Notes
Isolation of Natural Prodrug-Like Metabolite by Simulating Human Prodrug Activation in Filamentous Fungus
2022 Volume 70 Issue 4 Pages 304-308
Details
2022 Volume 70 Issue 4 Pages 304-308
Prodrugs have seen increased clinical applications as therapeutic agents, as they reduce undesirable side effects and improve the therapeutic potential of drugs. While microorganisms produce numerous secondary metabolites with useful medicinal properties, there are only a handful of naturally occurring prodrugs discovered to date. The techniques of isolating secondary metabolites with therapeutic potential from natural product producers have been developed extensively over the years. However, the methods of identifying prodrugs from microbes have not been examined in depth, partly because prodrug-type compounds inherently lack the biological activities that are often used to screen for therapeutically useful secondary metabolites. Therefore, we hypothesized that the difficulty in searching for natural prodrug-type compounds may be addressed by simulating human prodrug activation within natural product-producing microbes. We chose to introduce human CYP (hCYP) into natural product-producing filamentous fungi, because hCYPs are the key enzymes that activate prodrugs in human body, and filamentous fungi are known to be prolific producers of a wide variety of natural products. Here, we successfully identified a cytotoxic, antibiotic and potential anti-diabetic natural product leporin B from Aspergillus flavus that was previously not known to produce this compound. Through bioinformatic and metabolite analyses, we identified the prodrug-equivalent compound leporin C that is converted into leporin B by the action of the hCYP isoenzyme 3A4. By employing various prodrug-activating enzymes and microbes that biosynthesize diverse arrays of natural products, we should be able to probe wider biosynthetic space for identification of interesting prodrug-type natural products.
Humans are constantly metabolizing ex vivo substances in the body including drugs. Drug metabolism is important for maximizing the effects of drugs and reducing side effects and mostly handled by the enzymes in the liver. In fact, certain compounds called “prodrugs” are activated into functional drugs by being metabolized. The term prodrug that was first introduced in 1958 by Adrien Albert to describe compounds that undergo biotransformation prior to eliciting their pharmacological effects.1) Since then, the interest in preparing therapeutics as prodrugs to improve the drug effectiveness has increased substantially. For example, among the blockbuster drugs in the current pharmaceutical market, simvastatin, omeprazole, acyclovir and enalapril are all prodrugs.2) Use of prodrugs can reduce side effects and toxicity, improve bioavailability and prolong half-life of the drugs.3) Despite the wide range of medicinal compounds have been developed from microbial secondary metabolites, only a small number of prodrug-type compounds have been isolated from natural sources.4–6) One of the main reasons for the difference is that the methodology for identifying prodrugs among natural secondary metabolites has not been explored extensively. Therefore, many prodrug-type compounds may have been overlooked during the search for microbial secondary metabolites with therapeutic potentials. However, isolation of prodrug-type natural products with masked bioactivities is indeed a significant challenge. Hence, we attempted to simulate the process of human prodrug activation within microbes that biosynthesize secondary metabolites. By identifying bioactive natural products from microbes equipped with human prodrug activation capability, we would be able to identify prodrug-type secondary metabolites that are naturally produced by microbes. Here, we focused on human CYP (hCYP) as the prodrug-activating enzyme, because hCYPs play a significant role in xenobiotic metabolism and prodrug activation for a number of clinically relevant therapeutics.7,8) Furthermore, hCYPs have a unique ability to accept a variety of molecules as their substrates due to their unusually broad substrate specificity.9) Thus, we speculated that the process of hCYP-catalyzed prodrug activation may be replicated within natural product-producing microorganisms by having them express an hCYP gene. Specifically, we aimed at screening filamentous fungi transformed with an expression vector carrying an hCYP gene to identify bioactive secondary metabolites produced specifically upon incorporation of the hCYP gene. Through detailed analyses of the secondary metabolites produced by the fungi, which are known prolific natural product producers,10) we would also be able to identify the prodrug-type product that served as the substrate for the hCYP to yield the bioactive compound identified during the initial screening.
First, we isolated hCYP cDNAs from human liver cells, because CYPs are extensively expressed in human liver. In this study, we focused on the hCYPs 1A1, 1A2 and 3A4, which are known to exhibit an exceptionally broad substrate tolerance.7) Using these hCYP cDNAs, we constructed expressing vectors (Supplementary Fig. 1) and introduced the vector into several Aspergillus species, including A. flavus, A. parasticus, A. lentulus and A. novofumigatus. Each plasmid was also provided with a copy of the human reduced nicotinamide adenine dinucleotide phosphate (NADPH):CYP oxidoreductase gene POR to help maintain the hCYPs active within the fungus11) (Supplementary Fig. 1).
