Polyene macrolide natural product mandimycin biosynthetic gene cluster as well as natural product and application thereof

Abstract

The invention discloses a biosynthetic gene cluster of a polyene macrolide natural product mandimycin as well as a natural product and application of the biosynthetic gene cluster. The nucleotide sequence of the biosynthetic gene cluster of the mandimycin is shown as SEQ ID NO.1, and the compound structural formulas of the natural product mandimycin and mandimycin B are shown as a formula I and a formula II. The natural product mandimycin can target fungal cell membrane phospholipid molecules, especially phosphatidylinositol, so that important ions in fungal cells are discharged, and fungal cell death is caused. The natural products mandimycin and mandimycin B disclosed by the invention have strong in-vivo and in-vitro antibacterial activity and broad-spectrum antibacterial spectrum on multiple drug-resistant fungus key pathogens published by WHO (World Health Organization), such as candida, aspergillus, cryptococcus, mucor and fusarium bacteria.

Description

Technical Field

[0001] The present invention belongs to the field of microbial natural products, and particularly relates to a polyene macrolide natural product mandimycin biosynthetic gene cluster, its natural product and application. Background Art

[0002] With the widespread use of fungal drugs, the increasing number of immunodeficient and immunocompromised patients, and the limited variety of antifungal drugs, fungal infections have become a major security issue threatening human health. Therefore, there is an urgent need to find new antifungal drugs. Currently, the antifungal drugs used clinically mainly include echinocandins (such as caspofungin), polyene macrolides (such as amphotericin B), azoles (such as fluconazole), and 5-fluorocytosine (Nature reviews microbiology 2022; 20:9557 - 571). Among them, polyene macrolide antibiotics represented by amphotericin B have broad-spectrum and potent antifungal activities and low drug resistance. Since the 1950s, they have been the first-line drugs for clinical antifungal treatment and are included in the WHO's list of essential drugs for the treatment of fungal infections (Antibiot Annu. 1955; 3:587 - 91; Curr Opin Microbiol. 2022; 70:102208). Currently, various polyene antibiotics with different structural types have been isolated from nature. Their main skeletons mainly include macrocyclic structures with 26 rings, 28 rings, 36 rings, and 38 rings, 3 to 7 double bond structures, and 1 - 2 deoxysugar substituent structures (Mol Phylogenet Evol. 2018; 127:239 - 247). Currently known polyene antibiotics mainly bind to the sterol molecule ergosterol on the fungal cell membrane to form transmembrane channels, release important intracellular ions and small molecules, thereby killing fungal pathogens and exerting potent antifungal activities (Proc Natl Acad Sci US A. 2012; 109(7):2234 - 9; Proc Natl Acad Sci U SA. 2011; 108(17):6733 - 8).

[0003] Although the marketed polyene antifungal drugs (such as amphotericin B, nystatin, and natamycin) have significant bactericidal activities, broad-spectrum properties, and relatively low fungal pathogen drug resistance and are potent antifungal drugs, the continuous increase in fungal drug resistance, extremely low water solubility, and severe side effects caused by long-term use make the existing polyene antifungal drugs unable to meet the clinical needs. In addition to its broad-spectrum and potent activities, amphotericin B has the outstanding advantage of low drug resistance. Its disadvantages are that it is inactive against multi-drug resistant bacteria that appear clinically, has large side effects, and low oral bioavailability, thus limiting its clinical application scope. Summary of the Invention

[0004] Object of the Invention: To solve the problems existing in the prior art, the present invention aims to provide a polyene macrolide natural product mandimycin biosynthetic gene cluster. The nucleotide sequence of the mandimycin biosynthetic gene cluster is shown as SEQ ID NO.1. The gene cluster of the present invention can produce a novel polyene macrolide antifungal natural product. Further, by means of genetic manipulation, the key glycosyltransferase gene of the mandimycin biosynthetic gene cluster is knocked out to construct an engineering strain CPU002, which biosynthesizes a novel antifungal antibiotic natural product mandimycin B lacking AtratcynoseA by using the CPU002 engineering strain, thus solving the technical problems such as the small variety of antifungal drugs currently available clinically, large toxic and side effects, low oral bioavailability, and ineffectiveness against multi-drug resistant fungi, which limit their clinical application scope. The present invention also provides the polyene macrolide natural products mandimycin and mandimycin B and their applications.

[0005] Technical Solution: To achieve the above object, the present invention provides a polyene macrolide natural product mandimycin biosynthetic gene cluster, and the nucleotide sequence of the mandimycin biosynthetic gene cluster is shown as SEQ ID NO.1.

[0006] The preparation method of the polyene macrolide natural product mandimycin of the present invention comprises the following steps:

[0007] Preparing a seed solution from a strain containing the mandimycin biosynthetic gene cluster, performing fermentation culture of the seed solution, and extracting, separating and purifying to obtain the natural product mandimycin.

[0008] Among them, the strain containing the mandimycin biosynthetic gene cluster is Streptomyces netropsis DSM40259.

[0009] The application of the polyene macrolide natural product mandimycin biosynthetic gene cluster of the present invention in preparing the natural product mandimycin and the natural product mandimycin B.

[0010] The preparation method of the polyene macrolide antifungal natural product mandimycin B of the present invention comprises the following steps:

[0011] Knock out the MandQ gene in the strain containing the mandimycin biosynthetic gene cluster, obtain an engineered bacterium, prepare it into a seed solution, culture, ferment, extract, separate and purify to obtain the natural product mandimycin B; the nucleotide sequence of the MandQ gene is as shown in SEQ ID NO.2.

[0012] Among them, Streptomyces-Escherichia coli indirect conjugation transfer was used to knock out the MandQ gene in the mandimycin biosynthetic gene cluster of Streptomyces netropsis DSM40259 to obtain the engineered bacterium CPU002. The obtained engineered bacterium CPU002 lacking glycosyltransferase was fermented, extracted, separated and purified to obtain mandimycin B.

[0013] The present invention relates to a polyene macrolide natural product mandimycin or mandimycin B and a pharmaceutically acceptable salt thereof. The structural formula of the natural product mandimycin compound is as shown in Formula I, and the structural formula of the natural product mandimycin B compound is as shown in Formula II:

[0014]

[0015] Furthermore, the core mother ring of the structural formula of the polyene macrolide natural product mandimycin or the structural formula of mandimycin B is a 38-membered macrolide. Among them: C20-C29 is a conjugated pentadiene structure, C32 is a monoene structure; C19 is connected to a mycarose; C3, C7, C10, C13 are connected to a β-hydroxy group; C15 is connected to an α-hydroxy group; C13 and C15 are connected by an oxygen bridge; C1 is an ester group; C5 is a ketone group; C34, C36, C37 are connected to an α-methyl group; C16 is connected to a formic acid.

[0016] The application of the polyene macrolide natural product mandimycin or mandimycin B and a pharmaceutically acceptable salt thereof in the preparation of an antifungal drug.

[0017] Furthermore, the fungus is any one of key pathogenic fungi such as Candida, Aspergillus, Cryptococcus, Mucor and Fusarium.

[0018] Furthermore, the Candida is Candida albicans, Candida auris, Candida glabrata, Candida tropicalis or Candida parapsilosis.

[0019] The present invention provides a pharmaceutical composition for an antifungal drug, comprising the polyene macrolide natural product mandimycin or the natural product mandimycin B and a pharmaceutically acceptable carrier.

[0020] Furthermore, the pharmaceutical composition is a capsule, powder, tablet, granule, pill, injection, syrup, oral liquid, inhalant, ointment, suppository or patch.

[0021] Use of the pharmaceutical composition of the antifungal drug of the present invention in the preparation of an antifungal drug.

[0022] Furthermore, the fungus is any one of a variety of fungal key pathogens such as Candida, Aspergillus, Cryptococcus, Mucor and Fusarium.

[0023] The present invention discovers a biosynthetic function gene cluster mandimycin-BGC expressing mandimycin from the microbial secondary metabolite database (MiSM) by a systematic evolution-guided directed mining method;

[0024] mandimycin-BGC contains 6 core genes (mandD, mandE, mandF, mandL, mandM, mandN), encoding 19 polyketide synthase modules.

[0025] Combinatorial biosynthesis discovers a natural product mandimycin with potent broad-spectrum anti-multidrug-resistant fungal activity. Mandimycin has 3 deoxysugar substituents (including a mycosamine at C19, a dideoxysaccharide atratcynose A ((α-L-oleandropyranosoyl-(1→4)-β-D-digitoxopyranoside) at C35, a pentene structural unit at C20, a monene structural unit at C32, and a special chemical structure of a 38-membered ring with a keto group at the 5-position;

[0026] The mandimycin compound of the present invention is a natural product produced by fermentation of Streptomyces netropsis DSM40259. For the screening of this Streptomyces, a directed mining technique combining big data analysis, phylogenetic analysis and clustering analysis was used: with the conserved carbomycin glycosyltransferase as a sequence tag, all biosynthetic gene clusters expressing the carbomycin sequence tag were mined from the microbial secondary metabolite database (MiSM) using the Hidden Markov Model, and then a novel branched biosynthetic gene cluster, mandimycin-BGC, was obtained through phylogenetic analysis. The gene sequence is shown in SEQ ID NO.1. By analyzing the strains potentially containing this gene cluster, this special Streptomyces (DSM40259) was obtained, which includes the synthetic gene cluster mandimycin-BGC. Mandimycin is unstable under acidic conditions and is prone to dehydration and de-side chain glycosylation. Therefore, neutral or weakly basic organic solvents can be used to dissolve mandimycin.

