Various Streptomyces strains, including strain HUT6047, were cultured in YM medium (0.4% yeast extract, 1.0% malt extract, and 0.4% D-glucose, pH 7.3) at 28°C with 120 rpm (revolutions per minute) for 3 days, according to our standard protocol (Arakawa et al., 2005). The Streptomyces culture library was prepared as described below. Briefly, 100 ml of culture broth was extracted twice with EtOAc, and then the combined organic phase was dried (Na₂SO₄), filtered, and concentrated in vacuo. The residues were dissolved in MeOH (1 ml), and their aliquots (10 µl each) were collected for the bioassay.

The compounds (active component against A. coerulea polyps and synthetic MBC) were analyzed by electrospray ionization-mass spectrometry (ESI-MS) and nuclear magnetic resonance (NMR). ESI-MS was performed using an LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). NMR spectra were recorded on a JEOL ECA-600 spectrometer equipped with a field gradient accessory (JEOL, Ltd., Tokyo, Japan). The NMR chemical shifts were recorded as δ values in ppm. The coupling constants in ¹H-NMR were shown as J value in Hz. Dimethylsulfoxide-d ₆ (99.8 atom %; Kanto Chemical, Co., Inc., Tokyo, Japan) was used as the solvent for ¹H- and ¹³C-NMR, while tetramethylsilane (δ_H = 0) was used as the internal standard for ¹H-NMR and ¹³C-NMR.

Metabolite production in strain HUT6047 was analyzed using high-performance liquid chromatography (HPLC) and thin-layer chromatography (TLC). The EtOAc extract from 100-ml culture of strain HUT6047 (average 39 mg extracts from the 430-mg dry cell per 100-ml culture) was dissolved in MeOH (1 ml), and then an aliquot (10 µl) was passed through a COSMOSIL Cholester column (4.6 × 250 mm, Nacalai Tesque, Kyoto, Japan) and eluted with 40% aqueous acetonitrile containing 0.1% TFA at a flow rate of 1.0 ml/min. The eluate was monitored using a JASCO MD-2010 multi-wavelength photodiode array detector (JASCO Corporation, Tokyo, Japan), and active component was detected at 360 nm ( Figure 3 ). Purified natural MBC (see section 2.4) and synthetic MBC (see section 2.5) were also analyzed in the same manipulation. TLC analysis of the EtOAc extract of strain 6047 prepared as above-mentioned was performed using a mixture of CHCl₃ and MeOH (15:1, v/v) and exposed to iodine vapor.

The culture supernatant of strain HUT6047 (26 L) was extracted twice with an equal volume of EtOAc. The combined organic phases were dried (Na₂SO₄), filtered, and concentrated in vacuo. The crude extract was purified using Sephadex LH-20 (GE Healthcare, Chicago, IL, USA) gel filtration chromatography with MeOH. All fractions (1 ml each; total 50 fractions) eluted with MeOH were subjected to a bioassay using A. coerulea polyps (detailed protocol is described in section 2.5). The fractions containing the active component(s) were combined, and the resulting residue was further purified using silica gel chromatography with two different solvent systems, CHCl₃–MeOH = 50:1–10:1 (v/v) and hexane–EtOAc = 2:1 (v/v). All the fractions (3 ml each; total 50 fractions) were dried in vacuo, and redissolved in MeOH (1 ml) and also subjected to a bioassay described as above to obtain an active component, MBC (5.7 mg from 26-L culture broth). ¹H-NMR (DMSO-d ₆) δ = 3.84 (3H, s), 6.12 (1H, d, J = 1.5 Hz), 6.27 (1H, s), 6.75 (1H, brs), 6.91 (1H, brs), 9.30 (1H, s), 11.24 (1H, brs), 11.42 (1H, brs). ¹³C-NMR (DMSO-d ₆) δ = 57.8 (q; OCH₃), 90.9 (d), 108.2 (d), 109.3 (d), 117.3 (s), 120.4 (d), 123.4 (s), 133.2 (s), 158.6 (s), 171.6 (d).