Search for Natural Products Generated in the Presence of hCYPsThe transformed Aspergillus sps. were cultured for 2 weeks in MYS (malt extract, yeast extract, sucrose), potato dextrose broth (PDB) or oatmeal broth (OMB) liquid medium. Furthermore, each culture was incubated in a shake flask. The culture was analyzed for the presence of metabolites by liquid chromatography–mass spectrometer (LC–MS), and the metabolomic analyses of the LC–MS data were performed by the Compound Discoverer software (Thermo Fisher Scientific, MA, U.S.A.). The analysis identified that compound 1 was produced at an appreciable level only by A. flavus A1421 transformed with pKW18623, the vector harboring the hCYP3A4 gene (Supplementary Fig. 1), when cultured in MYS liquid medium. 1 had a m/z of 352.1907 [M + H]+ with the RT value of 5.43 min (Fig. 1(iv)). This m/z value indicated that the compound had the molecular formulas of C22H26NO3. To check whether the fungi were transformed with the hCYP-carrying vectors successfully, we performed PCR and RT-PCR analyses on the transformed strains. The analyses confirmed that each transformant harbored the intended vector and expressed the desired hCYP gene. Specifically, the A. flavus A1421 strain that produced 1 carried pKW18623 and expressed the hCYP3A4 gene (Supplementary Fig. 2). Furthermore, no production of 1 was observed when the hCYP3A4 gene was absent in A. flavus A1421 (Fig. 1(i–iii) vs. (iv)). Therefore, we concluded that 1 was produced specifically by the action of hCYP3A4 produced by the A. flavus strain. Therefore, we proceeded to purify 1 for further characterizations.
The HPLC traces of the metabolic extract from (i) A. flavus A1421 WT, (ii) A. flavus A1421 carrying the CYP1A1 expression vector pKW18611, (iii) A. flavus A1421 carrying the CYP1A2 expression vector pKW18613 and (iv) A. flavus A1421 carrying the CYP3A4 expression vector pKW18623. The traces were monitored at 280 nm.
Compound 1 was purified by liquid chromatography from a 2-L culture of A. flavus A1421 transformed with pKW18623 harboring the hCYP3A4 gene grown in MYS liquid medium for 3 weeks. The NMR data, including 1H- and 13C-NMR spectra, together with the MS/MS data (Supplementary Fig. 3) allowed determination of the chemical structure of 1, revealing its identity as leporin B, which was originally isolated from an unidentified fungal strain12) and later from A. flavus13) and Scytalidium cuboideum,14) only when grown on solid culture media.
Isolation and Characterization of the Natural Product 2Next, we set out to identify the precursor of 1, which would correspond to the prodrug-type compound that led to the formation of 1 in the presence of hCYP3A4. When all of the A. flavus A1421 strains examined in this study, including the wild-type strain, were analyzed, a peak with a m/z of 336.1959 [M + H]+, which corresponding to the calculated m/z of leporin C (2), was detected by LC–MS analysis in all strains grown in MYS liquid medium (Fig. 2C). 2 is the biosynthetic precursor of 1 that has no hydroxyl group on the 2-pyridone nitrogen.13) However, the yield of 2 that could be isolated from the wild-type and the hCYP-transformed strains was negligible, preventing us from confirming the identity of the constituent of the peak. To increase the production level of 2 in A. flavus A1421, we looked to a previous study in which the production of 2 was successfully increased by overexpression of lepE from the leporin biosynthetic gene cluster that was predicted to encode a putative Zn(2)-Cys(6) transcription factor13) (Fig. 2A). Thus, we constructed the vector pKW24012 that could overexpress lepE by placing it under the control of a glaA promoter,15) and introduced it into the wild-type A. flavus A1421 (Supplementary Fig. 4). Examination by RT-PCR indicated that the expression level of lepE was dramatically enhanced in the pKW24012-transformed A. flavus A1421 as compared to the wild-type strain (Supplementary Fig. 5). Correspondingly, the yield of 2 also improved substantially in the lepE-overexpressing A. flavus A1421 than the wild type or the three hCYP-expressing strains (Fig. 2C(v) vs. (i–iv)). Using the lepE-overexpressing strain, 2 was purified with a yield of 0.8 mg from a 4-L culture grown in CD-ST (Czapek Dox-starch, tryptone) liquid medium16) for 2 weeks using a liquid chromatographic method. The purified sample of 2 was used to collect the NMR data, including 1H- and 13C-NMR spectra, and MS/MS spectra (Supplementary Fig. 6), which allowed determination of the chemical structure of 2 to confirm that 2 is indeed leporin C by comparison to the reference data.13,17)
(A) The organization of the leporin biosynthetic gene cluster and the predicted functions of the enzymes encoded by the annotated genes in the cluster. The lepE gene is highlighted in grey in the gene cluster map. (B) The proposed biosynthesis of leporin B 2 from leporin C 1 by the human CYP3A4, hCYP3A4. (C) The HPLC traces of the metabolic extracts from (i) the WT A. flavus A1421, (ii) A. flavus A1421 carrying the hCYP1A1 expression vector pKW18611, (iii) A. flavus A1421 carrying the hCYP1A2 expression vector pKW18613, (iv) A. flavus A1421 carrying the hCYP3A4 expression vector pKW18623 and (v) A. flavus A1421 carrying the lepE overexpression (OE) vector pKW24012.
