Background

Nonribosomal peptides is a diverse family of natural products fall into the class of secondary metabolites with a diverse properties as toxins, siderophores, pigments, antibiotics, cytostatics, immunosuppressants or anticancer agents [1, 2]; and have a particularity: their synthesis is independent of ribosomal machinery. Soil-inhabiting microorganisms, such as Actinomycetes and Bacilli, and eukaryotic filamentous fungi are mostly producers of nonribosomal peptides, but marine microorganisms have also emerged as a source for such peptides [3]. These peptides have a structural features such as contain amino acids like ornithine or imino acids, and their structures are macrocyclic, branched macrocyclic, dimers or trimers of identical structural elements [4]. Usually nonribosomal peptides are synthesized on large nonribosomal peptide synthetase (NRPS) enzyme complexes, defined as modular multidomain enzymes; nevertheless more than half of the NRPS enzymes found in a genome-mining study of 2699 genomes by Wang et al. [1] are nonmodular NRPS enzymes. Nonmodular NRPS enzymes are found in siderophore biosynthetic pathways like EntE and VibH in enterobactin, and VibE in vibriobactin [5] or as a stand-alone peptidyl carrier protein such as BlmI from the bleomycin gene cluster [6]. The presence of NRPS enzymes has been reported in the three domains of life, being prevalent in bacteria, less frequent in eukarya and rare in archaea. Within bacteria domain, Proteobacteria, Actinobacteria, Firmicutes, and Cyanobacteria were the phyla with major abundance of these enzymes, and there has been observed a correlation between genome size and the number of NRPS’s clusters [1]. The key features of the microbial biosynthesis of nonribosomal peptides, the structure and regulation of NRPS enzymes, and some applications in industry are summarized below.

Biosynthesis of nonribosomal peptides

The biosynthesis of nonribosomal peptides is done by NRPSs which are modularly organized multi-enzyme complexes which serve as templates and biosynthetic machinery, via a thiotemplate mechanism independent of ribosomes [2]. A module is defined as a section of the NRPS enzyme that incorporate in a specific manner one amino acid into the peptide backbone, and in turn the modules can be divided into domains, which catalyze the individual steps of nonribosomal peptide synthesis. Each module consist of three domains, adenylation (A) domain, peptidyl carrier protein (PCP) or thiolation (T) domain, and condensation (C) domain which carry out the synthesis of nonribosomal peptides (Fig. 1) [7]. The order of modules is usually co-linear to the product peptide sequences [8, 9]. Nonribosomal peptides synthesis proceeds in a N- to C-terminal direction, producing peptides that are usually about 3–15 amino acids in length, and the released peptides can be linear, cyclic, or branched-cyclic [3]. There is a sequence motif conserved in domains that facilitates the identification of these using sequences search tools like BLAST, as in the study done by Etchegaray et al. [10] to identified NRPSs in the genome of Xanthomonas axonopodis and X. campestri. The first step in the biosynthesis is done by the A-domain, which recognizes and performs activation of amino acid substrate via adenylation using Mg-ATP, result in an aminoacyl adenylated intermediate [11]. The A-domain consist of ~550 amino acids, and has ten amino acid residues that can be considered as the “codons” of NRPS enzymes and are important to substrate specificity [12]. The substrates that can be recognized by A-domain can include D and L forms of the 20 amino acids used in ribosomal proteins synthesis, as well as non-proteino-genic amino acids such as ornithine, imino acids, and hydroxy acids such as α-aminoadipic and β-butyric acids [2]. The reaction carried out by the A-domain shares the same chemistry as that performed by aminoacyl-tRNA synthetases (AARSs), which act in the initial step of ribosome-driven protein synthesis [8, 13]. The second step is performs by the PCP-domain of ~80 amino acids [12], which covalently binds the activated amino acid to their cofactor 4′-phosphopantetheine (PP) arm via a thioester bond and transfer the activated substrate and the elongation intermediates to C-domain [14]. The final step is carried out by the C-domain of ~450 amino acids, which catalyzes peptide bond formation between the carboxyl group of the nascent peptide and the amino acid carried by the flanking module, allowing the translocation of the growing chain to the following module [12, 13]. After condensation step, the linear intermediate peptide is release with the help of thioesterase (TE) domain by either hydrolysis or internal cyclization in bacterias, and less frequent in NRPSs of fungi. In fungi the chain release is carried out by a variety of mechanisms, two of which are (1) a terminal C domain, which catalyzes release by inter- or intra-molecular amide bond formation, and (2) a thioesterase NADP(H) dependent reductase (R) domain which catalyzes reduction with NADPH to form an aldehyde [2]. The primary product of this synthesis may be post-synthetically modified to achieve its mature form by additional tailoring enzymes which are not part of NRPS by N-, C- and O-methylation, glycosylation, hydroxylation, acylation, halogenation or heterocyclic ring formation [3, 15, 16]. These tailoring enzymes and their reactions contribute to generate the structural diversity of nonribosomal peptides [10].

