Production and characterization of a broad-spectrum antimicrobial 5-butyl-2-pyridine carboxylic acid from Aspergillus fumigatus nHF-01

The present study aims at the production optimization, purification, and characterization of a potent broad-spectrum antimicrobial compound (AMC) produced by Aspergillus fumigatus nHF-01 (GenBank Ac. No. MN190286). The culture conditions were optimized for a higher amount of AMC. The AMC was solvent extracted and characterized by UV–Vis, FT–IR, ESI–MS, and 1H-NMR spectroscopy. The MIC, MBC and mode of action were determined against a set of Gram-positive and Gram-negative human pathogenic bacteria. Its antibiofilm, synergistic and cytotoxic effects were also tested. The putative target site of action was evaluated through in silico molecular docking study. The stain A. fumigatus nHF-01 produced the maximum AMC (5-butyl-2-pyridine carboxylic acid) in 2% MEB (w/v) and 4% YE (w/v) at pH 6.0 and 20 °C temperature with 100 rpm agitation for ten days. It caused complete lethality of the Gram-positive and Gram-negative human pathogenic bacteria at a 129 µg/mL dose by rupture and entire dissolution of cell integrity. It showed moderate antibiofilm activity and had a synergistic activity with streptomycin and additive effects with ciprofloxacin and vancomycin. It targets a respiratory enzyme, Quinol-Fumarate Reductase (1l0v), with the highest binding affinities. It had cytotoxicity against human lung carcinoma A549 cell line and was stable up to 100 °C. Thus, the study revealed that the strain A. fumigatus nHF-01 produces a potent broad-spectrum AMC 5-butyl-2-pyridine carboxylic acid that could be used against human food and topical pathogenic bacteria. This is the first report of such a compound produced from the A. fumigatus.


Results
The strain Aspergillus fumigatus nHF-01: a azole sensitive strain. The strain A. fumigatus nHF-01 could grow in different media like Malt extract broth (MEB), Czapek dox broth (CDB), and Cornmeal broth (CMB) at pH ranges from 3.0 to 10.0 and temperature ranges from 20 to 45 °C 13 . To check the health risk, an antimicrobial drug resistance study was done. The MIC of eight different antifungal drugs is shown in Table 1. The study shows that among these drugs, Luliconazole exhibited the best in vitro activity with MIC values of 0.25 µg/mL, while Fluconazole was the less active with MIC values of 10 mg/mL (Supplementary Table S1).
Optimum culture conditions for AMC production. Among the different culture media, the highest amount of AMC was produced in MEB and extractable in DCM solvent, producing the highest growth inhibition zone of 25 to 30 mm, at a 15 mg/mL concentration, comparable to streptomycin (20 µg/mL) against the tested bacteria (Supplementary Table S2; Supplementary Fig. S1). It produced suitable biomass of mycelia (an average 1.334 g/100 mL culture) and AMC (an average 10.5 mg/100 mL culture). In comparison, the other media produced a negligible antibacterial activity though they had a high biomass production. The suitable pH and temperature range were at pH 6.0, 20 °C for 10 days incubation in 100 rpm shaking ( Supplementary Fig. S2). A combinational study with MEB and yeast extract (YE) showed that a mixture of 2% MEB and 4% YE (w/v) produced the highest specific activity ( Table 2; Fig. 1a-c). Set-4 produced the highest amount of biomass among the variants, while Set-8 produced the highest extractable compound and specific activity. The Set-7 produced a good amount of biomass but a very low extractable compound with high specific activity. Moreover, Set-12, Set-14, Set-15, Set-16 were found to produce a low-moderate extractable compound with no specific activity. From the heat map (Fig. 1d), it was observed that a high concentration of YE produces poor results in terms   www.nature.com/scientificreports/ of the extractable compound and specific activity; according to the interest of this study, Set-4 and Set-8 was clustered together for producing nearly similar results. A significant difference was found between sets (1 to 16) for biomass (p = 0.0034, df = 15), extractable compound (p = 3.2e−14, df = 15) and specific activity (p < 2.2e−16, df = 15) ( Fig. 2a-c), indicating a prominent