No retraction, correction or concern found when last checked on Oct 17, 2026.
Overview of the Antimicrobial Compounds Produced by Members of the Bacillus subtilis Group
Simon Caulier, Catherine Nannan, Annika Gillis and 3 more
Frontiers in Microbiology · 2019 · Review · CC BY
Abstract
Over the last seven decades, applications using members of the Bacillus subtilis group have emerged in both food processes and crop protection industries. Their ability to form survival endospores and the plethora of antimicrobial compounds they produce has generated an increased industrial interest as food preservatives, therapeutic agents and biopesticides. In the growing context of food biopreservation and biological crop protection, this review suggests a comprehensive way to visualize the antimicrobial spectrum described within the B. subtilis group, including volatile compounds. This classification distinguishes the bioactive metabolites based on their biosynthetic pathways and chemical nature: i.e. ribosomal peptides (RPs), volatile compounds, polyketides (PKs), non-ribosomal peptides (NRPs) and hybrids between PKs and NRPs). For each clade, the chemical structure, biosynthesis and antimicrobial activity are described and exemplified. This review aims at constituting a convenient and updated classification of antimicrobial metabolites from the B. subtilis group, whose complex phylogeny is prone to further development.
Full text
From PubMed Central, U.S. National Library of Medicine, as current on Oct 17, 2026. Paragraph numbers are ours, so an answer can point at one.
Introduction
The genus Bacillus comprises 377 species1 (last update in January 2019) of Gram-positive, rod-shaped bacteria. Their ability to form endospores, their diversity in physiological properties, as well as their capacity to produce numerous antimicrobial compounds (AMCs) favor their ubiquitous distribution in soil, aquatic environments, food and gut microbiota of arthropods and mammals.
Bacteria from the Bacillus subtilis group consist of small vegetative cells (<1 μm-wide) for which the strain B. subtilis subsp. subtilis 168 is considered as model organism. They are usually mesophilic and neutrophilic, although some can tolerate high pH. The four original species of the group (B. subtilis, Bacillus licheniformis, Bacillus pumilus, and Bacillus amyloliquefaciens) were discovered more than 40 years ago. Since then, the evolution of their molecular, chemotaxonomic and physiological characterizations led to regular re-evaluations and (re-)description of numerous novel species and subspecies (see current taxonomy of the group in Figure 1).
The potential of B. subtilis group strains to produce a wide diversity of secondary metabolites mediating antibiosis was recognized for decades. For any given strain of the B. subtilis group, it is now estimated that at least 4–5% of its genome is devoted to antimicrobial compounds (AMCs) production. These molecules are mainly antimicrobial peptides (AMPs). Their structures are usually cyclic, hydrophobic and contain peculiar moieties such as D-amino acids (AA) or intramolecular thioether bonds. In addition to AMPs, volatile metabolites also constitute a large family of antimicrobials exhibiting numerous metabolic and functional roles.
Due to the wide diversity of these molecules, their classification is rather complex and can be based on several criteria such as their biosynthetic machinery, sources, biological functions, properties, three-dimensional structure, covalent bonding pattern or molecular targets. Here a classification of the B. subtilis group antimicrobial molecules is proposed, based on their biosynthetic pathways and their chemical nature as shown in Figure 2. This review will emphasize the biosynthesis pathway and the bioactivity of the main clades of AMCs within the B. subtilis group: i.e., the ribosomal peptides (RPs) (bacteriocins and enzymes), the polyketides (PKs), the non-ribosomal peptides (NRPs) and the volatiles. A full overview of this chart is provided as Supplementary Material (Supplementary Figure S1).
Ribosomal Peptides
Ribosomally synthesized peptides (RPs) are usually derived from short precursors (ca. 100 AA) and are processed to mature compounds through post-translational modifications. Various enzymes mediate these modifications and therefore generate a wide diversity of chemical structures. Most of these peptides were originally referred to as “bacteriocins,” characterized as low molecular weight molecules that exhibit inhibiting growth activities against bacteria closely related to the producing strain. In addition to bacteriocins, other types of enzymes exhibiting antagonistic activities are also ribosomally synthesized. However, those compounds display diverse metabolic activities such as quorum sensing (QS) mediation, cell lysis or induction of genetic competence. It should also be noted that molecules referred to as BLIS (bacteriocins-like inhibitory substances) include AMPs for which the ribosomal synthesis has not been confirmed yet.
B. subtilis Group Bacteriocins
It is estimated that 99% of the bacteria and archaea are able to produce at least one bacteriocin. Historically, lactic acid bacteria (LAB) were studied as main bacteriocin producers, mostly because of their long history of safe use in food fermentation. Nisin (Figure 3C), produced by Lactobacillus lactis subsp. lactis, was approved as a food additive in the 1960s and has since then been used in over 50 countries for its antimicrobial activity against Gram-positive pathogens such as Clostridium spp. and Bacillus spp.. However, the search for new bioactive molecules has rapidly expanded to other bacteriocin-producing genera, with a particular attention, in the late 1990s, to the GRAS (generally recognized as safe) Bacillus species whose bacteriocin antimicrobial spectra were broader than those of LAB.
The generic biosynthetic pathway of Bacillus species bacteriocins includes several post-translational modifications, including the proteolytic cleavage of the leader peptide at the N-terminal end. The modifications of active peptides, its secretion and the immunity to the bacteriocin (as described below) vary depending on the bacteriocin class.