Fig. 1
figure 1

Structural features of nonribosomal peptide synthetase enzymes. Nonribosomal peptide synthetase enzymes can be subdivided in modules, each incorporating one amino acid. Each module consist of three domains: adenylation (A) domain, peptidyl carrier protein (PCP) domain, and condensation (C) domain, which carry out the synthesis of nonribosomal peptides; the epimerization (E) domain and thioesterase (TE) domain is also represents. In this figure, surfactin biosynthesis was exemplified

Full size image

Structure of nonribosomal peptide synthetase (NRPS) enzymes

NRPSs are defined as multimodular enzymes, consisting of repeated modules with A-PCP-C domains [17]. Multimodular NRPS proteins are frequently in fungal genomes [3], and all modules that participate in the assembly of a peptide are connected within a single enzyme [10]. Tandem duplication and recombination is may be the origin of modules of multimodular NRPS enzymes, as in the case of SimA (Cyclosporin synthetase) enzyme of Tolypocladium inflatum and Enniatin synthetase of Fusarium equiseti, respectively [2]. Nevertheless, more than half of the NRPS enzymes finding in a genome-mining study of 2699 genomes are nonmodular NRPS enzymes, being common in bacterial systems where they are organized in clusters [1, 3, 10]. Nonmodular NRPS enzymes consisting of one or two A-PCP-C modules, or lacked complete A-PCP-C modules and consist of a single A-domain or an A-PCP unit followed by a variety of C-terminal domains [2]. In bacterial systems, mono or bi modular NRPSs can interact with others NRPS proteins and performs biosynthesis by first activating the amino acid and then transferred the activated substrate either to a C domain in the same NRPS or in a different NRPS, known as nonlinear biosynthesis [3]. Taxonomic distributions of mono/bi modular NRPS subfamilies suggest an ancient origin, possibly predating the divergence of eubacteria and fungi [2]. There is third structure of NRPSs which is fused to a polyketide synthase (PKS) unit, both types of synthetases are fused in a single polypeptide [10]. Of 3339 gene clusters encoding NRPS and PKS found by Wang et al. [1], one-third (1147) of gene clusters belonged to the hybrid type and encoded 462 hybrid proteins that contain both NRPS and PKS core domains. PKSs are large megasynthases related to fatty acid synthases, that biosynthesize small molecule polyketides with diverse natural function as nonribosomal peptides [2].