Prodrugs are valuable therapeutic agents that allow us to enhance the properties and mask the undesired characteristics of drugs.3) For example, prodrugs can be designed to have improved aqueous solubility, enhanced stability under physiological conditions, increased membrane permeability or reduced unpleasant taste. Prodrugs can also be engineered such that they can be absorbed or activated selectively to allow targeted administration of the active drug substances. Typically, prodrugs are prepared by chemical modification of active drug components. The modification is designed so that it can be removed to reveal the active drug moiety either spontaneously by processes such as hydrolysis or by enzymes within our body.3) Nevertheless, crafting of specific modifications to generate prodrugs may not be so straightforward, as prodrugs can be processed in an unintended manner to generate products that differ from the expected active drugs. One way to circumvent this problem is to identify secondary metabolites that would exert desirable biological activities upon modification within the human body. In another words, we would be looking for prodrug-like products in secondary metabolite-producing microorganisms. However, developing a method to isolate prodrug-like natural products with concealed biological activities from a pool of numerous metabolites produced by microbes would be a substantial challenge. Instead, we attempted to simulate the prodrug activation step for metabolites within the producer microbes by introducing a human enzyme commonly involved in prodrug activation and look for the formation of bioactive compounds in the microbe.
As an initial attempt to test the concept, we chose Aspergillus strains as our natural product producers and hCYPs as the metabolite-activating enzyme. Aspergillus sps. are known to produce a plethora of bioactive secondary metabolites,18) and hCYPs are one of the most common enzymes involved in prodrug activation in the human body.8) Through the exercise, we successfully identified the cytotoxic polyketide–nonribosomal peptide hybrid natural product leporin B 1 and its precursor C 2 as an equivalent of a prodrug–drug pair upon introduction of a hCYP3A4 gene into A. flavus A1421. 1 is a 2-pyridone-containing polyketide–nonribosomal peptide hybrid13) with antibiotic and cytotoxic activities.14) This compound was also found to be an iron chelator13) and also a transcription activator of the hexokinase II gene, making it potentially valuable in treating type 2 diabetes.12) Most recently, 1 along with leporin A were predicted as potential leads for inhibitors of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease.19) Furthermore, the pathway for the biosynthesis of leporins proposed based on the composition of the leporin biosynthetic gene cluster identified in A. flavus suggested that 2 was converted into 1 by LepD, a CYP encoded by the leporin biosynthetic gene cluster13,17) (Fig. 2A). Therefore, we concluded that 2 served as the prodrug that was converted into 1 by the hydroxylation activity of hCYP3A4 within A. flavus (Fig. 2B). It was also not surprising that we found hCYP3A4 to carry out the hydroxylation of 2, as hCYP3A4 is well known for its broad substrate specificity.20)
In comparison to the hCYP3A4-transformed strain, the wild-type A. flavus A1421 only produced a negligible amount of 2 and did not produce 1 at all (Figs. 1(i), 2(i)). Considering that there is no report of any significant bioactivity exhibited by 2, it would have been difficult to identify 2 as a metabolite worth examining further for drug development using conventional screening methods. Similarly, although isolation of leporins from several strains of fungi was previously reported when the fungi were grown on solid media, isolation of those compounds from the fungi grown in liquid media has never been reported.12–14,21) Thus, our study makes the first report of isolation of leporins from fungus grown in a liquid medium. Since liquid culture is more amenable than solid-state culture to efficient screening of bioactive metabolites produced by microbes, identification of this prodrug–drug pair would have been impossible without the incorporation of the hCYP gene into the system. Thus, our approach was also capable of altering the profile of discoverable compounds produced by a microbe. This could widen the search space for bioactive secondary metabolites to improve the productiveness of our drug discovery work.