effect on fungal culture due to mixture of MEB and YE. A significant positive correlation was found between MEB:EC (p < 0.001) and SA:EC (p < 0.05), and significant negative correlation was found between YE:EC (p < 0.05) (Fig. 2d). From surface plots ( Fig. 3a-c), the interaction between MEB and YE was determined. Higher concentrations of MEB and YE at equal proportions induces higher biomass production (Fig. 3a). The highest production of the extractable compound was produced between MEB concentrations 4 to 6; YE was found to contribute less in the production of the extractable compound and was controlled mainly by the concentration of MEB (Fig. 3b). MEB concentration 4% and YE concentrations between 2 to 6% were most suitable for achieving the highest specific activity (Fig. 3c). From the dot plots ( Fig. 3d-f) comparing predicted values with observed values, it can be observed that the models predicting response are fitted quite well with the observed values and can be used for further development. Although the models fitted well (Fig. 3d-f), none of the models (for biomass, extractable compound, specific activity) found significant (at 0.05 alpha level) ( Table 3). The R-squared and adjusted R-squared values turned out to be moderate to low; probably, a better experimental design can improve the design and R-squared values. Model details and coefficients are given below in Table 3.
Chromatographic purification of the AMC. The Fig. S3f). Figure 2. Box plots (a-c) represent the data distribution range of biomass (BM), extractable compound (EC), specific gravity (SA) and variation between experimental sets. Correlation matrix (d) represents the correlation between parameters [Colour scale bar at right representing correlation from negative (− 1) to positive (+ 1). Significance stars representing significance level of correlation (*p < 0.05, **p < 0.01, ***p < 0.001)].     www.nature.com/scientificreports/ respectively it decreased the viability sharply within 15 min of the incubation period ( Fig. 4a). Thus, a rapid decrease in the growth and viability of treated bacterial cells indicated that the active compounds had a strong bactericidal mode of action against the tested pathogens 30,31 . A comparative growth inhibition study with standard antibiotics, ciprofloxacin, streptomycin and vancomycin at MICx2 and MICx50 doses showed that all three antibiotics caused a rapid loss of viability within 30 min of treatment (Fig. 4b, and Supplementary Fig. S6). However, surprisingly the CFU/mL of the tested bacteria regained viability after 60 min. This observation indicates that the organisms either developed transient resistance to the antibiotics by changing the target site of action or enhancing the cellular transport or the efflux mechanism of the antibiotics. On the contrary, these results emphasize the potency of the new active compound from the A. fumigatus nHF-01 could be a potential lead drug in future antimicrobial therapy. Lactate dehydrogenase (LDH) is an essential cytoplasmic enzyme of all living cells. In the presence of the pure active compound, the cellular integrity had lost with the gradual increase in LDH activity (Fig. 4c). A subsequent gradual decrease in the colony-forming unit was observed in all the strains tested. The positive control (sonication) showed that it had high LDH with the lowest CFU values. The time-kill curve showed that after 12 h of treatment, the compound at a MIC × 2 dose caused zero viability from the initial Log CFU values (Fig. 4d). This indicates that the compound had a bactericidal activity with absolute lethality in the treated cells. The effect of the active compound on B. cereus and E. coli showed remarkable changes in the morphology (Fig. 5). It was found that in comparison to the untreated cells (Figs. 5a and 4d), the treatment after 30 min caused minute cell wall to rupture (Fig. 5b,e), and at 3 h of treatment, it caused drastic changes in cell morphology like the formation of blebbing, notch, rupture of the entire cell walls, and entire dissolution of cell integrity (Fig. 5c,f). This indicates that the compound is causing lysis of the bacterial cell wall resulting in a rapid decrease in cell viability.

