniformis-pumilus; c) the group of B. clausii-halodurans; and d) the group that includes Bacillus sp. NRRLB-14911-coahuilensis (^{Alcaraz et al., 2010}; ^{LPSN, 2016}).

Ecology

The Bacillus species are widely distributed worldwide due to their ability to form endospores, a characteristic that provides them with resistance and boosts its isolation in several habitats, both water and terrestrial ecosystems, and even in environments under extreme conditions (^{Tejera-Hernández et al., 2011}). However, soils are considered the main reservoir of this bacterial genus, since most Bacillus species are saprophytic and are able to use the large diversity of organic substrates in the soil, making this a complex matrix for the establishment of a large genetic and functional diversity of microbial species (^{McSpadden, 2004}). For this reason, the multiple Bacillus species can grow in soil, the cultivable counts are found in the range of log 3 to log 6 per gram of fresh weight of soil, essentially in genetically similar species to the group of B. subtilis and B. cereus (^{Vargas-Ayala et al., 2000}). Regardless of rRNA studies in soil contradicting the relative abundance of cultivable and un-cultivable species of this bacterial genus (^{Kumar et al., 2011}).

Under a focus aimed at agricultural sustainability, few investigations have aimed at understanding the specific population diversity and dynamics of Bacillus in rhizospheric soils, in which the bacterial communities that inhabit the rhizosphere respond particularly to the soil fertility and the exudates of plant roots, which vary with the phenology and genotype of the plant (^{de Souza et al., 2015}), and therefore bacteria that interact with plants and present associative or endophytic capacities, or symbiotic processes to adapt to the rhizospheric conditions are acknowledged as potential microbial inoculants. ^{Rudrappa et al. (2008)} reported that the production of B. subtilis FB17 biofilm is a rhizosphere colonization mechanism, due to its attraction to the L-malic acid secreted by the roots of Arabidopsis thaliana and induced by the foliar pathogen Pseudomonas syringae pv tomato. On the other hand, ^{Kumar et al. (2011)}, ^{López-Fernández et al. (2016)} and ^{Selim et al. (2016)} point out that diverse species of Bacillus may reside in internal tissues of grape (Vitis vinífera) and cotton plants (Gossypium barbadense L). These characteristics play a crucial role in the development, colonization and function of Bacillus, stimulating their relationship with the host plant, the biological control characteristics of which are strengthened.

Main characteristics

Some of the main characteristics of the Bacillus genus are its optional aerobic or sometimes anaerobic growth, Gram-positive, bacillary morphology, flagellar motility, and variable size (0.5 to 10 μm), its optimal growth being in a neutral pH, with a wide interval of growth temperatures, although most species are mesophilic (temperature between 30 and 45 °C), its metabolic diversity related to the promotion of plant growth and control of pathogens (^{Tejera-Hernández et al., 2011}); other features that stand out are its capacity to produce endospores (oval or cyllindrical) as a mechanism of resistance to diverse types of stress (^{Calvo and Zúñiga, 2010}; ^{Layton et al., 2011}; ^{Tejera-Hernández et al., 2011}) (Figure 1).

Figure 1 Morphological characteristics of Bacillus sp. TE3 belonging to the Collection of Native Edaphic and Endophytic Microorganisms (www.itson.edu.mx/COLMENA). A) Bacillary cells, Gram positive; B) endospores (white arrow) and bacillary cells (black arrow); and C) macroscopic morphology of strain TE3. 

The presence of endospores confers the Bacillus genus its capacity to spread and its prevalence in ecosystems. These form during the second phase of its life cycle, which is composed of a phase of vegetative growth and a phase of sporulation (Figure 2). During the first stage, the bacteria grows exponentially by binary fission, since it is found in a medium with favorable conditions for its growth. The second phase begins as a strategy for survival in the presence of some type of stress (high population density, lack of nutrients, external factors such as salinity, temperature, pH, and others). In this way, the vegetative cell begins the formation of the endospore, which implies an asymmetric cell division, giving way to the formation of two compartments: a mother cell and the immersion of a prespore. Later, the prespore is swallowed, forming a cell within the mother cell. During the later stages, the prespore is covered by protective layers (protein components, peptidoglycan and a wall that resides below it formed by germ cells), followed by the dehydration and the final maturation of the prespore. Finally, the mother cell is lysed by programmed cell death, releasing the endospore. The endospore can remain viable in the environment until the conditions become favorable to begin their metabolic processes and generate a vegetative cell (^{Errigton 2003}; ^{Tejera-Hernández et al., 2011}; ^{CALS; 2016}). Due to this, the formation of endospores resistant to heat and desiccation is an important characteristic for the formulation of biotechnological products based on strains of this bacterial genus (^{Pérez-García et al., 2011}).