HRMS (positive ESI): m/z calculated for C₁₀H₁₀N₂O₂Na: 213.0640 [M+Na]⁺; observed: 213.0631.

A clonal polyp strain KH1A, which was established from A. coerulea jellyfish caught in the Seto Inland Sea, Japan (Tsujita et al., 2015), was used for the bioassay. Polyps were reared in filtered seawater (FSW) at 22–25°C.

For screening and purification, strobilation was induced by lowering the culture temperature to 10°C from 22–25°C [cold shock (CS)]. Strobilae at the earlier segmentation phase with one to five segments were collected 46–60 days after the temperature drop and kept in a 14-cm dish at 10°C until transferred to 24-well microtiter plates. Three strobilae were placed in individual wells of a 24-well microtiter plate and then cultured in 1 ml of FSW containing 10 µl of the aliquots of either the culture extracts or fractions separated by chromatography at 22°C. Control strobilae were cultured in FSW at 22°C. Their strobilation was monitored over a 7-day period. For the dose–response analysis, strobilation was induced via incubation with 10 µM IM at 22°C. After a 7-day incubation period, segmentation-phase strobilae with one to three segments were collected, and then rinsed with 50 ml of FSW to remove the IM. Three to five strobilae were incubated in individual 9-cm dishes in 40 ml of FSW containing 1.6, 3.1, 6.3, 12.5, 25, or 50 µM MBC at 22°C. Control strobilae were cultured in FSW containing 0.025% DMSO at 22°C. Strobilation was monitored over a 7-day period.

Genomic DNA of strain HUT6047 was subjected to paired-end sequencing using an Illumina NextSeq sequencing system (San Diego, CA, USA) according to the manufacturer’s protocol. De novo assembly of the raw genome sequencing data was performed using SPAdes 3.13.0 (Bankevich et al., 2012), and Illumina read data were deposited as Bioproject: PRJDB15399, Biosample: SAMD00585629. The MBC biosynthetic gene cluster was identified using antiSMASH ver. 6.0.1 (Blin et al., 2021), and its sequence was also deposited (GenBank Accession number: LC760459).

Previously, we revealed that IM induces strobilation in a dose-dependent manner through a chemical library screening using the 456 substances provided by RIKEN Natural Products Depository (RIKEN NPDepo, RIKEN Advanced Science Institute, Wako, Japan) (Kuniyoshi et al., 2012). In addition, we found that the lysosomal acidification inhibitors, chloroquine and bafilomycin A1, partially inhibited strobilation (Tsujita et al., 2017). Owing to these findings, we proceeded to further investigate the unique chemical substances with notable biological activity against strobilation in A. coerulea. Recently, we independently constructed a Streptomyces culture library and performed a pilot screening to explore various biological activities, including antimicrobial, antitumor, and other activities, of these crude extracts.

Strobilation is induced by CS at 10°C, and CS-induced strobilae metamorphose into ephyrae even at 25°C (Tsujita et al., 2015). Similarly, IM-induced strobilae metamorphose into ephyrae in the absence of IM (Kuniyoshi et al., 2012). Thus, once initiated, strobilation is autonomously completed independent of temperature and chemicals. In this study, we aimed to identify substances that inhibit this autonomous process of strobilation.

To determine the effect of the Streptomyces culture library on the autonomous process of strobilation, the culture extracts were administered to earlier segmentation-phase strobilae. In the control experiments, strobilae with three segments ( Figure 1B ) became fully segmented strobilae ( Figure 1C ) after 1 day and then morphogenesis-phase strobilae ( Figure 1D ) after 2 days. Finally, the ephyrae were detached after 6 days ( Figure 1E ). Among the 38 culture extracts tested ( Table S1 ), three strains displayed remarkable inhibition of autonomous process of strobilation. Hereafter, we focused on the strain HUT6047, which exhibited the highest strobilation-inhibiting activity among the three strains.