While many classifications have been suggested over the years, one reasonable way to cope with the diversity of the Bacillus bacteriocins is to sort them on the basis of their biosynthetic pathway as previously reported for Streptococcus spp. and Enterococcus spp. bacteriocins and reviewed in. Accordingly, three main classes subdivided into several subclasses can be distinguished for the B. subtilis group. As detailed in Table 1, Class I includes the post-translationally modified peptides such as the lantibiotics whereas the non-modified peptides are grouped in Class II; Class III involved bacteriocins larger than 10 kDa. Supplementary Table S1 summarizes the different RPs produced by the strains belonging to the B. subtilis group, as well as their reported antimicrobial activities.
Class I includes small AMPs (19–38 AA) with extensive post-translational modifications. Subclasses I.1, I.2, and I.3 have in common their lantibiotic structure, which refers to inter-residual thioester bonds made of modified AA residues. As illustrated in Figure 3, lantibiotics involve 2,3-didehydroalanine (Dha) and (Z)-2,3-didehydrobutyrine (Dhb), resulting from the dehydration of serine and threonine residues, respectively. The intra-molecular addition of Dha or Dhb on a cysteine residue leads to the respective formation of lanthionine and methyllanthionine bridges. Subtilin (Figure 3B), from subclass I.1, is one of the most studied bacteriocins from the B. subtilis group. Its structure shares several similarities with nisin A lantibiotics, shown in Figure 3C. Peptides from subclass I.4 undergo other types of modifications. For instance, subtilosin A is a head-to-tail cyclic peptide with unusual inter-residue linkages (i.e., Cys-Phe bond).
Class II bacteriocins include small (<10 kDa), linear and non-modified peptides, resistant to heat and acido-basic treatments. They are divided in three subclasses based on a conserved AA motif near their N-terminus. The YGNGVXC (X is any AA) motif is associated to pediocin-like peptides from subclass II.1 whereas DWTXWSXL is specific to thuricin-like peptides from subclass II.2. Subclass II.3 comprises the small non-modified AMPs without any typical motif in their AA sequence. Finally, class III bacteriocins consist into large and heat labile molecules, generally characterized by a phospholipase activity.
Because of their wide diversity, bacteriocins display different modes of action such as protoplasm vesicularization, pore formation or cell disintegration. They are generally bactericidal with some exceptions that exhibit bacteriostatic activities. For most class I and II bacteriocins, the target of their activity is the bacterial envelope due to their amphiphilic or hydrophobic properties. For instance, lantibiotics from subclass I.1 have a dual mode of action. On the one hand, they can inhibit the cell wall synthesis of the targeted bacteria through binding to lipid II, the major transporter of peptidoglycan subunits across the inner cell membrane. On the other hand, lipid II can be used as a docking molecule to insert the lantibiotic in the membrane leading to pore formation and ultimately to cell death as well described in and. This duality has been reported for subtilin, a class I bacteriocin which is active against a broad range of Gram-positive bacteria such as Staphylococcus simulans, B. subtilis, and Bacillus stearothermophilus.
Many regulation systems mediate bacteriocin production, secretion and immunity. Bacteriocin production is usually linked to particular cellular events such as stress responses. For instance, subtilin production depends on cell density and is increased under starvation conditions. Lantibiotic production is also mediated by QS. For subtilin, it has been demonstrated that the peptide itself acts as an auto-inducer of its own production. The export of bacteriocins is generally ensured by a dedicated membrane-associated ATP-Binding Cassette (ABC) transporter. For some lantibiotics, the cleavage of the leader peptide often occurs in a proteolytic domain present in the ABC transporter as described in and. The immunity of the producing strains to its own active bacteriocin(s) can be achieved by several mechanisms like the secretion of immunity proteins sequestering the peptide, the bacteriocin re-export through an ABC transporter system or the alteration of the targeted peptidoglycans bonds (e.g., modification of the cell wall or cytoplasmic membrane charge).
B. subtilis Group AMP Enzymes
Among the B. subtilis group, two major types of enzymes exhibit antagonistic activities (Supplementary Table S1): the lytic enzymes and those involved in quorum quenching (QQ). Several strains from the B. subtilis group have indeed been identified as capable to produce lytic enzymes with biocontrol potential. They include cellulases, glucanases, proteases and chitinases and are generally referred to as cell wall degrading enzymes (CWDE). They are particularly active against fungi since chitin and glucan are the major constituents of their cell wall where various glycoproteins are embedded.
Quorum quenching is able to silence or block QS which is generally defined as the cell-to-cell communication mechanism through the production of signal molecules. N-acyl-homoserine lactones (AHLs), composed of a fatty acid side chain and a homoserine lactone (Figure 4) are the most characterized signal autoinducers in Gram-negative bacteria. When a bacterial population proliferates, concentration of AHLs increases so that all the cells coordinate their metabolic activities (e.g., biofilm formation, sporulation, virulence factors or antibiotic production). As the QS system brings ecological advantages to a coordinate population, QQ is able to counteract QS. Four types of enzymes (i.e., lactonase, decarboxylase, acylase, and deaminase) are able to inactivate AHLs, as illustrated in Figure 4. B. subtilis AHL-lactonases have for instance attracted interest for biocontrol since they affect the growth of deleterious microbial pest such as Pectobacterium carotovorum subsp. carotovorum causing potato soft rot.