Figure 2 Reproduction cycle of the genus Bacillus. Modified from ^{Errigton, (2003)}. 

The Bacillus genus as a Biological Control Agent

Large diversity of species of the Bacillus genus have proven to have antagonistic activity against diverse phytopathogenic microorganisms in agricultural crops, such as maize, rice, fruit trees, and others (^{Wang et al., 2014}; ^{Li et al., 2015}). The study of this ability of Bacillus began with the discovery of insecticidal activity of Cry proteins produced by B. thuringiensis; currently, several species of the Bacillus genus (B. subtilis, B. pumilus, B. amyloliquefaciens and B. licheniformis) are widely studied to mitigate the incidence of diseases of importance to agriculture (Raaijmakers and Mazzola, 2012). Some of the main ways in which these strains avoid the establishment and development of phytopathogenic organisms is through different mechanisms, which include A) the excretion of antibiotics, B) siderophores, C) lytic enzymes, D) toxins and E) inducing the systemic resistance of the plant (ISR) (Figure 3) (^{Layton et al., 2011}; ^{Tejera-Hernández et al., 2011}).

Figure 3 Main biological control mechanisms of the genus Bacillus. Production of A) lipopeptides, B) siderophores, C) lytic enzymes, D) δ-endotoxins, E) induction to the systemic response. 

Main mechanisms of biological control of the Bacillus genus

Production of lipopeptides

One of the most important characteristics of the Bacillus genus is its capacity to produce a large variety of antibiotics capable of inhibiting the growth of phytopathogenic agents, including non-ribosomal cyclic peptides, which have been the most widely studied. Lipopeptides (LPs) structurally consist of a cyclic peptide joined to a chain of β-hydroxy or β-amino fatty acids (Figure 4-A), classified into 3 different families (iturines, fengicines and surfactines), according to their amino acid sequences and length of the fatty acid (^{Ongena and Jaques, 2008}; ^{Falardeau et al., 2013}). These molecules are synthesized by multienzymatic complexes, known as nonribosomal peptide synthetase (NRPS), which are independent to messenger RNA (^{Chowdhury et al., 2015}). The families of iturines, fengicines and surfactines have been widely studied for their antibacterial and antifungal activity (^{Meena and Kanwar, 2015}). The antimicrobial activity of these LPs occurs because of their interaction with the cytoplasmic membrane of bacterial or fungal cells, causing the formation of pores and an osmotic imbalance, which triggers the cell death of the phytopathogenic microorganisms (Figure 4-B) (^{Aranda et al., 2005}; ^{Gong et al., 2006}). However, recent reports claim that some lipopeptides can have multiple action mechanisms, altering cell processes such as intracellular calcium homeostasis, energetic metabolism and RNA processing (^{Zhang et al., 2016}).

Figure 4 Lipopeptides as metabolites involved in the biological control of plant pathogens. A) Representatives of the family of lipopeptides, B) Action mechanism of lipopeptides. Modified from ^{Ongena and Jaques (2008)}; ^{Bayer (2016)}. 

The role of the LPs has become evident in the growth inhibition in phytopathogenic microorganisms. ^{Touré et al. (2004)} identified several isoforms of lipopeptides in extracts produced from liquid cultures of B. subtilis GA1. Later, in a test for co-inoculation of strains GA1 and Botrytis cinerea in apple fruits, they identified the presence of fengicines and iturines in inhibitory concentrations, showing the activity of these lipopeptides in-situ. Likewise, genetic expression and mutagenesis analyses have contributed significantly to the understanding the role of lipopeptides in the microbial activity of the Bacillus genus against phytopathogenic microorganisms (^{Xu et al., 2013}). For example, ^{Li et al. (2014)} mutated the gene of the phosphopantetheine transferase (sfp) in the strain SQR9 of B. amyloliquefaciens, crucial for the functioning of non-ribosomal peptide synthetase (NRPS), and therefore the synthesis of lipopeptides. This mutation resulted in a phenotype lacking antifungal activity against diverse agriculturally important phytopathogens, including Verticillium dahliae, Sclerotinia sclerotiorum, Fusarium oxysporum, Rhizoctonia solani, Fusarium solani, Phytophthora parasiticaThe efficient control of diseases by the biological control agents requires these to be able to establish in, and interact with, their host plant (^{Pérez-García et al., 2011}). Likewise, along with antibiosis, lipopeptides have an influence on the establishment of Bacillus with the regulation of cell processes such as motility and the formation of biofilms (^{Choudhary and Jhori, 2009}; ^{Xu et al., 2013}). The latter has been observed in the mutant of B. subtilis 6051, deficient in the production of surfactines, producing an irregular biofilm during colonization, as well as the reduction in its capacity to control infections caused by Pseudomonas syringae pv. tomato in Arabidopsis thaliana (^{Bais et al., 2004}). Similarly, the role of lipopeptides in the motility of Bacillus has been proven with the study of deficient mutants in the production of lipopeptides, i.e. Bacillus subtilis 168 deficient of the gen sfp gene, displayed a loss in motility in semi-solid agar, although it was restored when purified lipopeptides were added to the culture medium (^{Kinsinger et al., 2003}). Due to the role of the lipopeptides in several cell processes (antibiosis, formation of biofilms and motility), its isolated study on the motility in situ of Bacillus is complex.

