Genomic, biological, and chemical studies of Streptomyces sp. LaBMicrA B280 isolated from the rhizosphere of Inga edulis Martius in the Amazon

TL;DR

This study explores a new Streptomyces species from the Amazon, showing it produces compounds with antifungal, antimalarial, and cytotoxic properties.

Contribution

The discovery of a novel Streptomyces species with a broad range of biotechnological potential and the identification of pentamycin as its active compound.

Findings

LaBMicrA B280 is a novel Streptomyces species with 39 biosynthetic gene clusters and 314 metabolic subsystems.

Pentamycin isolated from FR3 showed strong antimalarial and cytotoxic effects.

The compound demonstrated 100% larvicidal efficacy against Aedes aegypti at 250 µg/mL.

Abstract

In this study, we investigated the taxonomic identity and biotechnological potential of the actinobacterium Streptomyces sp. LaBMicrA B280, isolated from the rhizosphere of Inga edulis (Mart.). Our approach combined a phylogenomic approach, annotation of metabolic subsystems and biosynthetic gene clusters (BGCs), biological assays to assess antifungal, antimalarial, cytotoxic, and larvicidal activities, classical molecular networking analysis, and characterization of the major compound in the most active fraction (FR3). The digital DNA–DNA hybridization values, using the d2 (GGDC) and d4 (TYGS) formulas, average nucleotide identity, and average amino acid identity (< 70% and ≤ 95–96%, respectively) revealed LaBMicrA B280 as a novel species within the Streptomyces genus. Our genomic analysis revealed 39 BGCs, including the cluster responsible for pentamycin biosynthesis, 314 metabolic subsystems, and 24 gene categories. In the molecular networking analysis, pentamycin was the only compound identified. Biological assays with fraction FR3 at a concentration of 50 µg/mL showed 96.59% ± 0.03 and 72.39% ± 0.18 inhibition against Plasmodium falciparum strains W2 and 3D7, respectively, as well as 91.45% ± 0.55 cytotoxicity against the HEp-2 cancer cell line. Larvicidal activity assays demonstrated 100% efficacy against Aedes aegypti (FO) at 250 µg/mL after 48 h. Pentamycin, a known antifungal compound, was successfully isolated from the highly active FR3 fraction. These findings highlight that LaBMicrA B280 exhibits a broad biotechnological potential and represents a promising source of important secondary metabolites for future application.

The online version contains supplementary material available at 10.1007/s10482-026-02268-z.

Figures

Map of functional CDS annotated in the genome of LaBMicrA B280 and their roles in metabolic subsystems. At the right bar, their whole percentage is in green

Analysis of core genes between LaBMicrA B280 and *S. murinus* NRRL B2286 and synonymous strains

Types and numbers of BGCs in LaBMicrA B280 and the type strains that showed > 50% similarity in the phylogenomic analysis based on the dDDH formula (d4, in %): *S. costaricanus* (murinus) DSM 41827 (63.6%), *S. graminearus* JCM 6923 (62.7%), *S. phaeogriseichromatogenes* DSM 40710 (61.6%), *S. griseofuscus* NRRL B-5429 (61.4%), *S. griseofuscus* DSM 40191 (61.3%), and *S. malaysiense* MUSC 136 (51.1%). The side bar indicates absolute abundance

Region 52.1: Synteny between the genes of LaBMicrA B280 and those of *Streptomyces* sp. S816 (MIBiG-BGC0002032) associated with pentamycin synthesis

Classical molecular network of fraction FR3, derived from the culture broth extract of the LaBMicrA B280 strain. Edges represent correlations between ions inferred from their MS^2^ profiles and cosine similarity values (0–1). A Detail of the cluster containing the major ion of fraction FR3 (693.382 [M + Na]^+^), along with ions exhibiting similar MS2 fragmentation patterns (cosine > 0.7). B Overall structure of the molecular network, with the cluster corresponding to the major ion of fraction FR3 highlighted by a orange circle

Pentamycin produced by LaBMicrA B280 strain and its ^13^C shifts, observed in its ^13^C NMR spectrumTable 4^1^H, ^13^C, and 2D NMR data of pentamycin1H/13C*δ* 1H^a^ *Jδ* 13C^a^HMBC^a^COSY^a^NOESY^a^NOE^a^1172.9s22.49 *dd* (*J* = 8.4, 7.4 Hz) 1H60.6 *d*1, 3, 4, 1′, 2′3, 1′3, 2′b2′a/b, 3, 3-OH34.02* m* 1H72.2 *d*1, 5, 2w, 1′w2, 4, 3-OH241.39 m42.0* t*2f, 53, 53-OH, 5-OH53.91 m 1H72.1 *d*no4, 6, 5-OH6a6b1.30* m*1.39 m45.8* t*521, 5-OH73.90 m 1H71.8 *d*7-OH5-OH8a8b1.33 m1.41 m45.7* t*21, 7-OH, 9-OH2193.89 m 1H73.0 *d*10a/b, 9-OH7-OH, 9-OH10a10b1.29* m*1.41* m*44.7 *t*9, 11, 12w9, 1110b10a, 1113113

Full text

Introduction

In a recent review, we reported numerous studies published in the current century demonstrating the potential of actinomycetes, both known and novel, for the production of antibiotics, agricultural applications, and various other biotechnological uses (Souza Rodrigues et al. 2024a). Explorations in underexplored ecosystems (oceans, seas, deserts, mangroves, and insects, among others) have not only led to the discovery of new natural products (NPs) but also new species (Donald et al. 2022; Komaki 2023). Several screenings have shown that actinomycetes, particularly those from the genus Streptomyces, remain the primary sources of NPs with antimicrobial activities. However, there has been an increasing number of reports on the production of NPs by these microorganisms with anticancer, antiparasitic, larvicidal, and biocidal activities, among others (Donald et al. 2022; Oyedoh et al. 2023a; Souza Rodrigues et al. 2024a). Bacteria of the genus Streptomyces, which account for the largest number of reported NPs and are the most abundant genus within the phylum actinomycetes, have genomes ranging in size from 6 to 10 million base pairs (bp), are rich in G + C content, and may harbor between 8 and 80 Biosynthetic Gene Clusters (BGCs) (Barka et al. 2016; Belknap et al. 2020). These BGC-rich genomes, on average, occupy 13% of the genomes of these filamentous bacteria, most of which are silent, retaining a potentially unknown wealth of secondary metabolites yet to be transcribed. Data on many of these BGCs and their related NPs are accessible in public databases and can be utilized in NP prospecting or genomic mining studies, as well as for extract sample dereplication (Belknap et al. 2020; Donald et al. 2022). This approach is highly valuable for addressing challenges in various fields that rely on the discovery of new bioactive NPs (Souza Rodrigues et al. 2024a).

Many problems related to human health and agriculture that have been prominent in recent decades may find alternative solutions in the metabolites produced by actinomycetes. Among these issues, antimicrobial resistance and multidrug resistance stand out as major concerns for the World Health Organization (WHO), having caused millions of deaths. For instance, data from 2019 revealed 4.95 million deaths attributed to antimicrobial resistance (Murray et al. 2022). A typical example is the resistance of Plasmodium falciparum to the antimalarial drug artemisinina (WHO 2022a). This protozoan is one of the primary agents of malaria, a highly aggressive parasitic disease transmitted by vectors of the genus Anopheles. It has caused significant devastation in the Amazonian public health and was responsible for thousands of deaths in 2020, particularly on the African continent (Chan et al. 2022). In addition to combating the malaria protozoan, actinomycetes have proven to be a viable, economical, and ecological alternative for controlling mosquitoes of the genera Anopheles and Aedes, which are well-known vectors of tropical diseases such as parasitic diseases (malaria) and viral infections (dengue, chikungunya, among others) that cause hundreds of thousands of deaths annually (Katak et al. 2023). Another human health issue for which actinomycete metabolites may offer alternative solutions is cancer. Long recognized as a major global health problem, cancer is the second leading cause of death worldwide, with 19.3 million cases recently reported globally. The most prevalent cancers are female breast cancer (11.7%), lung cancer (11.4%), colorectal cancer (10.0%), prostate cancer (7.3%), and stomach cancer (5.6%) (Parkin 2001; Sung et al. 2021; WHO 2022b). In turn, agriculture faces a significant challenge in controlling phytopathogens that destroy crops worldwide, resulting in losses amounting to billions of dollars annually (Raymaekers et al. 2020; WHO 2016). Fungi of the genera Colletotrichum, Corynespora, and Sclerotium are among the most significant contributors to these losses, causing infestations in dozens of plant species and even infections in humans, as is the case of subcutaneous infection caused by Corynespora cassiicola (Gasparotto et al. 2019; Huang et al. 2010; Liotti et al. 2018; Ma et al. 2018; O’Connell et al. 2012). To control phytopathogens and face problems such as those mentioned above and others, there are several reports of how actinobacteria can be useful (Souza Rodrigues et al. 2024a).