Polyketides
Among the bioactive compounds produced by microorganisms, PKs are well known from the human health sector for their broad spectrum of activity encompassing antibacterial, immunosuppressive, antitumor and many more antagonistic abilities. Typical PKSs structures from the B. subtilis group are presented in Figure 5. They are synthetized from acyl CoA precursors such as malonate and methyl malonate. Their biosynthesis depends on multifunctional polyketide synthases (PKSs). Their structure was first extrapolated from fatty acid synthases (FASs) that share similarities in terms of chain extension mechanisms, precursors and overall architecture design. As shown in Figure 6A, PKS are composed of a succession of elongation modules, flanked by initiation and termination modules. The reactive mechanism of these three PKS domains is illustrated in Figure 7A and is well summarized in. The initiation module is composed of two domains: an acyltransferase (AT) domain that recruits and catalyzes the binding of a monomer substrate to an acyl carrier protein (ACP) domain. The ACP then acts as an arm with a second catalytic domain located on the next elongation module. This domain, a β-ketoacyl synthase (KS), catalyzes the chain-elongation reaction that occurs through a decarboxylative Claisen thioester condensation. In addition to the three core domains, auxiliary domains can also be present on elongation modules (gray domains in Figure 6A). These auxiliary domains mediate ketoreduction (KR), dehydration (DH), or enoylacyl reduction (ER) occurring before the chain-elongation reaction. These modifications considerably enrich the structural complexity and diversity of mature PKs. Finally, a termination module harboring an additional thiosterase (TE) domain catalyzes the macrolactonization and the release of the mature PK.
Polyketide synthases have been classified in three canonical types based on the structural organization of their functional domains. Type I PKSs involve large multifunctional enzymes housing several domains linearly arranged and covalently bonded. Type II PKSs are multienzyme complexes composed of separate monofunctional enzymes combined during the PK synthesis. Type III PKSs are chalcone synthase-like PKSs that operate the acid CoA thioesters directly without any ACP domain. Beside these structural differences, PKSs are classified as iterative or non-iterative depending on how many KS domains are used in the biosynthetic process. Within prokaryotes, the non-iterative type I PKSs is the most represented. They produce PK compounds that harbor a one-to-one correspondence with the PKS modular architecture. This conservation of collinearity is used for PKS discovery via genome mining.
Due to the diversity of PKSs, many exceptions and transition states between the three main types are observed. In some cases, mixed PKs pathways combine different types of PKSs or can even be associated with FASs or NRP synthetases (NRPSs) to form PK-peptide hybrid metabolites such as bacillaene, compactin, fusarin C or salinosporamide A.
To date, seven PKs families have been recognized based on their carbon skeletons and typical structures, as summarized in Table 2. However, to our knowledge, only three antimicrobial PKs and their variants are produced within the B. subtilis group: bacillaene, difficidin, and macrolactin. These compounds exhibit antibacterial activities through selective inhibition of protein synthesis (Table 3). Bacillaene is a polyene PK resulting from a hybrid synthesis by a type I PKS and a NRPS bae operon (baeJ, baeL, baeM, baeN and baeR). Its exhibits antimicrobial activity against various bacteria (e.g., Myxococcus xanthus or Staphylococcus aureus) and fungi (e.g., Trichoderma spp. or Fusarium spp.). Difficidin, and its oxidized form oxydifficidin, are polyenes synthesized by a type I PKS encoded in the dif operon. They both inhibit bacterial pathogens such as Clostridium perfringens, Erwinia amylovora, Escherichia coli or Xanthomonas oryzae. Finally, macrolactins and their 7-O-succinyl- or 7-O-malonyl-derivatives are synthetized via a type I PKS. They show antibacterial and antifungal activities against Burkholderia cepacia, Ralstonia solanacearum, S. aureus or Fusarium oxysporum. Some macrolactins, such as the macrolactin A, apparently also displays antiviral properties (e.g., against Herpes simplex viruses).
Non-Ribosomal Peptides
Non-ribosomal peptides form a versatile family of secondary metabolites with growing interest in many industrial fields as antibiotics, siderophores, surfactants, pigments, immunosuppressors or antitumor molecules. NRPs show a broad structural diversity, from linear to cyclic or branched structures. As illustration, the Norine database counts almost 1.200 NRP molecules, including their structure, synthesis and evolution2 (last update in January 2019).
Two categories of NRPs can be distinguished whether they are synthetized through a multi-enzyme thio-template mechanism or not. The first ones usually result in structures with two to ca. 50 residues and other moieties such as fatty acid chains [i.e., lipopeptides (LPs) and siderophores] whereas the second ones are generally smaller. Figure 8 shows the chemical structures of typical NRPS from the B. subtilis group.
Thiotemplate NRPs – Lipopeptides
Lipopeptides are usually synthetized through a NRPS sequential addition of AA residues, either in an iterative or non-iterative way. Similarly to PKSs, NRPSs have a modular organization implementing the initiation, elongation, and termination modules (Figure 6B). Each module is subdivided in core domains whose catalytic and carrier domains slightly differ from PKSs, as shown in Figure 7B. The biosynthesis which was previously summarized in and starts with an adenylation domain (A domain) that recruits and phosphorylates an AA monomer into an aminoacyl adenylate intermediate. The intermediate is then linked to the corresponding peptidyl carrier protein or thiolation domain (PCP or T domain) through a thioester bond. The PCP acts as a bridge and ensures the link with the condensation domain (C domain) that forms the C–N bond between the recruited aminoacyl and the peptide acyl chain in formation. The termination module contains a thioesterase domain (TE) that catalyzes the release of the final peptide acyl chain. The elongation modules can be supplemented with accessory domains such as cyclization domain (Cy), epimerization domain (E) and methylation domain (M). Those domains are able to modify the growing peptide chain which leads to diverse mature compounds structure.
Since the LP biosynthetic pathways are highly flexible, the range of produced LPs is extremely heterogeneous. Among LPs produced by Bacillus spp., four main families have been distinguished: kurstakins, surfactins, iturins, and fengycins. Each family shares the same structural features based on the nature and organization of the peptide moiety or fatty acid tail, as summarized in Table 4. Strains from the B. subtilis group produce surfactins, iturins and fengycins whereas kurstakins are produced by B. thuringiensis strains. Among the three LP families produced by B. subtilis, at least eight fengycins, 13 surfactins and 14 iturins variants have been described so far, as detailed in Supplementary Table S2.
