The Journal of Antibiotics (2025) 78:542–551 https://doi.org/10.1038/s41429-025-00846-3 ARTICLE Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurella pulmonis, as an efficacious therapeutic agent against multidrug-resistant bacteria Sung-Jin Kawai1 Shumpei Asamizu Kouji Kimura6 Makoto Ojika7 ● 2,3,4 ● Hiroaki Suzuki1 Hiroyasu Onaka ● 2,3,5 ● Yoshichika Arakawa6,8 ● ● 1234567890();,: 1234567890();,: Received: 6 April 2025 / Revised: 4 June 2025 / Accepted: 15 June 2025 / Published online: 9 July 2025 © The Author(s) 2025. This article is published with open access Abstract Bacterial interactions can affect the production of secondary metabolites and, therefore, provide a promising approach to exploring new microbial compounds. In this study, we screened actinomycetes isolated from Hegura Island, Ishikawa Prefecture, Japan, to discover new antibiotics through combined-culture with Tsukamurella pulmonis TP-B0596. Three new phenazine-class antibiotics, griseoluteins T (1), C (2), and D (3), along with two known related metabolites, griseoluteic acid (4) and griseolutein A (5), were detected in both mono- and combined-cultures of Streptomyces seoulensis HEK131 with T. pulmonis at different production levels. Detailed spectroscopic analysis revealed that 1 contained a dihydrophenazine core, and was converted to 5 by accepting oxidation spontaneously. 1, containing a dihydrophenazine group, was relatively unstable under oxidative conditions, and the addition of ascorbate was required during the isolation of the compound. 2 and 3 were found to be cysteine-adducts analogous to 4, and their productivity was increased in the combined-culture. We further assessed the antibacterial activities of 1 against clinically significant Gram-positive pathogenic bacteria, including 30 methicillin-resistant Staphylococcus aureus (MRSA), 27 vancomycin-resistant Enterococci (VRE), and 17 Clostridioides difficile. Notably, 1 was found to possess higher antibacterial activity against these microorganisms than several clinically important antibiotics, while displaying lower cytotoxicity against HeLa-S3 cells. Introduction Filamentous actinomycetes are the source of more than 60% of the known natural product antibiotics [1]. The genomes of actinomycetes have been found to contain a number of These authors contributed equally: Sung-Jin Kawai, Shumpei Asamizu. Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41429025-00846-3. * Shumpei Asamizu shumpei.asamizu@port.kobe-u.ac.jp 1 New Field Pioneering Division, Toyota Boshoku Corporation, Kariya, Aichi, Japan gene clusters for secondary metabolite biosynthesis, including pharmaceutically important compounds [2, 3]. However, most of the biosynthetic gene clusters encoding secondary metabolites are silent in culture under normal laboratory conditions [4]. Bacterial co-culture has been found to activate the laboratory silent biosynthetic gene clusters for the production of secondary metabolites in Actinobacteria [5–9]. Combined-cultures, comprising a specific combination of actinomycetes and mycolic acid-containing bacteria, including Tsukamurella pulmonis TP-B0596, have been 4 Engineering Biology Research Center, Kobe University, Kobe, Japan 5 Department of Life Science, Faculty of Science, Gakushuin University, Tokyo, Japan 6 Graduate School of Medicine, Nagoya University, Nagoya, Japan Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan Present address: Department of Microbiology, Fujita Health University School of Medicine, Toyoake, Aichi, Japan 2 Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan 7 3 Collaborative Research Institute for Innovative Microbiology, The University of Tokyo, Tokyo, Japan 8 Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurellay successful in augmenting the production of cryptic secondary metabolites in monocultures, resulting in the isolation of 44 new compounds from 15 actinomycetes to date [5, 10, 11]. In combined-cultures, modifications of known compounds have also been observed. These include harundomycin A, a conjugate between platensimycin and enterobactin from Streptomyces hygroscopicus HOK021 [11]; desferrioxamines modified with an unusual N-containing five-membered ring structure, termed I2a, I1a, and I1b, from Streptomyces davawensis JCM 4913 [12]; conversion of [5,5]-spirohemiaminals to 5-alkyl-1,2,3,4tetrahydroquinolines in Streptomyces nigrescens HEK616, which may involve an additional oxidation step [13]; and variations in the cyclization patterns of macrolactams from similar backbones, such as sceliphrolactam to niizalactam [14], micromonolactam to dracolactam [15], and mirilactam A to mirilactam C, D, and E [16], which may involve additional oxidative pathways. Mechanisms of activation in combined-culture for secondary metabolism have been investigated using model actinomycetes [17–20], which may involve as yet unidentified mechanisms along with multiple cellular processes. Multidrug-resistant bacteria present an urgent and serious public health threat. It has been estimated that the number of deaths due to infectious diseases caused by antimicrobial resistance will exceed 10 million per year by 2050 [21]. In fact, infectious disease mortality due to multidrug-resistant bacteria is increasing year by year [22]. The development of novel antibiotics that operate via new antibacterial mechanisms is a promising approach to combat this increasing threat. We previously isolated 854 actinomycetes from environmental samples