We recently reported a study on the diversity and antifungal activity of actinomycetes from the rhizosphere of Inga edulis Martius on the campus of the Federal University of Amazonas (UFAM), an area spanning approximately 670,000 hectares of Amazonian forest, mostly native (Souza Rodrigues et al. 2024b). Among 64 strains isolated from the rhizospheres of I. edulis, 20 were preliminarily characterized through the 16S rRNA gene and their metabolites were tested against important phytopathogen strains. Among these 20 strains, Streptomyces sp. LaBMicrA B280 (OR724701) stood out for its antifungal activity against the tested phytopathogens and became a focus of more comprehensive studies reported in this work. These studies assessed its taxonomic identity, metabolic subsystems, types of BGCs associated with NP synthesis in its genome, and new biological activities of its extracts, including antifungal activity against phytopathogens, cytotoxicity against the human laryngeal carcinoma cell line (HEp-2), larvicidal activity against Aedes aegypti (FO), and antiplasmodial activity against Plasmodium falciparum strains (W2 and 3D7). The culture medium extract of this strain was also chemically studied to identify the bioactive compounds responsible for these activities, and the composition of the fraction showing the best results was investigated using the classical Molecular Networking GNPS approach.

Materials and methods

Origin of the strain

The Streptomyces sp. LaBMicrA B280 strain (GenBank OR724701) was isolated from the rhizosphere of a I. edulis plant on the UFAM campus in Manaus, Amazonas, Brazil, and is preserved at the Laboratory of Bioassays and Microorganisms of the Amazon (LaBMicrA) in the Analytical Center Division of the Multidisciplinary Support Center—UFAM.

DNA extraction, sequencing, genome assembly, deposit, and annotation

The LaBMicrA B280 strain was cultivated in 125 mL Erlenmeyer flasks containing 30 mL of BDL culture medium (Souza et al. 2004), at 28 ± 2 °C, and 120 rpm, for 72 h. Bacterial biomass was separated from the culture broth by centrifugation, and genomic DNA (gDNA) was extracted using the Zymo Research Fungal/Bacterial DNA MicroPrep™ kit (Zymo Research, Irvine, CA, USA). The gDNA was quantified by spectrophotometry (NanoDrop 2000, Thermo Scientific, Waltham, MA, USA), and its integrity was assessed on 0.8% (w/v) agarose gel. The gDNA sequencing was performed on the Illumina platform (HiSeqX—PE 150 cycle) with an estimated minimum coverage of 100x. Genome assembly was conducted using a combination of sequencing data and the SPAdes workflow (Prjibelski et al. 2020). The genome sequence of Streptomyces sp. LaBMicrA B280 has been deposited in the National Center for Biotechnology Information (NCBI) under the accession number JBMJCT000000000. Genome structure and quality, as well as subsystem predictions and annotations, were analyzed using the BV-BRC (Bacterial and Viral Bioinformatics Resource Center) (https://www.bv-brc.org/) and RAST (Rapid Annotation using Subsystem Technology) (https://rast.nmpdr.org/rast.cgi) platforms (Aziz et al. 2008), accessed on 02/20/2024.

Phylogenomic identification

Phylogenomic identification at the species level for LaBMicrA B280 was conducted using the Type Strain Genome Server (TYGS) (TYGS v.391; https://tygs.dsmz.de/user_requests/new) and Genome-to-Genome Distance Calculator (GGDC) 3.0 (GGDC v3.0; https://ggdc.dsmz.de/ggdc.php#) servers, accessed on 15/02/2025, with default parameters (dDDH calculation using d2 and d4 formulas) and a similarity threshold of < 70% for species delimitation (Meier-Kolthoff and Göker 2019). A phylogenomic tree was constructed based on the GBDP d5 formula using the TYGS platform. Complementary analyses through the Overall Genome Relatedness Index (OGRI) were performed using internationally validated metrics: AAI (average amino acid identity; http://enve-omics.ce.gatech.edu/aai/), OrthoANI (orthologous average nucleotide identity; https://www.ezbiocloud.net/tools/orthoani), ANI (average nucleotide identity; https://www.ezbiocloud.net/tools/ani), BLAST ANI (ANIb), and MUMmer ANI (ANIm) (https://jspecies.ribohost.com/jspeciesws/), accessed on 02/25/2024, with a similarity threshold of < 95–96% for species delimitation (Richter and Rosselló-Móra 2009; Richter et al. 2016; Yoon et al. 2017). Besides, an analysis was done in the platform OrthoVenn3 (https://orthovenn3.bioinfotoolkits.net/home; accessed em 02/15/2025) to achieve the quantities of central and unique genes of the LaBMicrA B280 strain in comparison to its closer type species Streptomyces murinus NRRL B-2286, S. costaricanus DSM 41827, S. phaeogriseichromatogenes DSM 40710, and S. griseofuscus NRRL B-5429 (Hulsen et al. 2008).

Prediction of clusters associated with natural product synthesis and synteny analysis

Biosynthetic Gene Clusters (BGCs) of the LaBMicrA B280 strain DNA were predicted using the Antibiotics and Secondary Metabolites Analysis Shell (antiSMASH) tool Bacterial version, version 7.0 (https://antismash.secondarymetabolites.org), accessed on15/02/2025, with the platform's default parameters (Blin et al. 2024). Additionally, an analysis of BGC diversity was conducted between LaBMicrA B280 and the type strains that exhibited > 50% similarity in the phylogenomic identification. A heatmap was constructed using the 'pheatmap' package in the R software environment (version 4.0.2). All genomes of the type strains were downloaded from the NCBI genome database (https://www.ncbi.nlm.nih.gov/datasets/genome/?taxon=1883: accessed on 15/02/2025), and their BGCs were predicted using the antiSMASH platform with default parameters. Hybrid clusters were disaggregated and individually counted; for example, "T1PKS, NRPS" was split into T1PKS and NRPS and counted separately. To identify proteins associated with the BGCs, the InterPro platform (https://www.ebi.ac.uk/interpro/) and Protein BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) were used, accessed on 03/10/2024. The Clinker tool (https://cagecat.bioinformatics.nl/tools/clinker), accessed on 03/16/2024, was employed for the synteny analysis (Gilchrist and Chooi. 2021).