For each LP family, the compounds production is mainly regulated by environmental factors such as carbon sources, oxygen availability, pH and temperatures. Warm temperature (≥37°C) and anaerobic conditions increase the production of surfactins while lower temperatures (25–37°C) and aerated bioreactors favor fengycins and iturins family metabolites. The production of surfactins by B. subtilis is also QS-dependent and involves ComX and PhrC. These pheromones trigger complex cascades regulating cell density-dependent processes such as sporulation and competence.
Iturins and fengycins are mainly known for their strong antifungal activity against several plant and human pathogenic fungi (Supplementary Table S2). In addition, iturin-like mycosubtilin, bacillomycin R, subtulene A and eumycin show antibacterial properties. Contrary to iturins and fengycins, surfactins mainly display antiviral and antibacterial activities. Their antiviral activity essentially targets enveloped viruses (e.g., herpes simplex or porcine epidemic diarrhea viruses). They also inhibit pathogenic bacteria such as Legionella pneumophila, Listeria monocytogenes, R. solanacearum or X. oryzae. However, some surfactins are able to control important fungal plant and human pathogens such as Botrytis cinerea, Candida albicans, F. oxysporum or Rhizoctonia solani.
The mere composition of LPs, where a peptide moiety is bound to a lipid tail, gives them an amphiphilic property. This nature makes them excellent surfactants and plays a significant role in their biological functions and antimicrobial properties. Indeed, LPs are able to destabilize the plasma membrane via a pore forming activity leading to the cell death of the target microbes. Their antiviral activity is the result of a similar disintegration of the bi-lipid envelope of virions explaining the weak LPs activity against plant viruses among which very few are enveloped.
Bacillus spp. LPs have many other biological and ecological functions as fully documented by. They are also known to impact other metabolic mechanisms such as biofilm formation, motility, virulence, plant root colonization, and plant defenses. Moreover, it has been suggested that their participation to the degradation of hydrophobic substrates could be used for polluted soils bioremediation. Although some lipopetides have already been exploited as food biopreservatives or crop protection products, the industrial interest for LPs in specific applications is unsurprisingly continuously growing.
Thiotemplate NRPs – Siderophore
Itoic acid is a mono-peptide composed of a 2,3-dihydroxybenzoate (DHB) molecule bound to a glycine. It is used as a precursor by trimodular NRPS machinery to produce bacillibactin which is obtained after a condensation of three units of DHB-glycine-threonine. The synthesis of the final hexapeptide is catalyzed by a terminal thioesterase domain leading to the production of a methylated trilactone ring link to three catecholates moieties. It is this cyclic structure that enables the sequestration of the metal atom. Itoic acid and bacillibactin are both catecholic siderophores that chelates iron reducing its bioavailability. This is limited access to iron that allows B. subtilis to antagonize the growth of other surrounding microbes such as, for instance, F. oxysporum f. sp. capsici.
Non-thiotemplate NRPs
Bacteria from the B. subtilis group are also able to synthesize other antimicrobial NRPs through non-thiotemplate mechanism. Rhizocticins are di- and tri-phosphono-peptides. They are constituted of a L-2-amino-5-phosphono-3-cis-pentenoic acid (APPA) linked to an arginine (rhizocticin A). They can be supplemented with an additional valine (rhizocticin B), isoleucine (rhizocticine C) or leucine (rhizocticine D). After their integration into the target microbes, their cleavage by host cell peptidases releases the fungitoxic L-APPA moiety that interferes with threonine metabolism in fungal cells. Interestingly, rhizocticin A has also an antagonistic activity against nematodes such as Caenorhabditis elegans.
In addition to rhizocticin compounds, two other dipeptide NRPs are produced by B. subtilis: bacilysin (also known as tetaine) and its chlorinated derivative, chlorotetain. They contain L-alanine (or chlorine-L-alanine) bound to the non-proteinogenic L-anticapsin. Despite their simple composition, these bioactive compounds display strong antibacterial activity mediated by the anticapsin moiety that inhibits the glucosamine-6-phosphate synthase. Its inhibition suppresses the biosynthesis of peptidoglycans that are the main constituents of bacterial cell wall. For the fungi, it has been proposed that because anticapsin is able to inhibit the production of chitin and fungal membrane mannoproteins, bacilysin and chlorotetain exhibit antifungal activity against Aspergillus fumigatus or C. albicans.
Finally, bacitracin and mycobacillin are two cyclic polypeptides produced by B. subtilis. Bacitracins are dodecapeptides containing a cyclic heptapeptide linked to a thiazoline ring. They are mostly active against Gram-positive bacteria where they inhibit the bacterial cell-wall biosynthesis by preventing the lipid carrier from re-entering in the reaction cycle of peptidoglycan synthesis. Besides this primary mode of action, bacitracin might also act through other mechanisms affecting membrane functions, hydrolytic enzymes and/or the biosynthesis of ubiquinone precursors. Mycobacillin is an antifungal cyclic tridecapeptide altering the membrane of fungi like Aspergillus niger. Interestingly, its biosynthesis is rather peculiar. Although it is catalyzed by a large NRPS complex, it is divided in three fractions (A, B, and C) and does not use a thio-template mechanism. Each fraction of the enzymatic complex contains a single enzyme polypeptide that catalyzes the polymerization of a first pentapeptide (A), a second nonapeptide (B) and the final tridecapeptide.