Antibiofilm and biofilm destabilization assay. Biofilm is a complex association of microorganisms
formed on solid and liquid systems. B. cereus and E. coli are found to be significant organisms residing on food commodities, on surfaces of package materials, and even inside the human body, forming a toughened matrix. Many potential AMCs are found ineffective in this state of the microbes. The present study revealed that the active compound showed moderate inhibitory properties on the biofilm-forming E. coli and B. cereus compared to the standard antibiofilm compound Usnic acid (Fig. 6a). The active compound at a concentration of 4 µg/ mL and 129 µg/mL showed 22.30% of biofilm inhibition against B. cereus and E. coli, respectively, compared to control. The percentage of such inhibition is concentration-dependent, and it was 45.38% and 65.18% at a concentration of 517 µg/mL. The standard antibiofilm drug usnic acid showed a notable antibiofilm activity at 10 µg/mL and 64 µg/mL against E. coli and B. cereus, which is comparable to 20 µg/mL, and 4 µg/mL of the nHF-01 active compound. So the study revealed that the active compound 5-butyl-2-pyridine carboxylic acid has the potential to inhibit both the planktonic and biofilm stages of the Gram-positive and Gram-negative bacterial strains. The biofilm destabilization assay also showed that it could destabilize the preformed biofilm of the target pathogens (Fig. 6b).
Cytotoxicity and synergistic effect with different antibiotics. The compound 5-butyl-2-Pyridinecarboxylic acid showed cytotoxicity against the tested human A549 cell line (Fig. 6c). The synergistic effects of 5-butyl-2-pyridine carboxylic acid with three conventional antibiotics (ciprofloxacin, streptomycin and vancomycin) showed that all combinations demonstrated synergistic, partiality synergistic and additive effects against the tested bacteria (Fig. 6d). It showed synergistic activity with streptomycin against B. cereus, whereas it has additive effects with ciprofloxacin and vancomycin. While it showed partial synergistic activity with streptomycin and ciprofloxacin in E. coli, an additive effect with vancomycin (Fig. 6d) 32 . To check such health risks, a drug resistance/sensitivity test of an isolate to lifesupportive drugs is very much crucial. Among these, azole resistance is one of the leading concerns. The results show that the present strain A. fumigatus nHF-01 is sensitive to such azole and other antifungal drugs; thus, it is assumed that the handling of the strain for large scale AMC production is less risky and safe (Supplementary Table S1). The culture conditions and incubation period is crucial consideration for AMC production. The AMC production by the strain showed that the MEB and YE (low-cost fermenting substrates) triggered a high amount of AMC when grown at low temperatures (20 °C) with mild acidic (pH 6.0) fermentation conditions. It did not produce AMC at neutral to alkaline pH (8.0-12.0) and strong acidic pH (3.0-5.0) though it has ample mycelial biomass. The results indicate that a unique H + balance might induce the associated molecules or genes to produce AMC. The influence of temperature on antibiotic production varies from strain to strain. Generally, fungi are grown at 28 °C at pH 5.6-6.5. Effect of incubation temperatures and pH showed that A. fumigatus nHF-01 produced the maximum stable and effective AMC with inhibition zone (30-32 mm) in pH 6.0 at 20 °C while at 28 °C, 37 °C and 45 °C significantly less inhibition zone (7-23 mm) was observed at the same concentration of extractable mass. Low temperature accelerated the metabolite production by the fungus while high temperature slowed down. The subsequent microscopic observation shows that at 20 °C, A. fumigatus nHF-01 had significantly more sporulation with restricted growth, while at > 20 °C, it had a puffy and velvety appearance less sporulation, no prominent vesicle was found in both liquid and solid media ( Supplementary Fig. S1), as reported earlier 13 . This indicates that the cultural conditions of this organism are pretty unusual, and restricted mycelial growth with more sporulation is ideal for this organism for AMC production. A similar observation was reported in a marine fungus A. ustus MSF3 that produced AMC in 45% Sabouraud dextrose broth (SDB) with carbon (glucose)-nitrogen (yeast extract) ratio of 3:2 at 20 °C temperature at 7 days in solid culture 33,34 . Compaore et al. 35 and Agastian et al. 36 reported that A. fumigatus produces fumagillin and gliotoxin optimally in a synthetic media condition (supplemented with yeast extract, lactose and other carbon sources) for 6-8 days of incubation at 37 °C, pH 7.0, at 150 rpm, and extractable in acetonitrile and methanol. It was also observed that A. fumigatus nHF-01 produced maximum AMC at 10th day with an average diameter of 19-23 mm inhibition www.nature.com/scientificreports/ zone at 15 mg/mL concentration, while the 8th and 12th days period produced significantly less antimicrobial activity at the same concentration ( Supplementary Fig. S2). Moreover, no inhibition zones were observed on the 5th, 15th and 20th