collected at Hegura Island, Ishikawa Prefecture, Japan [23], and discovered that combinedculture successfully induced the specific production of various metabolites [15, 24, 25]. To search for new antibiotics active against clinically hazardous drug-resistant bacteria, we performed mono- and combined-culture of the actinomycetes strains with T. pulmonis and tested their antibacterial activities. A bioassay-guided fractionation of the culture broths of a selected Streptomyces seoulensis HEK131 strain resulted in the discovery of promising metabolites with a phenazine core structure. Here, we report the structural elucidation and biological activities of the isolated metabolites. Materials and methods General Actinomycetes and T. pulmonis were cultured using ISP2 and A3M media. ISP2 agar medium contained glucose 543 (0.4%), malt extract (1.0%, Difco, BD Biosciences, CA, USA), yeast extract (0.4%, Bacto, BD Biosciences), and agar (1.5%) in distilled water (pH 7.2). A3M medium contained starch (2%), glucose (0.5%), glycerol (2%), pharmamedia (1.5%), yeast extract (0.3%), and Diaion HP20 (1%, pH 7.0). Cultures were grown in 96-well plates, shaken on a Maximizer MBR-420FL (TAITEC, Tokyo, Japan). Plates were sealed with breathable film (SureSeal RA Breathable; BMBio, Tokyo, Japan) during cultivation. For butanol extraction, plates were sealed with an aluminum seal (WATSON, Tokyo, Japan). Solvents were removed using a centrifugal concentrator equipped with a cold trap (EC500 and CS-110-4, respectively; Sakuma Co., Tokyo, Japan). Instruments and settings Optical rotations were recorded on a DIP-370 digital polarimeter (Jasco, Tokyo, Japan). UV and IR spectra were measured on a V-730 BIO spectrophotometer and an FT/IR-4100 (Jasco), respectively. NMR spectra were recorded on an Avance ARX400 (400 MHz for 1H, 100 MHz for 13C) or an Avance III HD 600 MHz Cryoprobe spectrometer (600 MHz for 1H, 150 MHz for 13C) (Bruker Japan, Yokohama, Japan) using the residual solvent peak of DMSO-d6 at δ 2.50 for 1H and δ 39.5 for 13C as internal standards. Preparative high-performance liquid chromatography (HPLC) was performed on a highpressure gradient system equipped with PU-2087plus pumps, a DG-2080-53 degasser, and a UV-2075plus detector (Jasco). LC/MS analysis was performed by a 6520 Accurate-Mass Q-TOF spectrometer (Agilent Technologies, Santa Clara, CA, USA) in the electrospray ionization mode (positive) and a 1100 HPLC system (Agilent Technologies). Bacterial strains used in this study Eight hundred and fifty-four strains of actinomycetes, collected from Hegura Island, Ishikawa Prefecture in Japan, were used for mono- and combined-culture with Tsukamurella pulmonis TP-B0596. Methicillin-susceptible Staphylococcus aureus ATCC29213 (MSSA), methicillinresistant S. aureus ATCC43300 (MRSA), Escherichia coli ATCC25922, E. coli ATCC BAA-2471 (NDM-1-producing reference strain), Enterococcus faecalis ATCC51299 (vanB positive, VRE), and Clostridioides difficile ATCC700057 were purchased from American Type Culture Collection for the antibacterial activity tests. E. coli E1346 is a clinically isolated strain with resistance to colistin as it harbors the mcr-1 gene [26]. Colistin and ampicillin were purchased from FUJIFILM Wako Pure Chemical (Tokyo, Japan) and used as controls. 544 Culture plate screening of antibacterial compounds For screening using 96-deep-well culture plates, first, T. pulmonis was streaked onto ISP2 agar medium and cultivated for 2 days at 30 °C. Then, for the seed culture, T. pulmonis was cultivated in a test tube containing 10 mL of ISP2 medium, which was shaken on a rotary shaker (200 rpm) at 30 °C for 2 days. Each frozen stock of actinomycetes (854 strains) was inoculated separately into a well of a 96-deep-well plate containing 100 µL of ISP2 medium for seed culture. The plates were shaken on a Maximizer (1500 rpm) at 30 °C for 3 days for the pre-culture. Then, for the production combined-culture, A3M medium (500 µL) with 1.5% (v/v) of T. pulmonis culture broth was added to the 96-deep-well plate containing the actinomycetes seed cultures. For monoculture, only A3M medium (500 µL) was added. Then, all plates were sealed with breathable seal and shaken on the Maximizer (1500 rpm) at 30 °C for 3 days or 7 days. To minimize evaporation of the culture medium, the outer 36 wells of a 96-well plate were filled with water, and cultivation was conducted in the inner 60 wells (arranged in 6 rows × 10 columns). Although evaporation could not be completely prevented, the remaining culture medium was used directly for butanol extraction. After cultivation, 500 µL of butanol was added to each well, sealed with an aluminum seal, and shaken on the Maximizer (1500 rpm) at 30 °C for 30 min for extraction. After centrifugation (3000 rpm, 10 min, room temperature), the supernatants (butanol layers) were transferred into a new microplate and dried in vacuo using a centrifugal concentrator equipped with a cold trap. The dried crude extracts were dissolved in 0.1 vol. of dimethyl sulfoxide (DMSO) and transferred to new microplates for an antibiotic assay. Extraction and isolation of active compounds from S. seoulensis HEK131 A 100 mL broth of S. seoulensis HEK131 (monoculture or combined-cultured with T. pulmonis) was diluted with EtOH (100 mL), and sodium ascorbate (10 mg) was added to prevent the autooxidation of compounds, because griseolutein T (1) was found to be susceptible to autoxidation. The oxidation product was formed without the addition of sodium ascorbate prior to the separation process. The mixture was shaken at 180 rpm for 1 h and centrifuged at 6000 rpm for 10 min to extract the metabolites. The supernatant was separated, and the precipitate was extracted again with 50% EtOH (20 mL). The first supernatant and the second cell extracts were combined and concentrated to approximately 30 mL. A portion (0.1%) of the extract was stored and used for LC/MS analysis. Two extracts obtained from the 5-day cultures without T. pulmonis (200 mL in total) were combined and chromatographed on an octadecylsilyl (ODS) silica gel column S.