Production and fractionation of the cultured medium extract

Extracts of the LaBMicrA B280 strain were obtained as described by Souza Rodrigues et al. (2024b). Briefly, after cultivation in ISP2 medium at 120 rpm and 28 ± 2 °C for ten days, the cells were separated by centrifugation and macerated with AcOEt/MeOH 1:1 and the culture medium was partitioned with AcOEt/2-propanol 9:1. After concentration, the AcOEt/2-propanol 9:1 extract was mixed with 63–200 μm silica gel to form a slurry and, after drying the solvent, was fractionated using an open chromatography column with C8 phase (30 × 150 mm), as follow. The column was activated with 50 mL of HPLC-grade MeOH (3x) and conditioned with H_2_O/MeOH 8:2 v/v (3x). The dried slurry was then added to the top of the column and eluted with 50 mL of H_2_O/MeOH 8:2 v/v (FR1); 50 mL of H_2_O/MeOH 1:1 v/v (FR2); 50 mL of H2O/MeOH 2:8 v/v (FR3); and 50 mL of MeOH 100% (FR4). After further chromatographic analyses and biological assays, fraction FR3 was re-fractionated in a similar manner using a 10g-C8 cartridge Strata™ (^©^2025 Phenomenex Inc., Torrance, California, USA), eluted with H_2_O/MeOH /Acetone 8:1:1 v/v (FR3-A); H_2_O/MeOH /Acetone 4:4:2 v/v (FR3-B); and MeOH 100% (FR3-C). Finally, after preliminary analyses by high-performance liquid chromatography (HPLC) using a C18 Luna™ (5 µm, 4.6 × 150 mm) (^©^Phenomenex Inc., Torrance, California, USA) analytical column, fraction FR3-B (5.4 mg) was fractionated using a semi-preparative column (C18 Luna™, 5 µm, 10 × 250 mm) (^©^Phenomenex Inc., Torrance, California, USA) on a Shimadzu UFLC system (Shimadzu, Columbia, MD, USA), with water (A) and methanol (B) as mobile phases. The elution gradient was 65–100% B (v/v) over 12 min, at a flow rate of 3.5 mL min⁻^1^, with UV monitoring at 315 nm. The major compound (2.3 mg) was obtained at a retention time of 7.7 min and subjected to Nuclear Magnetic Resonance (NMR) identification. The one-dimensional (1D) and two-dimensional (2D) NMR spectra were obtained using a Bruker AVANCE III HD 500 spectrometer (Bruker, Billerica, MA, USA), operating at 11.75 T, 500.13 MHz for ^1^H, and 125.76 MHz for ^13^C. DMSO-d6 was used as solvent and TMS as internal reference standard.

New biological assays

In our previous paper, we reported the antifungal assays of the extract AcOEt/2-Propanol 9:1 (Souza Rodrigues et al. 2024b). Here, we assayed the fractions of this extract using the same methodology. Briefly, 20 mg of each fraction (FR1, FR2, FR3, and FR4) were dissolved in 200 µL of dimethyl sulfoxide (DMSO), followed by the addition of 800 µL of autoclaved distilled water to obtain the stock solution. As a positive control, 20 mg of nystatin were suspended in 1 mL of water. Extract and control assays were performed in triplicate using 12-well cell culture plates containing PDAY medium (potato dextrose agar supplemented with yeast extract), with each fraction tested at a concentration of 1000 µg/mL. Negative controls consisted of PDAY and DMSO. Plates were incubated at 28 ± 2 °C for 96 h. For the MIC (Minimum Inhibitory Concentration) determination, the fractions were tested under the same conditions described above, starting at 1000 µg/mL and subsequently diluted to 500, 250, 125, 62.5, 31.25, 15.63, and 7.81 µg/mL (Souza Rodrigues et al. 2024b).

Cytotoxicity assay

The human laryngeal carcinoma cell line (HEp-2) was cultured in Dulbecco's Modified Eagle Medium (DMEM) (Gibco), supplemented with 10% inactivated fetal bovine serum (Gibco) and gentamicin antibiotic (50 μg/mL). The assays were performed in triplicate using the Alamar Blue method, according to Ahmed et al. (1994). The cells were trypsinized, resuspended in fresh medium, plated at a concentration of 1.0 × 10^4^ cells/well in 96-well plates, at 100 μL per well, and incubated for 24 h in a 5% CO₂ incubator at 37 °C. After solubilized in dimethyl sulfoxide (DMSO) and diluted in water, the extracts AcOEt/MeOH 1:1 (cell biomass), AcOEt/2-propanol 9:1 (cultured medium), and fractions FR1, FR2, FR3, and FR4 were individually diluted in a serial way in fresh-medium containing plates and, after, added to the wells with the cells to achieve concentrations of 50, 25, 12.5, and 6.3 μg/mL and a final volume of 200 μL. Final DMSO concentrations in the wells were ≤ 0.1%. After 72 h of incubation under the same conditions, 10 μL of 0.4% resazurin solution was added to each well, and the metabolization of Alamar Blue® (Sigma-Aldrich, Brazil) was monitored for 2 h. Fluorescence was measured using a microplate reader (GloMax^®^ Explorer) at an emission wavelength of 580–640 nm and an excitation wavelength of 520 nm. Doxorubicin (Sigma-Aldrich, Brazil) was used as the substance of reference. Cell growth in wells with no sample was used as the negative control and annotated as 100% cell growth. The percentage of cell viability was calculated using the formula: % Viability = Ft × 100/Fb, where Ft = (fluorescence of cells + medium + substance + resazurin) and Fb = (fluorescence of cells + medium + resazurin). The assay was performed in triplicate.

Antiplasmodial assay

P. falciparum strains (W2 and 3D7) were cultivated according to Trager and Jensen (1976), with final parasitemia and hematocrit values of 2%, being conducted in triplicate in flat-bottom 96-well plates. The extracts AcOEt/MeOH 1:1 (cell biomass), AcOEt/2-propanol 9:1 (cultured medium), and fractions FR1, FR2, FR3, and FR4 were dissolved in DMSO and diluted in water, following dilutions in a serial way in fresh-medium containing plates and, after, added to the wells with the cells to achieve concentrations ranging from 50 to 0.39 μg/mL. Quinine (Sigma-Aldrich, Brazil) was used as a positive control at the same concentrations. Final DMSO concentrations in the wells were ≤ 0.5%. Parasite growth in wells with no sample was used as a negative control (presence of the parasite), and uninfected erythrocyte growth as a positive control (absence of the parasite). After 72 h of incubation, the test wells were washed with 1X PBS buffer and ethidium bromide. Finally, the cells from W2 and 3D7 strains were resuspended in 200 μL of 1X PBS for analysis in a BD FACSCanto II flow cytometer (BD Biosciences, San Jose, USA) using the FL-1 channel, coupled with the software Getting Started with BD FACSDiva™ and FlowJo™ version 10. The fluorescence ranged from 0 to 10. The closer to 0 the fluorescence, the more active, and the closer to 10, the less active the sample.

Larvicidal assay

The preparation of A. aegypti (FO) larvae was performed as described by Oliveira et al. (2021). The mosquito eggs were from colonies maintained at the National Institute of Amazonian Research (INPA). Selective bioassays followed the criteria established by Dulmage et al. (1990) and the WHO (2005), with minor modifications, and were conducted under controlled conditions of temperature, humidity, and photoperiod. The selective bioassays were performed in triplicate using 50 mL plastic cups containing 10 mL of distilled water, ten third-instar larvae, powdered rat feed (Teklad Global 18%), and samples of the extracts AcOEt/MeOH 1:1 and AcOEt/2-propanol 9:1 and the fractions FR1, FR2, FR3, and FR4 solubilized in DMSO at concentrations of 500 μg/mL and 250 μg/mL (WHO 2005). Temephos (Pestanal Sigma-Aldrich, Brazil) larvicide was used as a positive control at the same concentrations. Mortality readings were recorded at 24, 48, and 72 h after exposure to the extracts and fractions (Danga et al. 2014).