Despite lipopeptides not being the only metabolites of the antibiome of the Bacillus genus involved in the biological control of phytopathogens, they have been proposed as the most efficient metabolites for this biological activity, due to its ecological role and antimicrobial capability, which explains their use in the search and selection of promissory BCA of this genus (^{Cawoy et al., 2015}; ^{Mora et al., 2015}).

Production of lytic enzymes

The production of enzymes involved in the degradation of the cell wall of phytopathogenic agents is one of the most widely reported biological control mechanisms, in particular against pathogens of fungal origins. The fungal cell is made up of glycoproteins, polysaccharides and other components that vary according to the fungal species (^{Bowman and Free, 2006}). The fraction of polysaccharides can comprise up to 80% of the cell wall of fungi, particularly chitin (~ 10 - 20%) and glucane (~ 50 - 60%), which are composed of residues of beta-1,3-glucose and beta-1,4-N-acetylglucosamine, respectively (^{Bowman and Free, 2006}; ^{Latgé, 2007}). These polymers play a crucial role in the firmness of the cell wall by a wide network of glycosidic bonds. Therefore, the interference in these bonds can deteriorate the cell wall of phytopathogenic fungi, causing their lysis and cell death.

The production of lytic enzymes such as chitinases (EC 3.2.1.14) and β-glucanases (EC 3.2.1.6 and EC 3.2.1.39) excreted by the BCA, including the Bacillus genus, have displayed an inhibitory effect against pathogens of fungal origins (^{Compant et al., 2005}). These enzymes are responsible for the degradation of the main polysaccharides that make up the cell wall of the fungi, by the hydrolysis of their glycosidic bonds (Figure 5). There are currently diverse scientific studies that report the role of these enzymes in the in vitro antifungal activities obtained from strains of the Bacillus genus (^{Kishore et al., 2005}; ^{Liu et al., 2010}; ^{Shafi et al., 2017}). For example, ^{Yan et al. (2011}) demonstrated the role a chitinase in the control of Rhizoctonia solani by B. subtilis SL-13, evaluating the inhibitory activity of the purified enzyme, confronted with this pathogen. ^{Chien-Jui et al. (2004)} cloned the gene chiCW of B. cereus 28-9 on Escherichia coli DH5α and showed that the purified products presented a high inhibitory activity in the germination of Botrytis elliptica spores. Likewise, ^{Martínez-Absalón et al. (2014)} observed that when placing liquid B. thuringiensis UM96 supernatants in the specific inhibitor of chitinases, alosamidine, they lost their ability to inhibit the growth of Botrytis cinerea, thus showing the importance of chitinases on the biological control activity of the strain. Similarly, different authors have reported β-glucanases as important components in the activity for the biological control strains of the Bacillus genus; Aktuganov et al. (2007), reported that β-1,3-glucanases are the main lytic enzymes involved in the in vitro control of Bipolaris sorokiniana by Bacillus sp. 739, and in studies with B. amyloliquefaciens MET0908, with the aid of scanning electronic microscopy, the lytic activity of a β -1,3-glucanases on the Colletotrichum lagenarium hyphae was observed. Also, genetic engineering studies focused on the increase of antifungal activity of Bacillus strains have successfully achieved the insertion of codifying genes into lytic enzymes, such as gene ChiA of B. subtilis F29-3 in B. circulans, obtaining a more aggressive phenotype against Botrytis elliptica (^{Chen et al., 2004}). In addition, ^{Zhang et al. (2012)} significantly increased the antifungal activity of Burkholderia vietnamiensis P418 in the face of different fungal phytopathogens by the chromosomal insertion of the gene Chi113, from a B. Subtilis strain.

Figure 5 Degradation of the cell walls of plant pathogenic fungi by lytic enzymes. Modified from ^{Moebius et al. (2014)}. 