-J. Kawai et al. (Cosmosil 75C18-PREP, 500 g, Nacalai Tesque, Kyoto, Japan) eluted with 20%, 30%, 40%, and 50% acetonitrile (MeCN) in H2O containing 0.1% trifluoroacetic acid and 0.01% sodium ascorbate (250 mL × 4 for each fraction batch). Fractions 7 and 8 eluted with 30% MeCN were combined and concentrated to give a brown material-I (229 mg). Material-I was purified by preparative HPLC [Develosil ODS-HG-5 (20 × 250 mm, Nomura Chemical, Aichi, Japan), 10%–40% MeCN-20 mM NH4OAc, 60 min linear gradient, 6 mL min−1, detected at 273 nm] to give 1 (1.04 mg, tR = 24.7 min). Fractions 11–15 eluted with 40%–50% MeCN were combined and concentrated to give a brown material-II (250 mg). Material-II was purified by preparative HPLC [Develosil ODS-HG-5 (20 × 250 mm), 25%–35% MeOH-20 mM NH4OAc, 60 min linear gradient, 5 mL min−1, detected at 265 nm] to give 2 (0.05 mg, tR = 57.3 min) and 4 (0.03 mg, tR = 50.7 min). To further obtain the purified compounds, the extracts obtained from the 4-, 5-, and 6-day combined-culture broths (300 mL in total) were combined, extracted, and chromatographed on an ODS column under the same conditions described above. Fractions 7 and 8 eluted with 30% MeCN were combined and concentrated to give a brown materialIII (245 mg). Material-III was purified by HPLC under the same conditions as material-I to obtain another purified 1 (0.83 mg, tR = 25.0 min). Fractions 10–14 eluted with 40%–50% MeCN were combined and concentrated to give a brown material-IV (290 mg). Material-IV was purified by HPLC (under the same conditions as material-II) to obtain another purified 2 (0.52 mg, tR = 58.1 min), 3 (0.17 mg, tR = 39.2 min), and 4 (0.03 mg, tR = 48.5 min). Some samples containing oxidized 1 were purified by HPLC (under the same conditions as material-I) to give 5 (0.08 mg, tR = 35.1 min). Griseolutein T (1): UV λMeOHmax nm (ε) 224 (23,000, sh), 267 (14,300), 341 (5700); IR (KBr) cm−1 3385, 1677, 1607, 1588, 1510, 1478, 1446, 1397, 1287, 1260, 1105, 1026, 999, 768; NMR (see Table 1), ESI-TOF MS m/z 327.0973 (calcd for C17H15N2O5 [M − OH]+ 327.0975), 343.0916 (calcd for C17H15N2O6 [M − H]+ 343.0925), 367.0892 (calcd for C17H16N2O6Na [M + Na]+ 367.0901). Griseolutein C (2): [α]14D –6 (c 0.016, MeOH), UV λMeOHmax nm (ε) 247 (31,600), 273 (41,700), 370 (13,500); IR (KBr) cm−1 3413, 3079, 1716, 1653, 1602, 1542, 1465, 1286,1134, 1096, 1025, 766; NMR (see Table 1), ESI-TOF MS m/z 430.1068 (calcd for C20H20N3O6S [M + H]+ 430.1067), 452.0878 (calcd for C20H19N3O6SNa [M + Na]+ 452.0887). Griseolutein D (3): [α]14D –32 (c 0.031, MeOH), UV MeOH −1 λ max nm (ε) 253 (60900), 370 (15100); IR (KBr) cm 3425, 1715, 1649, 1538, 1465, 1026, 999, 761; NMR (see Table 1), ESI-TOF MS m/z 400.0958 (calcd for C19H18N3O6S [M + H]+ 400.0962), 422.0777 (calcd for C19H17N3O6SNa [M + Na]+ 422.0781). Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurellay Table 1 NMR data for griseolutein T (1), C (2), and D (3) (in DMSO-d6)a No Griseolutein T (1) δC δH (mult., J in Hz) Griseolutein C (2) δC d 545 Griseolutein D (3) δH (mult., J in Hz) δCd δH (mult., J in Hz) 1 125.2 - 127.2 - 129.1 - 2 129.4 7.77 (d, 7.6) 134.8 8.63 (d, 6.8) 133.4 8.49 (d, 8.8) 3 118.2 6.84 (t, 7.6) 131.0 8.12 (dd, 8.4, 6.8) 130.4 8.08 (dd, 8.0, 7.2) 4 126.1 7.36 (d, 7.6) 133.8 8.53 (d, 8.4) 133.6 8.48 (m) 4a 123.0 - 141.8e - 141.7 - 5a 123.3 - 141.9e - 141.5 - 6 130.7 - 128.8 - 138.0 - 7 118.6 7.04 (d, 8.4) 131.3 7.94 (d, 8.0) 130.8 8.01 (m) 8 108.6 6.90 (d, 8.4) 108.9 7.40 (d, 8.0) 132.4 8.02 (m) 9 145.1 - 152.5 - 127.7 8.27 (m) 9a 129.4 - 133.2 - 141.1 - 10a 142.7 - 137.9 - 139.6 - 1’ 59.1 4.42, 4.45 (d, 14.6) 29.9 4.35, 4.40 (d, 13.6) 30.0 4.47, 4.53 (d, 13.2) 1” n.d.b - 172.0 - 172.1 - 2” 59.8 3.64, 4.37 (brd, 13.2) 52.6 4.37 (m) 52.2 3” - 1-CO 169.5 - 4.44 (m) 33.7 2.76 (dd, 13.5, 7.4) 33.4 2.94 (dd, 13.5, 4.8) 2.78 (dd, 13.6, 8.0) 2.95 (dd, 13.6, 5.0) - 165.9 - 166.5 9-OMe 56.0 3.84 (s) 56.5 4.12 (s) - 1’-OH - 4.91 (brs)c - - - 2”-OH - 4.73 (brt)c - - - NH - 11.54 (s) - 7.98 (m) 8.10 (m) Ac - - 168.9, 22.5 1.83 (s) 169.1, 22.4 1.84 (s) a - Measured at 400 MHz for 1H and 100 MHz for 13C b Not detected c Detected in another spectrum measured at 600 MHz d Determined by HMBC or heteronuclear single quantum coherence (HSQC) e Interchangeable signals Griseoluteic acid (4) [27, 28]: 1H NMR (400 MHz, DMSO-d6) δ 8.55 (1H, brd, J = 8.0 Hz), 8.48 (1H, d, J = 7.6 Hz), 8.12 (1H, m), 7.97 (1H, d, J = 8.0 Hz), 7.45 (d, J = 8.0 Hz), 5.20 (2H, s), 4.12 (3H, s); ESI-TOF MS m/z 285.0867 (calcd for C15H13N2O4 [M + H] + 285.0870) (See Supplementary Table S1 for the comparison with reported values). Griseolutein A (5) [29]: 1H NMR (400 MHz, DMSO-d6) δ 8.22 (1H, d, J = 8.0 Hz), 8.05 (1H, m), 7.97 (1H, m), 7.94 (1H, d, J = 8.0 Hz), 7.33 (1H, d, J = 8.0 Hz), 5.80 (2H, s), 5.39 (1H, m), 4.11 (3H, s), 4.09 (2H, m); ESI-TOF MS m/z 343.0905 (calcd for C17H15N2O6 [M + H] + 