Introduction

It is predicted that the antibiotic-resistant strains lead to 10 million deaths by 2050 and impose an economic cost of $100 trillion [1, 2]. Many antibiotics have been discovered from Actinobacteria as a rich source of natural compounds, and mining rare Actinobacteria are one of the most promising strategies to find new and structurally diverse microbial-derived compounds [3].

The time-consuming and labor-intensive process of purification and structure elucidation of isolated antibiotics and, in particular, the high percentage of the rediscovery of previously known compounds has dramatically declined the efficiency of new compound discovery over time [4]. Due to the high rate of rediscovery, major pharmaceutical companies have abandoned screening for novel antibiotics and have employed other alternative strategies such as bioinformatics-based approaches to reinvigorate antibiotic discoveries.

The genome analyses revealed that a large portion of the biosynthetic gene clusters (BGCs) of Actinobacteria is unexplored. It was envisaged that the currently available data on BGCs in microbial genomes would revolutionize the rate of discovery through genome-mining based approaches; however, it could not make acceleration as it was predicted [5].

For instance, genome mining of Actinobacteria shows that they have metabolite-encoding genes for thousands of unknown compounds in their genomes, which remain undiscovered since they are not transcribed under general laboratory conditions [6].

This mismatch between biosynthetic potential and isolable bioactive compounds in the fermentation process may result from week expression and/or translation of BGCs and disability of available detection methods.

These cryptic BGCs represent an enormous reservoir of new druggable leads to produce novel antimicrobials and other bioactive compounds. Inducing production of silent secondary metabolites can be encouraged through several techniques including, molecular (genetic manipulation and expression of BGC of interest in heterologous hosts or ribosome engineering), cultivation-based approaches (modifications in chemical constituents of culture media and cultivation parameters like pH, temperature, or aeration), and elicitation approach (using elicitors to induce the expression) [7,8,9].

The production of secondary metabolites can be increased in culture-based methods by altering the levels of substrates or activity of the involved enzymes affecting gene regulation, metabolism, and growth parameters. Although medium optimization can alter the metabolic profile of various microorganisms, it has disadvantages like laborious and resource-intensive need that limit its application in transcription induction of new metabolites [10].

The elicitation approach has been developed because the products of cryptic gene clusters may involve in competitiveness and responsiveness to environmental stress in a natural ecosystem. Therefore, in an alternative approach, microbial cells, which are secondary metabolite production prone, can be subjected to biological, physical, or chemical elicitation. These agents may mimic ecological and environmental stimuli.

In this regard, chemicals, including antibiotics (at sub-inhibitory levels) [11], polysaccharides or their monomers [12], bacterial or plant hormones [13, 14], exposure to metals (at low-level) [15], rare earth elements (REEs) (scandium, yttrium, lanthanides, lanthanum, lutetium) [16], inorganic or organic compounds (dimethyl sulfoxide (DMSO) and ethanol), molecules with modulating ability of fatty acid biosynthesis [14], and physical (various temperatures and irradiation) approaches can be applied to trigger the secondary metabolite pathways [17, 18].

It has been shown that elicitors can modify the transcription of clusters of secondary metabolite genes in Actinobacteria. Among these chemicals, inhibitors of histone deacetylases (HDAC) are well known to be involved in the transcriptional regulation of biosynthetic pathways in Streptomyces coelicolor A3(2) through modifying the nucleoid structure [19]. REEs can activate the expression of silent or poorly expressed genes in bacteria and consequently enhance the production of bioactive compounds, including antibiotics, enzymes, pigments, mycotoxins, and phytotoxins [20]. Ethanol and DMSO have also been applied to elicit biosynthesis of microbial compounds, perhaps by mistranslation or inducing the stress response [21, 22].