days of incubation. This indicates that the organisms' physiological status, like cell age and the media's nutritional status, might trigger the organism to produce such AMC. In addition, a sudden drop in antimicrobial content on the 12th day indicate that the organism could produce some degradative enzymes to impede the activity of the antimicrobial compounds. On the other hand, the compounds in cell-free supernatant or after extraction with DCM were stable for more than 36 months. So, the activity loss is not due to compound stability but degradative molecules' production. In addition to nutrient conditions, circulatory agitation also helps aeration in the culture medium. It produced a higher antimicrobial compound (0.688 mg/mL) at an optimum agitation speed of 100 rpm than non-agitation (0.38 mg/mL) condition. It is hypothesized that this fungal organism gets its optimum sheerness to spread or grow its mycelia and gain its optimum gas balance (O 2 /CO 2 ) inside the culture system to produce the metabolite optimally. It was also observed that AMC produced by the mycelia was exclusively released in the culture broth and no further extracts were recoverable from the dried mycelia. This is unique from the industrial point of view. Considering all these findings, for AMC production by A. fumigatus nHF-01, certain unique fermentation conditions are of the utmost need for the strain and all these parameters would guide designing the RSM model for large scale production at an industrial scale. A similar culture condition-dependent metabolite profiling of A. fumigatus with antifungal activity study was done by Kang et al. 37

AMC characterization and its structure-function relations on bacterial cells: Planktonic and biofilm stages.
The spectroscopic studies revealed that the molecular structure of the AMC is 5-butyl-2-pyridine carboxylic acid (also known as Fusaric acid, FA or 5-butyl-2-picolinic acid). The molecule and its analogues exhibited moderate antimicrobial activities 38 , including growth inhibitors of E. coli 39 , and act as quorum sensing 40 . It is also reported from species of Fusarium 41 and Gibberella fugikuroi 39 . Moreover, previous studies also show that it inhibits dopamine beta-hydroxylase enzymes that convert dopamine to norepinephrine and inhibits cell proliferation and DNA synthesis 40 anti-tumour activity on heme enzymes 39 . Structure-activity co-relation indicates that 2-pyridine carboxylic acid and its derivatives act as a bidentate chelating agent that effectively chelates metals in metal-containing protein complexes and enzymes required for growth replication or inflammatory response and thereby used to treat cancer. So the novelty of the present study is that the strain A. fumigatus would provide an easy biological source for large scale production of 5-butyl-2-pyridine carboxylic acid. The AMCs with such broad-spectrum activities is very limiting to the list of antimicrobial drugs in pharmaceutical industries. The study shows that the MIC and MBC values were more active against a broad range of food and waterborne pathogens that cause fatal food poisoning, typhoid fever, tuberculosis, and infections in scars and wounds ( Table 4). The greater efficacy of this compound against these strains finds its application against food and topical pathogenesis. Therefore, further subsequent drug safety and molecular action studies of this compound are an utmost need for its global use. The release of LDH enzyme with content with a concomitant gradual decrease in CFU value indicates that the compound affects cellular permeability and thus rendered a quick death of the cells 30,31 . Moreover, the compound is stable up to 100 °C, superior to the antifungal peptide that was stable up to 70 °C produced by A. clavatus 42 . Therefore, compared to the other antimicrobial compounds produced by Aspergillus spp., the present antimicrobial compound is more heat stable and could be used as antimicrobials in many processes involving thermal treatment up to 100 °C but not autoclaved processes. The drug efficacy is nowadays being trialled with combination mode [43][44][45] . Very often, it was observed that co-administration of more than one active compound might enhance or reduce the drug efficacy of the lead compound [43][44][45] . Therefore, synergistic/antagonistic study is very much essential for drug potentization. The study revealed that the compound 5-butyl-2-pyridine carboxylic acid has a synergistic effect with three conventional antibiotics (Fig. 6d) like ciprofloxacin which acts on bacterial topoisomerase II (DNA gyrase) and topoisomerase IV, streptomycin that binds irreversibly to the 16S rRNA and S12 protein within the bacterial 30S ribosomal subunit, and vancomycin that inhibits bacterial cell-wall biosynthesis (https:// go. drugb ank. com/ drugs/ DB005 12). Moreover, these combinations would reduce the application of the antibiotic dose to cure many challenging pathogens. The literature study suggested that the combined antimicrobial effect of antibiotics and extracted metabolites increase by increasing their bonding reaction [46][47][48][49] .