Antifungal bioautography assay of the Fraction FR3-B

For the bioautography assay of the fraction FR3-B, a solution with 5 mg in 0.5 mL of MeOH was prepared. C18 TLC plates measuring 4.5 × 6 cm were divided into three sections of 1.5 × 6 cm, and each section was spotted with approximately 5 µL of the fraction solution, applied five times. The experiment was performed in triplicate. The TLC plates were eluted with a mixture of H2O/MeOH/Acetone in a 4:4:2 ratio using chromatographic tanks. After elution, one section of each TLC plate replica with an elution spot was cut. The cut sections were visualized under UV light at 254 nm and 365 nm and stained with vanillin, while the remaining two sections were used for the bioautography assay. For the assay, the two remaining TLC sections of each replica were immersed in 70% ethanol for 30 s and gently heated to evaporate the solvent. The plates were then placed in the center of 1.5 × 90 mm Petri dishes and covered with a thin layer of ISP2 medium. Discs measuring 5 × 5 mm containing cultures of the pathogens were inoculated laterally on the plates, at mid-height, aligned with the bands visualized on the TLCs. The plates were incubated at 28 °C, and the growth of each pathogen was observed over 20 days. The controls were performed under the same conditions but without applying the extracts on the TLC plates. In this assay, we used the same phytopathogens reported in our previous work (Souza Rodrigues et al. 2024b): Colletotrichum sp. (ISO01), isolated from leaf lesions in habanero pepper (Capsicum chinense); Colletotrichum guaranicola (P01) and Pestalotiopsis sp. (3002R2), both isolated from leaf lesions in guarana plants (Paullinia cupana); Corynespora cassiicola (ISO079), isolated from diseased tomato leaves (Solanum lycopersicum); and Sclerotium coffeicola (M01), isolated from leaf lesions on mango trees (Mangifera indica). These cultures represent hosts of great economic importance to the Amazon region, particularly guarana, recognized as one of the primary natural sources of caffeine.

LC–ESI–MS/MS analysis of the fraction FR3

Fraction FR3 was prepared at a concentration of 1 mg mL^−1^ in HPLC-grade MeOH, centrifuged at 4,000 rpm for 15 min, and the supernatant was analyzed in a UHPLC-HRMS system consisted by a Dionex UltiMate 3000 UPLC coupled to a Q-Exactive Plus mass spectrometer (Thermo Scientific, San Jose, CA, USA). The analysis parameters were: positive and negative modes; acquisition type Full MS-ddMS2; resolution of 70,000 for MS1 and 17,500 for MS2 acquisition; collision energy of 10, 20, and 40 for ions with relative abundance above 20%. Ionization parameters: Sheet Gas: 55; Auxiliary Gas: 20; Sweep Gas: 0; Spray voltage: 3.5 kV; Capillary temperature: 380 °C; Gas temperature: 380 °C; mass range: m/z 100 to 1000. Chromatographic parameters: injection volume = 10 µL; runtime = 30 min; flow rate = 300 µL/min; oven temperature = 40 °C. Mobile phase for positive mode: A = water with 0.1% formic acid; B = methanol with 0.1% formic acid. Mobile phase for negative mode: A = water with 5 mM ammonium formate; B = methanol. Column: Agilent Zorbax C18 (2.1 × 50 mm, 1.8 µm). Elution was performed in gradient mode as follows: 40% B for 1 min, linear increase to 100% B in 27 min, and 100% B for 2 min.

Molecular networking (GNPS)

An optimized acquisition with MS1 and MS2 data from fraction FR3 was analyzed on the Global Natural Products Social Molecular Networking (GNPS2, https://gnps2.org/homepage) platform to construct a molecular network (https://gnps2.org/status?task=845d76827dba4890a05c3f7d694d24bb). Both the precursor and the MS/MS fragment ion mass tolerances were set to 0.002 Da. A network was then created, where edges were filtered to have a cosine score above 0.7 and more than 6 matched peaks. Additionally, edges between two nodes were retained in the network if and only if each node appeared among the respective top 10 most similar nodes. The maximum size of a molecular family was set to 100. A procedural blank was included as a control to ensure proper background subtraction and to avoid interference from signals not originating from fraction FR3. The spectra in the network were subsequently searched against GNPS2 spectral libraries. Data conversion for the analysis was performed using MSConverter software (ProteoWizard, Palo Alto, USA). The molecular network was visualized using Cytoscape (v. 3.10.4; Cytoscape Consortium, Seattle, USA).

Results

Genome assembly and metabolic system annotations

In its genome (Supplemental Material-1), the LaBMicrA B280 strain contains 8,257,865 bp, a G + C content of 72.0%, 96 contigs, an N50 of 229,434, and an L50 of 13. It includes 7,701 protein-coding sequences (CDSs), 66 tRNAs, and 6 rRNAs, with the genome quality classified as "Good." A total of 314 metabolic subsystem categories were annotated. Only 18% of the CDSs—1351 genes distributed across 27 subsystem resources—were classified as functional metabolic resources (Fig. 1). For example, 231 genes are associated with protein metabolism, 318 with carbohydrate metabolism, 26 with phosphorus metabolism, 170 with the metabolism of fatty acids, lipids, and isoprenoids, 182 with the metabolism of cofactors, vitamins, prosthetic groups, and pigments, 49 with iron acquisition and metabolism, 7 with potassium metabolism, and 21 with nitrogen metabolism, among other functions. In the Secondary Metabolism (9) category, no subcategory related to phytohormone synthesis was observed. However, a subcategory consisting of six genes associated with antibiotic synthesis, specifically Thiazole-oxazole-modified microcin (TOMM), was identified.Fig. 1. Map of functional CDS annotated in the genome of LaBMicrA B280 and their roles in metabolic subsystems. At the right bar, their whole percentage is in green

Phylogenomic identification and average amino acid identity

The phylogenomic identification of the LaBMicrA B280 strain was conducted using digital DNA–DNA hybridization (dDDH) values (Supplemental Material-2), ANI, and AAI. The dDDH values based on the d2 (GGDC) and d4 (TYGS) formulas presented in this study support the classification of LaBMicrA B280 as a likely novel species within the Streptomyces genus, with < 70% similarity to its closest phylogenetic relative, S. murinus and synonyms species (Table 1). The ANI and AAI values remained below or within the cutoff (< 95–96%) for species delineation, further supporting the d4 and d2 results (Table 1). In the core gene analysis, LaBMicrA B280 was found to contain 21 unique clusters (Fig. 2A) and share 6620 clusters (Fig. 2B) when its genome was compared with those of the four related strains. Among the unique clusters, 15 present unknown and six known functions, namely: (GO:0051762) sesquiterpene biosynthetic process, (2 × GO:0006313) transposition, DNA-mediated, (GO:0006635) fatty acid beta-oxidation, (GO:0006270) DNA replication initiation, and (GO:0006556) S-adenosylmethionine biosynthetic process. Table 1. Phylogenomic comparison of Streptomyces sp. LaBMicrA B280 with S. murinus NRRL B2286 and its synonymous speciesSpeciesStraindDDH (d2) (%)dDDH (d4) (%)ANI (%)ANIb (%)ANIm (%)OrthoANI (%)AAI (%)S. murinusNRRL B228663.5063.2095.5695.2595.6595.8194.71S. costaricanusDSM 4182763.7063.6095.4395.1695.7295.8794.92S. phaeogriseichromatogenesDSM 4071061.0061.6094.9794.6795.2895.2895.81S. griseofuscusNRRL B542960.7061.0495.0294.6095.2895.2594.67Values are expressed as percentages. dDDH digital DNA–DNA hybridization (d2: GGDC formula; d4: TYGS formula), ANI average nucleotide identity (b: BLAST; m: MUMmer), OrthoANI orthologous ANI, AAI average amino acid identityFig. 2Analysis of core genes between LaBMicrA B280 and S. murinus NRRL B2286 and synonymous strains