[0027] When exploring the mechanism of action of this compound, feeding experiments, scanning electron microscopy, ion concentration detection and isothermal titration calorimetry were used. During the preparation of mandimycin, it is necessary to explore the fermentation conditions to remove fermentation by-products, and carry out processes such as extraction, separation and purification of the crude product.

[0028] The present invention discovers the action target of a new antifungal drug. Mandimycin targets fungal cell membrane phospholipids (including phosphatidylinositol, phosphatidylglycerol, phosphatidylserine, phosphatidylethanolamine, phosphatidylcholine, sphingomyelin, cardiolipin), especially phosphatidylinositol, causing the efflux of important intracellular ions (including K + etc.) in fungi, leading to the death of fungal cells. This mechanism has not been reported in antifungal drugs.

[0029] Due to its unique multi-target mechanism of action, mandimycin does not develop drug resistance under laboratory conditions and has potent activity against drug-resistant strains of clinical antifungal drugs.

[0030] A novel biosynthetic gene cluster was discovered from the strain Streptomyces netropsis DSM 40259 by pan-genome mining. Further fermentation purification and knockout experiments verified that its product is mandimycin. Mandimycin is a polyene macrolide antifungal antibiotic with a completely new mechanism of action. It targets phospholipid molecules on the fungal cell membrane, especially the phosphatidylinositol structure, leading to the efflux of important intracellular ions and small molecules, resulting in fungal death. Mandimycin has potent broad-spectrum antifungal activity and shows strong activity against a variety of key fungal pathogens such as Candida, Aspergillus, Cryptococcus, Mucor, and Fusarium. Further, the present invention used Streptomyces-Escherichia coli indirect conjugation transfer to knockout the glycosyltransferase - MandQ responsible for the synthesis of atratcynose A in the mandimycin BGC by gene knockout technology, obtaining an engineered strain CPU002 lacking glycosyltransferase. The engineered strain was further fermented to biosynthesize a novel polyene macrolide natural product, mandimycin B. Mandimycin B of the present invention is a natural product produced by the fermentation of the engineered strain CPU002, that is, by knocking out the MandQ gene in the mandimycin biosynthetic gene cluster of Streptomyces netropsis DSM 40259 by gene knockout technology, and the obtained CPU002 engineered strain was cultured and fermented in the FS / 9 medium to produce it.

[0031] In the preparation of the synthetic natural product mandimycin B, the key enzyme MandQ responsible for the synthesis of mandimycin atratcynose A was discovered; by means of gene knockout, 1072 genes in MandQ were knocked out to disrupt the glycosyltransferase function of MandQ, and the engineered strain CPU002 lacking the MandQ enzyme was obtained; and the engineered strain CPU002 was used to synthesize a novel mandimycin derivative without atratcynose A, namely mandimycin B. It was found through experiments that mandimycin B also has potent broad-spectrum antifungal activity. Different from the mechanism of action of mandimycin, mandimycin B exerts its antifungal activity by binding to ergosterol on the fungal cell membrane. The novel mandimycin derivative mandimycin B provided by the present invention is different from the mandimycin compound. Although mandimycin B and mandimycin have the same parent nucleus, due to the lack of the disaccharide structure of the key atratcynose A, it does not have the disaccharide structural unit of atratcynose A. Mandimycin B has a mechanism of action different from that of mandimycin. It exerts its antifungal activity by binding to ergosterol on the fungal cell membrane and has no affinity for the target phospholipid molecule bound by mandimycin.

[0032] Advantages: Compared with the prior art, the present invention has the following remarkable advantages:

[0033] The present invention first proposed the biosynthetic gene cluster of the polyene macrolide natural product mandimycin. The strain containing the mandimycin biosynthetic gene cluster was prepared into a seed solution for culture, fermentation, extraction, separation and purification to obtain the natural product mandimycin.

[0034] The natural product mandimycin obtained in the present invention has potent antibacterial activity and a broad antibacterial spectrum against a variety of pathogenic microorganisms with multiple drug resistance of fungal priorities announced by the WHO, including Candida, Aspergillus, and Cryptococcus. The minimum inhibitory concentration (MIC) of the compound of the present invention against Candida, especially against multi-drug resistant bacteria including Candida albicans and Candida auris, is 1-2 times that of nystatin used clinically, and is equivalent to the activity of amphotericin B, between 0.25-1 μg / mL; the activity against multi-drug resistant bacteria of Cryptococcus neoformans reaches 0.125 μg / mL, which is 32 times that of caspofungin, 64 times that of fluconazole, and 128 times that of 5-fluorocytosine; the activity against multi-drug resistant bacteria of Aspergillus fumigatus reaches 2 μg / mL, which is 32 times that of fluconazole and 16 times that of 5-fluorocytosine; it shows good antibacterial activity against amphotericin B-resistant strains. It shows that mandimycin has a broad spectrum of antifungal activity and potent antibacterial activity against clinically antifungal drug-resistant strains. The natural product mandimycin B of the present invention has potent antibacterial activity and a broad antibacterial spectrum against a variety of pathogenic microorganisms with multiple drug resistance of fungal priorities announced by the WHO, including Candida, Aspergillus, and Cryptococcus. Among them, the minimum inhibitory concentration (MIC) of mandimycin B against Candida, especially against multi-drug resistant bacteria including Candida albicans and Candida auris, is between 1-2 μg / mL; the activity against multi-drug resistant bacteria of Cryptococcus neoformans reaches 0.5 μg / mL; the activity against multi-drug resistant bacteria of Aspergillus fumigatus reaches 2 μg / mL, which is 32 times that of fluconazole and 16 times that of 5-fluorocytosine.

[0035] The action target of the natural product mandimycin of the present invention is different from that of known antifungal drugs, but acts on the phospholipid components on the fungal cell membrane. Among them, phosphatidylinositol on the fungal cell membrane is the target with the best binding force, and the K d value is 21.9 μM; followed by the phosphatidylglycerol target, and the K d value is 28.2 μM; the phosphatidylserine target, and the K d value is 30.9 μM; the phosphatidylethanolamine target, and the K d value is 36.5 μM; the phosphatidylcholine target, and the K d value is 36.8 μM; the sphingomyelin target, and the K d value is 50 μM; the cardiolipin target, and the K d value is 63.6 μM. And under laboratory conditions, low-concentration mandimycin does not produce drug-resistant bacteria. Compared with amphotericin B, the compound of the present invention has better water solubility, and its water solubility is 9700 times that of amphotericin B.

[0036] In vitro antifungal activity assays showed that mandimycin exhibited significant antifungal activity against multi-drug resistant Candida albicans and multi-drug resistant Candida auris in a dose-dependent manner. At a dose of 5 mg / kg, the fungal cells of Candida albicans BNCC186382 decreased by 3 log10, and the fungal cells of Candida auris BNCC357785 decreased by 2 log10. At a dose of 10 mg / kg, the fungal cells of Candida albicans BNCC186382 decreased by 3.6 log10, and the fungal cells of Candida auris BNCC357785 decreased by 2.5 log10. At a dose of 20 mg / kg, the fungal cells of Candida albicans BNCC186382 decreased by 4.2 log10, and the fungal cells of Candida auris BNCC357785 decreased by 3 log10. In a pan-drug resistant Candida auris infection model resistant to amphotericin B, mandimycin still showed good in vivo activity. At a dose of 10 mg / kg, the strain content decreased by 2.3 log10. No acute toxicity was observed in all in vivo animal experiments with different concentrations of treatment. The natural product mandimycin of the present invention can be used to prepare antifungal drugs. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is the process of mining the mandimycin biosynthetic gene cluster;

[0038] Figure 2 is the biosynthetic gene cluster and structure of mandimycin;

[0039] Figure 3 is the predicted biosynthetic process of mandimycin;

[0040] Figure 4 is the ultraviolet spectrum of mandimycin;

[0041] Figure 5 is the comparison of HPLC and activity of the products of mandimycin gene knockout strains and wild strains;

[0042] Figure 6 is the high-resolution mass spectrum of mandimycin (A: positive ion, B: negative ion);

[0043] Figure 7 is mandimycin 1 H-NMR(DMSO-d 6 ) nuclear magnetic resonance spectrum;

[0044] Figure 8 is mandimycin 13 C-NMR(DMSO-d 6)Nuclear magnetic resonance spectrum;

[0045] Figure 9 is mandimycin HSQC-NMR (DMSO-d 6 )Nuclear magnetic resonance diagram;

[0046] Figure 10 is mandimycin HMBC-NMR (DMSO-d 6 )Nuclear magnetic resonance spectrum;

[0047] Figure 11 is mandimycin COSY-NMR (DMSO-d 6 )Nuclear magnetic resonance spectrum;

[0048] Figure 12 is mandimycin ROESY-NMR (DMSO-d 6 )Nuclear magnetic resonance spectrum;

[0049] Figure 13 is the structure of mandimycin and two-dimensional nuclear magnetic resonance signal spectrum;

[0050] Figure 14 is the chiral bioinformatics analysis of mandimycin polyketide reductase;

[0051] Figure 15 is the cytotoxicity of mandimycin to HepG2 cells;

[0052] Figure 16 is the hemolytic property of mandimycin;

[0053] Figure 17 is the in vivo antifungal activity of mandimycin, (A) Multidrug-resistant Candida albicans BNCC186382 infection model; (B) Multidrug-resistant Candida albicans BNCC357785 infection model; (C) Pan-drug-resistant Candida auris AMR05 infection model;

[0054] Figure 18 Disseminated candidiasis mouse model;

[0055] Figure 19 Invasive candidiasis mouse model;