Production of siderophores

Iron (Fe) is an essential nutrient for important cell functions, such as in redox reactions of proteins with cofactors (Fe-S), in the electron transportation chain, and catalyzing vital enzyme reactions, such as those involving hydrogen, oxygen and nitrogen (^{Faraldo-Gómez and Sansom, 2003}). Fe is found in nature mostly in a ferric form (Fe³⁺) with a low solubility, making its use impossible for some live beings (^{Aguado-Santacruz et al., 2012}). In response to the iron restriction in the environment, some microorganisms have developed diverse receptor protein structures with low molecular weights and high affinity to iron, called siderophores, facilitating the capturing of Fe⁺³ (^{Thyagarajan et al., 2017}). Siderophores are secondary metabolites that act as iron sequestrants or chelators, as a consequence of their high dissociation constant due to this metal, fluctuating between 10²² and 10⁵⁵. This allows for the formation of Fe^3+- siderophore complexes, so siderophore-producing microorganisms, using a specific receptor located in the membrane, can use it using two mechanisms: 1) directly with the Fe3^+- siderophore complex through the cell membrane, or 2) reducing extracellularly to Fe²⁺ complexes (^{Neilands, 1995}) (Figure 6-A). Siderophores, based on their main chelating group, are classified into: hydroxamates (using hydroxamic acids), catecholates (containing catechol rings), carboxylates, phenolates, and in a combination of two or more of these groups (^{Wilson et al., 2016}) (Figure 6-B). A wide diversity of strains with the capacity of biological control belonging to the Bacillus genus have shown the ability to synthesize siderophores, regulating the concentration of iron in the medium through its chelation (Fe³⁺-siderophore), causing this metal to become unavailable for pathogenic microorganisms, which depend highly on this element for growth (^{Scharf et al., 2014}). On the other hand, the formation of such complexes does not affect the development of the plants, since most of them can grow in lower iron concentrations than required by these biological control agents; likewise, some plants have the ability to use these microbial complexes, increasing their bioavailability to this element (^{Aguado-Santacruz et al., 2012}).

Figure 6 Siderophores as a mechanism for the inhibition of plant pathogens. A) Capture and solubilization of iron, B) main chelating structures of siderophores. Modified from ^{Neilands (1995)}; ^{Wilson et al. (2016)}. 

In this way, several species of the Bacillus genus have been reported for their ability to control plant diseases by the secretion of siderophores, limiting the growth and colonization of iron-dependent phytopathogenic microorganisms (^{Fgaier and Eberl, 2011}). ^{Yu et al. (2011)} showed, in co-culture tests in chrome azurol sulfonate (CAS) agar slides that B. subtilis CAS15 strongly antagonized the growth of 15 pathogenic fungi belonging to the genera of Fusarium, Colletotrichum, Pythium, Magnaporthe and Phytophthora, with inhibition rates in the interval of 19 to 94%, attributing this effect to the catecholate siderophores (Bacillibactin), identified by ESI-MS and DHB. Similarly, ^{May et al. (2001)} reported the potential of strains of the Bacillus genus on the production of siderophores by the analysis of genes involved in the synthesis of Bacillibactin (BB), particularly in B. subtilis, observing that the mutant strain JJM405 in gene dhb, presented a limited production of BB, while ATCC21332 (wild strain) presented a higher production of this siderophore.

Production of δ-endotoxins

δ-endotoxins, produced particularly by Bacillus thuringiensis (Bt), are protein parasporal bodies made up of polypeptide units of different molecular weights, from 27 to 140 kDa. There are currently 300 reported holotypes of Bt toxins, classified inti 73 Cry and 3 Cyt families (^{Porcar and Juárez, 2004}; ^{Xu et al., 2014}). Bt toxins are produced during the sporulation phase; the Cry (crystal) protein is known for its toxic effects in an objective organism (most belong to the order of insects); likewise, Cyt (cytolitic) protein has been related to toxic effects on a large variety of insects, particularly diptera; however, its cytotoxicity has been tested on mammal cells (^{Soberón and Bravo, 2007}).