Regulation of NRPS genes

Global regulators such as DegU and Comp/ComA two-component system can act at transcriptional level, regulating positively the expression of srfA, bac, and bmy cluster genes of Bacillus spp., which encodes surfactin (Fig. 1), bacilysin and bacillomycin, respectively [18–20]. In the case of srfA genes that encoding for the surfactin, the transcriptional initiation is through ComX pheromone that can activated ComP, causing ComP to autophosphorylate and, subsequently, donate a phosphate group to ComA. Upstream of the srfA promoter there are ComA boxes, which are recognized by phosphorylated ComA and binding to them as a tetramer initializing the transcription of srfA [18, 19, 21]. Another positive regulator of transcription of srfA is PerR protein, which positively modulates srfA expression by binding to regions located upstream of ComA boxes, known as PerR boxes [22]. Expression of the bac operon is dependent on a σA-dependent promoter, which is activated by interaction with DegU at the final stage of vegetative growth. Binding of DegU to the bac operon promoter occurred mainly at three sites, in a similar way that occur with bmy operon, in which DegU binds directly to two sites located upstream of the bacillomycin D promoter [20, 23]. In the opposite case, a down-regulation is exerted on the bac operon through ScoC protein, which binds at the bac promoter sequence in ScoC boxes located between positions −50 and −42 (ScoC box1) and between positions −12 and −4 (ScoC box2) [24]. While srfA is repressed by CodY at high concentrations of amino acids Ile, Leu, and Val. AbrB and Spx are two other negative regulators that turn off the transcription of srfA genes [19]. In the case of SrfA NRPS enzyme, is necessary a post-translational modification to become active; 4′-phosphopantetheinyl transferase (Sfp/PPTase) is required for the activation of SrfA enzymes by converting the inactive apo-forms to active holo-forms [25]. In the case of Streptomyces peucetius, the biosynthesis of non-ribosomal peptide doxorubicin, an antitumor drugs, is up regulated by the networking of transcriptional regulators dnrO, dnrN, and dnrI. The product of dnrO gene binds to the dnrN/dnrO promoter region and activates dnrN. DnrN activates the transcription of dnrI gene and, DnrI activates the transcription of the doxorubicin biosynthesis genes. Down regulation of doxorubicin is indirected controlled by doxR regulator, a gene belonging to the IclR family of transcripional regulators, which inhibits the dnrI expression, leading to blockade of doxorubicin production [26].