Many fungal secondary metabolites show important biological efficacies like antibacterial and antiviral activities mainly targeting different microbial proteins, like DNA-gyrase, topoisomerase IV, dihydrofolate reductase, transcriptional regulator TcaR (protein), and aminoglycoside nucleotidyltransferase. However, the metabolites acting on respiratory enzymes are very rare. The present study shows that the 5-butyl-2-pyridine carboxylic acid has a strong binding affinity (Supplementary Table S5) towards the quinol-fumarate reductase (QFR), succinate: quinone oxidoreductase (SQR, succinate dehydrogenase) and menaquinol: fumarate oxidoreductase (QFR, fumarate reductase), members of the integral membrane proteins Complex II family, that play a key role in the Krebs cycle. Hence, the molecular docking studies (Supplementary Fig. S5) revealed that the compound targetting QFR of E. coli inhibits the essential respiratory enzymes, thus leading to energy depletion and cellular viability. This observation is different from the novel anthraquinone, 2-(dimethoxymethyl)-1-hydroxyanthracene-9,10-dione, isolated from A. versicolor, that had efficacy against topoisomerase IV and AmpC β-lactamase enzymes 50 . Thus, molecular docking studies also revealed a novel target site of action of the 5-butyl-2-pyridine carboxylic acid that could be used as a future drug in combating many infectious and chronic diseases.

Conclusions
The species of Aspergillus are the leading microfungi that have wide use in different industries. They produce a diverse array of potential biomolecules like antibacterial, antifungal, immunodepressants, anti-AIDS drugs, etc. However, reports on the broad-spectrum antimicrobials from this organism are very limiting. The development of new, novel and high potential antimicrobials is a global challenge due to the upsurge in multidrug resistance among the food and topical pathogens. To this critical demand, the present study reports for the first time that A. fumigatus nHF-01 produces a broad-spectrum 5-butyl-2-pyridine carboxylic acid antibacterial compound that has activity against human pathogenic bacteria on both the planktonic and biofilm states. It has absolute lethality at 15 h treatment at a dose of MICx2. Moreover, the compound has a strong binding affinity towards the respiratory enzymes resulting in rapid depletion of energy and subsequent death of cells. This robust broadspectrum antibacterial compound could be produced in a very low-cost media. Further analysis with a detailed pharmacological mechanism of action study would decipher the antibacterial action more vividly and thus, be trialled as a potent drug contributing to human endeavour in future.

Materials and methods
Fungal strain and sensitivity towards antifungal drugs. The micro-fungus A. fumigatus nHF-01 (GenBank Acc. No. MN190286) was cultured and maintained in Potato Dextrose Agar (PDA) at 28 °C and subcultured in every 5-7 days interval 13 . The antifungal drug sensitivity of this strain was tested against many azole and systemic fungicide drugs (Supplementary Table S1) by agar well diffusion assay and MIC was determined following Mandal et al. 13 .
