Three naphthomycin type polyketide compounds isolated from strain Streptomyces sp. HKIB0008

1 Introduction

Ansamycins constitute a distinctive class of antibiotics produced predominantly by Streptomyces species, and they exert their antimicrobial activity by interfering with bacterial RNA synthesis [1]. Based on their structural features, ansamycins are generally classified into three major groups: benzenoid, naphthalenoid, and related derivatives [2]. Among these, naphthalenoid ansamycins have been extensively studied due to their wide distribution and structural diversity. They are produced by various bacterial strains isolated from different ecological niches [2], including soil (Streptomyces sp., Amycolatopsis sp., Micromonospora sp.), marine and freshwater environments (Streptomyces sp., Salinispora sp., Micromonospora sp.), as well as higher plants such as Manglietia hookeri [3] (Streptomyces sp. CS). The biosynthesis of naphthalenoid ansamycins involves the polyketide pathway [4], initiated from the starter unit 3-amino-5-hydroxybenzoic acid, and subsequently elongated by successive condensations of acetate and propionate units catalyzed by type I polyketide synthases. Within this group, naphthomycins represent a representative family, characterized by a C23 "ansa" chain bridging a naphthoquinone core [2]. This distinctive structural framework clearly differentiates them from benzenoid ansamycins, exemplified by rifamycins [3]. In terms of biological activity, naphthomycins exhibit pronounced inhibitory effects against various Gram-positive bacteria (e.g., Staphylococcus aureus, Bacillus subtilis), whereas their activity against Gram-negative bacteria is comparatively weaker [5]. They also demonstrate moderate antifungal activity (e.g., against Canidia Albicans and Aspergillus spp.) [6]. Certain members of this class display cytotoxic properties capable of suppressing tumor cell proliferation [7], while a few compounds additionally exhibit immunomodulatory and antiparasitic activities [8].

Literature surveys have revealed five distinct carbon skeleton variations within the naphthomycin family to date [3–16] (Fig. 1). Most members are differentiated by substitutions at the C-30 and C-2 positions, while a smaller subset exhibits additional ring closures at alternative sites. In certain cases, cleavage of the C–C bond at C-24 transforms the macrocyclic framework into a linear chain. Notably, the carbon–carbon architecture of the naphthoquinone core generally remains conserved.

Open image in new window

In our work, naphthomycins R-T were successfully isolated and identified from Streptomyces sp. HKIB0008. Within the naphthomycin family, the carbon skeleton of naphthomycin R was identified for the first time, and its biosynthetic pathway was subsequently proposed. The minimum inhibitory concentrations (MICs) of the three new compounds were determined to evaluate their antimicrobial activities, and they all showed weak antibacterial activity.

2 Results and discussion

2.1 Planar configuration elucidation of compounds 1–3

Compound 1 was obtained as a yellow oily substance, readily soluble in acetone, methanol, acetonitrile, and ethyl acetate. Its specific optical rotation was determined as [α] –10.20 (c 0.10, MeOH). The molecular formula was established as C₄₀H₄₉NO₁₁ by positive high-resolution electrospray ionization mass spectrometry (HR-ESI-MS) at m/z 720.3377 [M + H]⁺, (calcd. for C₄₀H₅₀NO₁₁⁺, 720.3384), corresponding to 17 degrees of unsaturation. Its ultraviolet (UV) spectrum exhibited an absorption maximum at 247 nm, indicating the existence of conjugated systems.