Cry proteins are widely used for their efficiency in the biological control of insects, the action mechanism of which begins once the Cry proteins are processes proteolytically through proteases present in the host’s middle intestine, separating a section of amino acids in the N-terminal region and in the C-terminal end (depending on the nature of the Cry protein), thus releasing active and toxic fragments that interact with the receptor proteins present in the intestine cells (Figure 7). These fragments are acknowledged by specific receptors in the membrane and inserted through the cadherin (1), giving rise to a series of signals for the formation of a pre-porous oligomeric structure (2-3), and consequently lytic porous (4), which triggers an osmotic imbalance, which destroys the intestinal epithelium and the consequent cell death (5) (^{Portela-Dussán et al., 2013}; ^{Xu et al., 2014}). Several authors have reported the potential of Cry proteins in the toxicity of agriculturally important pests. For example, ^{Niedmann and Meza-Basso (2006)} point out that the tomato leafminer (Tuta absoluta) causes losses of between 60 and 100% of the crops that are not treated with insecticides, thus showing the potential of Bt, highlighting that in tests run on tomato leaves with added Cry protein concentrates extracted from strains LM-11, LM-12, LM-14 and LM-33 of Bt, mortality was reduced to between 20 and 60% of the T. Absoluta larvae, which would suggest the reduction in losses of up to 12 and 60%. On the other hand, ^{Vázquez-Ramírez et al. (2015)} highlight the prevailing role Bt strains can take against the pest of the fall armyworm (Spodoptera frugiperda), on biotests carried out with Cry protein extracts obtained from strains LBIT-13, LBIT-44, LBIT-383, LBIT-418 and LBIT-428 of Bt, showed that they all had a toxic effect on the fall armyworm, although strains LBIT-13 and LBIT-418 showed a high toxicity towards S. frugiperda, with CL₅₀ of 137.2 and 197.2 ng cm^-2, respectively, in comparison with the commercial standard HD-1 (CL₅₀ of 142 ng cm^-2).

Figure 7 Action mechanism of Cry proteins in insects. Modified from ^{Xu et al. (2014)}. 

Induced systemic response

Throughout evolutionary history, plants have developed mechanisms to defend themselves against the invasion of pathogenic organisms (bacteria, fungi, nematodes, insects, etc.). These mechanisms are latent and are activated by stimuli during the interaction with pathogenic agents. In general terms, these mechanisms are known as systemic acquired resistance (SAR) and become activated, not only in the site of the infection, but systemically in other tissues (^{Pieterse et al., 2014}).

SAR is activated by stimuli perceived mainly by two receptors, the PRRs (pattern recognition receptors) and NB-LRRs (nucleotide-binding-leucine-rich repeat) (^{Pieterse et al., 2014}). The former perceives cell components such as fungal chitin or flagellins (PAMPs o MAMPs, pathogen o microbe associated molecular patterns) triggering the first line of defense, known as PTI (PAMP-triggered immunity). In pathogens with mechanisms for the evasion of PTI, a second line of defense is activated, which perceives virulence effector proteins via receptors NB-LRRs (^{Boller et al., 2009}). SAR depends on the signaling of salicylic acid (SA), which activates PR (pathogenesis-related) genes by codifying many of them to PR proteins with antimicrobial ability (e.i. PR1) (^{Vlot et al., 2009}).

Similarly to SAR, the systemic response in plants may be induced (ISR, induced systemic resistance) by chemical signals (elicitors) produced by beneficial microorganisms (^{Pérez-Montaño et al., 2014}) (Figure 3-E). Despite SAR and ISR being considered synonymous due to the similarity between the mechanisms (^{Pieterse et al., 2014}), signaling in ISR depends on the jasmonic acid and ethylene (^{Pieterse, 1998}The protection of agriculturally important crops (tomato, peppers, bean, rice, etc.) by inducing the systemic response using strains of the Bacillus genus has been documented (^{Akram et al., 2016}; ^{Choudhary and Johri, 2009}; ^{Wang et al., 2013}; Yi et al., 2013). Bacillus produces a large diversity of eliciting molecules that induce a systemic response in plants, including lipopeptides (^{Chowdhury et al., 2015}), phytohormones (^{Ryu et al., 2003}) and volatile compounds (^{Kim et al., 2015}). The latter activate PR genes, which protect from the invasion of pathogenic genes. This has been observed in tobacco plants, where PR2 codifies for one β-1,3 glucanase and PR3 codifies for one chitinase, they were activated in response to volatile compounds of Bacillus sp. JS, conferring resistance to Rhizocronia solani and Phytophthora nicotianae (^{Kim et al., 2015}). Along with PR genes, Bacillus activates other plant protection mechanisms, which include structural changes in the cell wall with the accumulation of lignin (^{Singh et al., 2016}) or the production or secondary metabolites such as flavonoids, phytoalexins, auxins or glucosinolates in general (^{Pretali et al., 2016}). Despite the plant protection mechanisms induced by elicitors being activated by different metabolically routes, a microorganism can activate multiple protection mechanisms. This has been observed in wheat plants, where after the inoculation of Bacillus amyloliquefaciens B-16, the production of multiple PR proteins and secondary metabolites such as gallic acid and ferulic acid was induced, conferring resistance against Bipolaris sorokiniana (^{Singh et al., 2016}).

The Bacillus genus and pesticides

The use of chemical pesticides as the main method for the control of pests and diseases in agriculture has helped increase agricultural productivity significantly in the past decades. However, their excessive use has created resistance to these compounds by phytopathogenic microorganisms, while having harmful effects on human health and the environment. This shows the need to generate efficient and environmentally friendly alternatives to reduce the use of synthetic products in order to achieve an efficient and sustainable control of diseases in agricultural crops (^{Pérez-García et al., 2011}).