Bacterial strains and the assessment of antibacterial efficacy. For the antimicrobial assay, the strain was grown in respective culture broth in a BOD incubator at 28 °C for 10 days. After profuse growth and sporulation, the culture aliquots were tested for their antibacterial efficacy by the agar well diffusion method on nutrient agar (NA) plates following the standard protocol 13,51 , against the human pathogenic Gram-positive and Gram-negative bacteria (Supplementary Table S2  Optimization of culture conditions. After screening the primary media constituents influencing AMC production, two critical factors, i.e. Yeast extract and Malt extract, were tested in different proportions. Therefore, the batch fermentation (100 mL) were conducted containing different % ME (0 to 8, w/v) and % YE (0 to 8, w/v) set at pH 6.0, 20 °C for ten days in shake condition, and the amount of AMC production was recorded as described above. Approximately 1 L of batch fermentation was conducted to harvest an ample amount of AMC for purification and characterization studies.
Chromatographic purification of the active compound. The analytical and preparative TLC (Thin layer chromatography) was performed using TLC silica gel 60 F 254 plates (Merck, Germany). The compounds of the crude extract were separated in n-hexane and ethyl acetate solvent system (9:10, v/v), visualized under 254 nm UV light, and the Rf values were recorded. The potent antibacterial fraction was determined through TLC zymogram assay by overlaying NA soft-agar seeded with sensitive bacteria. The potent antibacterial spot www.nature.com/scientificreports/ was marked by observing the growth inhibition zone and tallied with the Rf values. Similarly, several preparative TLC plates were run, and the potent AMC was scraped out from the TLC plates, extracted with DCM, and evaporated to dryness. Preparative HPLC analysis of the separated fraction was carried out to harvest the pure compound in a multidimensional RP-HPLC system (Water Alliance, 2695, MDLC, UV-Vis detector 2487) equipped with a C 18 column. Elution was done in a 0.5% H 3 PO 4 , 90% acetonitrile, and 0.1% trifluoroacetic acid as mobile phase with a flow rate of 1.0 mL/min at 30 ± 2 °C and the detected at 246 nm. The fractions were collected and tested for potential antibacterial activity, as mentioned above. Several runs were conducted to harvest the pure compound for further characterization. Phytochemical screening, solubility and thermo-stability studies. Chemical tests for active constituents of the pure compound were carried out following Mandal et al. 31 . The thermo-stability of the compound was checked at 60 °C, 80 °C, 100 °C in a water bath, and at 121 °C in an autoclave for 30 min. The residual antimicrobial activity was assayed against the above mentioned pathogenic bacteria by agar well diffusion assay. The solubility was checked in different organic solvents, viz. n-hexane, diethyl ether, DCM, ethyl acetate, methanol, DMSO and chloroform. The physical appearances and colour were also observed.

Characterization
Antimicrobial potentiality studies. Determination of MIC and MBC. Serial dilution in the range of 35.7 mg/mL to 0.069 mg/mL was prepared with DCM solvent and tested against the target bacteria. The MIC is the lowest concentration of the active extract treatment, where no visible growth is observed. 20 µL of different concentrations was applied by the agar well diffusion plates seeded with each bacterial strain at Mc Farland standard (0.6 at 610 nm), incubated at 37 °C for 48 h and observed for the zone of inhibition. The MBC was defined as the minimum concentration of the antibacterial compound) that produced no viable cells. The actively grown sensitive bacteria were treated with compounds higher than MIC values and incubated for 48 h. The number of viable cells was determined by counting the colony-forming units (CFU/mL) by decimal dilution plating 31 .