Environmental applications

Surfactants are amphipathic molecules with both hydrophilic and hydrophobic moieties. These properties render surfactants capable of reducing surface and interfacial tension and forming microemulsion. Such characteristics confer excellent detergency, emulsifying, foaming, and dispersing traits; they are the active ingredients found in soaps and detergents [27]. Surfactants currently in use are chemically derived from petroleum; however, interest in microbial surfactants has been steadily increasing in recent years due to their diversity, environmentally friendly nature, and their potential applications in the environmental protection, crude oil recovery, health care, and food-processing industries [28–30]. Biosurfactants are biological surface-active compounds largely produced by a wide variety of microorganisms, secreted either extracellularly or attached to parts of cells, predominantly during growth on water-immiscible substrates. Bacteria of various genera such as Pseudomonas, Bacillus, Acinetobacter, Arthrobacter, and Rhodococcus are able to produce biosurfactants during hydrocarbon oxidation [31]. Biosurfactants have several properties and advantages over the chemical surfactants, such as lower toxicity, higher biodegradability, better environmental compatibility, higher foaming, high selectivity and specific activity at extreme temperatures, pH and salinity [28]. Biosurfactants are capable of lowering surface and interfacial tensions effectively and thus are potential substitutes for widely used chemically synthesized surfactants, they have very low critical micelle concentrations (CMC), this means that are effective at low concentrations. In general, the structure of these molecules includes a hydrophobic portion commonly made up of fatty acids (saturated, unsaturated, or hydroxylated), whereas the hydrophilic portion is usually composed of peptides or mono-, di-, or polysaccharides [26]. The major classes of biosurfactants include glycolipids, lipopeptides, lipoproteins, phospholipids, polysaccharide-lipid complexes, hydroxylated and cross-linked fatty acids, and the complete cell surface [27]. One of the potential uses is in the oil industry, in which case whole-cell broth could be used with minimum purity specification and required in small quantities to oil recovery from underground sandstone. Another use of biosurfactants is in remediation of hydrocarbon and crude oil-contaminated soils (Fig. 2b); the addition of biosurfactant increase the bioavailability of petroleum hydrocarbon pollutants in soil to stimulate the indigenous bacterial population to degrade hydrocarbons at rates higher [32]. One example is the use of rhamnolipid biosurfactant from Pseudomona aeruginosa that removed oil from contaminated Alaskan gravel from the Exxon Valdez oil spill [33]. From studies conducted by Urum and Pekdemir [27], it is noted that biosurfactants were able to remove significant amount of crude oil from contaminated soil, for instance rhamnolipid removed up to 80 % oil and lecithin about 42 %. While in the studies done by Lai et al. [32], was shown that rhamnolipid from P. aeruginosa and surfactin from B. subtillis have a higher removal efficiency from a heavy oil-polluted site, than the chemicals surfactants Tween 80 and triton X-100. In the case of surfactin, has been reported that its production has reached concentrations of 10.26 g/L in a medium containing starch [34] 2933 times higher than that achieved by one report by Ponte Rocha et al. [35]. Rhamnolipid production using P. aeruginosa mutant strains grown in blackstrap molasses with or without supplementary nitrogen source was of 1.45 g/L after 96 h incubation [36]. While Silva et al. [37] using P. aeruginosa UCP0992 grown in 100 ml mineral medium (aeration of 80 %) supplemented with 3 % glycerol and 0.6 % NaNO3, as the nitrogen source, at 28 °C after 96 h reached a production of 8.0 g/L; similarly Wu et al. [38] using an indigenous strain P. aeruginosa EM1 originating from an oil-contaminated site located in southern Taiwan grown in inorganic nitrogen (NaNO3) obtained a productivity of 8.63 g/L. To avoid the inconvenience of working with opportunistic pathogen strains like P. aeruginosa, attempts have been made to express biosurfactants using non-pathogenic strains of bacteria. Wittgens et al. [39] using heterologous expression in Pseudomonas putida KT42C1, a strain certified as safety, produced up to 1.5 g/L of rhamnolipid; while Ochsner et al. [40] reached a production of 0.6 g/L in a recombinant P. putida strain KT2442. Escherichia coli has also been used to production of biosurfactants, as the case reported by Wang et al. [41], with the engineered E. coli TnERAB that produced 65–80 mg/L in MS plus glucose media, and 150–185 mg/L in LB plus glucose media, respectively.

Fig. 2
figure 2

Nonribosomal peptides and some applications. a Nonribosomal peptide molecules of surfactin and bacitracin. b Some applications of nonribosomal peptides in environmental remediation and human health