Genomic mining and synteny analysis

A total of 39 BGCs were annotated on the antiSMASH platform for the genome of the LaBMicrA B280 strain, with seven predictions showing 100% similarity to BGCs involved in the synthesis of the following NPs: geosmin, albusnodin, ectoine, flaviolin/1,3,6,8-tetrahydroxynaphthalene, leupeptin Pr/leupeptin Ac, pentamycin, desferrioxamine B, and desferrioxamine E (Table 2). Among the other predictions, five showed similarity ≥ 50%, 22 below 50%, and five had no BGC annotation, making most predictions (69.23%) with similarity levels below 50%. In addition, the predictions were largely of hybrid BGCs (41.02%). The predominant types of BGCs are “non-ribosomal peptide synthetases” (NRPS) and “polyketide synthase” (PKS I, II, and III). A comparative analysis of types and numbers of BGCs between LaBMicrA B280 and six type strains with > 50% similarity in the phylogenomic analysis was also conducted (Supplemental Material-3). In common, they have prevalence of NRPS and PKS BGC types, with S. griseofuscus NRRL B-5429, S. murinus NRRL B-2286, and S. graminearus JCM 6923 having the highest number of copies (Fig. 3). Table 2. Predicted BGCs for LaBMicrA B280 and their potential natural products as reported by the antiSMASH platformRegionBGC typePredictionSimilarity1.1RiPP-like––1.2TerpeneGeosmin100%1.3NI-siderophoreEnduracididin23%2.1TransAT-PKS-like, NRPS-like, T1PKSCinnabaramide A18%4.1LassopeptideAlbusnodine100%5.1Lanthipeptide-class-iii––6.1EctoineEctoine100%9.1hglE-KS, T1PKSHexacosalactone A13%9.2T3PKSFlaviolin/1,3,6,8 tetrahydroxynaphthalene100%10.1NRPSOmnipeptin15%11.1T1PKS, NRPS, butyrolactoneSalinomycin26%11.2T2PKS, aminocoumarinSpore pigment83%14.1T1PKSTetronasin3%14.2NRP-metallophore, NRPSMyrubactin78%15.1Triceptide–21.1TerpeneKitacinamycin A/Kitacinamycin B/Kitacinamycin C/Kitacinamycin D/Kitacinamycin E/Kitacinamycin F4%22.1NRPS, terpeneLeupeptin Pr/leupeptin Ac100%22.2Terpene, thiopeptide, LAP, NI-siderophore, aminopolycarboxylic-acidHopene92%26.1NRPS, lanthipeptide-class-i, lanthipeptide-class-iiBleomycin12%28.1NRPS, NRPS-likeNapyradiomycin A80915C/napyradiomycin 2/napyradiomycin 3/napyradiomycin 49%33.1Butyrolactonegaburedin A/gaburedin B/gaburedin C/gaburedin D/gaburedin E/gaburedin F20%34.1Lanthipeptide-class-iii, butyrolactoneHydroxystreptomycin18%36.1NRPSWS 79089B/benaftamycin/WS 79089D/WS 79089A/WS 79089C5%37.1NRPS, NAPAAStenothricin13%38.1RiPP-likeInformatipeptin28%42.1Lanthipeptide-class-iii–44.1CrocaginFoxicin A/Foxicin B/Foxicin C/Foxicin12%52.1T1PKS, NRPSPentamycin100%52.2NRPSDiisonitrile antibiotic SF276850%52.3NRPS-like, T1PKSBorrelidin11%53.1T1PKS, terpene, other, lanthipeptide-class-ivA-503083 A/A-503083 B/A-503083 E/A-503083 F7%53.2TerpeneEbelactone5%54.1NI-siderophoreDesferrioxamine B/deferoxamine E100%54.2Melaninmelanin60%55.1NI-siderophoreQuinamycin16%55.2T2PKS, ectoineCosinostatin47%55.3Other, T1PKS, PKS-likeTambjamine BE-1859121%56.1TerpeneJulichrome Q3-3/julichrome Q3-525%59.1Butyrolactone––Fig. 3. Types and numbers of BGCs in LaBMicrA B280 and the type strains that showed > 50% similarity in the phylogenomic analysis based on the dDDH formula (d4, in %): S. costaricanus (murinus) DSM 41827 (63.6%), S. graminearus JCM 6923 (62.7%), S. phaeogriseichromatogenes DSM 40710 (61.6%), S. griseofuscus NRRL B-5429 (61.4%), S. griseofuscus DSM 40191 (61.3%), and S. malaysiense MUSC 136 (51.1%). The side bar indicates absolute abundance

In the synteny analysis of region 52.1 (Fig. 4), 15 genes from LaBMicrA B280 and Streptomyces sp. S816 exhibit similarity and are associated with the synthesis of the polyketide pentamycin (Fig. 6), isolated from LaBMicrA B280 in this study. Darker tones between the BGCs indicate higher levels of similarity. In the central region of the synteny, where PKS I products are located, the similarity is lower (approximately 50% between some gene products in this BGC region). The 15 genes (described from left to right) are associated with the following functions: thioester bond hydrolysis (GrsT); cholesterol oxidation (Cholesterol_Oxidase); transcriptional activation (PAS/CitB); transcriptional activation (AfsR-Dnrl-RedD_regulator); electron transfer (3Fe-S_cluster_ET); hydroxylation of C-26 (CypX (Cytochrome P450)); hydroxylation of C-1′ (CypX (Cytochrome P450*)); carboxylation and reduction (Crotonyl-CoA_carboxylase/reductase*); polyketide synthesis (Polyketide_synth (PksD)); and, at the end of the synteny, gene products (highlighted in red boxes) related to pentamycin synthesis (hydroxylation of C-14 (ptnJ—CypX (Cytochrome P450) and electron transfer (*ptnI—*3Fe4S_ferredoxin) (Zhou et al. 2019), which exhibit a high degree of similarity. Additionally, within the LaBMicrA B280 BGC, there are other gene products, such as non-ribosomal peptide synthetase (NRPS), among others, that complete its structure.Fig. 4. Region 52.1: Synteny between the genes of LaBMicrA B280 and those of Streptomyces sp. S816 (MIBiG-BGC0002032) associated with pentamycin synthesis

Biological activities

In our previous paper, we reported antifungal activities of crude extracts of the LaBMicrA B280 strain against Corynespora cassiicola (ISO079), Colletotrichum sp. (ISO01), C. guaranicola (P01), Pestalotiopsis sp. (3002R2), and Sclerotium coffeicola (M01) (Souza Rodrigues et al. 2024b). Now, after fractioning the AcOEt/2-propanol 9:1 extract (culture medium), these activities were concentrated in its fraction FR3, which showed fungicidal activity against all pathogens at a MIC lower than that of the extract (Table 3). In the evaluation of cytotoxic, larvicidal, and antiplasmodial activities of the AcOEt/MeOH 1:1 (cell biomass) and AcOEt/2-propanol 9:1 extracts, effectiveness was again observed in the AcOEt/2-propanol 9:1 extract and in its fraction FR3, which showed the best cytotoxic, larvicidal, and antiplasmodial results among the four fractions (Table 3). For the W2 strain, the 50 µg/mL concentration of fraction FR3 exhibited a stronger antiplasmodial effect (95.7% ± 0.03) than the positive control quinine (85.5% ± 0.04), which; however, this effect was not observed at the lower concentrations. In contrast, for the 3D7 strain, FR3 showed lower antiplasmodial inhibition than quinine at all concentrations. Furthermore, the IC50 values of FR3 were greater than the respective controls in the antiplasmodial and cytotoxic assays (Table 3). Table 3. Biological activities of fraction FR3 from LaBMicrA B280Biological activities and effective concentrations of FR3Antifungal activities (MIC)62 µg/mL62 µg/mL 62 µg/mL62 µg/mL125 µg/mLColletotrichum* sp. (ISO01)Colletotrichum guaranicola (P01)Conynespora cassiicola (Iso079)Pestalotiopsis sp. (3002R2)Sclerotium coffeicola (M01)Nystatin control16.5 µg/mL16.5 µg/mL16.5 µg/mL16.5 µg/mL31.25 µg/mLCytotoxic activity against the human laryngeal epithelioma cancer cell line (HEp-2)—IC50 14.27 µg/mL50 µg/mL25 µg/mL12.5 µg/mL6.3 µg/mL91.45 % ± 0.5576.13 % ± 0.5337.80 % ± 0.8824.28 % ± 0.88Doxorubicin control—IC50 3.11µg/mL 88.94 % ± 0.6686.41 % ± 0.3581.24 %± 0.3979.19 % ± 0.36Larvicidal activity against Aedes aegypt (FO)i24 h24 h48 h500 µg/mL250 µg/mL250 µg/mL100 % de M83.3 % M100 % MTemephos Control500 µg/mL250 µg/mL250 µg/mL100 % de M100 % de M100 % M Antiplasmodial activity against P. falciparum (W2 and 3D7)50 µg/mL25 µg/mL12.5 µg/mL6.3 µg/mL3.12 µg/mL1.56 µg/mL0.78 µg/mL0.39 µg/mLStrain W2—IC50 23.84 µg/mL 95.7 % ± 0.0356.1 % ± 0.0946.3 % ± 0.0837.0 % ± 0.5332.1 % ± 0.0330.1 % ± 0.2427.1 % ± 0.2723.0 % ± 0.26Positive control (quinine)—IC50 2.397 µg/mL85.7 % ± 0.0484.6 % ± 0.1182.2 % ± 0.279.5 % ± 0.279.2 % ± 0.1378.5 % ± 0.3177.2 % ± 0.0971.3 % ± 0.29Strain 3D7—IC50 28.13 µg/mL72.3 % ± 0.1839.0 % ± 0.0134.9 % ± 0.0433.3 % ± 0.1425.2 % ± 0.1526.0 % ± 0.1223.4 % ± 0.1622.1 % ± 0.12Positive control (quinine))- IC50 3.501 µg/mL87.7 % ± 0.0587.5 % ± 0.0687.4 % ± 0.0086.8 % ± 0.0486.6 % ± 0.0286.4 % ± 0.0486.1% ± 0.085.7 % ± 0.03MIC* = minimum inhibitory concentration. M = Mortality. Fractions FR1, FR2, and FR4 did not show significant results