[0056] Figure 20 Mouse fungal skin infection model;

[0057] Figure 21 Mouse vaginal infection model;

[0058] Figure 22 In vivo pharmacokinetic study of mandimycin;

[0059] Figure 23 Evaluation of the nephrotoxicity of mandimycin in vivo;

[0060] Figure 24 For the drug resistance results of mandimycin and the change multiples of its activity against other polyene drug-resistant bacteria, where (A) amphotericin B-resistant bacteria, (B) natamycin-resistant bacteria, (C) nystatin-resistant bacteria;

[0061] Figure 25 Is the bactericidal curve of mandimycin;

[0062] Figure 26 Is the state of the bacteria observed by scanning electron microscopy after treating Candida albicans with mandimycin;

[0063] Figure 27 Is the change in K+ concentration after treating Candida albicans with mandimycin;

[0064] Figure 28 Is the inhibition curve of the activity of mandimycin by different components of fungal cells;

[0065] Figure 29 Is the exothermic curve of mandimycin by phospholipid small molecules;

[0066] Figure 30 Is the 1072bp fragment of MandQ in the mandimycin gene cluster knocked out by gene editing;

[0067] Figure 31 Is the determination of mandimycin B by HPLC;

[0068] Figure 32 Is for mandimycin B 1 1H NMR spectrum;

[0069] Figure 33 Is for mandimycin B 13 13C NMR spectrum;

[0070] Figure 34 Is the HSQC spectrum of mandimycin B;

[0071] Figure 35 Is the HMBC spectrum of mandimycin B;

[0072] Figure 36 Is the COSY spectrum of mandimycin B;

[0073] Figure 37 Is the TOCSY spectrum of mandimycin B;

[0074] Figure 38 Correlations of the key HMBC and COSY spectra of mandimycin B;

[0075] Figure 39 Determination of the target of mandimycin B by UV-vis;

[0076] Figure 40 Determination of the target of mandimycin B by ITC. Detailed implementation mode

[0077] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings. The materials, reagents, etc. used in the following examples can be obtained from commercial channels without special instructions. The experimental methods without specific conditions noted in the examples are usually carried out under conventional conditions or according to the conditions recommended by the manufacturer.

[0078] Among them, the strain Streptomyces netropsis DSM40259 was purchased from the DSMZ platform. Casein tryptic digest (purchased from Guangdong Huankai Microbial Technology Co., Ltd., product number 024048). Soybean digest (purchased from Beijing Hongrun Baoshun Technology Co., Ltd., product number Y030A220816). Beef extract (purchased from Beijing Hongrun Baoshun Technology Co., Ltd., product number Y014C 230809).

[0079] Example 1

[0080] Discovery of the mandimycin biosynthetic gene cluster

[0081] As Figure 1 shown, for the mining process of the mandimycin biosynthetic gene cluster, based on the carbonmycin transferase sequence information of known polyene macrolide natural products (Table 1), the conserved protein structure sequence was analyzed and obtained, a specific hidden Markov model of polyene carbonmycin transferase was constructed, and this model was used to perform a sequence similarity scan on the microbial secondary metabolite database. For protein sequences with a similarity value less than 1e -181 were regarded as polyene natural product-related carbon enzyme sugar sequences, and the functional gene cluster where they were located was the functional gene cluster encoding potential novel polyene natural products. Through this method of retrieval, 280 deduplicated candidate sequences were obtained, phylogenetic trees of these sequences were constructed, and the phylogenetic relationships between different sequences were analyzed. As Figure 2 shown, a new evolutionary branch was found, encoding a brand-new polyene natural product, and we named this BGC mandimycin-BGC, and the sequence is shown in SEQ ID NO.1. As Figure 3As shown, the predicted biosynthetic process of mandimycin, where the functional annotations of the biosynthetic genes are shown in Table 2. Further screening revealed that the fermentation of Streptomyces netropsis DSM40259 contains the biosynthetic gene cluster mandimycin - BGC, which can be used for the fermentation of the natural product mandimycin.

[0082] Table 1 Carbon enzyme sugar sequence information of known polyene macrolide natural products

[0083]

[0084] Table 2 Functional annotations of the biosynthetic gene cluster of mandimycin

[0085]

[0086]

[0087] Example 2

[0088] Biological fermentation of mandimycin

[0089] (1) Preparation of spore suspension

[0090] The strain Streptomyces netropsis DSM40259 containing the mandimycin biosynthetic gene cluster was spread on ISP4 solid medium (containing 10.0 g soluble starch, 1.0 g dipotassium hydrogen phosphate, 1.0 g magnesium sulfate, 1.0 g sodium chloride, 2.0 g ammonium sulfate, 2.0 g calcium sulfate, 0.001 g ferrous sulfate, 0.001 g manganese chloride, 0.001 g zinc sulfate, 15.0 g agar, pH = 7.2) per liter of medium, and cultured at 30 °C for 5 days to prepare a spore suspension.

[0091] (2) Preparation of seed liquid:

[0092] Take 1 mL of the spore suspension and inoculate it into 50 mL of TSB medium (containing 17.0 g casein tryptic digest, 3.0 g soybean digest, 5.0 g sodium chloride, 2.5 g dipotassium hydrogen phosphate, 2.5 g glucose monohydrate, pH = 7.3) per liter of medium, and place it on a shaker (conditions: rotation speed 200 rpm, temperature 30 °C) and culture for 2 days to prepare the seed liquid.

[0093] (3) Preparation of fermentation broth:

[0094] Transfer 0.5 mL of the seed solution prepared in step (2) of this example to 50 mL of F2 fermentation medium (containing per liter of medium: 69.0 g glucose, 25.0 g beef extract, 9.0 g CaCO 3and 0.1 g KH 2 PO 4 ) and cultured for 10 days (shaker rotation speed 200 rpm, 30 °C) to prepare the fermentation broth.

[0095] Example 3

[0096] Extraction, separation and purification of mandimycin

[0097] After fermentation, add n-butanol to the fermentation flask (volume ratio of fermentation broth to n-butanol is 1:1), stir overnight for extraction (100 rpm), and after the n-butanol extract is dried by rotary evaporation, dissolve it in methanol solution to obtain the crude extract. The purification of the crude extract adopts a two-step method. First, isocratic elution is used, and column chromatography separation is carried out using a YMC-GEL C 18 powder (12 nm × 50 μm) packing column. The column is equilibrated with 2 cv of 10% methanol (H 2 O:CH 2 OH 9:1), add the methanol solution for dissolution, and elute with different concentrations of methanol aqueous solutions (10%, 30%, 50%, 70%, 90%, 100%), 100 mL per bottle. The components are detected by UPLC-MS, and the fractions containing mandimycin (eluate of 90% methanol aqueous solution) are collected and dried by rotary evaporation. The UPLC-MS conditions are as follows: C18 chromatographic column (Waters, T3-1.8 μm, 2.1 × 100 mm), mobile phase: phase A - pure water (0.1% formic acid), phase B - acetonitrile (0.1% formic acid), flow rate, 0.6 mL / min, the gradient of mobile phase B is 30% - 90%, increasing by 10% per minute. The MS detection range is 200 - 2000, and both positive and negative modes are carried out simultaneously. Then, semi-preparative reversed-phase high performance liquid chromatography (RP-HPLC) is used to further purify the semi-pure product. The RP-HPLC conditions are as follows: C18 chromatographic column (Shimadzu, ShimNet HE C18-AQ, 5 μm OBD, 19 × 250 mm chromatographic column), solvent A - deionized water; solvent B - acetonitrile. The flow rate is 3 mL / min, the gradient of solvent B is 30% - 90%, increasing by 1.5% per minute, and full wavelength detection (190 nm - 800 nm). As Figure 4 shown, the UV spectrum of mandimycin contains peaks with characteristic UV absorption of conjugated pentaene (320, 335, 352 nm). The mandimycin natural product (yellow powder) with a purity exceeding 95% is collected, which is the purified natural product mandimycin, and the yield is 4 - 6 mg / L of fermentation broth.

[0098] Example 4

[0099] In vivo knockout verification of mandimycin

[0100] To study the relationship between mandimycin - BGC and conjugated pentaene polyene natural products, genetic manipulation at the gene level was carried out on the Streptomyces strain that produces this compound. Homologous arms of 1 kb each upstream and downstream of mandL were amplified by PCR (the primer sequences are shown below), and cloned into the pKC1139 vector to construct the pKC1139 - MandL_KO knockout vector. Through the Streptomyces conjugation transfer experiment, the key 692 - base sequence from - 334 to + 358 of MandL was knocked out (shown in Table 2). Comprehensive comparative analysis of secondary metabolites was performed on wild - type and knockout strains, as Figure 5 shown, to confirm that the conjugated pentaene natural product is a biosynthetic product of mandimycin - BGC, and this compound was named mandimycin.