343.0925). Antibacterial activity test E. coli BAA-2471 (NDM-1-producing strain), E. coli E1346 (MCR-1-producing strain), E. coli ATCC25922, and S. aureus (MRSA) were used for an antibiotic assay. MRSA was cultivated overnight at 30 °C, and other E. coli strains were cultivated overnight at 37 °C. Colonies of each strain were suspended in saline and adjusted to McFarland 0.5 by a MicroScan Turbidity Meter (Siemens, Bayern, Germany). Then, the adjusted suspension was diluted (200×) with Mueller Hinton II Broth (MHB, cation-adjusted) (Nippon Becton Dickinson Company, Ltd, Tokyo, Japan). In each well of the microtiter plate, a crude extract in DMSO solution (10 μL) was added to 200 μL of bacterial suspension diluted with MHB. The plates were incubated at 37 °C or 30 °C for 16–20 h. Ampicillin and colistin were used as positive controls. ATCC strains of S. aureus (MRSA), S. aureus (MSSA), E. faecalis, and C. difficile were used as standard strains. Clinical isolates of MSSA (30 isolates), MRSA (30 isolates), VRE (27 isolates), and C. difficile (17 isolates) were supplied by the Department of Bacteriology, Nagoya University School of Medicine (Supplementary Tables S2–S6) and used in an antimicrobial susceptibility test. MIC50 values were determined by the microdilution method according to the CLSI guidelines [30]. The following commercial antibiotics were selected for 546 comparison of the MIC50 values with compound 1. Vancomycin (FUJIFILM Wako Pure Chemical Corporation, Tokyo, Japan), which is one of the anti-MRSA drugs; fidaxomicin (Tokyo Chemical Industry Co., Ltd [TCI], Tokyo, Japan), which is one of the drugs used to treat C. difficile infections; tigecycline (TCI, Tokyo, Japan), which is used to treat Gram-negative and -positive bacterial infections; mupirocin (TCI, Tokyo, Japan), which is used as a topical treatment for MRSA infections; and phenazine-1-carboxylic acid (tubermycin B; Toronto Research Chemicals, Toronto, Canada) as an analog of compound 1. Ascorbate was used as an additive during the purification stage to ensure an effective yield of 1. However, because such additives are not appropriate for biological assays, the bioactivity of 1 was evaluated with consideration of its inherent instability. Cytotoxicity assay The HeLa-S3 cells were obtained from RIKEN BRC (Tsukuba, Ibaraki, Japan). The medium consisted of Eagle’s Minimal Essential Medium (445 mL) (EMEM, FUJIFILM Wako Pure Chemical, Osaka, Japan), fetal bovine serum (50 mL) (heat inactivated, Thermo Fisher Scientific, Waltham, MA, USA), and an antibiotic mix (5 mL) (10,000 units mL−1 penicillin, 10,000 µg mL−1 streptomycin, Thermo Fisher Scientific). The cells (1 × 104 cells per well) in 99 µL of the medium were incubated in the wells of a 96-well plate at 37 °C for 24 h in a 5% CO2 incubator (MCO-96, Sanyo Electric, Osaka, Japan). Then, 1 µL of a compound in 50% DMSO, staurosporine (Fujifilm, Osaka, Japan) or paclitaxel (Fujifilm) in 50% DMSO as positive controls, or 50% DMSO as the negative control, was added to each well. After incubating for 48 h, the cells were treated with 10 µL of 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) in phosphate buffer saline (5 mg mL−1) and incubated for an additional 3 h. The medium was removed, and the cells were dissolved in 100 µL of DMSO and shaken for 5 min. Finally, the absorbance values were measured at 595 nm using a Multiscan FC microplate reader (Thermo Fisher Scientific). The IC50 values were calculated by sigmoid fitting curves generated from the mean values from three experiments at each concentration. Because additives such as ascorbate are not suitable for biological assays, the bioactivity of 1 was evaluated with consideration of its inherent instability. S.-J. Kawai et al. [23]. They were grown in both mono- and combinedcultures with T. pulmonis using 96-well culture plates, and the butanol extracts were used for the antibacterial assay. As a result, extracts of combined-culture from 37 strains (out of the 854 strains) showed more than 10% growth inhibition specifically against three drug-resistant bacteria: a Gramnegative Escherichia coli strain BAA-2471 (NDM-1) exhibiting carbapenem resistance, an E. coli strain E1346 (MCR-1) exhibiting colistin resistance, and a Gram-positive methicillin-resistant Staphylococcus aureus (MRSA) strain ATCC43300 exhibiting multidrug resistance. We selected these 37 actinomycete strains for large-scale culture. When the culture volume was scaled up to flask culture (100 mL in a K-1 flask), seven strains (out of the 37 strains) showed consistent antimicrobial activity against the three drugresistant bacteria tested. We first isolated the active metabolites against E. coli strains from these seven strains. However, most of their growth inhibition activities against E. coli strains were found to be derived from nocardamine, a siderophore well-known as an iron-chelating natural product (data not shown) [23, 31]. We then sought active metabolites against MRSA. Among the seven strains, the crude extracts from Streptomyces seoulensis HEK131 exhibited strong antibacterial activity against MRSA. After chromatographic separation (Supplementary Figs. S1 and S2), the metabolites in the active fraction were found to contain unknown chemical properties. We therefore isolated these metabolites for detailed structural elucidation. Discovery of bioactive phenazine metabolites from S. seoulensis HEK131 Results and discussion S. seoulensis HEK131 was grown in a 100 