The ¹H NMR data (Table 1) revealed characteristic signals for a carboxylic acid proton at δ_H 11.7 (COOH, 1H, s), eleven olefinic protons at δ_H 7.12 (H-4, 1H, t, J = 11.4 Hz), δ_H 7.03 (H-27, 1H, s), δ_H 7.00 (H-3, 1H, dd, J = 1.8 Hz and J = 9.6 Hz), δ_H 6.56 (H-6 and H-13, 2H, m), δ_H 6.22 (H-5, 1H, t, J = 11.4 Hz), δ_H 5.92 (H-21, 1H, d, J = 8.4 Hz), δ_H 5.85 (H-30, 1H, s), δ_H 5.58 (H-7, 1H, dd, J = 10.2 Hz and J = 15.0 Hz), δ_H 5.54 (H-16, 1H, dd, J = 5.4 Hz and J = 15.0 Hz), δ_H 5.37 (H-17, 1H, dd, J = 5.4 Hz and J = 15.0 Hz), three oxygenated alkyl protons at δ_H 4.01 (H-15, 1H, m), δ_H 3.73 (H-9, 1H, m), δ_H 3.12 (H-19, 1H, dd, J = 1.8 Hz and 9.6 Hz), six aliphatic protons (CH, CH_₂) at δ_H 2.98 (H-10, 1H, dd, J = 3.4 Hz and J = 16.2 Hz), δ_H 2.79 (H-10, 1H, d, J = 5.4 Hz and J = 16.2 Hz), δ_H 2.65 (H-20, 1H, m), δ_H 2.41 (H-14, 1H, m), δ_H 2.36 (H-8 and H-14, 2H, m), δ_H 2.19 (H-18, 1H, m), twenty-one protons from methyl substituents at δ_H 2.32 (H-26a, 3H, s), δ_H 2.21 (H-2a, 3H, s), δ_H 1.98 (H-22a, 3H, s), δ_H 1.77 (H-12a, 3H, s), δ_H 1.17 (H-8a, 3H, d, J = 6.6 Hz), δ_H 0.95 (H-18a, 3H, d, J = 6.6 Hz), δ_H 0.87 (H-20a, 3H, d, J = 6.6 Hz). The ¹³C NMR, DEPT, and HSQC data of 1 revealed 40 carbon signals, comprising a carboxylic acid carbon at δ_C 164.5 (C-31), an amide carbonyl at δ_C 166.1 (C-1), an aliphatic ketone at δ_C 203.9 (C-11), and two aromatic ketones at δ_C 200.3 (C-23) and 196.2 (C-28). The data also indicated 20 olefinic carbons (a benzene ring, four trisubstituted double bonds, and three disubstituted double bonds), along with three alkoxy carbons (CH) at δ_C 73.2 (C-9), δ_C 72.5 (C-15), and δ_C 77.6 (C-19). The remaining 12 aliphatic carbons were assigned to three tertiary carbons, two secondary carbons, and seven primary carbons. Collectively, these features indicate that compound 1 contains seven double bonds and one phenyl ring, and that the 17 degrees of unsaturation calculated from its molecular formula account for one additional ring.

The ¹H-¹H COSY spectrum of 1 indicated two proton spin–spin systems, namely H-3/H-4/H-5/H-6/H-7/H-8 (/H3-8a) /H-9/H-10 and H13/H-14/H-15/H-16/H-17/H-18 (/H3-18a) /H-19/H-20 (/H3-20a) /H-21 (Fig. 3). The HMBC correlations (Fig. 3) between H-3 and C-1/C-2/C-2a, H3-2a and C-1/C-2/C-3, H-10 and C8/C9/C11, H3-12a and C11/C12/C13, H-13 and C-11/C-12/C-12a/C-14/C-15, H-21 and C-19/C-20/C-22/C-23, H3-22a and C-21/C-22/C-23/C-24, H3-26a and C-24/C-25/C-26/C-27/C-27a, H-27 and C-25/C-26/C-26a/C-27a/C-31a, H-27 and C-25/C-26/C-26a/C-27a, H-30 and C-28/C-29/C-30/C-31, H-COOH and C-30/C-31 collectively established the structural connectivity. The molecular formula of compound 1 suggested the presence of an unassigned NH proton signal. The relatively downfield chemical shift of C-29 (δ_C 143.7) is consistent with a carbon bonded to nitrogen, which undergoes reduced electron density and an approximate 10 ppm deshielding effect. In addition, C-1 corresponds to an amide group. Taken together, these observations confirm that the NH proton is connected to both C-29 and C-1. In summary, the planar structure of 1 is shown in Fig. 2.