[0101]

[0102] Example 5

[0103] Structure identification of mandimycin

[0104] The purified mandimycin in Example 3 was a yellow powder. As Figure 6 shown, OrbiTrap high - resolution analysis showed that its protonated ion was m / z 1198.6344 [M + H] + ; m / z 1196.6228 [M - H] - , indicating its molecular formula C 60 H 95 NO 23 (Δppm - 1.97), containing 14 double - bond equivalents (DBE). As Figure 4 shown, the ultraviolet - visible spectrum of mandimycin in methanol showed maximum absorption wavelengths at 320, 335, and 352 nm, indicating the presence of a conjugated pentaene structure. The mandimycin isolated and purified in Example 4 was dissolved in deuterated DMSO, and using a 700M nuclear magnetic resonance instrument, as Figures 7 - 12 shown, the 1H spectrum, 13C spectrum, and two - dimensional spectra such as HSQC, HMBC, and COSY were obtained to determine the structure of mandimycin. The 1 1H and 13 13C NMR spectral data of mandimycin showed 12 allylic protons (δ H 5.60 - 6.30, δ C 129.0 - 136.0), multiple oxygen - containing methylenes (δ C 63.0 - 87.0) and three carbonyls (δ CThe NMR signals of 208.4, 174.8, and 170.1) indicated that mandimycin has a highly oxidized polyene macrolactone skeleton. A series of 2D NMR techniques (HSQC, HMBC, COSY, ROSEY) were used to analyze and determine a 38-membered macrolactone skeleton. In addition, through the HMBC correlations of H-3 (δ H 4.29), H-4 (δ H 2.53; 2.59), and H-6 (δ H 2.42, 2.47) to C-5 (δ C 208.4), it was further confirmed that C5 was substituted by a keto group, which is different from the cases where C5 is substituted by a hydroxyl group or a methylene group in other 38-membered macrolactone polyene antibiotics.

[0105] By corresponding to the NMR signals of three acetal carbons (δ H-1′ 4.49, δ C-1′ 96.8; δ H-1″ 4.42, δ C-1″ 99.5; δ H-1″′ 4.61, δ C-1″′ 100.0), the presence of three glycosyl groups in the mandimycin structure was determined. COSY and HMBC spectral analyses revealed the presence of a mycosamine, a digitoxose, and a 3-O-methyldigitoxose in mandimycin. By observing the HMBC correlation from H-1′ (δ H 4.49) to C-19 (δ C 74.7), the connection of mycosamine to C-19 of the macrolactone was determined. By the support of the HMBC correlation from H-1″ (δ H 4.42) to C-35 (δ C 84.1), the connection of digitoxose to C-35 was determined. In addition, through the confirmation of the HMBC correlation from H-1″′ (δ H 4.61) to C-4″ (δ C 86.8), it was found that a glycosidic bond was formed between C-1″′ and C-4″ between 3-O-methyldigitoxose and digitoxose. This is the first report of the presence of trisaccharide substitution in polyene macrolide antibiotics. 1 H and 13 C spectrum (DMSO-d6) NMR data are shown in Table 3.

[0106] Table 3 1 H and 13 C spectrum (DMSO-d6) NMR data of mandimycin

[0107]

[0108]

[0109] *The assignment of some carbon signals was supported by HSQC and HMBC correlations. # The coupling constants for proton signals were not provided as most signals are highly overlapped or broad.

[0110] Furthermore, based on the H spectrum, C spectrum, and two-dimensional spectra such as HSQC, HMBC, and COSY ( Figures 7 - 13 ) and the chiral bioinformation of mandimycin polyketide synthase ( Figure 14 ), it was analyzed that the specific configuration of the natural product mandimycin of the present invention is 2R, 10R, 11S, 12S, 13R, 14R, 16R, and 18R. The natural product mandimycin is shown as formula I:

[0111]

[0112] Example 6

[0113] The water solubility of mandimycin

[0114] Prepare methanol stock solutions of amphotericin B (1.2 mg / mL) and mandimycin (12 mg / mL) purified in Example 3. A 100 μL portion of each stock solution was lyophilized under vacuum for 24 hours. To the lyophilized samples, 50 μL (for amphotericin B) or 2 μL (for mandimycin) of 10 mmol Tris-HCl buffer (pH = 7.0) was added, and the solution was saturated by vortexing for 15 minutes. The resulting supernatant was centrifuged at 12,000 rpm for 3 min and then diluted in Tris-HCl buffer (amphotericin B was diluted 10-fold; mandimycin was diluted 1000-fold). 10 μL of the sample was injected, and the compound concentration in the solution was determined by measuring the UV absorption spectrum using a Shimadzu PDA detector spectrophotometer. Specifically, amphotericin B and mandimycin were accurately weighed and dissolved in DMSO to prepare a 2 mg / mL stock solution. The peak area of the characteristic heptaene UV peak at the maximum absorption wavelength of 384 nm was used for the quantification of amphotericin B; the peak area of the characteristic pentaene UV peak at the maximum absorption wavelength of 334 nm was used for the quantification of mandimycin. Standard curves of the relationship between the UV absorption peak area and mass of mandimycin and amphotericin B were plotted respectively, and the solubility of mandimycin and amphotericin B in aqueous solution was calculated. As shown in Table 4: the solubility of mandimycin was 8.07 mg / mL, which was 9700 times that of amphotericin B.

[0115] Table 4 Solubility test of mandimycin

[0116]

[0117] Example 7

[0118] Biological activity analysis of mandimycin

[0119] (1) In vitro anti-multidrug resistant bacteria activity detection

[0120] Using the CLSI standard, the antibacterial activity of mandimycin against the priority fungal pathogens announced by the WHO was determined. The results are shown in Table 5 below. Mandimycin has potent and broad-spectrum anti-multidrug resistant fungal activity.

[0121] Table 5 Antifungal activity of mandimycin

[0122]

[0123] Note: CPF R , Caspofungin resistance; FCA R,Fluconazole resistance;TBF R ,Terbinafine resistance;5-FU R ,5-Fluorocytosine resistance;Amp R ,Amphotericin B resistance.All MIC values were measured in duplicate and repeated three independent times with consistent results.

[0124] Meanwhile, under the same experimental conditions as above, the minimum inhibitory concentration (MIC) of the natural product mandimycin against multidrug-resistant bacteria including Candida albicans and Candida auris was 1-2 times that of nystatin A1 used clinically, equivalent to the activity of amphotericin B, between 0.25-1 μg / mL; the activity against Cryptococcus neoformans multidrug-resistant bacteria reached 0.125 μg / mL, which was 32 times that of caspofungin, 64 times that of fluconazole, and 128 times that of 5-fluorocytosine; the activity against Aspergillus fumigatus multidrug-resistant bacteria reached 2 μg / mL, which was 32 times that of fluconazole and 16 times that of 5-fluorocytosine; it showed good antibacterial activity against amphotericin B-resistant strains.

[0125] (2) Cytotoxicity of mandimycin

[0126] The MTT (2-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was used to determine the cytotoxicity of mandimycin. HepG2 cells cultured in DMEM (containing 10% fetal bovine serum) were seeded into 96-well flat-bottomed microplates (2,500 cells per well) and cultured at 37 °C and 5% CO 2 conditions. After 24 hours, the medium was aspirated, and 100 μl of fresh medium containing serial concentrations of mandimycin (maximum DMSO concentration less than 0.25%) was added. After incubation at 37 °C for 48 hours, the medium was removed, and 110 μl of MTT solution (10 μl of 5 mg / ml MTT premixed with 100 μl of DMEM in mandimycin) was added to each well. At 37 °C and 5% CO 2After culturing for 3 hours under the conditions, 100 μl of lysate (40% DMF, 16% SDS and 2% acetic acid aqueous solution) was added to dissolve the precipitate. Then, the absorbance of each well was measured at a wavelength of OD570nm using a microplate reader (Epoch microplate spectrophotometer, BioTek). Amphotericin B was used as a positive control. The calculation of the IC50 value (Prism 7.0) refers to the concentration required for each compound to inhibit cell growth by 50% relative to the compound-free control. The IC50 of mandimycin against Hepg2 cells = 57.56 μM, the IC50 against HK2 cells = 88.67 μM, the IC50 against PANC-1 cells = 48.28 μM, and the IC50 against SK-Hep-1 cells = 76.70 μM. As Figure 15 shown, mandimycin has low toxicity.

[0127] (3) Hemolytic activity of mandimycin

[0128] According to the previously reported method (Kelvin JYWu et al., 2024, Science, 383(6684), 721 - 726), the hemolytic activity of mandimycin was detected. Fresh sterile defibrinated sheep blood was centrifuged at 3,000 rpm for 10 minutes at 4°C to separate the precipitated blood cells, and then they were resuspended in PBS solution (pH 7.4) to prepare a suspension with a concentration of 1×10 9 cells / mL. Test compounds with concentrations ranging from 0.39 μM to 100 μM were prepared and mixed with the blood cells, with a final volume of 500 μL. 0.5% DMSO and 1% Triton X100 were used as negative (0% hemolysis) and positive (100% hemolysis) controls respectively. After incubation at 37°C for 3 hours, the mixture was centrifuged at 3,000 rpm for 20 minutes, and the supernatant was collected and then transferred to a 96-well polypropylene plate. The absorbance of the supernatant at OD 540 nm was measured using an enzyme-linked immunosorbent assay (ELISA) reader to determine the degree of hemolysis. As Figure 16 shown, the results showed that mandimycin did not show hemolytic activity even at a high concentration of 100 mM, while amphotericin B showed severe hemolysis at 12.5 μM, indicating that mandimycin has better safety.