Biological control is an important part of the management of pests and diseases, and it consists of the use of living organisms to reduce and maintain the abundance of a pest or pathogen below the levels of economic damage. The potential of this alternative is based on an efficient control of a pest or disease in the middle and long terms, compatible with a low environmental risk and a sustainable production. In this way, the management of pests and diseases can be carried out by several methods: the use of synthetic pesticides, crops modified genetically to resist pests, biological control or the combination of one or more of these strategies. Biopesticides are a particular group of tools for the protection of crops used in the MIPE. Although there is not a formal definition for this term, a biopesticide refers to an agent produced at a massive scale, from a living microorganism or a natural product, and sold for the control of pests or diseases in plants (this definition covers most of the products classified as biopesticides, within the countries of the Organization for the Economic Co-operation and Development, ^{OCDE, 2009}). Biopesticides can be classified into three groups, depending on the active substance: i) biochemical products, that mainly comprise secondary metabolites of plants and microorganisms; ii) semi-chemical products, mostly made up of pheromones; and iii) microorganisms, which include bacteria, viruses, fungi, and protozoa (^{Chandler et al., 2011}).

Of the commercially available microbial pesticides, Bacillus is the most widely exploited in agricultural biotechnology, with 85% of the bacterial products, due to its large metabolic versatility that allow it to perform a biological control of pests and diseases through diverse mechanisms. Also, this bacterial genus is able to produce endospores, which are the main active ingredient of the formulae, and confer to them - as a property - a greater viability in time (^{Ongena and Jaques, 2008}).

One of the biological control mechanisms that has been most exploited in the biopesticides market is the ability of the Bacillus genus of producing δ-endotoxins. This mechanism was the milestone in the development of the first microbial biopesticide for the biological control of lepidoptera, produces from B. thuringiensis var. kurstaki HD-1. As mentioned earlier, B. thuringiensis (Bt) produces Cry proteins (Bt- δ endotoxins) during spore formation, which is capable of producing lysis in cells of the digestive tract when consumed by susceptible insects. Strain HD-1 is one of the most studied strains, since it characteristically carries varieties of cry antilepidoptera genes: cry1Aa, cry1Ab, cry1Ac, cry2Aa, cry2Ab and cry1Ia (^{Höfte and Whiteley, 1989}; ^{Sauka and Benintende, 2008}).

Currently, B. Thuringiensis-based products account for 75% of the biopesticides sold globally (^{Olson, 2015}), with a substantial impact on the Mexican biopesticide market since 1980. In Mexico, the use of Bt-based formulae is an efficient formula for the control of insects and it accounts for between 4 and 10% of the total of insecticides used for maize, cotton and vegetable crops (^{Tamez et al., 2001}). Likewise, other species of the Bacillus genus, such as: B. subtilis, B. pumilus, B. licheniformis and B. amyloliquefaciens, stand out for their successful implementation in commercial formulations, and those developed mainly for the control of fungal diseases (Table 1). For example, ^{Galindo et al., (2015)} developed the first Mexican wide-spectrum biofungicide “Fungifree AB”, produced with viable B. subtilis 83 spores. This bacteria is a natural antagonist to diverse phytopathogens, used to prevent at least 8 pathogens of different etiologies: Colletotrichum, Erysiphe, Leveillula, Botrytis, Sphaerothecamacularis, in over 20 agriculture crops; They even indicate that the success of its formula resides, not only in the support provided by scientists, but also in the publication of its results in a journal read by professionals in agribusiness, thus allowing the linked that bonded crop exporting companies in search for sustainable alternatives that could allow them to control phytopathogens (i. e. Colletotrichum gloeosporioides).

In recent years, the chemical pesticide production market has declined 2% every year, while the production of biopesticides presents an annual increase of 20% (^{Cheng et al., 2010}). There are several reasons for the increasing interest towards microbial biopesticides, including the limited development of resistance of pathogenic organisms to them, a reduction in the rate of discovery of new insecticides, a greater public perception of the dangers related to synthetic pesticides, the highest number of studies on the specificity of microbial pesticides, improvements in production, the technology of formulation and dissemination, as well as the interaction with producers and regulation bodies. In this way, biopesticides account for a small fraction of the global market focused on crop protection: 5% (approximate value of $3 billion USD). However, biopesticides are expected to have a Compound Annual Growth Rate (CAGR) of at least 8.64% by 2023, estimating a value of over $4.5 billion USD (^{Olson et al., 2013}; ^{Olson, 2015}). It is worth mentioning that the transition and integration of the use of biopesticides in current agricultural practices should comply with the following requirements: a) effectiveness against the pest or disease; b) compatibility with other control methods; c) low or no environmental impact; d) long-lasting effect on the medium; e) economy, from a cost/ benefit viewpoint; f) technical feasibility of its use; and g) acceptance by producers and society in general. Therefore, the use of biopesticides offers an opportunity to stimulate the development and modernization of current agricultural practices, with the aim of contributing to food security under the scope of biosecurity.