In conclusion, we successfully demonstrated that expression of a hCYP gene in a filamentous fungus led to the identification of a bioactive natural product from a strain previously not reported to produce the compound and a prodrug-like secondary metabolite that can be activated by an hCYP. Our approach can be developed further by employing other types of CYPs and prodrug-activating enzymes in combination with microbes with differing secondary metabolite biosynthetic potentials to explore a wider biosynthetic space for identification of interesting prodrug-type natural products.
Escherichia coli DH5α (TaKaRa Bio Inc., Shiga, Japan) was used for plasmid propagation. DNA restriction enzymes were used as recommended by the manufacturer (Thermo Fisher Scientific). PCR was performed using PrimeSTAR GXL DNA polymerase (TaKaRa Bio Inc.) The A. flavus genomic DNA was used as the template to amplify the target gene using the oligonucleotide primer sets listed in Supplementary Table 1. Sequences of the PCR products were confirmed through DNA sequencing (Macrogen Japan Corporation, Tokyo, Japan).
Spectroscopic AnalysesNMR spectra were obtained with a Bruker BioSpin AVANCE III HD 500 MHz spectrometer. 1H-NMR chemical shifts are reported in parts per million (ppm) using the proton resonance of residual solvent as reference: CDCl3 δ 7.26.22) 13C-NMR chemical shifts are reported relative to CDCl3 δ 77.16.22) Mass spectra were recorded with a Thermo SCIENTIFIC UltiMate 3000 Q Exactive Focus LC–MS using both positive and negative electrospray ionization (ESI). Samples were separated for analysis on an ACQUITY UPLC 1.8 µm, 2.1 × 50 mm C18 reversed phase column (Waters, U.S.A.) using a linear gradient of 5–100% (v/v) CH3CN in H2O supplemented with 0.05% (v/v) formic acid at a flow rate of 0.5 mL/min.
Fungal Strains and Culture ConditionsFor isolation of 1, the selected fungal strain was cultured in 5 L of MYS medium (10 g/L malt extract, 4 g/L yeast extract and 4 g/L starch), while isolation of 2 was performed with the fungus cultured in 5 L of CD-ST (Czapek Dox-starch, tryptone) liquid medium.16) In both cases, the culture was incubated at 30 °C for 3 weeks with shaking at 200 rpm. The culture was filtrated to separate the liquid medium from the mycelia. Subsequently, the cleared supernatant was extracted with ethyl acetate (2 × 5 L). The ethyl acetate extract was concentrated in vacuo to give an oily residue, which was then fractionated by silica gel column chromatography with CHCl3/CH3OH (1 : 0→0 : 1). The fractions eluted with CHCl3/CH3OH (100 : 1, 1) and (50 : 1, 2) were further purified by reversed-phase HPLC using COSMOSIL 5C18 MS-II, 20 × 250 mm (Nacalai Tesque Inc., Kyoto, Japan). For purification of 1, an isocratic elution system of 60% CH3CN (v/v) in H2O at a flow rate of 8.0 mL/min was applied, resulting in 1.2 mg of 1. For purification of 2, an isocratic elution system of 55% CH3CN (v/v) in H2O at a flow rate of 8.0 mL/min was used, resulting in 1.0 mg of 2.
Construction of the Plasmids for the Expression of Human Genes in Aspergillus sp.The plasmid pKW18601 (Supplementary Fig. 1) was constructed for the expression of the human NADPH:CYP oxidoreductase (POR) gene hPOR in Aspergillus sp. The hPOR gene was cloned from the human liver cell cDNA library using the pKW18601-F/pKW18601-R primer set (Supplementary Table 1). The hPOR gene was cloned into the parent plasmid pKW505415) and placed under the control of a tef1 promoter23) to create pKW18601. The plasmids pKW18613, pKW18615 and pKW18623 (Supplementary Fig. 1) were constructed for the expression of hCYP1A1, hCYP1A2 and hCYP3A4, respectively, in Aspergillus sp. The hCYP genes were cloned from the human liver cell cDNA library using the following primer sets: pKW18597-F/pKW18597-R for hCYP1A1, pKW18598-F/pKW18598-R for hCYP1A2 and pKW18600-F/pKW18600-R for hCYP3A4 (Supplementary Table 1). All three plasmids were constructed by inserting a copy of the hCYP gene under the control of a glaA promoter15) into the hPOR-carrying plasmid pKW18601.