[0129] Example 8

[0130] In vivo antibacterial activity evaluation of mandimycin

[0131] (1) Neutropenic mouse thigh muscle infection model

[0132] Six-week-old, specific pathogen-free female ICR mice weighing 23 - 27 g were used. Four days (150 mg / kg) and 1 day (100 mg / kg) before pathogen infection, the mice were intraperitoneally injected with cyclophosphamide (Sigma-aldrich). On the first day of infection, 0.05 mL of the inoculum (the inoculum was 2 x 10 7 CFU / mL of multi-drug resistant Candida albicans BNCC186382 / multi-drug resistant Candida auris BNCC357785 / pan-drug resistant Candida auris AMR05 resistant to amphotericin B) was injected into the thigh muscle of each mouse. Two hours after infection, the mice were treated by subcutaneous injection of the compound mandimycin prepared in Example 3 at different concentrations (20 mg / kg, 10 mg / kg, 5 mg / kg), injected once every 8 hours, and the treatment period was 24 hours. The mice were sacrificed by cervical dislocation and the number of colonies in the thigh muscle tissue was counted. As Figure 17 shown, all the mandimycin treatment groups had significant therapeutic effects. At a dose of 20 mg / kg, the fungal cells of Candida albicans BNCC186382 decreased by 4.2 log10, and the fungal cells of Candida auris BNCC357785 decreased by 3 log10; at the lowest dose of 5 mg / kg, the fungal cells of Candida albicans BNCC186382 also decreased by 3 log10, and the fungal cells of Candida auris BNCC357785 decreased by 2 log10. And this effect was better than that of amphotericin B in the same group. Against pan-drug resistant Candida auris resistant to amphotericin B, mandimycin showed good in vivo activity. At 10 mg / kg, the fungal cells of Candida auris AMR05 decreased by 2.4 log10. At the same time, neither group of mice showed acute toxicity at a dose of 20 mg / kg

[0133] (2) Disseminated candidiasis mouse model

[0134] Pathogen-free female ICR mice (Hangzhou Medical College, China), 6 weeks old, weighing 23 - 27 g, were used. The mice were randomly caged, four in each cage, and underwent three days of adaptation training before the experiment. A single colony of Candida albicans BNCC 186382 was inoculated into 5 mL of YPD liquid medium and shaken overnight at 30 °C and 220 rpm. The overnight fungal culture was washed three times with 0.9% sterile saline and then diluted to a final concentration of 2 x 10 7 CFU / mL. Subsequently, 50 μL of the diluted fungal suspension was subcutaneously injected via the tail vein to inoculate approximately 1 x 10 7CFU. Six hours after infection, mice were subcutaneously injected with single doses of 1 mg / kg, 3 mg / kg, 5 mg / kg, and 10 mg / kg of mandimycin (formulated with 10% DMSO and 10% Tween 80). Mice in the oral group were orally administered 10 mg / kg of mandimycin, and the formulated solution contained 5% DMSO and 10% Tween 80. Twenty-four hours after infection, the mice were euthanized, and the kidney and lung tissues were aseptically removed, weighed, homogenized, and then cultured on YPD agar and incubated at 30 °C for fungal burden counting by CFU counting. All chart data are represented by individual data points per group, and the results are as Figure 18 For doses of 1, 3, 5, and 10 mg / kg of mandimycin, the cell numbers of MDR Beauveria bassiana in kidney sections decreased by 0.93, 2.34, 2.52, and 3.43 log10 within 24 hours, respectively, and statistical analysis was performed using GraphPad Prism 9.

[0135] (3) Mouse model of invasive candidiasis

[0136] Pathogen-free female ICR mice (Hangzhou Medical College, China), 6 weeks old, weighing 23 - 27 g, were used. The mice were randomly caged, 6 mice per cage, and underwent 3 days of adaptive training before the experiment. To induce immunosuppression, the mice were intraperitoneally injected with cyclophosphamide (200 mg / kg) on day -2 and subcutaneously injected with cortisone acetate (500 mg / kg) on day +3. To prevent cross-infection, the mice were orally administered enrofloxacin at a concentration of 50 μg / mL in drinking water from day 1 to day 3, and then subcutaneously injected with ceftazidime (5 μg / dose) from day 0 to day 9. Invasive candidiasis was induced by intravenous injection of 1×10 6 CFU of Candida albicans BNCC 186382. Treatment began 16 hours after infection, including subcutaneous injection of cefotaxime, and single doses of 1 mg / kg, 5 mg / kg, 10 mg / kg, 20 mg / kg of mandimycin (formulated with 10% DMSO and 10% Tween 80) and 10 mg / kg of amphotericin B were injected daily for four consecutive days. The mice were monitored for a total of 20 days, and the survival rate was plotted using GraphPad Prism 9, and the results are as Figure 19 When the dose of mandimycin was 10 mg / kg, the survival rate of the mice reached 100%. In contrast, the survival rate of the amphotericin B group was only 80%, indicating that mandimycin has a more superior efficacy in improving the survival rate.

[0137] (4) Mouse model of fungal skin infection

[0138] A mouse skin infection model was established to evaluate the efficacy of mandimycin in the treatment of cutaneous fungal infections. BALB / c mice weighing 20 - 22 g (Hangzhou Medical College, China) were used. The mice were randomly housed in cages of four and underwent a three-day adaptation training before the experiment. Neutropenia was induced by intraperitoneal injection of 50 mg / kg cyclophosphamide on the third and first days before infection. Subsequently, the mice were anesthetized by intraperitoneal injection of 50 mg / kg pentobarbital sodium, and a full-thickness skin punch with a diameter of 0.8 cm was used to create a skin perforation on the dorsal skin. A suspension of Candida albicans BNCC 186382 (1×10 8 CFU / ml, 50 μL per mouse) was inoculated into the circular wound, and then gently blown until the skin was moist but without excess fluid. One day after infection, the wounds were locally treated with mandimycin (2.5 mg / kg or 7.5 mg / kg), amphotericin B (2.5 mg / kg or 7.5 mg / kg), or vehicle (PBS containing 10% dimethyl sulfoxide and 10% Tween 80). Mice in which wounds were created but not infected with fungi served as the negative control group and received only drug treatment. All compounds were administered once a day for 5 consecutive days. On days 1, 5, 9, and 11 after infection, the wounds were photographed and their sizes were measured. On day 11, the fungal counts of the wound specimens were recorded, and the wound specimens were collected. The results were as Figure 20 , mandimycin showed significant activity in the skin infection model. When the dosage was 2.5 mg / kg, the fungal burden was reduced by more than 2 log10, and the wound size and inflammation were also significantly reduced.

[0139] (5) Mouse vaginal infection model

[0140] Female BALB / c mice, weighing 19 - 21 g, were acclimatized for three days before the experiment. The mice were randomly housed in cages of four and acclimatized for three days before the experiment. On day 1, the mice were subcutaneously injected with 10 mg / kg estradiol benzoate once a day for 5 consecutive days to induce estrus (40, 41). On day 6, 50 μL of a suspension of Candida albicans BNCC 186382 (1×10 10(CFU / ml), invert the mice for 5 minutes after vaginal inoculation. After continuous infection for 3 days, the mice were normally fed for 1 day, and then subcutaneously injected with mandimycin (10 mg / kg), amphotericin B (10 mg / kg), rezafungin (10 mg / kg) or isavuconazole (10 mg / kg) once a day for 5 consecutive days. Mice infected with Beauveria bassiana but not treated with the compound were used as the control group, and they were injected with PBS containing 10% DMSO and 10% Tween 80. On the day after the last administration, the vagina was repeatedly rinsed with sterile PBS (20 μL), the vaginal lavage fluid was aspirated with a pipette, and the samples were cultured on MRS agar plates to count the Beauveria bassiana colonies. Finally, all animals were anesthetized and euthanized with ether, and the vaginal tissues were collected. The results are as Figure 21 , in the treatment of vaginal candidiasis, mandimycin also showed significant efficacy. After 5 days of treatment (10 mg / kg, intravenous injection, once a day), the vaginal fungal burden was significantly reduced by 2.51 log10. This efficacy is comparable to that of other well-known antifungal antibiotics (including amphotericin B, rezafungin and isavuconazole). In addition, the inflammation of the mice treated with mandimycin was significantly reduced, and the vaginal mucosa almost completely recovered

[0141] It can be seen from Examples 7-8 that the natural product mandimycin of the present invention not only exhibits excellent in vitro and in vivo antifungal activities, but also has no hemolytic property and no acute toxicity, and can be used to prepare antifungal drugs.

[0142] Example 9

[0143] Pharmacokinetic study of mandimycin in vivo

[0144] The study used pathogen-free male Sprague-Dawley rats (180 - 220 g, 7 - 8 weeks old, 3 rats per group). The mice were randomly caged, three per cage, and acclimated for three days before the experiment. The rats were injected subcutaneously with 25 mg / kg of mandimycin. Blood samples (about 0.15 mL) were collected from the jugular vein catheter into tubes containing sodium heparin at 5, 10, 20, and 30 minutes before drug administration, and at 1, 2, 4, 6, 8, 12, and 24 hours after drug administration. After blood sample collection, it was placed on ice and then centrifuged (8000×g, 5 minutes) to separate the plasma. Then the plasma was transferred and immediately frozen (-70 °C or below) until analysis. Mandimycin in rat plasma was analyzed by liquid chromatography - electrospray ionization - tandem mass spectrometry (LC-ESI-MS / MS), which consisted of an ABSCIEX Triple Quad 6500 system and an HPLC system equipped with a Quaternary Solvent Manager-R solvent distribution device and a Sample Manager FTN-R autosampler. Diazepam was used as the internal standard (IS). Mass quantification of mandimycin from m / z 1198.30 to m / z 725.20 and diazepam from m / z 285.00 to m / z 193.00 was performed using multiple reaction monitoring (MRM) in the positive ion mode. Mandimycin and IS were separated by high performance liquid chromatography (Waters ACQUITY C18 column, 1.9 μm, 100×2.1 mm). The isocratic mobile phase consisted of 80% acetonitrile and 20% (v / v) 5 mM ammonium acetate and passed through the mass spectrometry electrospray ionization chamber at a rate of 0.4 mL / min for 3 minutes. The relationship between plasma concentration and time was fitted using GraphPad Prism 9. The maximum plasma concentration (Cmax), time to reach Cmax (tmax), apparent elimination half-life (t1 / 2), mean residence time (MRT), area under the plasma concentration-time curve (AUC), clearance (CL), and volume of distribution (V) values were estimated by non-compartmental analysis using Phoenix WinNonlin 8.3. The formula for bioavailability was (AUCs.c. / AUCi.v.)×(Dosei.v. / Doses.c.)×100%. The results are as Figure 22 , mandimycin had a good kinetic curve, with a half-life of 3.84 hours, a maximum concentration (Cmax) of 55168.20 ng / mL, and an area under the curve (AUC0-∞) of 541692.11 hour-ng / mL. These in vitro and in vivo experiments together indicated that mandimycin had a broad therapeutic window.