Volatiles
Besides RPs, NRPs and PKs, strains from the B. subtilis group are able to produce a wide diversity of volatile compounds encompassing important roles especially in soil, one of the major habitats of this group (Supplementary Figure S1). Volatiles are notably involved in the bioconversion of the food chain, in the biogeochemical cycles of essential elements, in many physiological and metabolic reactions (e.g., nitrification, nitrogen mineralization, electron acceptor or donor reactions) as well as in communication signals triggering QS/QQ or defense mechanisms well reviewed in. Volatile compounds are generally classified into inorganic (VICs) and organic (VOCs) categories.
Volatile Inorganic Compounds (VICs)
Volatile inorganic compounds synthesized by microorganisms are mainly by-products of primary metabolism. They are carbonated, hydrogenated, sulfur or nitrogen-containing compounds such as CO2, CO, H2, HCN, H2S, N2, NH3 and NO. Nitrogen-containing compounds are mostly released in aerated upper sediments layers by denitrifying bacteria. In this process, nitric oxide is enzymatically produced by the nitric-oxide reductase or the nitric-oxide synthase. The range of antimicrobial activities exhibited by VIC nitrogen-containing compounds from the B. subtilis group is wide. For instance, NO is able to induce systemic acquired resistance (SAR) in plants against bacterial pathogens such as R. solanacearum. A contrario, ammonia, a secondary metabolite from the catabolism of the amino acids L-aspartate, is known to be active against soil-borne Oomycetes such as Pythium spp.. Hydrogen cyanide, derived from the glycine catabolism, shows a direct antagonistic activity against aerobic microorganisms by inhibiting metal-containing enzymes such as the cytochrome c oxidase active in the respiration chain.
Deeper in the soil, under low oxygen concentration, bacteria tend to produce different VICs such as H2 or H2S. Those compounds can serve as electron acceptors, AA precursors or antimicrobial metabolites. Hydrogen sulfide could be produced by B. subtilis from sulfate reduction or as a by-product of L-methionine and L-cysteine catabolism via a direct cleavage of L-methionine or a transamination followed by reductive demethiolations. It is known to exhibit antifungal activity against several plant pathogens such as A. niger or Penicillium italicum but also against some food-borne bacteria or human pathogens. Curiously, it is also known to act as a bacterial defense mechanism against antibiotics. Interestingly, ammonia increases the resistance of several Gram-negative and Gram-positive bacteria to antibiotics too.
Volatile Organic Compounds (VOCs)
Volatile organic compounds are small compounds with fewer than 20 carbon atoms and are characterized by low molecular mass (100–500 Da), high vapor pressure, low boiling point and a lipophilic moiety. These features ensure an easy evaporation and a long distance distribution which is convenient in a complex matrix like soil. Their diffusion and production by soil-borne microbes are strongly dependent on various factors such as nutrient and oxygen availability, temperature, pH, physiological state of microorganisms, soil moisture, texture and architecture. The majority of VOCs derives from glucose oxidation involving glycolysis and the subsequent cycles such as the tricarboxylic acid cycle (TCA) as it has been well summarized in and. However, their production can also result from various other pathways such as aerobic heterotrophic carbon metabolism, fermentations, AA degradation, terpenes synthesis or sulfur reduction. Based on previous reviews presented in; and, five categories of VOCs can be distinguished: (1) fatty acids and derivatives, (2) terpenoids, (3) nitrogen-containing VOCs, (4) sulfur-containing VOCs, and (5) metalloid- or halogenated-containing VOCs. To date, about 2,000 compounds produced by almost 1,000 species of microorganisms have been listed in the mVOC 2.0 database. According to this database, almost 70% of recorded Bacillus VOCs are fatty acids derivatives (alcohols, ketones, alkanes, aldehydes, alkenes, and acids) followed by sulfur- and nitrogen-containing compounds. Supplementary Table S3 displays the VOCs produced within the B. subtilis group and their antimicrobial activity.
Since many volatile fatty acids and their derivatives result from the glucose metabolism, their precursors mostly derive from the Embden-Meyerhof (glycolysis), Entner-Doudoroff, heterolactic and homolactic fermentation pathways. B. subtilis bacteria, for instance, ferment pyruvate to produce ketone compounds such as acetoin (3-hydroxy-2-butanone) or 2,3-butanedione under anaerobic conditions. Other intermediates coming from fatty acid biosyntheses or their β-oxydations are also used as precursors by microbes and transformed into VOCs through a decarboxylation reaction or a reduction of their carboxyl group. They provide essential hydrocarbons but also other fatty acid derivatives. An oxidative deamination of several amino acids can lead to the production of aldehyde, ketone or alcohol volatile too. For instance, the degradation of L-phenylalanine or L-tyrosine can be the first step of the aromatic volatile compounds synthesis such as benzene or its carbohydrate derivatives. Finally, benzenoid volatiles can also be synthesized by microbes through the shikimate pathway that leads to the formation of chorismate, a natural precursor of aromatic amino acids. Degradation of intermediates from the shikimate pathway or aromatic amino acids can also lead to the production of benzenoid volatiles.
This wide variety of volatile fatty acids and their derivatives make them the most important group of VOCs produce by microbes and represent up to 87% of known antimicrobial VOCs produced by B. subtilis bacteria (Supplementary Table S3). They can be divided in two main categories: hydrocarbons (alkanes, alkenes, alkynes) or carbohydrates (acids, alcohols, aldehydes, esters, furans, ketones, lactones, benzenoids). Among them, benzenoids is the most represented sub-category followed by alkanes, aldehydes, ketones, acids, and alcohols. Even though benzenoids could be considered as an individual category, they can also be seen as fatty acids derivatives because a large majority of antimicrobial benzenoid volatile produced by B. subtilis harbor a benzene core linked to a fatty acid derivatives.