Discussion and perspectives of biosecurity and biodiversity in the use of the Bacillus genus in agro-systems

The species of the Bacillus genus have a great metabolic and functional diversity, promoting its wide use in agriculture. In this sense, the groups of Bacillus cereus and Bacillus subtilis are the most widely used. However, in terms of biosecurity, they should be studied broadly before being used as biological control agents in the field. The group of Bacillus subtilis, which includes important species for agriculture, such as B. subtilis, B. licheniformis and B. pumilus, are not traditionally considered as pathogens for humans, and B. subtilis has even been granted the status of QPS (Qualified Presumption of Safety) by the European Food Security Authority (^{EFSA, 2015}). However, there are some isolated cases of intoxications form digestive manifestations, such as the one reported by ^{Pavic et al. (2005)}, pointing out B. subtilis and B. licheniformis as causal agents of the intoxication outbreak in a kindergarten caused by the consumption of powdered milk, which contained these bacterial species. On the other hand, in the group of Bacillus cereus made up of species such as B. cereus, B. thuringiensis, B. anthracis, B. mycoides, B. seudomycoides, B. cytotoxicus and B. weihenstephanensis, some strains have been identified as pathogens for humans (^{Ceuppens et al., 2013}; ^{Kim et al., 2016}). Among these, B. cereus has been identified in a large diversity of foods (dairy products, fresh vegetables and others), causing important worldwide epidemiological crises, and in some cases, even death by emetic and diarrheal infections (^{Oh et al., 2012}; ^{Kim et al., 2016}; ^{Glasset et al., 2016}). Also, B. thuringiensis strains have recently been related to intoxications caused by the ingestion of contaminated foods (^{Oh et al., 2012}).

The virulence of these species has been mainly related to the presence of two toxins, hemolysin BL (HBL) and the nonhemolytic enteric toxin (NHE), which form a protein complex (^{Kim et al., 2016}). Other toxins have also been identified in pathogenic strains, including cytotoxin K (cytK), enterotoxin FM (entFM), enterotoxin S (entS) and enterotoxin T (bceT). In strains that produce the emetic toxin (toxin of high resistance to thermal treatments, extreme pH values and the activity of proteases), virulence has been related to the presence of the dodeca depsipeptide synthesized by non-ribosomal peptide synthases (NRPS) codified by genes ces, found in type pXO1 plasmids. Likewise, products of other genes, such as hemolysin A (hlyA), hemolysin II and III (hlyI, hlyII), cereolysin A and B (cerA, cerB), and the pleitropic transcription factor (pclR) are involved in the pathogenicity of these strains (^{Ceuppens et al., 2013}).

Historically, the classification and differentiation of species in the Bacillus cereus group has been carried out using gene 16S RNAr and other characteristics such as i) virulence (B. cereus), ii) content of plasmids (B. anthracis and B. thuringiensis), iii) growth conditions (B. cytotoxicus and B. weihenstephanensis) and iv) morphological characteristics (B. mycoides and B. seudomycoides). However, in widely related species such as B. cereus, B. anthracis and B. thuringiensis, the differentiation using virulence factors and contents of plasmids is limited, due to its loss and transference during the evolutionary history of these species (^{Hoffmaster et al., 2006}; ^{Liu et al., 2015}). Recent comparative studies with complete genomes using dDDH (digital DNA: DNA hybridization) showed the distribution of cry genes and type pXO plasmids in members of this group, showing the low correlation between the phylogenetic position and the presence or absence of these plasmids (^{Liu et al., 2015}). The above study also showed the low resolution of the multilocus sequence typing (MLST) for the differentiation at the level of species.

In this way, due to the high metabolic versatility with agricultural application shown by members of the B. Cereus group, particularly B. cereus and B. thuringiensis, the correct identification and the determination of its virulence for the human being is decisive for the selection and commercialization of biological control agents of these species. Although the study of comparative genomics using complete genomes is the only accurate alternative for the classification and determining its virulence, it is important to consider that they are costly tools for the discrimination of pathogenic strains during the primary process of selection of potential biological control agents.