[0145] Example 10

[0146] In vivo nephrotoxicity evaluation of mandimycin

[0147] Specific pathogen-free female ICR mice, 6 weeks old and weighing 23 - 27 g (Hangzhou Medical College, China) were used. The mice were randomly placed in cages and divided into 12 groups of 4 mice each. The mice were given three days of acclimation training before the experiment. Amphotericin B or a solvent without any antibiotics was used as the positive control and placebo, respectively. Compound mandimycin was formulated in a solution containing 10% DMSO and 10% Tween 80. Subsequently, each group of mice was injected subcutaneously with 1 mg / kg, 5 mg / kg, 10 mg / kg, and 20 mg / kg of the compound or placebo once a day. Then, a commercial kit (Cloud-Clone Corp., China) was used to measure the concentrations of toxicity-related biomarkers according to the protocol provided by the kit, including kidney injury molecule-1 (KIM-1), tissue inhibitor of metalloproteinase-1 (TIMP-1), neutrophil gelatinase-associated lipocalin (NGAL), and osteopontin (OPN). Finally, all animals were euthanized, and the kidney tissues were collected, fixed, dissected, and stained with H&E. Pathological changes such as tubular degeneration, necrosis, cellular casts, dilation, congestion, and protein casts were blindly evaluated and scored by a clinical pathologist. As Figure 23 shown, after administration of mandimycin, the in vivo nephrotoxicity-related indicators were not significantly different from those of the placebo group, while the nephrotoxicity indicators in the amphotericin B group increased significantly. The conclusion drawn from the section score analysis was that mandimycin caused almost no kidney injury. All animal research procedures have been approved by the Animal Ethics Committee of China Pharmaceutical University.

[0148] Example 11

[0149] Study on the mechanism of action of mandimycin

[0150] (1) Drug resistance test of mandimycin

[0151] Different Candida and other fungi were cultured overnight with shaking (200 rpm, 30 °C) in YPD broth for 16 h. A bacterial suspension with a concentration of approximately 10 10 CFU / mL was prepared, and 0.1 mL of the bacterial suspension was spread on YPD agar plates containing 8 MIC of mandimycin, amphotericin B, nystatin, and natamycin, respectively. These inoculated plates were incubated at 30 °C for 2 days to identify resistant colonies. The number of colonies growing in the presence of the drug was counted, as Figure 24As shown, mandimycin did not produce resistant strains, while amphotericin B produced an average of 6 resistant Auricularia auricula-judae strains, 5 resistant Candida albicans strains, 12 resistant Cryptococcus neoformans strains, 560 resistant Candida tropicalis strains, and 5 resistant Candida parapsilosis strains per agar plate; nystatin produced an average of 2 resistant Auricularia auricula-judae strains, 24 resistant Candida albicans strains, 3 resistant Cryptococcus neoformans strains, 55 resistant Candida tropicalis strains, 2 resistant Candida parapsilosis strains, and 23 resistant Candida glabrata strains per agar plate; natamycin produced an average of 13 resistant Auricularia auricula-judae strains, 4 resistant Candida albicans strains, 34 resistant Cryptococcus neoformans strains, 40 resistant Candida tropicalis strains, 8 resistant Candida parapsilosis strains, and 5 resistant Candida glabrata strains per agar plate. It can be seen from the results that, different from the polyene antifungal drugs used clinically, mandimycin still does not produce drug resistance under the induction of high bacterial liquid concentration, suggesting that it is not easy to produce drug-resistant bacteria during its further clinical application.

[0152] (2) Bactericidal curve of mandimycin

[0153] The turbidity of the fresh Candida albicans BNCC186382 colonies grown at 30 °C and 220 rpm for 16 hours was adjusted to OD600 = 1 with normal saline, that is, the bacterial liquid concentration (6.6×10 7 CFU / mL), and then diluted 66 times to obtain a standard suspension (1×10 6 CFU / mL). This suspension was diluted with YPD liquid medium and standard compound stock solution to obtain an initial inoculum of approximately 10 5 CFU / ml. Mandimycin with concentrations of 0.5 μg / mL, 1 μg / mL, 2 μg / mL, 4 μg / mL, 8 μg / mL, and 16 μg / mL was added to test tubes containing 4 mL of culture (the initial bacterial liquid was 10 5 CFU / ml), and then continuously shaken and cultured at 30 °C and 220 rpm. At 0, 2, 4, 6, 8, 24, and 48 hours, 0.1 mL of appropriately diluted samples was taken out and spread on three YPD agar plates. After culturing for 24 hours, the number of monoclonal colonies was determined by viable counting. In the absence of antifungal agents, growth controls for each organism were prepared and carried out simultaneously. As Figure 25 shown, mandimycin is a potent bactericide. At a high concentration (8 μg / mL), the number of bacteria began to decline after 2 hours, and Candida could be completely killed in 8 hours.

[0154] (3) Observation of the cell morphology of Candida albicans treated with mandimycin by scanning electron microscopy

[0155] The cell samples of Candida albicans BNCC186382 were treated with 8×MIC for 2 h, 4 h, and 8 h, and the cell morphology after treatment at different times was observed using a scanning electron microscope. The results are as Figure 26 shown. Under the action of mandimycin, the cell membrane of Candida appeared to lyse at 2 h and completely collapse at 8 h.

[0156] (4) Determination of changes in in vitro and in vivo ion concentrations after treatment of cells with mandimycin

[0157] Candida albicans BNCC186382 was cultured overnight in 50 mL of YPD liquid medium in a shaker (rotation speed 200 rpm, 30 °C). The culture was washed three times with 10 mM Tris-acetate buffer containing 100 mM NaCl and pH 7.4. The washed Candida albicans (6.6×10 8 CFU / ml (OD600 = 1.0)) was resuspended in this buffer to prepare a 20 mL cell suspension. 1×MIC (0.5 μg / mL), 4×MIC (2 μg / mL), and 10×MIC (5 μg / mL) of mandimycin were added to the cell suspension, and the change in K + concentration was measured using an Orion SensorLink PCM-700 pH / ISE meter (the electrode had been calibrated with standard solutions containing 0.01, 0.1, or 1.0 mM KCl in 100 mM NaCl). The change in K ion concentration of the cell suspension treated with the solvent without mandimycin in the same volume was used as a control group. As Figure 27 shown, Candida albicans treated with mandimycin began to excrete a large amount of K + , while there was no change in the control group, and the drug concentration and the ion efflux concentration showed a dose-dependent relationship.

[0158] (5) Determination of the binding target of mandimycin to the fungal cell membrane

[0159] Candida albicans BNCC186382 and the broth dilution method were used to evaluate the effects of adding eukaryotic cell membrane phospholipid components (phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, phosphatidylglycerol, lecithin, sphingomyelin, cardiolipin), ergosterol, and β-1,3-glucan and mannan on the cell wall on the antibacterial activity of mandimycin. The various membrane components used were dissolved in 10% DMSO and prepared into different concentrations required for the experiment, and added to the mandimycin MIC determination medium to observe the effect of fungal cell membrane components on the mandimycin MIC. As Figure 28As shown, the ergosterol target of known polyene macrolide antifungal antibiotics has no inhibitory effect on the antibacterial activity of mandimycin. Another class of antifungal drugs, the target of echinocandins, β-1,3-glucan, also has no inhibitory effect on mandimycin, indicating that mandimycin may have different mechanisms of action. Further research found that phospholipid small molecules in the fungal cell membrane (including lecithin, cardiolipin, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, sphingomyelin, phosphatidylglycerol) have a strong inhibitory effect on the activity of mandimycin, and the inhibitory activity shows a dose-effect curve relationship. Among them, the inhibitory activity of lecithin on mandimycin increases with the increase of lecithin concentration, and the change multiple of the MIC of mandimycin against Candida albicans increases from 1 to 128; the inhibitory activity of cardiolipin, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, and sphingomyelin on mandimycin increases with the increase of their concentration, and the change multiple of the MIC of mandimycin against Candida albicans increases from 1 to 64; the inhibitory activity of phosphatidylglycerol on mandimycin increases with the increase of phosphatidylglycerol concentration, and the change multiple of the MIC of mandimycin against Candida albicans increases from 1 to 32. This experiment shows that, different from the known polyene natural products that act on the ergosterol target, mandimycin acts on the phospholipid small molecules in the fungal cell membrane.

[0160] (6) Determination of the action target of mandimycin by isothermal titration calorimetry

[0161] Dilute 20 mM mandimycin and amphotericin B to 1 mM with 5.0 mM HEPES (pH = 7.4) containing 5% DMSO. Dissolve different fungal cell membrane phospholipid components (lecithin, cardiolipin, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, sphingomyelin, phosphatidylglycerol) and ergosterol in 5.0 mM HEPES (pH = 7.4) solution containing 5% DMSO to prepare a 600 μM mixture. Then, hydrate DOPC phospholipids (Avanti Polar Lipids, 610014-1Ea) with 5.0 mM HEPES (pH = 7.4) and use an Avanti Mini extruder to prepare 100 nm liposomes.