Molecular networking and bioautography

The molecular networking analysis of fraction FR3 revealed a total of 36 molecular families (Supplemental Material-4), none of which were observed in the procedural blank. The major ion of fraction FR3 at m/z 693.381 ([M + Na]⁺) (Fig. 5, orange circle) was putatively identified as pentamycin (Fig. 5A). Within the molecular family, the ions m/z 693.381 and m/z 692.370 display nearly identical MS2 spectra (cosine = 0.97), indicating that they represent the same compound, with the latter ion corresponding to the loss of a hydrogen radical. In contrast, the ~ 30 Da difference between m/z 693.381 and m/z 663.371 suggests the presence of a structural analogue, potentially arising from the removal or absence of a CH_2_O unit (e.g., –CH_2_OH x –H). Likewise, the ion m/z 707.397, which differs by ~ 14 Da from the main precursor, is consistent with the addition of a CH_2_ group, a hallmark of homologous series. Overall, the molecular network reveals a set of closely related analogues, whereas the ~ 1 Da differences observed between certain ions reflect typical ionization-related variations rather than genuine structural changes.Fig. 5. Classical molecular network of fraction FR3, derived from the culture broth extract of the LaBMicrA B280 strain. Edges represent correlations between ions inferred from their MS^2^ profiles and cosine similarity values (0–1). A Detail of the cluster containing the major ion of fraction FR3 (693.382 [M + Na]^+^), along with ions exhibiting similar MS2 fragmentation patterns (cosine > 0.7). B Overall structure of the molecular network, with the cluster corresponding to the major ion of fraction FR3 highlighted by a orange circle

The global molecular network (Fig. 5B) shows additional families; however, only four annotations were observed (https://gnps2.org/status?task=f15519c5f65e448e9796f4031e82cbab). These results indicate the possibility of new substances in fraction FR3. Furthermore, the bioautography analysis (Supplemental Material-5) revealed that the biological activity spectrum of fraction FR3-B (Table 3) may be associated with pentamycin (which absorbs UV light at 365 nm) or with other compounds that were not annotated within their respective molecular families.

Structural Characterization of the Macrolide Pentamycin.