The presence of polysaccharides, antibiotics, and bacterial hormones (at low concentrations) in the culture medium can also be used to stimulate the production of bioactive metabolites [23, 24]. Hormones, e.g., gamma-butyrolactones play a critical role in bacterial communication, known as quorum sensing promote antibiotic production in many Streptomyces spp. [25]. In this communication, bacteria synchronize their gene expression via producing and detecting the autoinducing agents [26].

It has been revealed that a family of molecules that modulate fatty acid production referred to as antibiotic remodeling compounds (ARCs) can elicit cryptic genes via changing fatty acid pool by inhibition of key biosynthetic enzymes like FabI. Since there is a competition among a synthesis of fatty acid and secondary metabolism for the common substrate, acetyl-CoA, these compounds may act through partial inhibition of FabI and therefore make a preferential flow of acetyl-CoA for the synthesis of antibiotic. Remarkably, triclosan, a well-known biocide and inhibitor of fatty acid biosynthesis, can stimulate polyketide antibiotic synthesis by this mechanism [27].

According to these findings, several studies have been conducted on the elicitation of cryptic genes of Streptomyces, while the activation, production, and identification of natural products from rare Actinobacteria are less investigated. Here, we aimed to de-silence the hidden secondary metabolite biosynthetic cluster of a soil-driven rare Actinobacterium, Promicromonospora kermanensis [28], which shows no antimicrobial activity under reasonable lab condition by chemical (DMSO, triclosan, antibiotics, lanthanum chloride, HDAC inhibitors, alkaline pH) and biological (live and dead cell mass) approaches.

Material and Methods

Chemicals and Bacterial Strains

All chemicals were of analytical grade. Sodium butyrate, valproic acid, lanthanum chloride, benzamide, triclosan, and DMSO were obtained from Merck (Germany). Streptomycin, bacitracin, and vancomycin were purchased from Sigma-Aldrich (USA). Promicromonospora kermanensis DSM 45485, Nocardia sp. UTMC 557, Pseudomonas aeruginosa UTMC 1404, and Methicillin-resistant Staphylococcus aureus (MRSA) UTMC 1401 were provided from University of Tehran Microorganisms Collection (UTMC).

Treatments with Antibiotics as the Inducing Agents

To evaluate the efficiency of antibiotics on provoking the expression of silent pathways, ISP2 broth medium (4 g L−1 of glucose, 4 g L−1 yeast extract, 10 g L−1 malt extract, pH 7.2 ± 0.2) was prepared and autoclaved at 121 °C with 1.2 bar. After cooling, antibiotics including vancomycin (5 and 10 mg mL−1), bacitracin (5 and 10 mg mL−1), and streptomycin (25 and 50 mg/mL) were added to the medium following 0.45 μm after filtration of the stock solutions.

Evaluation of Chemical Factors as the Inducing Agents

To assign the provoking ability of chemical inducers, compounds, including sodium butyrate (25 and 50 mM), valproic acid (0.05 and 1 mM), triclosan (1 and 5 μM), DMSO (1 mM and 2 mM), benzamide (2.5 mM and 5 mM), lanthanum chloride (50 and 100 μM), were supplemented to the ISP2 broth medium. To analyze the inducing activity of alkaline condition, the pH of ISP2 broth medium was adjusted in pH 9 using 5 M NaOH.

Microbial Inoculation and Incubation Condition

The antibiotic and chemical supplemented media, and also alkaline ISP2 medium, were inoculated by 107–108 spores mL−1 and were incubated at 28 °C and 170 rpm for 5 days. The uninoculated ISP2 broth medium and ISP2 broth medium without inducing agents (pH 7.2 ± 0.2) were used as the control to evaluate the antimicrobial activity of medium and Promicromonospora kermanensis, respectively. The growth of Promicromonospora kermanensis in supplemented and control media was assessed via the colony counting.