[0162] Use a PEAQ-ITC isothermal titration calorimeter to detect the exothermic binding relationship between mandimycin and phospholipids. At 25 °C, place 1 mM (40 μl) of mandimycin solution in the automatic injection needle, and a 600 μM liposome (250 μl) suspension in the sample cell. The volume of the first injection is 0.23 μl. Next, 18 injections of 2 μl of the above compound are carried out, with an interval of 80 seconds between each injection and a stirring speed of 500 revolutions per minute. AsFigure 29 As shown in Table 6, small phospholipid molecules showed good binding exothermic curves to mandimycin. Among them, phosphatidylinositol had the strongest binding ability, with a Kd value reaching 21.9 μM; followed by the phosphatidylglycerol target, with a Kd value of 28.2 μM; the phosphatidylserine target, with a Kd value of 30.9 μM; the phosphatidylethanolamine target, with a Kd value of 36.5 μM; the phosphatidylcholine target, with a Kd value of 36.8 μM; the sphingomyelin target, with a Kd value of 50 μM; and the cardiolipin target, with a Kd value of 63.6 μM. Amphotericin B and nystatin did not bind to phospholipid components. Through ITC experiments, it was further confirmed that mandimycin did not bind to traditional ergosterol, but strongly bound to 7 phospholipid molecules through a new target, especially phosphatidylinositol binding.

[0163] Table 6 Affinity values of mandimycin and phospholipid components

[0164]

[0165]

[0166] As can be seen from Example 8, the natural product mandimycin of the present invention has a target different from known polyene antifungal drugs. Instead, it specifically binds to 7 phospholipid molecules on the fungal cell membrane, especially phosphatidylinositol, resulting in the efflux of important ions in the fungal cell and the death of fungal pathogens. This multi-target binding mode not only enables the natural product mandimycin of the present invention to have potent broad-spectrum anti-multidrug-resistant fungal activity, but also has the excellent property of not generating drug resistance. Based on the novel trisaccharide polyene skeleton of mandimycin, potent broad-spectrum antibacterial activity against drug-resistant bacteria, a new mechanism of action, and the excellent property of not generating drug resistance, mandimycin is expected to be developed into a new generation of antifungal drugs.

[0167] Example 12

[0168] Construct the engineered strain CPU002 with MandQ knockout

[0169] To obtain mandimycin derivatives lacking atratcynose A, a bioinformatics analysis of the mandimycin biosynthetic gene cluster in Streptomyces netropsis DSM 40259 was performed, and it was determined that MandQ (SEQ ID NO.2) played a crucial role in the formation of atratcynose A. To knockout MandQ, using the genomic DNA of Streptomyces netropsis DSM40259 as a template, the upstream and downstream fragments of MandQ were amplified by PCR with the primer pairs MandQ_KOUF / R and MandQ_KODF / R respectively, and cloned into the pKC1139 vector digested with Xba I to obtain the plasmid pKC1139-MandQ_KO. Further, the vector was introduced into Streptomyces netropsis DSM40259 by Escherichia coli - Streptomyces indirect conjugation transfer to knockout MandQ. The knockout strain was verified using MandQ_TestF / R. The results showed that, as Figure 30 shown, a 1072-base fragment in MandQ was successfully knocked out, and the engineered strain CPU002 with MandQ knockout was obtained. A comprehensive comparative analysis of secondary metabolites of the wild-type and knockout strains was carried out. As Figure 31 shown, it was confirmed that mandimycin was knocked out and a new conjugated pentenyl glycoside knockout natural product was generated. This compound was named mandimycin B.

[0170] Primer sequences;

[0171] MandQ_KOUF: gggctgcaggtcgactcacacccgaatcgaccact;

[0172] MandQ_KOUR: atggcttcgacggggctcgggatcatcag;

[0173] MandQ_KODF: gccccgtcgaagccatgcgggagatg;

[0174] MandQ_KODR: cgcggccgcggatcctcgggcagtcatcacaccatc;

[0175] MandQ_TestF: cgacgagtccatggtccg;

[0176] MandQ_TestR: cgccgatgtccaggatcac.

[0177] Example 13

[0178] Biological fermentation of the transformed strain

[0179] (1) Preparation of seed liquid:

[0180] Add 50 mL of TSB medium (per liter of ddH 2 O contains 17.0 g of casein tryptic digest, 3.0 g of soybean digest, 5.0 g of sodium chloride, 2.5 g of dipotassium hydrogen phosphate, 2.5 g of glucose monohydrate, pH = 7.3)) to a 250 mL Erlenmeyer flask with a notch, and inoculate the CPU002 strain cultured on an ISP4 (per liter of ddH 2 O contains 10.0 g of soluble starch, 1.0 g of dipotassium hydrogen phosphate, 1.0 g of magnesium sulfate, 1.0 g of sodium chloride, 2.0 g of ammonium sulfate, 2.0 g of calcium sulfate, 0.001 g of ferrous sulfate, 0.001 g of manganese chloride, 0.001 g of zinc sulfate, 15.0 g of agar, pH = 7.2)) agar plate into the TSB medium, and place it on a shaker (conditions: rotation speed 200 rpm, temperature 30 °C) for 2 days to prepare the seed liquid.

[0181] (2) Preparation of fermentation broth:

[0182] Transfer 0.5 mL of the seed solution to 50 mL (250 mL Erlenmeyer flask with a notch) of FS / 9 (add 40.0 g of glucose, 30.0 g of soybean powder, 10.0 g of CaCO 2 O) fermentation medium and culture for 5 days (shaker rotation speed 200 rpm, 30 °C). After fermentation, add 1:1 n-butanol to the fermentation flask and stir overnight (100 rpm) for extraction. The n-butanol extract is dried by rotary evaporation and dissolved in methanol to obtain the crude extract. 3 )

[0183] Example 14

[0184] Isolation and purification of mandimycin B

[0185] Purification is carried out in 2 steps. First, isocratic elution is used. The crude extract obtained in Example 10 is separated by column chromatography using a YMC-GEL C18 powder (12 nm × 50 μm) packing column: use 2 cv of 10% methanol (H 2 O:CH 2OH 9:1) Balance, add methanol solution, and elute with methanol aqueous solutions of different concentrations (10%, 30%, 50%, 70%, 90%, 100%), 100 mL per bottle. Use UPLC-MS to detect the components, collect the fractions containing mandimycin B (eluate with 90% methanol aqueous solution) and evaporate to dryness. The UPLC-MS conditions are as follows: C18 chromatographic column (Waters, T3 - 1.8μm, 2.1×100mm), mobile phase: phase A - pure water (0.1% formic acid), phase B - acetonitrile (0.1% formic acid), flow rate, 0.6 mL / min, the gradient of mobile phase B is 30% - 90%, increasing by 10% per minute. The MS detection range is 200 - 2000, and both positive and negative modes are carried out simultaneously. Then use semi-preparative reverse-phase high-performance liquid chromatography (RP-HPLC) to further purify the semi-pure product. The RP-HPLC conditions are as follows: solvent A, deionized water; solvent B, acetonitrile. The flow rate is 3 mL / min, the gradient of solvent B is 30% - 90%, increasing by 1.5% per minute, and full-wavelength detection (190nm - 800nm). Ultraviolet analysis can collect mandimycin B with a purity exceeding 95%, and the yield is 3 - 4 mg / L fermentation broth.

[0186] Example 15

[0187] Structure identification of mandimycin B

[0188] Mandimycin B is an amorphous yellow powder. High-resolution analysis by HR-ESI-MS shows that its protonated ion is m / z 924.4957 [M + H]+; m / z 922.4792 [M - H]-, indicating its molecular formula C 47 H 73 NO 17 , The ultraviolet-visible spectrum of mandimycin B in methanol shows that the maximum absorption wavelengths are 320, 335, and 352 nm respectively, indicating the presence of a conjugated pentadiene structure. Dissolve mandimycin B in deuterated DMSO, use a 700M nuclear magnetic resonance instrument to obtain 1H spectrum, 13C spectrum, and two-dimensional spectra such as HSQC, HMBC, COSY, and TOCSY, as shown in Figures 32 - 37As shown below. After comparing the one-dimensional and two-dimensional nuclear magnetic resonance data of mandimycin B with that of mandimycin, it was found that the main difference between the two lies in the absence of resonances of dimethoxysugar and 3-O-methyldimethoxysugar, which is consistent with the fact that mandimycin B is produced by the glycosyltransferase gene knockout strain CPU002. The significant upfield shifts of C-35 (ΔδC -7.3) and C-37 (ΔδC -5.6) further verified this conclusion. In addition, by comparing the 1H and 13C NMR data of the mycarose molecule in PAC-G10 and mandimycin B, it was found that their chemical shifts were almost the same, indicating that the mycarose molecule was retained in mandimycin B. The key HMBC and COSY NMR correlations of mandimycin B, such as Figure 38 As shown, the structure of mandimycin B was determined. The 1H NMR (700 MHz, DMSO-d6), 13C NMR (700 MHz, DMSO-d6), HMBC, HSQC, and COSY NMR spectra are shown in Table 8.