Full size image

Human health applications

Since the discovery of penicillin in 1928 by Alexander Fleming to 1940, the efforts to produce penicillin have conducted the biotechnology sector into a billion dollar industry with deep-tank fermentations at its core [42]. The fungi Penicillium chrysogenum is the organism utilized to produce penicillin at industrial scale. Penicillins are formed from the amino acids valine, cysteine and α-aminoadiapate and include residues such as pheniylacetyl [43]. The penicillin biosynthetic pathway encompasses δ-(L-α-aminoadipyl)-L-cysteinyl-d-valine synthetase (ACVS), isopenicillin N synthase (IPNS), isopenicillin N acyl transferase (IAT) and phenylacetyl CoA ligase (PCL), with the ACVS belonging to a class of NRPSs that exclusively occurs in certain filamentous fungi and bacteria (Actinomycetes, Bacilli) [44]. The production of penicillin have been reported in bioreactors of 100,000 L, achieving 36 g/L at 250 h [42], although 50 g/L of penicillin can be produced [45]. Therefore, the penicillin fermentation process is a good case of a development strategy model to follow into a large scale of nonribosomal peptide process production. However, the residual concentrations obtained of many NRP are from one to three orders of magnitude below compared to that of penicillin. For example, the cyclic undecapeptide Cyclosporin A is synthetized by cyclosporine synthetase one of the most complex and largest modular enzymes described [46]. Cylcosporine A is produced by fungus Tolypocladium inflatum, Beauveria nivea, Fusarium roseum, and Tolypocladium niveum and have anti-inflammatory, inmunosupressive, antifungal and antiparasitic properties [47]. Maximum Cyclosporin A production of 1274 mg/L with Tolypocladium inflatum was reported by Survase et al. [48] in submerged fermentation. In case of the Echinocandins, novel antymicotics produced by ascomycota fungi have a cyclic lipo-hexapeptide structure and act as β-1,3-glucan synthase inhibitors [49]. Echinocandin B the precursor of anidulafungin, is produced by Aspergillus nidulans and had reached 1.5 g/L in potato dextrose broth (PDB) [50]. The pneumocandin B0 precursor of caspofungin have reached about 2 g/L in the fermentation broth of Glarea lozoyensis [51]. Kanda et al. [52] reported the screening of a mutant of Coleophoma empetri and improved medium conditions for production of the antibiotic FR901379, the precursor of micafungin. The mutant strain had a 30-fold higher productivity compared to the wild type which produced 1 U/mL. In 2010, Kanda et al. [52] demonstrated the production of FR901379 with optimal conditions in a fermenter of 15,000 l reaching 50 U/mL. In the other hand, the bacteria Actinomycetes are known for produce novel bioactive compounds; more than 10,000 compounds were described only from genus Streptomyces [53]. Many of these compounds are synthesized by polyketide synthases (PKSs). In case of NRP, bleomycin is a glycopeptide produced by Streptomyces verticillus with antibacterial and antitumor properties [54] and is produced at 10 mg/L [55]. Daptomycin is a lipopeptide produced by Streptomyces roseosporus and consist of 13 aminoacids and have been approved because is effective for treatment of skin and skin structure infections caused by gram-positive pathogens [56]. A production of 812 mg/L has been reached through fed-batch fermentation with feedback control of dextrin [57]. In case of the peptide antibiotics, bacitracins (Fig. 2a, b) are produced by some species of Bacillus licheniformis and Bacillus subtilis, in addition contains at least 10 distinct dodecapeptides that differ by one or two aminoacids [58]. The bacitracin A production has reached approximately 900 mg/L in cultures of B. licheniformis NCIMB 8874 with the addition of oligosaccharides as elicitor [59]. Polymixns consist of ten amino acids with a characteristic polycationic heptapeptide ring and an N-terminal fatty acid modification and are produced by B. subtilis and Peanibacillus polymyxa [60]. Although this antibiotic is commercial, information about its level of concentration reached in production is vague. The issues of toxicity of certain NRPs have been described. For example, the family of polymixins such as polymyxin B and polymyxin E (colistin) that are cyclic lipopeptides produced by P. polymyxa were introduced into clinical medicine in the late 1950s but its use waned in the 1970s because the adverse effects in nephrotoxicity [61–63]. Nonetheless, its use has recently increased because the colistin is one of the antibiotics used for multidrug-resistant infections of Gram-negative bacteria such as P. aeruginosa, Acinetobacter baumannii, and Klebsiella pneumoniae [61, 64]. Schneditz et al. [65] reported that the NRP tilivalline produced by Klebsiella oxytoca have pathophysiological effects on human epithelial cells through the induction of apoptosis and disruption of epithelial barrier function. Because of the toxicity that can present some NRPs such as lipopeptides, efforts have been made to reduce cytotoxicity. For example, Jiang and colleagues [66] develop a novel class of antibacterial lipopeptides from surfactin that have a reduced cytotoxicity with no significant diminution in antimicrobial activity; another case is the mentioned by Robbel and Marahiel [67] with the daptomycin, a branched cyclic lipopeptide antibiotic, which was subjected to a deacylation to reduce its cytotoxicity.