Elicitation via Filtrated Supernatant of Bacterial Culture

Pseudomonas aeruginosa UTMC 1404 and Nocardia sp. UTMC 557 were cultured in liquid ISP2 medium and incubated at 28 °C, at 170 rpm for 4 days. At the end of the incubation period, the fermentation broth was centrifuged at 2500 g for 15 min. Supernatants were autoclaved at 121 °C with 1.2 bar and then used as a stimulant for Promicromonospora kermanensis metabolite production. The flasks were incubated at 28 °C, shaken at 170 rpm for 5 days.

Evaluation of Dead Bacteria Debris as Induction Agents

Suspensions of Nocardia sp. UTMC 557 (108 cell mL−1, 2.5% v/v) and Pseudomonas aeruginosa UTMC 1404 (108 cells ml−1, 5% v/v) were separately added to ISP2 medium and then were autoclaved at 121oC with 1.2 bar. The media were inoculated with Promicromonospora kermanensis and incubated at 28 °C, shaken at 170 rpm for 5 days.

Elicitation Using Cell Wall Component

The N-acetylglucosamine-supplemented (2 mg/mL) ISP2 broth medium was inoculated with Promicromonospora kermanensis and incubated at 28 °C, shaken at 170 rpm for 5 days.

Evaluating the Chemical Characteristics of Elicitors in the Supernatant of Pseudomonas aeruginosa Culture

The stimulating activity of compounds in the supernatant of Pseudomonas aeruginosa UTMC 1404 culture was evaluated after their exposure to different types of stress agents, including H2O2, SDS, and acidic (pH 4) and basic pH (pH 9) conditions. Treated supernatants were added to Promicromonospora kermanensis–inoculated ISP2 broth media and incubated at 28 °C, at 170 rpm for 5 days. Untreated supernatant of Pseudomonas aeruginosa UTMC 1404 culture and unelicited Promicromonospora kermanensis were used as negative controls.

Evaluation of the Hydrophobicity of Eliciting Agents in the Supernatant of Pseudomonas aeruginosa Culture

To evaluate the polarity or hydrophobicity of eliciting agents in the supernatant of Pseudomonas aeruginosa UTMC 1404 culture, the supernatant was passed through the solid state RP column and silica gel columns. The eluted samples were added to fermentation ISP2 broth of Promicromonospora kermanensis and incubated at 28 °C, 170 rpm for 5 days.

Extraction of Secondary Metabolites

The fermented broth was centrifuged at 2500 g for 10 min to remove the biomass and cell debris. The extracellular metabolites were extracted from supernatant using an equal volume of ethyl acetate by vigorous shaking for 1 h. The resultant organic phase was separated and concentrated at low pressure and temperature (Heidolph, Germany). The fermentation broth extracts of Promicromonospora kermanensis were stored at − 20 °C until use.

Qualitative Evaluation of the Extracts Using Thin-Layer Chromatography

The obtained extracts were qualitatively analyzed using the thin-layer chromatography (TLC). The extracts (10 μL with a concentration of 5 mg mL-1) were spotted on TLC silica gel 60F254 (Macherey-Nagel) and developed using the solvent system of dichloromethane/methanol (93:7 v/v). The developed TLC plates were observed under UV illumination of 254 nm and 366 nm and reacted by anisaldehyde reagent.

Antibacterial Activity Evaluation of the Extracts

The antibacterial activity of extracts was determined using the disc diffusion assay in triplicate experiments. In brief, the extracts were loaded on blank discs (50 and 150 μg per disc). The loaded discs were placed on inoculated Mueller-Hinton agar plates with Methicillin-resistant Staphylococcus aureus (MRSA) UTMC 1401. The vancomycin disc (30 μg/per disc) was used as a positive control. The plates were left at 4 °C for 3 h for diffusion before the bacterial growth. Then, the growth inhibition zones were measured after incubated at 37 °C for 18–24 h. The process of triggering and evaluation are summarized in Fig. 1.