mL culture for antibiotic production, and the active compounds were isolated and identified as described in the Materials and Methods section (Supplementary Fig. S1 and S2). Five compounds [griseoluteins T (1), C (2), D (3), griseoluteic acid (4), and griseolutein A (5), Fig. 1] were isolated to homogeneity and the chemical structures were predicted using high-resolution mass spectrometry and a series of nuclear magnetic resonance (NMR) spectra (described later). Notably, among the isolated metabolites, 1 was found to be susceptible to oxidative transformation after several rounds of purification. Therefore, the chromatographic processes were performed in the presence of sodium ascorbate, except for the final purification, to prevent autoxidation. Screening of actinomycetes producing antibacterial compounds Structural analysis of griseolutein T (1) isolated from S. seoulensis HEK131 The 854 previously reported actinomycete strains were isolated from Hegura Island, Ishikawa Prefecture, Japan Griseolutein T (1) has the molecular formula C17H16N2O6 based on the exact ion masses determined by an Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurellay 547 Fig. 1 Structures of griseolutein T (1), griseolutein C (2), griseolutein D (3), and known related metabolites griseoluteic acid (4) and griseolutein A (5) isolated in this study Fig. 2 Selected COSY (thick bonds) and HMBC correlations (arrows) for 1, 2, and 3 electrospray ionization time-of-flight mass spectrometer (ESI-TOF-MS). The ions of m/z 327.0973 (obs.) and 367.0892 (obs.) were assigned as [M − OH]+ (calcd 327.0975) and the sodium adduct form [M + Na]+ (calcd 367.0901), respectively, of 1 (Supplementary Fig. S3). Another ion of m/z 343.0916 (obs.) corresponding to [M − H]+ (calcd 343.0925) was observed (Supplementary Fig. S3). Observation of this non-typical ion may indicate that 1 is susceptible to oxidation (dehydrogenation) to form the known metabolite griseolutein A (5) (C17H14N2O5). Taken together, 1 was predicted to be a dihydro analog of 5. The structure of 1 was then analyzed by NMR experiments (Supplementary Figs. S4–S8). A correlation spectroscopy (COSY) experiment revealed two unsaturated hydrocarbon structures, =CH–CH = CH– and –CH = CH– (Fig. 2). Other detected substructures were two hydroxymethyl groups and one methoxy group. A notable 1H signal was observed at δ 11.54 (s), which was estimated to be NH based on the predicted molecular formula. The 13C NMR spectrum indicated the presence of two carbonyls and seven sp2 quaternary carbons. These partial structures were finally connected by a heteronuclear multiple bond correlation (HMBC) experiment (Fig. 2). Although the HMBC correlation between H-2” and C-1” of the hydroxyacetyl group was not observed, the presence of the group and its position were automatically determined from the molecular formula. The structure of griseolutein B [32], which was revised in 1964 [33], was analogous to that of 1 (Fig. 3a). The NMR of 1 in our study showed a similar spectrum to that of griseolutein B, which is a cyclized isomer of 1 (Fig. 3a). Since we observed clear COSY correlations between the protons of two hydroxy groups and each connected methylene (2″ and 1′) (Fig. 2), it was possible to discriminate 1 from griseolutein B (Fig. 3a). Although a theoretically proposed identical structure has been reported in the literature [34], we believe that this is the first study to perform isolation and provide spectral evidence to reveal the chemical structure of 1. 1 was susceptible to autoxidation to form 5, which was the first griseolutein antibiotic to be discovered in 1959 [29]. This instability is due to the dihydrophenazine structure. We predicted that 1 first undergoes hydride removal to form an acyl ammonium ion intermediate 548 S.-J. Kawai et al. Fig. 3 Proposed structure of griseolutein B and its structural relationship to 1 (a), and plausible mechanism of the conversion of 1–5 (b) followed by an intramolecular acyl shift to the 1′-OH group, resulting in the aromatization of the tricyclic system (Fig. 3b). In conclusion, we believe that 5 may be an oxidized shunt product of 1, and that 1 is a true biosynthetic metabolite. Based on its chemical properties, 1 may play some role in the redox system, as predicted for the phenazine-class antibiotic, pyocyanine from Pseudomonas aeruginosa [35]. Structural analysis of griseolutein C (2) and D (3) isolated from the combined-culture of S. seoulensis HEK131 and T. pulmonis The molecular formula of griseolutein C (2) was determined as C20H19N3O6S by HR ESI-TOF MS (Supplementary Fig. S3). The M + 2 isotopic peak intensity of 2 was theoretically consistent with the natural isotope pattern of sulfur, suggesting the presence of a sulfur atom in the molecule (Supplementary Table S7). The NMR spectrum (Table 1, Supplementary Figs. S9–S12) showed similarity to that of 1, but significant differences were the lack of the hydroxyacetyl group and the low-field NH signal in 1, and the presence of an extra substructure that contains a –CH2–CH< system. The 1H–13C long-range connectivity was determined by an HMBC experiment (Fig. 2, Supplementary Fig. S12), revealing a novel griseolutein-type structure for 2, which contains a cysteine substructure. The molecular formula of griseolutein D (3) was determined as C19H17N3O5S by HR ESI-TOF MS (Supplementary Fig. S3). The M + 2 isotopic peak intensity of 3 was theoretically consistent with the natural isotope pattern of