[0189] Table 8 1H and 13C spectral (DMSO-d6) data of mandimycin B

[0190]

[0191]

[0192] Based on the H spectrum, C spectrum, HSQC, HMBC, COSY and other two-dimensional spectra, as well as the chiral bioinformatics analysis of the polyketide synthase of mandimycin B, the specific configuration of the natural product mandimycin B of the present invention is 2R, 10R, 11S, 12S, 13R, 14R, 16R, and 18R. The natural product mandimycin B is shown in Formula II:

[0193]

[0194] Example 16

[0195] In vitro anti-multidrug-resistant bacteria activity detection of mandimycin B

[0196] Using the CLSI standard, the antibacterial activity of mandimycin B against the priority fungal pathogens announced by the WHO was determined by the method of Example 7. The results are shown in Table 9 below.

[0197] Table 9 Antifungal activity of mandimycin B

[0198]

[0199] As can be seen from Table 9, mandimycin B has potent and broad-spectrum anti-multidrug-resistant fungal activity. The minimum inhibitory concentration (MIC) of mandimycin B against Candida, especially against multidrug-resistant strains including Candida albicans and Candida auris, is between 1-2 μg / mL; the activity against Cryptococcus neoformans multidrug-resistant strains reaches 0.5 μg / mL; the activity against Aspergillus fumigatus multidrug-resistant strains reaches 2 μg / mL, which is 32 times that of fluconazole and 16 times that of 5-fluorocytosine.

[0200] Example 17

[0201] Mechanism of action of mandimycin B

[0202] (1) Binding curves of mandimycin B with sterol molecules or phospholipid molecules

[0203] In this example, the UV-vis method was used to study the binding of mandimycin B with sterol molecules or phospholipid molecules (Maji, A. Nature 2023, 623, 1079–1085). Mandimycin B was dissolved in DMSO, and 1 mM mandimycin B was mixed with different ratios of sterol solutions (cholesterol, ergosterol) or phospholipid molecule solutions (lecithin, cardiolipin, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, sphingomyelin, phosphatidylglycerol). The final volume was 1 mL, and the mixed solution was allowed to stand at room temperature for 30 min. Then, the mixture was scanned by UV-Vis (310-400 nm) using a microplate reader FlexA-200, and the obtained values were plotted using Origin plotting software. The results are shown in the appendix Figure 39 When different concentrations of phospholipids were mixed with mandimycin B, the UV absorption value of mandimycin B was not affected, indicating that different phospholipid molecules do not bind to mandimycin B; while when different concentrations of sterols (including cholesterol or ergosterol) were mixed with mandimycin B, the UV absorption value of mandimycin B changed significantly, indicating that mandimycin B binds to sterol molecules (cholesterol or ergosterol).

[0204] (2) Determination of the binding force of mandimycin B with sterol molecules or phospholipid molecules by isothermal titration calorimetry

[0205] In this example, 20 mM mandimycin B (using DMSO as the solvent) was diluted to 1 mM with 5.0 mM HEPES (pH = 7.4) containing 5% DMSO by isothermal titration. Different phospholipid components (lecithin, cardiolipin, phosphatidylinositol, phosphatidylserine, phosphatidylethanolamine, sphingomyelin, phosphatidylglycerol) and ergosterol and cholesterol were dissolved in 5.0 mM HEPES (pH = 7.4) solution containing 5% DMSO to prepare a 600 μM composition. Then, the purchased liposome membrane was hydrated with 5.0 mM HEPES (pH = 7.4), and different 600 μM phospholipid components and ergosterol suspensions were passed through a 100 nm polycarbonate filter six times using an Avanti Mini Extruder to prepare LUV. A PEAQ-ITC isothermal titration calorimeter was used to detect the exothermic binding relationship between mandimycin B and phospholipids or sterols. At 25 °C, a 1 mM (40 μl) mandimycin B solution was placed in the automatic injection needle, and a 600 μM LUV (250 μl) suspension was placed in the sample cell. The volume of the first injection was 0.23 μl. Next, 18 injections of 2 μl were made for the above compounds. The interval between each injection was 80 seconds to ensure that the instrument returned to a stable baseline before the next injection. The stirring speed for each experiment was 500 revolutions per minute. The results are shown in the appendix Figure 40 It shows that phospholipid molecules have no binding curve for mandimycin B, while sterols have an obvious binding curve for mandimycin B. Among them, the Kd value of the binding ability of mandimycin B to ergosterol reaches 25.9 μM; the Kd value of the binding ability to cholesterol is 28.6 μM, proving that the mechanism of action of mandimycin B is different from that of mandimycin and it has a strong binding ability to sterol molecules.

[0206] In the present invention, the nucleotide sequence SEQ ID NO.1 of the mandimycin biosynthetic gene cluster in the claims and the specification is split into 9 in the nucleotide and amino acid sequence list and corresponds to the sequence after combination with SEO ID NO.1-9 in the sequence list. At the same time, the nucleotide sequence SEQ ID NO.2 of the key gene MandQ in the claims and the specification corresponds to SEQ ID NO.10 in the sequence list.

[0207]

[0208] The nucleotide sequence SEQ ID NO.2 of the key genes MandQ is shown below (1072bp):

[0209] gtcacccacctcatgcccttggtccccctggcatgggcactgcgctccgcgggccacgagctcctcgtcgtcggacagccggacctg

[0210] atgggcgtggcccggcaggccgggctgaacgccgtgagcatcggcgaccggttcggcatggaggaggtcttccacggaatgctgg

[0211] aaccgggcaagcgccccatcgagctgtggggccggctccaccccgatcacctgaagcacttccccccggtctggaaggaccacgg

[0212] cgagcgcgtactgcccgcctacctggagctcgcccgcgcgtaccgccccgacctgatcgtggccgatccgatggagttcaacaccct

[0213] cgtggtgggcgggctgctgggcgtcccggtcctgcaccaccggttcggtgtcgacgcggtgtccgagccggtgcgcgcggccgcg

[0214] cggggcgcgatgcgggattcctgctgggccctgggcctcgacgagctgcccgatcccgacattcagctcgacccctgccccccgag

[0215] cctgcaactgcccagcctcgatgaggcccttcccatccgctacgtgcccttcaacggcagtggcgaggtgcccgcctggctccgcga

[0216] ggagcgaccgtcggccacggggaagcggcgcgtcgtggtctcgctggggacccgtacgctcgcgctcaacggagtgcccttcgtg

[0217] cgcggcctgttgcgggccttcgacggtctgcgggacgtcgaggccgtcgccaccgtcccggaggcgttccggggcgagatcggag

[0218] ccgtgccgggcaacgtgcgcatgaccgacccggtgccgctccacctgctcgtggagacctgcgacgcggtcgtccaccacggagg

[0219] gtcgggcacggtgctgaccgccgtgtccgccgggctcccgcacctggtactgccgcagatggccgaccagttcgggcacgccgac

[0220] cagctggtcgcggcgggggcgggcctcgcgatcgacgacgccgcggggcaggacgacacggtgcgactgcggtgcgcgctgga

[0221] ggaactgctgtcggagcccggctacgccaaggcggcgtgggaactgc

Claims

1. A polyene macrolide natural product mandimycin biosynthetic gene cluster, characterized in that: The nucleotide sequence of the mandimycin biosynthetic gene cluster is shown in SEQ ID NO.

1.

2. A method for preparing the polyene macrolide natural product mandimycin according to claim 1, characterized in that: The steps include: The strain containing the mandimycin biosynthetic gene cluster is prepared into a seed liquid for cultivation, fermentation, extraction, separation and purification to obtain the natural product mandimycin.

3. The method for preparing mandimycin, a natural product of olefin macrolides according to claim 2, characterized in that: The strain containing the mandimycin biosynthesis gene cluster is Streptomyces netropsis DSM 40259.

4. A method for preparing mandimycin B, a polyene macrolide antifungal natural product, characterized in that: The steps include: The MandQ gene in the strain containing the mandimycin biosynthetic gene cluster according to claim 1 is knocked out to obtain an engineered bacterium, which is prepared into a seed liquid for cultivation, fermentation, extraction, separation and purification to obtain the natural product mandimycin B; the nucleotide sequence of the MandQ gene is shown in SEQ ID NO.

2.

5. The preparation method according to claim 4, characterized in that: Mandimycin B was obtained by knocking out the MandQ gene in the mandimycin biosynthetic gene cluster in Streptomyces netropsis DSM 40259 using indirect transfer between Streptomyces and Escherichia coli to obtain the CPU002 engineered bacterium. The obtained glycosyltransferase-deficient engineered bacterium CPU002 was fermented, extracted, separated and purified to obtain mandimycin B.

6. A polyene macrolide natural product mandimycin or mandimycin B and a pharmaceutically acceptable salt thereof, characterized in that: The structural formula of the natural product mandimycin compound is shown in Formula I, and the structural formula of the natural product mandimycin B compound is shown in Formula II:

7. Use of the polyene macrolide natural product mandimycin or mandimycin B and pharmaceutically acceptable salts thereof according to claim 6 in the preparation of antifungal drugs.

8. The use according to claim 7, characterized in that: The fungus is preferably any one of a variety of key fungal pathogens such as Candida, Aspergillus, Cryptococcus, Mucor and Fusarium.

9. A pharmaceutical composition of an antifungal drug, characterized in that: The invention comprises the polyene macrolide natural product mandimycin or the natural product mandimycin B as claimed in claim 6 and a pharmaceutically acceptable carrier.

10. The pharmaceutical composition according to claim 9, characterized in that: The pharmaceutical composition is a capsule, powder, tablet, granule, pill, injection, syrup, oral solution, inhalant, ointment, suppository or patch.

11. Use of the antifungal drug composition according to claim 9 in the preparation of antifungal drugs.

12. The use according to claim 11, characterized in that: The fungus is any one of a variety of key fungal pathogens such as Candida, Aspergillus, Cryptococcus, Mucor and Fusarium.

Images