Fig. 1
figure 1

Schematic presentation of the secondary metabolite induction of Promicromonospora kermanensis using chemical and biological elicitors

Full size image

Results

The Ability of Tested Antibiotics on Improving Antimicrobial Activity

The spiked vancomycin, streptomycin, and bacitracin had no inducing effect on antibacterial metabolite production of Promicromonospora kermanensis at two levels of sub-MIC concentrations as summarized in Table 1, and no growth inhibition zone was observed around loaded discs with extracts of P. kermanensis. The streptomycin at the applied concentration even inhibited the growth of P. kermanensis.

Table 1 Antibacterial activity of Promicromonospora kermanensis extracts following the stimulation using sub-MIC concentration of antibiotics
Full size table

Evaluating the Effect of Chemical Agents as Inducers

The categories of compounds, including HDACs inhibitors (sodium butyrate, valproic acid, benzamide), rare earth element (lanthanum chloride), fatty acid synthesis inhibitors (triclosan)and miscellaneous (DMSO and high hydroxide concentration) were applied as chemical-inducing factors. All applied HDACs inhibitors except benzamide induced the expression of antimicrobial genes of P. kermanensis. It was observed that the extract of supplemented medium with 5 μM triclosan had higher antibacterial activity than that of supplemented medium with 1 μM triclosan. In contrast, lanthanum chloride showed its evoking activity at the lower applied concentration (50 μM), and the inducing activity was not observed at a higher concentration of 100 μM.

The extracts of sodium butyrate containing ISP2 broth media showed antibacterial activity. According to the low growth rate of P. kermanensis in these media, it is deduced that the antibacterial growth is due to the antimicrobial activity of sodium butyrate, not its eliciting activity (Table 2). The results showed that the production of the antibacterial secondary metabolite of Promicromonospora kermanensis could be stimulated in the alkaline condition of the growth medium in liquid ISP2. The appearance of anti-MRSA activities (17 and 16 mm) is attributed to the induction of the expression of antibacterial compound encoding genes.

Table 2 Antibacterial activity of the extracts of Promicromonospora kermanensis induced by HDAC inhibitors, rare earth elements, inhibitors of fatty acids synthesis, DMSO and high concentration of hydroxide ion
Full size table

Biological Induction Approach

Induction of cryptic or low yield secondary metabolite biosynthetic gene clusters of P. kermanensis via biotic elicitors was also effective. The result of this investigation is summarized in Table 3. Among the applied biological inducers, just live P. aeruginosa cells could induce the antibacterial activity against MRSA, while its culture supernatant had no antibacterial activity on the tested MRSA (Fig. S1). According to TLC patterns, new metabolite bands were appeared in elicitation using dead cells, alive cells, and cell wall component; however, only induced metabolite by the alive cells of P. aeruginosa had antibacterial activity (Fig. S2).

Table 3 Biotic elicitation of cryptic or low-expressing antimicrobial encoding genes of Promicromonospora kermanensis via dead and supernatant of a Gram-positive (Nocardia sp. UTMC 557) and Gram-negative (Pseudomonas aeruginosa UTMC 1404)
Full size table

Chemical Stability of Pseudomonas aeruginosa Elicitating Agent

According to the stability experiments, metabolites of Pseudomonas aeruginosa with provoking activity on P. kermanensis metabolites were sensitive to oxidative stress as they lost their inducing ability in the presence of H2O2, at the same time they were resistant to SDS, acidic, and basic pH (Table 4).

Table 4 The provoking activity of treated metabolites of Pseudomonas aeruginosa UTMC 1404 compared to the controls.
Full size table

Hydrophobicity Features of Eliciting Factors in Supernatants of Pseudomonas aeruginosa

Eluted samples from RP and silica gel columns could not induce the antibacterial metabolites of Promicromonospora kermanensis with anti-MRSA activity. Therefore, it can be concluded that eliciting agents have both nonpolar and polar identity (Table 5).