Effect of the active fraction on growth and viability of bacterial strains. The effect of the AMC was studied on sensitive human pathogenic bacterial strains such as B. cereus, E. coli, S. epidermidis, and S. enterica serovar Typhimurium. The 50 µL of purified active compound (MIC × 2 dose) was added to the 210 µL of exponentially growing cells and incubated at 37 °C for 33 h. The growth was monitored at OD 620 nm in a spectrophotometer (UV-Vis 1800, Shimadzu, Japan) and viability by the decimal diluted plate-count method. For the plate count method, the treated and control cells were serially decimal diluted and 500 µL of the dilution (10 -2 to 10 -8 dilution) were plated on NA plates and incubated at 37 °C. The viable colonies (CFU) were counted from respective treatment plates at 48 h and calculated as log CFU/mL. The values were compared with the standard antibiotics, viz. ciprofloxacin, streptomycin and vancomycin at MICx2 and MICx50 doses against the bacterial strains B. cereus and E. coli.
Determination of time-kill kinetics and time-kill endpoint. The

Evaluation of synergistic/antagonistic effects between AMC and antibiotics. The synergy
between two antimicrobial compounds is often expressed in terms of the fractional inhibitory concentration (FIC) which is expressed by the MIC of the compound in combination divided by the MIC of antibiotic acting alone. To determine the FIC, the target organisms B. cereus and E. coli were fresh cultured, harvested, and suspended in sterile NB to produce a McFarland value of 0.5 (OD 610 nm ). The suspension was diluted in fresh NB to achieve a final CFU of 4 × 10 6 to 5 × 10 6 from which 10 µL was inoculated into the microwell plates and incubated at 37 °C for 16-20 h. A checkerboard microdilution technique was used to examine the synergism between the antibiotics and AMC against test organisms. All tests were carried out in a duplicated manner. The values were inferred based on the nature of the interaction like FICI < 0.5 = synergy, 0.5 ≤ FICI < 1 = partial synergy, FICI = 1 additive, 2 ≤ FICI < 4 = indifferent, and > 4 FICI = antagonism 56 . FICI was calculated as follows: FIC of AMC = MIC of AMC in combination/MIC of AMC alone 57 . The calculated FIC index was used to detect the nature of the interaction between the two test agents and the interaction either synergism or indifference or antagonism type.

Molecular modelling.
The crystal structure of many AMC target proteins/enzymes were obtained from a protein data bank (http:// www. rcsb. org). The structures were then cleaned using Autodock tools by removing heteroatoms and by adding necessary hydrogen atoms. The structure of the 5-butyl-2-pyridine carboxylic acid molecule was obtained from PubChem (https:// pubch em. ncbi. nlm. nih. gov/). Using UCSF Chimera 58 , the PDB files of the 5-butyl-2-pyridine carboxylic acid were created for docking. Autodock Vina 59 package was used for docking between the best binding sites of enzyme and proteins and 5-butyl-2-pyridine carboxylic acid. All the docked compounds were subjected to further selection for ADMET property analysis based on Lipinski's five (Ro5) rule, and compounds with any Ro5 violations were eliminated. Ro5 includes molecular weight, lipophilicity, molar refractivity, number of hydrogen bond donors and acceptors. The Molinspiration server was used for calculating the physicochemical properties of compounds (http:// www. molin spira tion. com/ cgi-bin/ prope rties).
Statistical analysis. Exploratory analysis was done to find the best media composition 60,61 . Bar plots and a heatmap clustered based on similar performance were used to understand better the effects of media composition on biomass, extractable compound and specific activity. One-way ANOVA was used to analyse the variation between experimental sets. The correlation matrix was used for understanding the correlation between parameters. Response Surface Modeling (RSM) was used to determine the interaction between parameters 62 . The concentration of MEB and YE were treated as a predictor variable, biomass, extractable compound and specific activity were treated as the response variable for creating the RSM model; details of the experimental design and www.nature.com/scientificreports/ results were given in Table 2. We created three models for three response variables, including all the first order, second-order, and interaction terms; details of the model and coefficients are given in Table 3. Predictions were made from the model and compared with observed variables. Clustering and production of heatmap were done using Orange 3 (https:// orang edata mining. com/) software. The rest of the analysis, modelling and plotting were done using R 3.6.3 (Cran.r-Project.Org) software. All the codes of R were run in Rstudio environment 1.2.5042 (www. rstud io. com).

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.