Open image in new window

Open image in new window

Compound 2 was obtained as a yellow oily substance, readily soluble in acetone, methanol, acetonitrile, and ethyl acetate. Its specific optical rotation was determined as [α] + 323.60 (c 0.10, MeOH). The molecular formula was established as C₅₁H₆₇N₃O₁₃S by positive high-resolution electrospray ionization mass spectrometry (HR-ESI-MS) at m/z 962.4468 [M + H]⁺, (calcd. for C₅₁H₆₈N₃O₁₃S⁺, 962.4472), corresponding to 20 degrees of unsaturation. Its ultraviolet (UV) spectrum exhibited an absorption maximum at 247 nm, indicating the existence of conjugated systems.

The ¹H NMR data (Table 2) revealed characteristic signals for two amide proton at δ_H 8.85 (NH, 1H, s), δ_H 8.61 (NH, 1H, s), ten olefinic protons at δ_H 8.01 (H-27, 1H, s), δ_H 7.03 (H-4, 1H, t, J = 10.4 Hz), δ_H 6.97 (H-13, 1H, br t, J = 6.4 Hz), δ_H 6.88 (H-3, 1H, d, J = 11.2 Hz), δ_H 6.81 (H-6, 1H, t, J = 11.2 Hz), δ_H 6.63 (H-21, 1H, d, J = 10.4 Hz), δ_H 6.27 (H-5, 1H, t, J = 10.4 Hz), δ_H 6.06 (H-17, 1H, dd, J = 5.6 Hz and J = 15.2 Hz), δ_H 5.97 (H-16, 1H, dd, J = 5.6 Hz and J = 15.2 Hz), δ_H 5.63 (H-7, 1H, dd, J = 9.6 Hz and J = 14.4 Hz), six oxygenated alkyl protons at δ_H 4.56 (H-38, 1H, s), δ_H 4.43 (H-15, 1H, br q, J = 6.4 Hz), δ_H 4.25 (H-9, 1H, m), δ_H 3.93 (H-40, 1H, d, J = 10.4 Hz), δ_H 3.85 (H-40, 1H, d, J = 10.4 Hz), δ_H 3.51 (H-19, 1H, dd, J = 2.4 Hz and J = 9.6 Hz), fifteen alkyl protons (CH, CH₂) at δ_H 3.87 (H-36, 1H, m), δ_H 3.81 (H-36, 1H, m), δ_H 3.76 (H-33, 1H, m), δ_H 3.61 (H-32, 1H, m), δ_H 3.57 (H-33, 1H, m), δ_H 3.30 (H-10, 1H, dd, J = 2.4 Hz and J = 16.8 Hz), δ_H 3.25 (H-32, 1H, m), δ_H 2.91 (H-20, 1H, m), δ_H 2.87 (H-10, 1H, dd, J = 6.4 Hz and J = 16.8 Hz), δ_H 2.68 (H-14 and H-35, 3H, m), δ_H 2.57 (H-14, 1H, m), δ_H 2.51 (H-8, 1H, m), δ_H 2.44 (H-18, 1H, m), twenty-seven protons from methyl substituents at δ_H 2.37 (H-26a, 3H, s), δ_H 2.19 (H-22a, 3H, s), δ_H 2.05 (H-2a, 3H, s), δ_H 1.87 (H-12a, 3H, s), δ_H 1.33 (H-8a, 3H, d, J = 6.4 Hz), δ_H 1.31 (H-39a, 3H, s), δ_H 1.28 (H-39b, 3H, s), δ_H 1.19 (H-18a, 3H, d, J = 6.4 Hz), δ_H 1.04 (H-20a, 3H, d, J = 6.4 Hz). The ¹³C NMR, DEPT, and HSQC spectra of 2 revealed 51 carbon resonances, comprising a cycloalkanone at δ_C 201.8 (C-11); three aromatic ketones at δ_C 198.5 (C-23), 180.6 (C-28), and 183.3 (C-31); and three amide carbonyls at δ_C 171.5 (C-1), 172.3 (C-34), and 174.5 (C-37). In addition, the spectrum showed the presence of 20 olefinic carbons, corresponding to a benzene ring, one tetrasubstituted double bond, three trisubstituted double bonds, and three disubstituted double bonds. Five oxygenated carbons were also observed, consisting of four tertiary carbons (δ_C 72.4 for C-9, δ_C 72.8 for C-15, δ_C 78.2 for C-19, and δ_C 76.7 for C-38) along with one secondary carbon (δ_C 70.1 for C-40). The remaining 19 aliphatic carbons were accounted for by one quaternary carbon, three tertiary carbons, six secondary carbons, and nine primary carbons. Collectively, these data indicate seven double bonds and one phenyl moiety, accounting for 20 degrees of unsaturation, which implies the presence of two additional rings within the structure of compound 2.