On the other hand, in agro-systems, soil is a dynamic matrix that houses a large amount (~ 1x10⁹ cells/gram of soil) and diversity (1x10⁴ species/ gram of soil) of microorganisms (^{Curtis et al., 2002}). This edaphic microbiota plays a very important ecological part, offering several eco-systemic services, such as i) social and ecological sustainability, ii) adaptation to, and mitigation of, climate change, iii) biotechnological resource for humanity, iv) cycling of water and nutrients, and v) food security, mainly by the cycling of nutrients (^{van der Heijden et al., 2008}), and the boosting of plant growth through the production of phytohormones, solubilization of nutrients (^{Hayat et al., 2010}) and avoiding the establishment of phytopathogenic agents (^{Compant et al., 2005}). In this way, the use of biopesticides has acquired great relevance in the agricultural sector, offering a sustainable alternative, focused on increasing crop production. This generally implies the application of large populations of the microorganism of interest with the aim of promoting its establishment and colonization. However, this practice may cause disturbances in the microbial communities of the agro-systems (^{Trabelsi et al., 2013}), particularly when inoculating biological control agents, since their biological activity is not specific or selective for the phytopathogenic agent in question, which may cause unpredictable changes in the microbial structure of said agro-systems. It is therefore important to evaluate the impact of the inoculation of biological control agents on the structure and composition of the microbial communities in agro-systems, to guarantee the ecological balance, as well as the desired biological effect.

Different studies have shown the impact of the inoculation of strains with the ability for biological control on microbial communities in different crops. For example, ^{Li et al. (2015)} evaluated the effect of the B. subtilis strain B068150 on the microbial communities in the rhizosphere of cucumber plants using the DGGE (denaturing gradient gel electrophoresis) technique. The study was carried out using three different soil types, with no significant changes in the microbial diversity related to the cucumber rhizosphere, after the inoculation of strain B068150. On the other hand, ^{You et al. (2016)} reported that with the inoculation of B. subtilis Tpb55, not only was the Phytophthora parasitica infection reduced, but significant changes were also observed in the microbial community related to the rhizosphere of a tobacco plantation, with an ANDRA (Amplified ribosomal 16S rDNA restriction analysis), highlighting that the relative abundance of some communities was favored, essentially in those belonging to the dominant fila: Acidobacteria and Proteobacteria, in an increase of 2 and 10% respectively in regard to the control, yet reducing by at least 50% the relative abundance of the communities belonging to Planctomycetes, Nitrospirae, Bacteroidetes and Chloroflexi, which may be involved in an important activity for plant development. In this way, each BCA is a particular organism that performs its action in a specific manner, in which the studies of each microbial strain chosen must be explored in depth in order to acquire further knowledge on the way to strengthen its biological control with an effective formulation, considering aspects of ecological risk and biosafety for the agro-system.

Conclusions

The negative impacts of chemical pesticides on the environment has been widely documented, include health damages, resistance to compounds by phytopathogens, soil and water pollution, and produce the need to develop sustainable alternatives to protect agricultural soils against pathogens, for example, biological control agents.

The Bacillus genus presents a large metabolic diversity involved in the biological control of phytopathogens; given this, the academic and industrial sectors have concentrated on generating commercial formulae for their use in the field, and in the description of the main action mechanisms involved in this effect (competition for space and nutrients, antibiosis, production of lytic enzymes, secretion of toxins, inducing host resistance). However, it is unusual for a single action mechanism to be used by said antagonist for the suppression of phytopathogens in situ. In this way, the knowledge of the mechanisms with which the antagonist exerts its action is decisive for both the guarantee of its effect and for the development of commercial formulations, the success of which lies in the creation of microenvironments that boost their biological activity without stimulating the growth of the pathogen. Currently, several commercial formulations consider strains of the Bacillus genus as an active ingredient, due to its ability to colonize, reproduce easily and its high persistence related to the formation of endospores, with the latter being a characteristic of interest, since it allows it to live under conditions of abiotic stress, making its production and long-term storage easier. On the other hand, after the inoculation of biopesticides, it is possible to observe a dual effect on crops due to the action of biological control agents, mitigating phytopathogens and indirectly promoting plant growth with the improvement of the plant’s health. However, it is necessary to carry out basic studies in depth, which integrate other MIPE strategies (cultural practices, chemical control), as well as in the correct identification of pathogenic strains for humans, such as Bacillus cereus NS Bacillus anthracis, which are even potentially effective strains for the control of phytopathogens. The potential risk of these species may be detected with taxonomy studies, β-hemolytic activity, detection with defined virulence molecular markers, guaranteeing the use of biologically safe strains for agriculture. In addition, studies on the ecological impact of the introduction of a BCA to agro-systems must be developed, since under certain circumstances, they can cause changes in the microbial communities with agro-ecologically unpredictable results. Finally, the success of the use of biopesticides will depend largely on the innovation, research, marketing strategies and dissemination of results amongst producers and the bodies in charge of making decisions related to regulation and the use of this type of technologies.