The large-scale fermentation of strain HUT6047 was performed and its metabolites were purified using Sephadex LH20 and silica gel chromatography homogeneity ( Figure 3A ). All fractions were screened through strobilation-inhibiting activity using A. coerulea polyps. The active compound appeared as a gray green spot on TLC after iodine staining at Rf = 0.55 [CHCl₃–MeOH = 10:1 (v/v)] ( Figure 3A ), and displayed a distinct molecular ion peak at m/z 213.0631 [M+Na]⁺ (calcd. for C₁₀H₁₀N₂O₂Na, m/z 213.0640) on high-resolution ESI-MS. In the ¹³C-NMR spectrum of this compound, one methyl, five methine, and four quaternary carbons were identified. All spectral data including ¹H-NMR spectrum agreed well with the reported data for MBC ( Figure 2 ) (Dairi et al., 2006; Rastogi et al., 2013). MBC is a well-characterized biosynthetic intermediate of bipyrrole-containing natural products, including prodigiosins and tambjamines ( Figure S1 ) (Cerdeño et al., 2001; Stanley et al., 2006; Burke et al., 2007; Hu et al., 2016; Grenade et al., 2023).

To confirm its structure and prepare a considerable amount for the bioassay, MBC was synthesized from the commercially available 4-methoxy-3-pyrolin-2-one in two steps ( Supplementary Materials and Figure S2 ) according to previous reports (Dairi et al., 2006; Rastogi et al., 2013). HPLC chromatogram of both purified and synthetic MBC showed a distinct peak at 5.8 min, which was also detected in culture extract of strain HUT6047 ( Figure 3B ). The spectral data of natural MBC were in good agreement with those of synthetic MBC (¹H-NMR in Figure 3C and ESI-MS in Figure S3 ), supporting its chemical structure ( Figure 2 ).

The genetic feature of strain HUT6047 was assessed using next-generation sequencing. A biosynthetic gene cluster for MBC (mbc) was identified in the genome of strain HUT6047, and it showed a significant similarity to the biosynthetic gene cluster (tab cluster) for tambjamine BF-18591 ( Figure 2 ) (Grenade et al., 2023) as shown in Table S2 and Figure S4 , supporting the production of MBC in this strain.

The strobilation-inhibiting activity of the purified natural MBC was examined. As described above, CS-induced strobilae autonomously metamorphose into ephyrae at 22°C. When purified natural MBC was administered to strobilae with one to five segments, the autonomous process of strobilation was arrested ( Figures 1F, G ). In contrast, the strobilation proceeded until the end of 6 days in the control experiment ( Figures 1B–E ). Remarkably, abnormal tentacle-like structures were observed around these constrictions ( Figure 1F , arrowheads). These abnormal strobilae did not undergo morphogenesis stage during the 7-day observation period.

To confirm the strobilation-inhibiting activity of MBC, the synthetic MBC was subjected to an Aurelia bioassay. As shown in Figures 1H, I , the synthetic MBC reproduced the same activity as purified natural MBC: arrest of the autonomous process of strobilation and generation of abnormal tentacle-like structures. Hence, MBC was confirmed as the active component produced by strain HUT6047. Dose–response analysis using synthetic MBC revealed that the minimum activity was 6.3 µM concentration ( Table 1 ). No apparent death was observed in the bioassay. Despite the abnormal morphology, the MBC-treated strobilae remained alive for more than 1 month, even at a concentration of 50 µM ( Table 1 ), suggesting that MBC had no remarkable cytotoxic activity against A. coerulea. In our independent analysis, MBC also showed no remarkable cytotoxicity against brine shrimp, a model marine organism (unpublished results).

MBC halts the autonomous process of strobilation without causing cytotoxicity, and generates abnormal tentacle-like structures. Abnormal tentacle-like structures appeared to protrude from each constriction between strobilae segments ( Figure 1F , arrowheads). However, the developmental origin of the abnormal tentacle-like structures and their similarity with the normal tentacles of polyps remain unknown. Detailed histological observations are required to clarify these issues.

In a comparative assay, the culture extract of Streptomyces coelicolor M145, a notable producer of undecylprodigiosin, one of a prodigiosin derivative, had no effect on A. coerulea, indicating that the strobilation-inhibiting activity was caused by the bipyrrole structure and not by the accessory structures including a hydrocarbon chain and an additional pyrrole ring in undecylprodigiosin.