sulfur, suggesting the presence of a sulfur atom in the molecule (Supplementary Table S7). The NMR data (Table 1, Supplementary Figs. S13–S17) were similar to those of 2, except for the lack of the methoxy group at the C-9 position. Although 13C data were inadequate due to low yield, the structure was determined by its limited 2D NMR correlations (Fig. 2) and comparison of the chemical shifts between 2 and 3. Griseoluteic acid (4) was isolated as a minor component, and the structure was determined by comparison with the reported MS and 1H NMR data (Supplementary Table S1) [32]. Using LC/MS, the amounts of metabolites 1–5 were quantified to analyze the production profiles in mono- and combined-culture (Supplementary Fig. S18). The standard curves of each metabolite [1–5] for quantification were generated using the isolated metabolites. In the monoculture from day 5, 2 was undetectable, and the yield of 3 was 2.3 mg L−1. By contrast, in the combined-culture from day 5, the yield of 2 was 1.5 mg L−1 and the yield of 3 was 11 mg L-1, indicating the increased productivity of 2 and 3 in the combined-culture. Similar to compounds 2 and 3, cysteinylated streptophenazines have also been isolated from Streptomyces sp. ID63040 [36]. It is speculated that the formation of cysteinylated streptophenazines occurs through the addition of mycothiol, which is used to maintain cellular redox balance and is known to play a role in detoxifying xenobiotics [37]. In combined-culture, it is hypothesized that these cellular processes were enhanced, leading to the increased production of compounds 2 and 3. Although 4 was originally reported as a degradation product of griseolutein B [38], it was subsequently detected in the culture broth of S. griseoluteus and regarded as a biosynthetic precursor of griseoluteins based on the observation that 4 was converted to 5 and griseolutein B following treatment with cell homogenates [28]. Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurellay Table 2 Cytotoxicity of 1–5 against HeLa-S3 cells Compounds 1 2 3 4 5 STA PAC IC50 (µM) 1.6 96 78 8.1 2.5 0.049 0.0094 STA staurosporine, PAC paclitaxel. STA and PAC were generally used as the positive controls Table 3 MIC50 of griseolutein T (1) and known antibiotics against clinical isolates MIC50 (mg L−1) 1 PCA VAN FDX TGC MUP MRSA 0.12 >8 1 4 1 0.25 MSSA 0.12 >8 1 8 0.25 0.25 VRE 0.12 >8 16 2 0.06 0.5 C. difficile 0.06 8 0.25 <0.016 <0.016 >8 MIC50 the lowest minimum inhibitory concentration at which at least 50% of the isolates in a test population are inhibited [44], MRSA methicillin-resistant Staphylococcus aureus, MSSA methicillinsusceptible Staphylococcus aureus, VRE vancomycin-resistant enterococci, C. difficileClostridioides difficile, PCA phenazine-1-carboxylic acid (tubermycin B), VAN vancomycin, FDX fidaxomicin, TGC tigecycline, and MUP mupirocin Antibacterial activities and cytotoxicity of isolated compounds Cytotoxicity (IC50) against the HeLa-S3 cell line was evaluated for 1–5 using the MTT assay (Table 2, Supplementary Fig. S19). 1 (1.6 µM) exhibited the highest potency, followed by 5 (2.5 µM) and 4 (8.1 µM). By contrast, the cysteinebound analogs 2 and 3 showed lower cytotoxicity (96 and 78 µM, respectively). Staurosporine and paclitaxel were included as positive controls. Compared with these standards, 1 showed lower cytotoxicity (30-fold lower than staurosporine and 170-fold lower than paclitaxel). Four clinically isolated Gram-positive bacterial strains— methicillin-susceptible S. aureus (MSSA), methicillinresistant S. aureus (MRSA), vancomycin-resistant Enterococci (VRE), and Clostridioides difficile—were used to assess antibacterial efficacy (Supplementary Tables S2–S6). In our preliminary tests, 3 was found to exhibit no antimicrobial activity (data not shown). Therefore, 2, which is structurally similar to 3, was also expected to lack antibacterial activity. Thus, we evaluated 1 further, comparing it with four commonly used antibiotics [vancomycin (VAN), fidaxomicin (FDX), tigecycline (TGC), and mupirocin (MUP)] and structurally similar phenazine-1-carboxylic acid (PCA) as a control. We determined the lowest minimum inhibitory concentration at which at least 50% of the isolates in a test population are inhibited (MIC50) [30] for each bacterial species (Table 3). Remarkably, overall 1 showed relatively high antibacterial activity (0.06–0.12 mg L−1), while PCA showed no activity at the concentration employed 549 (8 mg L−1). Given its relatively low cytotoxicity and potent antibacterial effects, 1 is a promising drug candidate for the oral or injectable treatment of various infections. If any issues regarding the metabolism or excretion of 1 arise, it may alternatively be evaluated as a candidate topical drug. Comparing the antibacterial activity with the structurally similar dehydrophenazines, griseolutein B from Streptomyces griseoluteus P-37 was reported to possess an MIC50 of 0.2–0.4 µg mL−1 against Gram-positive Streptococcus pyogenes and Bacillus anthracis [39]. Additionally, compared with the 5,10-dehydrophenazines with the N-hydroxyacetyl group from Kitasatospora sp. HKI714 reported by Heine et al. showed a high MIC50 (100 µg mL−1) comparable to that of PCA for methicillin-resistant S. aureus SG 511 [40]. These results suggest that additional functional groups (6-hydroxymethyl and/or 9-methoxy groups) in 1 may play important roles in the antibacterial activity. Notably, 1 exhibited higher growth inhibition activity (0.12 