The ¹H-¹H COSY spectrum of 2 indicated four proton spin–spin systems, namely H-3/H-4/H-5/H-6/H-7/H-8 (/H3-8a) /H-9/H-10, H13/H-14/H-15/H-16/H-17/H-18 (/H3-18a) /H-19/H-20 (/H3-20a) /H-21, H32/H33/NH (δ_H 8.85) and H35/H36/NH (δ_H 8.61) (Fig. 3). The HMBC correlations (Fig. 3) between H-3 and C-1/C-2/C-2a, H3-2a and C-1/C-2/C-3, H-10 and C8/C9/C11, H3-12a and C11/C12/C13, H-13 and C-11/C-12/C-12a/C-14/C-15, H-21 and C-19/C-20/C-22/C-23, H3-22a and C-21/C-22/C-23/C-24, H3-26a and C-24/C-25/C-26/C-27/C-27a, H-27 and C-25/C-26/C-26a/C-27a/C-31a, H-27 and C-25/C-26/C-26a/C-27a, H32 and C-30/C-33, NH (δ_H 8.85) and C-33/C-34, H-35 and C-34/C-36, NH (δ_H 8.61) and C-36/C-37, H-38 and C-37/C-39/C-40/C-39a/C-39b, H-40 and C-38/C-39/C-39a/C-39b collectively established the structural connectivity. The molecular formula of compound 2 suggested the presence of an unassigned NH proton signal. The relatively downfield chemical shift of C-29 (δ_C 143.8) is consistent with a carbon bonded to nitrogen, which undergoes reduced electron density and an approximate 10 ppm deshielding effect. In addition, C-1 corresponds to an amide group. Taken together, these observations confirm that the NH proton is connected to both C-29 and C-1. In summary, the planar structure of 2 is illustrated in Fig. 2.

Compound 3 was obtained as a yellow oily substance, readily soluble in acetone, methanol, acetonitrile, and ethyl acetate. Its optical rotation was determined as [α] + 519.20 (c 0.10, MeOH). The molecular formula was established as C₅₀H₆₅N₃O₁₃S by positive high-resolution electrospray ionization mass spectrometry (HR-ESI–MS) at m/z 948.4315 [M + H]⁺, (calcd. for C₅₀H₆₆N₃O₁₃S⁺, 948.4316), corresponding to 20 degrees of unsaturation. Compared with 2 (C₅₁H₆₇N₃O₁₃S), 3 contains one fewer carbon atom and two fewer hydrogen atoms, which is consistent with the loss of a methyl group. Analysis of the ¹H and ¹³C NMR data in combination with HSQC correlations (Table 2) showed that the chemical shift of C-2 in compound 3 is lower than that in compound 2. In the ¹H–¹H COSY spectrum (Fig. 3), H-2 (δ_H 6.19) correlates with H-3 (δ_H 7.28), and in the HMBC spectrum (Fig. 3), H-2 showed correlations with C-1 (δ_C 168.0) and C-3 (δ_C 138.2). These observations confirm that C-2 is a methine carbon lacking a methyl substituent, thereby supporting our structural inference and establishing the planar structure of 3 as shown in Fig. 2.

2.2 Stereochemical discussion of compounds 1–3

To further ascertain the stereochemical configurations of compounds 1–3, we integrated analyses of NMR coupling constants and NOESY correlations, and compared their ECD spectra with those of previously reported analogues. These complementary approaches enabled the stereochemical assignments.