Table 5 The antibacterial activity of the fractions derived from the biological inducer
Full size table

Discussion

Drug discovery is entering a new era using the available data on genome sequences of various microorganisms. The discovery of new microbial small molecules can be contributed by the induction of metabolite production. Increased synthesis of rare metabolites is another critical usage of this procedure, and also increased production of well-known metabolites via elicitors has been reported.

The knowledge of the existence of new biosynthetic pathways in the bacterial genomes enables us to work on activating the silent or poorly expressed genes via various stimulants that act as transcription activators and reinforce their expression [20].

The expression of cryptic metabolites in fungi and bacteria can be successfully elicited by small exogenous compounds, like growth-inhibitory molecules, e.g., antibiotics. It seems that these antibiotics at sub-inhibitory concentrations are involved in inter- and intra-species interactions, probably via stress response mechanisms, which can modify the activity of the ribosome and thus evoke the silence genes. Goadsporin, bacitracin, kanamycin, streptomycin, thiostrepton, paromomycin, tetracycline, and gramicidin D are some antibiotics from which eliciting activities at sub-inhibitory levels are reported. Some of these antibiotics exhibit hormonal characteristics at low concentrations and can play a stimulatory role to induce synthesis of antibiotic while at high concentrations exhibit their inherent antibacterial properties. Trimethoprim is known as a general activator of secondary metabolism in Burkholderia thailandensis [29]. However, none of the applied antibiotics in this study could exhibit eliciting activity on antibiotic encoding genes of Promicromonospora kermanensis.

Stimulatory activity of triclosan, a potent synthetic antibiotic via inhibiting lipid synthesis, was observed at low concentration (1 and 5 μM). Moreover, triclosan (1 μM) has efficiently acted as an eliciting agent in increasing the production of actinorhodin and salinomycin by S. coelicolor A3(2) and S. albus likely via reinforcement of substrates like acetyl-CoA and malonyl-CoA [30, 31].

Transcriptional inhibition of secondary metabolites could be a strategy to preserve bacteria from autotoxicity. Production of metabolically pricey and potentially autotoxic secondary metabolites is likely only provoked whenever a suitable environmental trigger is present [32, 33]. The use of small-molecule elicitors, in some cases, non-random targeted elicitors such as methyltransferase and histone deacetylase inhibitors, allows researchers to dominate the transcriptional inhibition of considerable gene clusters in the lab.

According to the results, Promicromonospora kermanensis DSM 45485 showed significant anti-MRSA activity when inoculated in an ISP2 broth medium containing valproic acid and DMSO (Fig. 2). They are agents that probably affect histone acetylation and subsequently modify the structure of chromatin and may lead to the activation of biosynthetic clusters for natural compounds [34]. HDAC proteins are rather extensive, and a lot of them exist in bacteria, e.g., S. coelicolor A3(2) possesses three HDAC-like genes [19]. Although there are significant differences in structural DNA organization of eukaryotes and Actinobacteria, inhibitors of HDAC like valproic acid and suberanilohydroxamic acid can provoke antibiotic production in several Streptomyces strains [35]. Another investigated HDAC inhibitor, benzamide, could not stimulate the expression of antimicrobial encoding biosynthetic gene clusters in Promicromonospora kermanensis.

Fig. 2
figure 2

The anti-MRSA activity of the elicitated Promicromonospora kermanensis. Two concentrations of the extracted metabolites (50-μg/disc and 150-μg/disc) of each elicitation treatment were assessed. Only inducible elicitors are shown. Benzamide, antibiotics, the supernatant of Nocardia, P. aeruginosa, and Nocardia debris could not induce the antibacterial metabolites of Promicromonospora kermanensis in the applied concentrations