The pentamycin (Fig. 6) sample was obtained as a yellow amorphous solid (2.3 mg), corresponding to the major peak (tR = 7.7 min at 315 nm) from the HPLC purification of FR3 (Supplemental Material-6). High-resolution mass spectrometry (HRMS) analysis confirmed its molecular mass and formula of 670.393 (calculated) u and C_35_H_58_O_12_, respectively, through the ion observed at m/z 671.573 ([M + H]^+^) (671.471, calculated), confirming the ions at m/z 693.4 ([M + Na]^+^) and 709.5 ([M + K]^+^) in the low-resolution spectrum. Its structural identification as pentamycin was first verified through the interpretation of the 1D and 2D NMR data (Supplemental Material-7) (Table 4). In the ^1^H NMR spectrum, multiple signals consistent with this molecule were observed in the region of carbinolic hydrogens, at δ 3.1–4.0, and in the olefinic region, at δ 5.9–6.5, as well as signals from CH, CH_2_, and CH_3_ groups at δ 0.8–2.5. Additional signals of hydrogen of OH were detected at δ 4.8–5.3, which typically exhibited lower intensities when the experiment was conducted with water signal suppression and varied slightly in position across different repetitions of the ^1^H NMR spectrum. In the ^13^C NMR spectrum, the 35 expected signals for pentamycin were observed, including two overlapping signals at δ 73.0 (Table 4). Using DEPT-135 and 2D NMR (Supplemental Material-7) spectra, the multiplicities and correlations coherent with the structure of this compound were determined (Table 4). The hydrogen H-2 at δ 2.49 (dd, 8.4, 7.4 Hz) correlates in the COSY spectrum (Supplemental Material-7) with hydrogens H-3 and H-1′, and in the HMBC spectrum (MS7) with carbons C-1, C-3, C-4, C-1′, and C-2′, consistent with its central position relative to the ester functionality, the side chain, and the polyhydroxylated portion of the molecule. H-27 shows COSY correlations with H-26 and H-28 and HMBC correlations with C-1, C-25, C-26, and C-28, in coherence with the ring closure through the ester oxygen and its proximity to the beginning of the olefinic sequence. At the opposite end of this sequence, the methyl protons H-29 display a weak COSY correlation with H-17 and HMBC correlations with C-15, C-16, and C-17. These correlations, together with those of H-14—correlating in HMBC with C-12, C-13, C-15, and C-16 and in COSY with neighboring H-13 and H-15—confirm the junction between the olefinic sequence and the three adjacent carbinolic carbons. In turn, the connection between this region and the end of the sequence of alternating methylene and carbinolic carbons is observed by the correlations of H-12a and H-12b in HMBC with C-10 and C-11, and the COSY correlations of H-12b with H-12a and neighboring H-11 and H-13. Correlations of the hydroxyl protons within the alternating methylene/carbinolic sequence were also important, as they allowed the assignment of the positions of their respective carbons. For example, the signal at δ 4.90 for the 1′-OH proton correlates with H-1′ in COSY and with C-2, C-1′, and C-2′ in HMBC. The proximity of this hydroxyl proton to 3-OH is supported by a NOESY cross-peak, and additional NOESY correlations confirmed the spatial proximities and orientation of neighboring protons along the sequence up to 11-OH. All spatial correlations observed in the NOESY and NOE experiments are also consistent with the reported data for pentamycin (Table 4). Notably, the spatial correlations of H-2 with H-2′ and 3-OH, and those of the olefinic protons H-17, H-21, and H-25 with protons of the alternating methylene/carbinolic sequence, were clearly observed (Supplemental Material-8). The multiplicity and chemical-shift data from the ^1^H and ^13^C NMR spectra closely match those previously reported for pentamycin (also known as fungichromin) (Babczyk and Menche 2023; Noguchi et al. 1988), despite the fact that those authors used 25 mol% DMSO-d₆/MeOH-d₄ for ^1^H NMR and MeOH-d₄ for ^13^C NMR, while we used only DMSO (Supplemental Material-9). The use of DMSO as the sole solvent provided the advantage of enabling the observation of hydroxyl-proton correlations, reported here for pentamycin for the first time (Table 4).Fig. 6. Pentamycin produced by LaBMicrA B280 strain and its ^13^C shifts, observed in its ^13^C NMR spectrumTable 4^1^H, ^13^C, and 2D NMR data of pentamycin1H/13Cδ 1H^a^ J**δ 13C^a^HMBC^a^COSY^a^NOESY^a^NOE^a^1172.9s22.49 dd (J = 8.4, 7.4 Hz) 1H60.6 d1, 3, 4, 1′, 2′3, 1′3, 2′b2′a/b, 3, 3-OH34.02* m* 1H72.2 d1, 5, 2w, 1′w2, 4, 3-OH241.39 m42.0* t2f, 53, 53-OH, 5-OH53.91 m 1H72.1 dno4, 6, 5-OH6a6b1.30 m1.39 m45.8 t521, 5-OH73.90 m 1H71.8 d7-OH5-OH8a8b1.33 m1.41 m45.7 t21, 7-OH, 9-OH2193.89 m 1H73.0 d10a/b, 9-OH7-OH, 9-OH10a10b1.29 m1.41 m44.7 t9, 11, 12w9, 1110b10a, 1113113.80 m 1H71.1 d12b, 10b, 11-OH12a12b1.39 m1.60 ddd (J = 14.3, 10.8, 3.9 Hz) 1H40.5 t10, 1110, 1112b11, 12a, 1313, 17,11-OH133.14 d (J = 10.5 Hz)70.2 d12b, 13-OH, 1410b, 12a10b, 12a, 15, 17, 29143.50 d (J = 9.2 Hz)78.4 d12, 13, 15, 16w13, 15153.71 d (J = 8.9 Hz) 1H80.0 d14, 16, 17, 2914, 15-OH, 17f1716140.6s175.96 dd (J = 11.2, 1.0 Hz) 1H129.0 d15, 18, 19, 2918, 15f, 2912a, 15, 1910b, 13, 18, 19, 29186.48 dd (J = 14.4, 11.4 Hz) 1H129.8 d16, 17, 2017, 192917, 29196.28 m 1H134.9 d17, 2118, 2017206.35 m 1H134.2 d1819216.37 m* 1H130.9 d19226b, 8a/b, 25226.26 m 1H133.1 d20, 2421, 23236.40 dd (J = 14.0, 10.7 Hz) 1H135.11 d22, 2522, 2425246.36 m 1H135.13 d22, 2623, 25256.06 dd (J = 14.4, 4.2 Hz) 1H136.9 d23, 2624, 2621, 23, 264, 23, 24, 26, 27, 28264.01 m 1H73.0 d25, 2725, 26-OH, 2725, 27, 28274.65 dq (J = 7.3, 6.4 Hz) 1H75.0 d1, 25, 26, 28w26, 2826, 283/26, 24, 25, 28281.22 d (J = 6.3 Hz) 3H19.6 q26, 272726, 27291.71 s 3H13.5 q15, 16, 17, 19w171813, 181′3.69 m 1H71.4 d2, 1′-OH, 2′a2′a1.28* m36.0 t4′1′22′b1.38 m4′w3′a3′b1.26 m1.43 m26.4 t41.25 m33.1 t3′, 5′51.27 m*23.9 t4′, 6′6′6′0.88 t (J = 7.0 Hz) 3H15.8 q5′, 4′5′1′-OH4.90 s2, 1′, 2′1'3-OH3-OH5.19 d (J = 2.6 Hz)2, 332, 3, 4, 5-OH, 1′OH5-OH5.07 s4, 5, 654, 3-OH, 6b, 7, 7-OH7-OH4.97 s6, 7, 875-OH,8a, 9, 9-OH9-OH5.22 s8, 9, 1097-OH, 8a, 9, 11-OH11-OH4.91 s10w, 11, 12w119-OH, 12a13-OH4.42 d (J = 5.0 Hz)13w1314-OH4.57 sno15-OH15-OH4.85 s1514-OH26-OH5.33 d (J = 4.8 Hz)25, 26, 272626

Pent NMR spectra from the Streptomyces sp. LaBMicrA 280 strain, with ^1^H NMR at 500.13 MHz and ^13^C NMR at 125.76 MHz, in DMSO-d6 with TMS as internal reference.

Discussion

The Streptomyces sp. LaBMicrA B280 strain was isolated from the rhizosphere of I. edulis, an endemic plant of the Amazon region used as an agroecological vector in the cultivation of other economically important crops in the Amazon (Iglesias et al. 2011; Nichols and Carpenter 2006). In a previous study, scanning electron microscopy (SEM) revealed that the micromorphological structures of LaBMicrA B280, with spiral and septate filamentous hyphae, in coherence to the Streptomyces genus (Souza Rodrigues et al. 2024b). Furthermore, the 16S rRNA gene of this strain exhibited more than 98% similarity to the type strains S. graminearus NBRC 15420^T^ and S. murinus NBRC 12799^T^ (Souza Rodrigues et al. 2024b). With the genus confirmed, identifying the species remained unresolved, which was one of the motivations for sequencing the complete genome in this study. N50 and L50 values, visualized on the BV-BRC platform (3.35.5), indicated the high quality of the LaBMicrA B280 draft genome structure, since the N50 value exceeded the L50 value, facilitating a more comprehensive understanding of the genomic structure and composition of the gDNA sample. As observed, the dDDH values (d2 and d4, in %) together with other metrics, that is, ANI, OrthoANI, AAI, ANIb, and ANIm, observed in the comparison to S. murinus DSM 41827^T^ and its synonymous type species (Table 1), identified as the closest species group, suggest the LaBMicrA B280 strain as a potential novel species within the Streptomyces genus, in line with internationally accepted validation standards (Meier-Kolthoff and Göker 2019; Meier-Kolthoff et al. 2022; Komaki 2021; Lee et al. 2016; Prasad et al. 2022; Richter and Rosselló-Móra 2009; Richter et al. 2016; Yoon et al. 2017). Whether or not it is a novel species, the LaBMicrA B280 strain represents a novelty in its DNA sequence, a hallmark of its evolutionary development. Together with BGCs of known valuable metabolites, such as the antibiotic Pentamycin, it codes unique BGCs that may signal its adaptation and differentiation in the Amazonian environment and represent a source of helpful new metabolites, including antibiotics.

Among the 27 functional CDSs annotated using the RAST platform (SEED Viewer version 2.0) in the LaBMicrA B280 genome, some may be associated with functions critical for soil and plant health. For example, the Phosphorus Metabolism category suggests that LaBMicrA B280 may act as an environmental regulator of phosphorus bioavailability, an essential nutrient for plants, as it promotes root development and water absorption (Oyedoh et al. 2023a, b). In the Secondary Metabolism category, the ability to synthesize the non-proteinogenic amino acid lanthionine, utilized in the production of peptide antibiotics and microcin-class peptide antibiotics, highlights LaBMicrA B280's potential as a biocontrol agent (Oyedoh et al. 2023a, b). In the Nitrogen Metabolism category, the strain's ability to metabolize nitrogen compounds relates to nitrogen bioavailability, indispensable for amino acid synthesis (Oyedoh et al. 2023a, b). Additionally, the Iron Acquisition and Metabolism category indicates the capacity to synthesize siderophores, such as Siderophore assembly kit (26) and Siderophore Desferrioxamine E (6), which facilitate iron uptake by solubilizing and transporting it. Iron is a vital plant nutrient, contributing to chlorophyll synthesis, photosynthesis, nitrogen metabolism, and cellular respiration (Oyedoh et al. 2023a, 2023b). Moreover, siderophores can form stable complexes with heavy metals (e.g., U, Np, Al, Cu, Cd, Ga, Zn, and Pb), increasing their soluble concentrations (Schütze et al. 2015). Along these, other potential metabolic activities identified in the functional CDSs of LaBMicrA B280 underscore its importance for soil and plant health, positioning it as a promising agent for bioremediation, fertilization, and biocontrol (Oyedoh et al. 2023a, b).