Culture Media and Transformation of A. flavus, A. parasticus, A. lentulus and A. novofumigatusTransformation of the fungal strains with a plasmid was performed following the protocol described below. A. flavus, A. parasticus, A. lentulus and A. novofumigatus strains were initially grown on Czapek Dox (CD) agar plates at 30 °C for five days for sporulation. Approximately 1 × 108 to 1 × 109 of conidia that were collected from a single plate were used to inoculate 200 mL of CD liquid medium. The culture was shaken for 24 h at 30 °C. The grown mycelia were collected by filtration and washed with 0.8 M sodium chloride. The mycelia were incubated with 1 mL of 10 mM sodium phosphate buffer (pH 6.0) containing 0.8 M sodium chloride, 50 mg mL−1 lysing enzyme (Sigma-Aldrich, U.S.A.) at room temperature for 3 h. The resulting protoplasts were filtered and subsequently centrifuged at 2500 × g for 5 min at room temperature. The collected protoplasts were washed with 0.8 M sodium chloride and centrifuged to remove the wash solution. The protoplasts were suspended in 200 µL of STC buffer at pH 8.0 (0.8 M sodium chloride, 10 mM calcium chloride, and 10 mM Tris–HCl pH 8.0). Then 40 µL of PEG solution at pH 8.0 (400 mg/mL polyethylene glycol 4000, 50 mM calcium chloride and 50 mM Tris–HCl) was added to the protoplast suspension. The mixture was subsequently combined with 5 µg of the expression vector with which the strain was to be transformed. The mixture was incubated on ice for 20 min to allow the transformation to proceed. After incubation on ice, 1 mL of the PEG solution was added to the reaction mixture, and the mixture was incubated at room temperature for an additional 5 min. The treated protoplasts were plated on appropriate selection plates to select for the desired mutants.
PCR Analyses of the Transformed Aspergillus StrainsFor standard PCR analyses, the genomic DNAs from the A. flavus, A. parasticus, A. lentulus and A. novofumigatus strains were used as the templates. The gDNAs were prepared using the isolation buffer at pH 8.0 (10 mM Tris–HCl, 100 mM ethylenediaminetetraacetic acid (EDTA) and 0.5% (w/v) SDS).24) For RT-PCR analyses, cDNAs were synthesized from the Aspergillus strains being analyzed to be used as the templates. Total RNA was prepared using the RNAqueous™ Total RNA Isolation Kit (Thermo Fisher Scientific) from the CYP3A4- and lepE-transformed A. flavus A1421 strains. Briefly, when 1 or 2 was produced by the strain being investigated, the mycelia were taken out of the culture medium and freeze-dried overnight. The lyophilized mycelia were ground, and the total RNA was isolated using the protocol recommended by the manufacturer of the isolation kit. For cDNA synthesis, 1 µg of the total RNA was treated with TURBO™ DNase (Thermo Fisher Scientific), and cDNAs were synthesized using the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) following the method provided by the manufacture. For both types of PCR analyses, the reaction was performed for 35 cycles with PrimeSTAR GXL DNA Polymerase (TaKaRa Bio Inc.) using the cDNA as the template. The primer set CYP3A4_RT-F and CYP3A4_RT-R was used for detection of CYP3A4, and the primer set Afla_Act1_F and Afla_Act1_R was used for measuring the expression level of Actin-1 as a control. Refer to Supplementary Table 1 for the sequence information of the primers used.
We would like to thank the financial support from the Development of Innovative Research on Cancer Therapeutics from Japan Agency for Medical Research and Development (AMED) (K.W., 16ck0106243h0001; 19ck0106475h0001), Innovative Areas from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan (K.W., 16H06449), the Japan Society for the Promotion of Science (JSPS) (K.W., 19H02898; S.K., 19K15757; T.O., JP20K22568), the Takeda Science Foundation (K.W.), the Institution of Fermentation at Osaka (K.W.), the Princess Takamatsu Cancer Research Fund (K.W., 16-24825), Kobayashi Foundation for Cancer Research (K.W.), the Yakult Bio-Science Foundation (K.W.) and SECOM Science and Technology Foundation (K.W.).
The authors declare no conflict of interest.
This article contains supplementary materials.