mg L−1) than commercially available antibiotics, such as VAN, TGC, and MUP, against both MSSA and MRSA, which are significant pathogens in bloodstream infections and opportunistic epidermal infections [41]. C. difficile causes severe enteric infections, and FDX is the main antibiotic used for treatment. VRE is a common cause of bloodstream and surgical site infections, and it is noteworthy that 1 exhibited greater efficacy against VRE than VAN, FDX, and MUP. Although VAN remains an effective agent against multidrug-resistant MRSA and C. difficile infections, the emergence of diverse VRE strains worldwide poses a serious threat to human health [42]. Since 1 surpasses VAN in antibacterial potency against both MRSA and C. difficile, it may offer a promising novel therapeutic option. FDX is a well-established drug to treat infections caused by C. difficile [43]. In comparison to FDX, 1 exhibited slightly weaker antibacterial activity against C. difficile, but the MIC50 of 1 indicates sufficient clinical efficacy. In cases of co-infection with multiple pathogens, a combination of antimicrobial agents may be preferable to using 1 alone. Similar to TGC, 1 exhibited broad efficacy against multidrug-resistant pathogens, albeit with species-specific variations. While TGC is administered intravenously, its use in bacteremia is limited because of its extensive tissue distribution and low blood concentration [44]. Future investigations should explore whether the distribution of 1 is selective and organ-specific, potentially positioning it as an alternative to TGC for the treatment of bacteremia. Data availability All data generated or analyzed during this study are included in this published article and its Supplementary Information files, which are available at the Journal of Antibiotics website. 550 Acknowledgements We are thankful to Rika Izumi for conducting the antimicrobial activity tests. We thank Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript. Author contributions SJK, SA, HS, HO, YA, and MO designed the research, and analyzed and interpreted data. SA, HO, YA, KK, and MO supervised the study. SJK performed screening of active compounds and antibacterial tests. HS performed antibacterial tests. MO isolated the compounds, conducted cytotoxicity tests, and determined the structure. SJK prepared the initial draft of the manuscript, and SA wrote the main text based on the draft. SA and MO prepared figures and tables. HS, HO, YA, KK, and MO revised the manuscript. Compliance with ethical standards Conflict of interest The authors declare no competing interests. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/. References 1. Hutchings MI, Truman AW, Wilkinson B. Antibiotics: past, present and future. Curr Opin Microbiol. 2019;51:72–80. 2. Belknap KC, Park CJ, Barth BM, Andam CP. Genome mining of biosynthetic and chemotherapeutic gene clusters in Streptomyces bacteria. Sci Rep. 2020;10:2003. 3. Navarro-Munoz JC, Selem-Mojica N, Mullowney MW, Kautsar SA, Tryon JH, Parkinson EI, et al. A computational framework to explore large-scale biosynthetic diversity. Nat Chem Biol. 2020;16:60–8. 4. van der Heul HU, Bilyk BL, McDowall KJ, Seipke RF, van Wezel GP. Regulation of antibiotic production in Actinobacteria: new perspectives from the post-genomic era. Nat Prod Rep. 2018;35: 575–604. 5. Onaka H, Mori Y, Igarashi Y, Furumai T. Mycolic acidcontaining bacteria induce natural-product biosynthesis in Streptomyces species. Appl Environ Microbiol. 2011;77:400–6. 6. Traxler MF, Watrous JD, Alexandrov T, Dorrestein PC, Kolter R. Interspecies interactions stimulate diversification of the Streptomyces coelicolor secreted metabolome. mBio. 2013;4:e00459–13. 7. Derewacz DK, Covington BC, McLean JA, Bachmann BO. Mapping microbial response metabolomes for induced natural product discovery. ACS Chem Biol. 2015;10:1998–2006. 8. Adnani N, Chevrette MG, Adibhatla SN, Zhang F, Yu Q, Braun DR, et al. Coculture of marine invertebrate-associated bacteria and interdisciplinary technologies enable biosynthesis and discovery of a new antibiotic, keyicin. ACS Chem Biol. 2017;12:3093–102. 9. Lee N, Kim W, Chung J, Lee Y, Cho S, Jang KS, et al. Iron competition triggers antibiotic biosynthesis in Streptomyces S.-J. Kawai et al. coelicolor during coculture with Myxococcus xanthus. ISME J. 2020;14:1111–24. 10. Hoshino S, Onaka H, Abe I. Activation of silent biosynthetic pathways and discovery of novel secondary metabolites in actinomycetes by co-culture with mycolic acid-containing bacteria. J Ind Microbiol Biotechnol. 2019;46:363–74. 11. Asamizu S, Pramana AAC, Kawai SJ, Arakawa Y, Onaka H. Comparative metabolomics reveals a bifunctional antibacterial conjugate from combined-culture of Streptomyces hygroscopicus HOK021 and Tsukamurella pulmonis TP-B0596. ACS Chem Biol. 2022;17:2664–72. 12. Hagihara R, Katsuyama Y, Sugai Y, Onaka H, Ohnishi Y. Novel desferrioxamine derivatives synthesized using the secondary metabolism-specific nitrous acid biosynthetic pathway in Streptomyces davawensis. J Antibiot. 2018;71:911–9. 13. Ozaki T, Sugiyama R, Shimomura M, Nishimura S, Asamizu S, Katsuyama Y, et al. Identification of the common biosynthetic gene cluster for both antimicrobial streptoaminals and antifungal 5-alkyl1,2,3,4-tetrahydroquinolines. Org Biomol Chem. 2019;17:2370–8. 14. Hoshino S, Okada M, Wakimoto T, Zhang H, Hayashi F, Onaka H, et al. Niizalactams A-C, multicyclic macrolactams isolated from combined culture of Streptomyces with mycolic acidcontaining bacterium. J Nat Prod. 2015;78:3011–7. 