There is an important diversity of benzenoids, sometimes linked with carbohydrate chains containing nitrogen, sulfur or both. Most of these antimicrobial volatile exert fungicidal activities but some have been characterized for their antibacterial or nematicidal abilities, too. Their mode of action is rarely fully characterized. For instance, morphological abnormalities on fungal and bacterial cells have been documented after an exposition to B. subtilis VOCs. Volatile such as 1,3-butadiene or 2,3-butanediol are also known to induce modifications in the expression of genes linked to the pathogenicity of R. solanacearum and Pectobacterium carotovorum. In addition to direct antimicrobial activities, fatty acids volatile have also several other biological functions. For instance, acetoin and 2-butanone have the ability to stimulate plant defenses or to induce plant stress tolerance which then promote plant growth. They are essentially produced by strains of B. amyloliquefaciens, B. velezensis or B. subtilis.
Terpenes and their derivatives (also known as terpenoids or isoprenoids) are among the most abundant secondary metabolites found in living systems. They originate from two main precursors: isopentenyl pyrophosphate (IPP) and its allylic isomer the dimethylallyl pyrophosphate (DMAPP). IPP and DMAPP are also the end-products of the deoxy-xylulose phosphate pathway (DOXP) starting with pyruvate and glyceraldehyde-3-phosphate originating from the glucose metabolism. Terpenoids can be synthesized from isoprene molecules too. showed that, in B. subtilis, isoprene is not formed by the MVA or DOXP pathways but, as in plant systems, might be a product of the methylerythritol phosphate (MEP) pathway.
Isoprenoid compounds are produced by all living organisms for essential physiological functions such as electron transport, membrane fluidity, light harvesting, photoprotection, anchoring of molecules to specific membranes and signaling. The signaling ability is particularly important and is associated with several antagonistic, mutualistic or multi-trophic interactions. More than 25,000 terpenic compounds have been listed and, for the vast majority, their biological functions and roles remain unknown. Volatile terpenes are generally recognized for their ability to inhibit bacteria, fungi, nematodes or insects. They can be classified in three categories: isoprene, monoterpenes (C10) and sesquiterpenes (C15).
The mode of action of these compounds might be linked to their lipophilic nature allowing them to destabilize the cell membrane integrity. To our knowledge, only two terpenes produced by B. subtilis show antimicrobial abilities: isoprene and monoterpene α-terpineol exhibit antagonistic activities against cyanobacteria and nematodes.
Little is known about the biosynthetic pathways of nitrogen-containing VOCs. Nevertheless, it is accepted that two main routes can be used: a non-enzymatic amination of acyloins, that can lead to the formation of pyrazines or derived from α-aminoketone intermediates resulting from AA catabolism.
Nitrogen-containing VOCs can be distinguished based on their cyclization rate. Within non-cyclic compounds, three groups are identified (amides, amines and imines) while there are five categories of cyclic compounds (azoles, pyrazines, pyridines, pyridazines, and pyrimidines). Pyrazines are strongly represented among microbial volatile and are separated in two classes: lower-alkylated and higher-alkylated pyrazines. These compounds are characterized by a strong odor and several B. subtilis coming from the rhizosphere or from food fermentations have already been recognized as pyrazines producers. Pyrazines from B. subtilis strains are known to exhibit antifungal and nematicidal activities. For instance, tetramethylpyrazine inhibits the growth of Moniliophthora perniciosa and F. oxysporum f. sp. lactucae. Additionally, it acts on sporulation and elongation of the germ-tube of B. cinerea. It is interesting to note that B. subtilis pyrazines can also exhibit antibacterial activities such as pulcherriminic acid which inhibits the growth of S. aureus, E. coli and Proteus vulgaris. Beside pyrazines, strains from the B. subtilis group are able to produce other nitrogen VOCs such as 1H-imidazole,1-ethyl showing antifungal activities against numerous soil-borne phytopathogens.
Microbial VOCs containing sulfur (VSCs) derive from two main pathways originated from inorganic or organic sources: inorganic sulfate reduction in methylated inorganic sulfides compounds or, for some microbial VSCs, originate from catabolism of AA such as L-methionine or more rarely, L-cysteine. Some VSCs are produced as secondary volatiles via the production of hydrogen sulfide or methanethiol. Indeed, these two compounds are important precursors for subsequent VSCs synthesis. Within the B. subtilis group, multiple VSCs such as dimethyl disulfide (DMDS), dimethyl trisulfide (DMTS), S-methyl thioacetate or S-methyl butanethioate have been characterized for their antifungal and nematicidal activities. A putative antibacterial effect of DMDS is not to exclude. Indeed, DMDS is known to affect the bacterial cell-to-cell communications through a decrease in the amount of N-acyl homoserine lactone (AHL) mediating QS.
Other volatile organic compounds such as halogenated, metalloids, tellurium or selenium compounds have also been described. However, at the time of writing, no B. subtilis strains have been proved to produce these type of VOCs, although related bacteria, like Bacillus arsenicoselenatis, have been shown to generate them.
Conclusion and Perspectives
The B. subtilis group offers a plethora of antagonistic compounds displaying a broad range of biological functions. This huge versatility increases the industrial and environmental interest of B. subtilis strains, especially when considering their range of action against foodborne or phytopathogenic flora as well as their history of safe use in food. The present review on known AMCs from the B. subtilis group proposes a consistent classification frame based on their biosynthetic pathways (i.e., RPs, PKs, NRPs, volatiles) and chemical nature.