Commercial and market opportunities for NRPs

Due to their environment-friendly properties, low toxicity and higher biodegradability, the NRPs such as biosurfactants have had an increase in their demand for use in biotechnological applications, for example, in extraction of petroleum, environmental restoration, foods, beverages, cosmetics, detergents or medicine [68, 69]. The total quantity of chemical and biological surfactants produced in the US is estimated at more than 10 billion pounds, and worldwide at 25 billion pounds [70]. Regarding only to biosurfactants, the reports for the values of the world market and its poduction in tonnes varies. Reis et al. [71] reports that the global market had a value of USD 1.7 billion in 2011 and is expected to reach USD 2.2 billion in 2018, based on a growth rate of 3.5 % per annum, and the global biosurfactants market volume is expected to reach 476,512.2 tons by 2018. While Campos et al. [72] reports that revenues from the biosurfactants market were USD 6.5 billion in 2012, and the market volume was 3.5 million tons. But regardless of the variation in the reports of market value, both the volume and the value of the biosurfantes is increasing. Currently the cost of biosurfactants in the market is high, compared with chemical surfactants. For example, surfactin (98 % purity) from B. subtilis available from Sigma Chemical Company has a cost of $191 for a 10 mg vial, while the cost of the rhamnolipids (95 % purity) from P. aeruginosa is $379 for 10 mg vial; in contrast, chemical surfactants as Alkanol® XC (Sigma Chemical Company), have a market cost of $72 for 500 g. Although at first glance the cost of chemical surfactants is much lower than the biosurfactants, due to environmental damage that they can cause, eventually their cost is much higher. The use of expensive substrates, limited product concentrations, low yields and formation of product mixture rather than pure compounds, are some reasons for limited use microbial surfactants and their high cost [73, 74]. Despite its high costs, biodegradability and low ecotoxicity of biosurfactants are features that draw the attention of companies as Ecover, which is a Belgian manufacturer of ecological detergents and cleansing agents, which use the biosurfactants sophorolipids in hard surface cleaners such as multisurface cleaner, floor soap, and window cleaner; or the Japanese company Saraya Co. LTD, which commercialized a dish washer containing sophorolipids as cleaning agent. Sophorolipids are also used in cosmetics products, for example, the French company Soliance produces sophorolipid-based cosmetics for body and skin; the Korean MG Intobio Co. Ltd commercializes Sopholine cosmetics, which is functional soap specific for acne treatment; or the Japanese company Kao Co. Ltd., which uses sophorolipids as humectants for cosmetic makeup brands such as Sofina [75, 76]. Biosurfactants also have been applied in the food industry, for example, rhamnolipids can be found as active substance in the fungicide Zonix™, produced by the company Jeneil Biotech Inc and approved by FDA for use on vegetables, legumes, and fruits crops [77, 78]. In the medical field there are examples of use of NRPs, such as Cyclosporin A and bleomycin A2. Cyclosporine A is an immunosuppressive agent, which has its application in the aftercare of organ transplantations; while bleomycin A2 exhibit cytostatic activity, which makes it suitable for cancer therapy [79]. These NRPs have high selling prices in the market. The cost of these molecules available from Sigma Chemical Company is $107 for 25 mg of Cyclosporin A (98 % purity) extracted from T. inflatum, and $847 for 20 mg of bleomycin A2 (70 % purity) extracted from S. verticillus. The use of biosurfactants increases, as well as investigations that result in patents for commercial use [77], but still need to reduce their production costs in order to be competent in terms of their prices.

Conclusions

Regarding the information of some commercial NRP described above, the concentration reached in production of them is still poor. Many studies have been made of production of NRP that are candidates to be utilized in different applications. However, its production also is deficient. Hence, the bioprocess engineering approaches must work together with the approach of bioinformatics genome mining, the heterologous production of NRP, the improvement of biosynthetic pathways as well as the physiology of the producer cells. Taken examples like process development penicillin, we will capable to produce many NRPs in larger quantities in order to overcome many problems in human health, crop protection, food industry as well as in environmental applications.