15. Hoshino S, Okada M, Awakawa T, Asamizu S, Onaka H, Abe I. Mycolic acid containing bacterium stimulates tandem cyclization of polyene macrolactam in a lake sediment derived rare actinomycete. Org Lett. 2017;19:4992–5. 16. Hoshino S, Ozeki M, Wong CP, Zhang H, Hayashi F, Awakawa T, et al. Mirilactams C-E, novel polycyclic macrolactams isolated from combined-culture of Actinosynnema mirum NBRC 14064 and mycolic acid-containing bacterium. Chem Pharm Bull. 2018;66:660–7. 17. Asamizu S, Ozaki T, Teramoto K, Satoh K, Onaka H. Killing of mycolic acid-containing bacteria aborted induction of antibiotic production by Streptomyces in combined-culture. PLoS ONE. 2015;10:e0142372. 18. Yanagisawa M, Asamizu S, Satoh K, Oono Y, Onaka H. Effects of carbon ion beam-induced mutagenesis for the screening of RED production-deficient mutants of Streptomyces coelicolor JCM4020. PLoS ONE. 2022;17:e0270379. 19. Lei YK, Asamizu S, Ishizuka T, Onaka H. Regulation of multidrug efflux pumps by TetR family transcriptional repressor negatively affects secondary metabolism in Streptomyces coelicolor A3(2). Appl Environ Microb. 2023;89:e0182222. 20. Lei Y, Onaka H, Asamizu S. Transcriptionally induced nucleoidassociated protein-like ccr1 in combined-culture serves as a global effector of Streptomyces secondary metabolism. Front Microbiol. 2024;15:1422977. 21. O’Neill J. Tackling drug-resistant infections globally: final report and recommendations. Rev Antimicrob Resist. London, United Kingdom: Wellcome Trust; 2016; p. 1–80. 22. WHO global action plan on antimicrobial resistance. Geneva, Switzerland: World Health Organization; 2016. 23. Kato M, Asamizu S, Onaka H. Intimate relationships among actinomycetes and mycolic acid-containing bacteria. Sci Rep. 2022;12:7222. 24. Sugiyama R, Nishimura S, Ozaki T, Asamizu S, Onaka H, Kakeya H. 5-Alkyl-1,2,3,4-tetrahydroquinolines, new membrane-interacting lipophilic metabolites produced by combined culture of Streptomyces nigrescens and Tsukamurella pulmonis. Org Lett. 2015; 17:1918–21. 25. Sugiyama R, Nishimura S, Ozaki T, Asamizu S, Onaka H, Kakeya H. Discovery and total synthesis of streptoaminals: antimicrobial [5,5]-spirohemiaminals from the combined-culture of Streptomyces nigrescens and Tsukamurella pulmonis. Angew Chem Int Ed Engl. 2016;55:10278–82. Griseolutein T from Streptomyces seoulensis, newly identified via combined-culture with Tsukamurellay 26. Kusumoto M, Ogura Y, Gotoh Y, Iwata T, Hayashi T, Akiba M. Colistin-resistant mcr-1-positive pathogenic Escherichia coli in Swine, Japan, 2007-2014. Emerg Infect Dis. 2016;22:1315–7. 27. Nakamura S. Structure of griseolutein B. J Antibiot. 1959;12:26–7. 28. Yagishita K. Production of phenazine compounds by Stretomyces griseoluteus. J Antibiot Ser A. 1960;13:83–96. 29. Nakamura S, Wang EL, Murase M, Maeda K, Umezawa H. Structure of griseolutein A. J Antibiot 1959;12:55–8. 30. CLSI. Performance standards for antimicrobial susceptibility testing. 26th Edition. Wayne, PA: Clinical and Laboratory Standards Institute; 2016. 31. Charbon G, Klitgaard RN, Liboriussen CD, Thulstrup PW, Maffioli SI, Donadio S, et al. Iron chelation increases the tolerance of Escherichia coli to hyper-replication stress. Sci Rep. 2018;8:10550. 32. Nakamura S. Studies on structure of griseolutein-B, a Streptomyces antibiotic. III. The complete structure. Chem Pharm Bull. 1958;6:547–50. 33. Nakamura S, Maeda K, Umezawa H. The structure of griseolutein B. J Antibiot. 1964;17:33–6. 34. Challand SR, Herbert RB, Holliman FG. A new phenazine synthesis. The synthesis of griseoluteic acid, griseolutein A, and methyl diacetylgriseolutein B. J Chem Soc D Chem Commun. 1970:1423–5. 35. Price-Whelan A, Dietrich LE, Newman DK. Rethinking ‘secondary’ metabolism: physiological roles for phenazine antibiotics. Nat Chem Biol. 2006;2:71–8. 36. Vind K, Maffioli S, Fernandez Ciruelos B, Waschulin V, Brunati C, Simone M, et al. N-acetyl-cysteinylated streptophenazines from Streptomyces. J Nat Prod. 2022;85:1239–47. 551 37. Newton GL, Buchmeier N, Fahey RC. Biosynthesis and functions of mycothiol, the unique protective thiol of Actinobacteria. Microbiol Mol Biol Rev. 2008;72:471–94. 38. Nakamura S. Studies on structure of Griseolutein B, a Streptomyces antibiotic. I. Characterization and degradation. Chem. Pharm. Bull. 1958;6:539–43. 39. Ogata Y. Biological studies on the antibiotics produced by Streptomyces griseoluteus. II. The effect of griseolutein B in vitro and in vivo. Jpn J Med Sci Biol. 1953;6:493–501. 40. Heine D, Martin K, Hertweck C. Genomics-guided discovery of endophenazines from Kitasatospora sp. HKI 714. J Nat Prod. 2014;77:1083–7. 41. Tong SY, Davis JS, Eichenberger E, Holland TL, Fowler VG Jr. Staphylococcus aureus infections: epidemiology, pathophysiology, clinical manifestations, and management. Clin Microbiol Rev. 2015;28:603–61. 42. Ahmed MO, Baptiste KE. Vancomycin-resistant enterococci: a review of antimicrobial resistance mechanisms and perspectives of human and animal health. Microb Drug Resist. 2018;24: 590–606. 43. Johnson AP, Wilcox MH. Fidaxomicin: a new option for the treatment of Clostridium difficile infection. J Antimicrob Chemother. 2012;67:2788–92. 44. Yaghoubi S, Zekiy AO, Krutova M, Gholami M, Kouhsari E, Sholeh M, et al. Tigecycline antibacterial activity, clinical effectiveness, and mechanisms and epidemiology of resistance: narrative review. Eur J Clin Microbiol Infect Dis. 2022;41: 1003–22.