Full size image

There are some reports on the eliciting activity of sodium butyrate as an HDAC inhibitor. For instance, the production of actinorhodin by S. coelicolor A3(2) strain M145 was stimulated in the presence of sodium butyrate (25 mM sodium butyrate) while inhibited on R5 medium (rich medium) [19]. Also, sodium butyrate (25 mM) induced five recognized cryptic pathways with end products of deoxysugar synthetase, sesquiterpene cyclase, hopanoids, germicidin, and coelibactin in S. coelicolor A3(2) [36]. However, sodium butyrate imposed inhibitory effect on Promicromonospora kermanensis, and the observed antimicrobial activities might be due to its inhibitory effect on MRSA growth, not eliciting its antimicrobial encoding biosynthetic gene clusters. It is also possible that sodium butyrate can impose its eliciting activity on minimal media. It has been reported that the eliciting ability of HDAC inhibitors like sodium butyrate and valproic acid is due to a created general stress response via these molecules [19, 37].

The eliciting activity of DMSO on various Actinobacteria (S. venezuelae and S. azureus) has been reported. DMSO enhanced the production of thiostrepton (by S. azureus) [38], chloramphenicol (by S. venezuelae), and the polyketide tetracenomycin C (by S. glaucescens) [22]. The eliciting activity of a derivate of DMSO, dimethyl sulfone, has also been reported [38]. Accordingly, DMSO exhibited eliciting activity on the expression of antibiotic coding gene clusters in Promicromonospora kermanensis in this study. Its promising eliciting ability could be as the consequences of translation perturbation or creation of stress responses.

Another most important eliciting agents involved in expression or overproduction of silent or poorly expressed secondary metabolites coding gene clusters are rare earth elements. It has been shown that scandium (10–100 μM) enhances the production of actinorhodin, actinomycin, and streptomycin by S. coelicolor, S. antibioticus, and Streptomyces sp. [39]. Scandium imposes its effect at the transcription level of regulatory genes in S. coelicolor. Scandium or lanthanum promoted the expression of nine secondary metabolite biosynthetic clusters of S. coelicolor A3(2) at low concentrations [40]. Also, Xu reported that the anti-microbial activity of 15 actinobacterial strains were induced or enhanced by the addition of LaCl3 [16]. Similarly, the eliciting activity of lanthanum (50 μM) was observed on Promicromonospora kermanensis by vanishing the inducing activity at the higher applied concentration (100 μM).

In addition to chemical elicitors, extreme changes in the acidity of culture media can provoke the expression of cryptic or poorly expressed biosynthetic gene clusters. Eliciting activity of alkaline pH can be due to lowering the concentrations of the proton or increasing the level of the common stress sigma factors [17]. Hayes et al. showed that S. coelicolor A3(2) produces more methylenomycin when it was influenced by an acidic condition [41]. According to our results, alkaline pH also plays an eliciting impact on genes in Promicromonospora kermanensis which are involved in the production of antibacterial agents.

The external stimuli can trigger the involved regulators in signal transduction networks [42]. Small molecule metabolites or hormone-like compounds of the bacteria can have a regulatory effect on gene expression of other bacteria or even eukaryotes [43].

The secondary metabolite pathway of Promicromonospora kermanensis was triggered by the diffusible metabolites produced by Pseudomonas aeruginosa. This probable signal or hormone-like molecule had amphipathic and oxidation sensitive nature, while it was stable in pH 4–9.

The monomers of polysaccharides and alkaline conditions were first applied for their elicitation effect on Actinobacteria in this study. This triggering effect on Promicromonospora can be investigated on other Actinobacteria for the same purpose.

In conclusion, eliciting the expression of cryptic or poorly expressed biosynthetic gene clusters via biological and chemical elicitors is considered as a promising approach in the enhancement of drug discovery yield due to some of its advantages, including feasibility, cost-effective, and lack of requirement to comprehensive knowledge on physiological and genomic information of the strains of interest. Since the complicated secondary metabolism of Actinobacteria challenges the favorable changes in the whole metabolic system through genetic manipulation, stimulating the expression of such silent BGCs can make it possible to significantly enhance the chemical diversity of microbial-derived origin compounds and improve the yield of drug discovery.