Corroborating the results described above, the high number of orphan gene clusters present in the LaBMicrA B280 genome (≥ 69.23%) not only confirms it as a novel species but also reveals a potential wealth of unknown natural products (NPs) that can be explored through currently available methodologies (Souza Rodrigues et al. 2024a). Among the predictions with 100% similarity, notable examples include: geosmin, a volatile secondary metabolite responsible for the characteristic earthy odor (Jiang et al. 2007) and environmental chemical signaling (Garbeva et al. 2023); ectoine, a secondary metabolite with significant biological activities, such as anticancer properties (Sheikhpour et al. 2019), and a potential drug candidate for Alzheimer's disease and rhinoconjunctivitis (Liu et al. 2021); pentamycin, an antifungal used to treat vaginal candidiasis (Zhou et al. 2019) [56] and trichomoniasis (Kranzler et al. 2015); and desferrioxamine B and E, non-peptidic siderophores, commonly synthesized by species of the genera Streptomyces, Nocardia, and Micromonospora, involved in iron chelation, an activity essential for microorganisms and plants (Arulprakasam and Dharumadurai 2021). Desferrioxamine B, marketed as Desferal, is minister to iron overload in humans (Barona-Gómez et al. 2004; Chiani et al. 2010). While having similarity levels < 100%, several biosynthetic gene clusters (BGCs) are potentially associated with the synthesis of unknown antibiotics, anticancer agents, siderophores, and other NPs (Table 2), as for example: the regions 14.2 (78%) and 44.1 (12%), potentially linked to the synthesis of mirubactin (Giessen et al. 2012) and foxicin (Greule et al. 2017) siderophores types; the regions 1.3 (Han et al. 2015), 9.1 (Duan et al. 2023), 37.1 (Liu et al. 2014), 52.3 (Olano et al. 2004; Schulze et al. 2014), 56.1 (Dong et al. 2020), and 28.1 (Snyder et al. 2009), potentially related to antibiotics; and the regions 2.1 (Stadler et al. 2007), 11.1 (Huczynski 2012), 21.1 (Shi et al. 2019), 26.1 (Hecht 2000), 55.1 (Wang et al. 2018), 55.2 (Rambabu et al. 2015), 55.3 (Carbone et al. 2010), and 36.1 (Xu et al. 2020), possibly related to antimicrobial and anticancer metabolites. Along these, other BGC regions with low similarity levels suggest that LaBMicrA B280 could be a source of unknown NPs, particularly those synthesized by PKS and NRPS-type BGCs, widely distributed among Streptomyces spp. (Belknap et al. 2020).

As observed, pentamycin-containing fraction FR3 not only confirmed the activity of the AcOEt/2-propanol 9:1 extract from the LaBMicrA B280 strain against the phytopathogens Colletotrichum sp. (ISO01), C. guaranicola (P01), Pestalotiopsis sp. (3002R2), C. cassiicola (CD: Iso079), and S. coffeicola (M01) (Souza Rodrigues et al. 2024b), but concentrated and overcame the extract activities as seen by its MIC values against these pathogens and, even, by its bioautographic analysis, where the observed activity against C. cassiicola corresponds to position of the diluted pentamycin’s spot. Its great activity against most strains above (MIC 62 µg/mL), larvicidal activity against A. aegypti (FO), the primary vector of dengue, yellow fever, Zika, and Chikungunya, good antiplasmodial activity against two strains of P. falciparum (W2 and 3D7), and cytotoxicity against the HEp-2 cancer cell line underscore the potential of pentamycin and/or another metabolite in fraction FR3. The molecular networking analysis of fraction FR3 revealed a compact cluster of ions structurally related to the m/z 693.381 ion, identified as pentamycin, which represents the major compound and the central node of the highlighted molecular family. Ions differing by approximately 1 Da (e.g., m/z 692.370) displayed nearly identical MS2 spectra (cosine = 0.97), indicating that they correspond to the same molecule in different ionic states—a typical behavior for polyoxygenated macrolides. Additional nodes exhibiting mass differences of ~ 30 Da or ~ 14 Da suggest the presence of minor structural analogues, possibly arising from small biosynthetic variations, such as modifications involving CH_2_O or CH₂ units. Although the chemical diversity within the highlighted family is relatively limited, the GNPS analysis also revealed other molecular families composed of ions without available annotations in public libraries. The absence of spectral matches in these clusters suggests the presence of poorly characterized or potentially novel metabolites in fraction FR3, which is consistent with the low representation of uncommon macrolides and polyketide-derived compounds in MS/MS databases (Wang et al. 2016). Altogether, these findings indicate that, although pentamycin dominates the chemical profile of the fraction, FR3 also contains other potentially novel metabolites that warrant further investigation.

These unidentified compounds in addition to pentamycin could be related to the biological activities reported in this study, supporting the genomic mining results. Macrolides represent an important class of natural products (NPs) with medicinal and agro-industrial relevance, exhibiting various biological activities (e.g., antimicrobial, anticancer), and are synthesized by actinobacteria isolated from diverse ecosystems (Al-Fadhli et al. 2022; Li et al. 2021; Wang et al. 2017; Zhang et al. 2023). A significant limitation of this class is the high chemical instability of some compounds, rendering them unsuitable for clinical use, as seen with pentamycin (Zhou et al. 2019). However, characterizing BGCs of biotechnological interest has paved the way for the biosynthetic engineering of macrolide analogs to overcome chemical instability in addition to improving activity. For example, Payero et al. (2015) demonstrated that deleting genes encoding CYPs responsible for hydroxylation at C-26 (FilC) and C-1′ (FilD) in filipin III biosynthesis resulted in analogs with stronger antifungal activity. This approach could be applied to evaluate whether pentamycin analogs might exhibit improved activity and chemical stability for clinical applications (Zhou et al. 2019). Another limitation is the low bacterial production of microbial macrolides, a problem that has been addressed by synthesis and biotechnological approaches. In an interesting example, Wan et al. (2023) reported efforts to enhance the synthesis of the macrolide fungichromin (pentamycin) by Streptomyces sp. WP-1. By overexpressing regulators ptnF (PAS-CitB), ptnR (AfsR-Dnrl-RedD_regulator), and ptnB (crotonyl-CoA reductase/carboxylase), fungichromin yield increased significantly to 8.5 g/L. Pentamycin is known as an agriculturally relevant antifungal due to its activity against key phytopathogens such as Rhizoctonia solani, Phytophthora infestans, and Pseudoperonospora cubensis (Shih et al. 2003). Our study highlights that its activity spectrum might extend to additional phytopathogen genera. In addition, the richness of yet-unknown compounds observed in the genome and secondary metabolism of LaBMicrA B280 ratifies the immense biotechnological potential of the Amazon rainforest.

Conclusion

Streptomyces sp. LaBMicrA B280, isolated from the rhizosphere of I. edulis, is a potential novel species based on the set of its dDDH, ANI, and AAI values compared to S. murinus, its closest known relative. Genome analysis revealed the presence of multiple BGCs associated with typical actinomycete-derived metabolites, such as geosmin, ectoine, pentamycin, and desferrioxamine B and E, as well as numerous uncharacterized clusters with potential for the production of novel metabolites. Metabolic subsystem analysis revealed the strain’s potential for environmental and agricultural applications, supported by CDSs involved in phosphorus, nitrogen, and iron and lanthionine biosynthesis. Furthermore, biological assays revealed LaBMicrA B280 as a promising source of bioactive compounds with strong antimalarial, larvicidal, cytotoxic, and antifungal activities. Pentamycin was identified as the major compound in the fraction exhibiting the most significant results; however, it remains unclear whether the observed effects were due to pentamycin alone or to its combined action with other metabolites in the fraction, underscoring the need for further studies to address this question.

Supplementary Information

Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 3859 KB)