The present classification suggests to establish systematic approaches for novel molecules discoveries and characterizations (biosynthesis, chemical nature and activity). Indeed, most current publications report antimicrobial activity of partially purified fractions which can involve mixtures of bioactive compounds. To assess the activity of an unique compound, implementations of genetic confirmation such as knockout strategy are needed. Besides, very few studies have focused on the putative synergistic effects within these bio-active mixtures. Also, the concentration of purified or semi-purified compound(s) often remains uncharacterized or biologically irrelevant. Finally, there is no doubt that novel AMCs originating from B. subtilis bacteria remain to be identified, characterized and properly classified.
Author Contributions
SC, CN, and FL conducted the bibliographic search. SC and CN wrote the manuscript. AG, CB, and JM edited and reviewed the manuscript. All authors have read and approved the final version.
Conflict of Interest Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References 246
- 1.
Abriouel H. Franz C. M. Omar N. B. Gálvez A. (2011 ). Diversity and applications of Bacillus bacteriocins. FEMS Microbiol. Rev. 35 201–232.
Not in our library10.1111/j.1574-6976.2010.00244.x
- 2.
Adak S. Aulak K. S. Stuehr D. J. (2002 ). Direct evidence for nitric oxide production by a nitric-oxide synthase-like protein from Bacillus subtilis. J. Biol. Chem. 277 16167–16171.
Not in our library10.1074/jbc.m201136200
- 3.
Agrios G. N. (1988 ). “1 - INTRODUCTION TO PLANT PATHOLOGY,” in Plant Pathology, 3rdEdition Edn, ed. Agrios G. N. (New York, NY: Academic Press), 3–39.
Not in our library10.1016/b978-0-12-044563-9.50005-0
- 4.
Alamri S. A. (2015 ). Enhancing the efficiency of the bioagent Bacillus subtilis JF419701 against soil-borne phytopathogens by increasing the productivity of fungal cell wall degrading enzymes. Arch. Phytopathol. Plant Prot. 48 159–170.
Not in our library10.1080/03235408.2014.884671
- 5.
Aleti G. Sessitsch A. Brader G. (2015 ). Genome mining: prediction of lipopeptides and polyketides from Bacillus and related Firmicutes. Comput. Struct. Biotechnol. J. 13 192–203.
Not in our library10.1016/j.csbj.2015.03.003
- 6.
An J. Zhu W. Liu Y. Zhang X. Sun L. Hong P. (2015 ). Purification and characterization of a novel bacteriocin CAMT2 produced by Bacillus amyloliquefaciens isolated from marine fish Epinephelus areolatus. Food Control 51 278–282.
Not in our library10.1016/j.foodcont.2014.11.038
- 7.
Arguelles Arias A. Ongena M. Devreese B. Terrak M. Joris B. Fickers P. (2013 ). Characterization of amylolysin, a novel lantibiotic from Bacillus amyloliquefaciens GA1. PLoS One 8:e83037.
Not in our library10.1371/journal.pone.0083037
- 8.
Ariffin H. Abdullah N. Md Shah U. K. Shirai Y. Hassan M. A. (2006 ). Production and characterization of cellulases by Bacillus pumilus EB3. Int. J. Eng. Technol. 3 47–53.
Not in our library
- 9.
Arrebola E. Sivakumar D. Korsten L. (2010 ). Effect of volatile compounds produced by Bacillus strains on postharvest decay in citrus. Biol. Control 53 122–128.
Not in our library10.1016/j.biocontrol.2009.11.010
- 10.
Audrain B. Farag M. A. Ryu C.-M. Ghigo J.-M. (2015 ). Role of bacterial volatile compounds in bacterial biology. FEMS Microbiol. Rev. 39 222–233.
Not in our library10.1093/femsre/fuu013
- 11.
Bais H. P. Fall R. Vivanco J. M. (2004 ). Biocontrol of Bacillus subtilis against Infection of Arabidopsis Roots by Pseudomonas syringae is facilitated by biofilm formation and surfactin Production. Plant Physiol. 134 307–319.
Not in our library10.1104/pp.103.028712
- 12.
- 13.
Béchet M. Caradec T. Hussein W. Abderrahmani A. Chollet M. Leclère V. (2012 ). Structure, biosynthesis, and properties of kurstakins, nonribosomal lipopeptides from Bacillus spp. Appl. Microbiol. Biotechnol. 95 593–600.
Not in our library10.1007/s00253-012-4181-2
- 14.
Begley M. Cotter P. D. Hill C. Ross R. P. (2009 ). Identification of a novel two-peptide lantibiotic, lichenicidin, following rational genome mining for LanM proteins. Appl. Environ. Microbiol. 75 5451–5460.
Not in our library10.1128/aem.00730-09
- 15.
Bentley R. Haslam E. (1990 ). The shikimate pathway — a metabolic tree with many branche. Crit. Rev. Biochem. Mol. Biol. 25 307–384.
Not in our library10.3109/10409239009090615
- 16.
Bernier S. P. Létoffé S. Delepierre M. Ghigo J.-M. (2011 ). Biogenic ammonia modifies antibiotic resistance at a distance in physically separated bacteria. Mol. Microbiol. 81 705–716.
Not in our library10.1111/j.1365-2958.2011.07724.x
- 17.
Besson F. Peypoux F. Michel G. Delcambe L. (1976 ). Characterization of iturin A in antibiotics from various strains of Bacillus subtilis. J. Antibiot. 29 1043–1049.
Not in our library10.7164/antibiotics.29.1043
- 18.
Besson F. Peypoux F. Michel G. Delcambe L. (1979 ). Antifungal activity upon Saccharomyces cerevisiae of iturin A, mycosubtilin, bacillomycin L and of their derivatives; inhibition of this antifungal activity by lipid antagonists. J. Antibiot. 32 828–833.
Not in our library10.7164/antibiotics.32.828
- 19.
Bowman S. M. Free S. J. (2006 ). The structure and synthesis of the fungal cell wall. Bioessays 28 799–808.
Not in our library10.1002/bies.20441
- 20.
Brötz H. Bierbaum G. Markus A. Molitor E. Sahl H. G. (1995 ). Mode of action of the lantibiotic mersacidin: inhibition of peptidoglycan biosynthesis via a novel mechanism? Antimicrob. Agents Chemother. 39 714–719.
Not in our library10.1128/aac.39.3.714
- 21.
Buckingham J. (1997 ). Dictionary of Natural Products. London: CRC press
Not in our library10.1007/978-1-4899-6850-0
- 22.
Caboche S. Pupin M. Leclère V. Fontaine A. Jacques P. Kucherov G. (2008 ). NORINE: a database of nonribosomal peptides. Nucleic Acids Res. 36 D326–D331.
Not in our library10.1093/nar/gkm792
- 23.
Cane D. E. Walsh C. T. (1999 ). The parallel and convergent universes of polyketide synthases and nonribosomal peptide synthetases. Chem. Biol. 6 R319–R325.
Not in our library10.1016/s1074-5521(00)80001-0
- 24.
- 25.
- 26.
Challis G. L. (2008 ). Genome mining for novel natural product discovery. J. Med. Chem. 51 2618–2628.
Not in our library10.1021/jm700948z
- 27.
Challis G. L. Naismith J. H. (2004 ). Structural aspects of non-ribosomal peptide biosynthesis. Curr. Opin. Struct. Biol. 14 748–756.
Not in our library10.1016/j.sbi.2004.10.005
- 28.
Chatterjee C. Paul M. Xie L. van der Donk W. A. (2005 ). Biosynthesis and mode of action of lantibiotics. Chem. Rev. 105 633–684.
Not in our library10.1021/cr030105v
- 29.
Chaves-López C. Serio A. Gianotti A. Sacchetti G. Ndagijimana M. Ciccarone C. (2015 ). Diversity of food-borne Bacillus volatile compounds and influence on fungal growth. J. Appl. Microbiol. 119 487–499.
Not in our library10.1111/jam.12847
- 30.
Chen H. Du L. (2016 ). Iterative polyketide biosynthesis by modular polyketide synthases in bacteria. Appl. Microbiol. Biotechnol. 100 541–557.
Not in our library10.1007/s00253-015-7093-0
- 31.
Chen H. Xiao X. Wang J. Wu L. Zheng Z. Yu Z. (2008 ). Antagonistic effects of volatiles generated by Bacillus subtilis on spore germination and hyphal growth of the plant pathogen, Botrytis cinerea. Biotechnol. Lett. 30 919–923.
Not in our library10.1007/s10529-007-9626-9
- 32.
Chen X. Scholz R. Borriss M. Junge H. Mogel G. Kunz S. (2009 ). Difficidin and bacilysin produced by plant-associated Bacillus amyloliquefaciens Dare efficient in controlling fire blight disease. J. Biotechnol. 140 38–44.
Not in our library10.1016/j.jbiotec.2008.10.015
- 33.
Chen X. Vater J. Piel J. Franke P. Scholz R. Schneider K. (2006 ). Structural and functional characterization of three polyketide synthase gene clusters in Bacillus amyloliquefaciens FZB42. J. Bacteriol. 188 4024–4036.
Not in our library10.1128/jb.00052-06
- 34.
Cherif-Silini H. Silini A. Yahiaoui B. Ouzari I. Boudabous A. (2016 ). Phylogenetic and plant-growth-promoting characteristics of Bacillus isolated from the wheat rhizosphere. Ann. Microbiol. 66 1087–1097.
Not in our library10.1007/s13213-016-1194-6
- 35.
Chernin L. Toklikishvili N. Ovadis M. Kim S. Ben-Ari J. Khmel I. (2011 ). Quorum-sensing quenching by rhizobacterial volatiles. Environ. Microbiol. Rep. 3 698–704.
Not in our library10.1111/j.1758-2229.2011.00284.x
- 36.
Chopra L. Singh G. Choudhary V. Sahoo D. K. (2014 ). Sonorensin: an antimicrobial peptide, belonging to the heterocycloanthracin subfamily of bacteriocins, from a new marine isolate, Bacillus sonorensis MT93. Appl. Environ. Microbiol. 80 2981–2990.
Not in our library10.1128/aem.04259-13
- 37.
Chopra L. Singh G. Jena K. K. Verma H. Sahoo D. K. (2015 ). Bioprocess development for the production of sonorensin by Bacillus sonorensis MT93 and its application as a food preservative. Bioresour. Technol. 175 358–366.
Not in our library10.1016/j.biortech.2014.10.105
- 38.
Cleveland J. Montville T. J. Nes I. F. Chikindas M. L. (2001 ). Bacteriocins: safe, natural antimicrobials for food preservation. Int. J. Food Microbiol. 71 1–20.
Not in our library10.1016/s0168-1605(01)00560-8
- 39.
Coosemans J. (2005 ). Dimethyl disulphide (DMDS): a potential novel nematicide and soil disinfectant. Acta Hortic. 698 57–64.
Not in our library10.17660/actahortic.2005.698.6
- 40.
Cosby W. M. Vollenbroich D. Lee O. H. Zuber P. (1998 ). Altered srf expression in Bacillus subtilis resulting from changes in culture pH is dependent on the Spo0K oligopeptide permease and the ComQX system of extracellular control. J. Bacteriol. 180 1438–1445.
Not in our library10.1128/jb.180.6.1438-1445.1998
Cited by papers we hold 2
Reads
Read by people on this page
None yet.
Checked or read by AI through naulon
None yet.
A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.
Are you an author?
Is this your paper?
Claim it to add your note and be paid when agents cite it. Free.