Based on NMR spectroscopic analysis, the configurations of the key double bonds in compounds 1–3 were determined. The highly conjugated polyene system from H-3 to H-7 exhibited complex coupling characteristics: in addition to the observation of a large coupling constant (J ≈ 15.0 Hz) corresponding to a trans double bond, a continuous set of triplets with J ≈ 11–12 Hz indicated potential conformational constraints or a mixture of cis/trans configurations within this system. Specifically, for the Δ^{6, 7} double bond, the coupling constant between H-6 and H-7 was 15.0 Hz in compound 1 and 14.4 Hz in compounds 2 and 3, which are characteristic of trans (E) olefins, leading to its assignment as the E configuration. For the Δ^{4, 5} double bond, strong NOESY correlations between H-4 and H-5 were detected in all three compounds, indicating that these two protons reside on the same side of the double bond, thereby confirming its Z configuration. The Δ^{16, 17} double bond showed a coupling constant between H-16 and H-17 was 15.0 Hz in compound 1 and 15.2 Hz in compounds 2 and 3, and no NOESY correlation was observed between these protons, supporting its assignment as the E configuration.

Compound 1 contains chiral centers at C-8, C-9, C-15, C-18, C-19, and C-20. Analysis of the NOESY spectrum revealed a correlation between H-8a and H-9 (Fig. 4), indicating that these protons are spatially close and lie on the same face, thereby confirming the assignments of 8α and 9α configurations. The coupling constant between H-17 and H-18 is 5.6 Hz, accompanied by a corresponding NOESY correlation. Although the coupling constant between H-18 and H-19 is 9.7 Hz, H-19 and H-20 show a small coupling constant of 1.8 Hz together with a clear NOESY correlation. These data indicate that H-19 and H-20 reside on the same side, confirming the stereochemical assignments of 18α, 19α, and 20β. Similarly, the coupling constant of 5.6 Hz and the distinct NOESY correlation between H-15 and H-16 indicate that these protons are spatially proximate. Moreover, the biosynthetic gene cluster (BGC) of compound 1 is highly similar to that responsible for naphthomycin E production in Streptomyces sp. CS [17]. Specifically, within the BGC of compound 1, the ketoreductase ctg00856 located in module 7 of the natC´ gene reduces the C-15 carbonyl group, consistent with the function of the corresponding ketoreductase (KR) in the naphthomycin E BGC. Consequently, compound 1 and naphthomycin E share the same relative configuration at C-15, both exhibiting the 15α configuration.

Open image in new window

The chemical shifts and coupling constants of compound 2 at C-8, C-9, C-15, C-18, C-19, and C-20 are identical to those of compound 1, and the NOESY correlations are fully consistent between the two structures. In addition, both compounds originate from the same strain, share the same biosynthetic gene cluster, and are generated by the same polyketide synthase responsible for constructing the naphthomycin backbone. Taken together, these lines of evidence strongly indicate that 1 and 2 possess the same stereochemical configurations at the above positions.

The relative configuration of compound 3 was inferred from its proposed biosynthetic pathway, together with analyses of NMR chemical shifts and coupling constants. Structurally, its only difference from 2 is the absence of a methyl group at C-2. The ¹H and ¹³C NMR data for C-8/C-8a, C-9, C-15, C-18/C-18a, C-19, and C-20/C-20a are highly consistent with those of compound 2, indicating that these stereocenters share the same configuration. Therefore, the relative configuration of compound 3 was assigned as 8α, 9α, 15α, 18α, 19α, and 20β.

To further substantiate the above conclusions, a comparative analysis was performed using the reported electronic circular dichroism (ECD) data of known compounds. The results showed that the ECD spectra of naphthomycins B, I, J and 1–3 exhibit highly consistent profiles in terms of overall shape, inflection points, and the directions of the Cotton effects. This spectral consistency supports the similarity in their overall three-dimensional configurations, particularly with respect to the chiral macrocyclic environment. Consequently, these findings provide strong auxiliary evidence for assigning the stereochemistry of 1–3 in alignment with their known analogues, as illustrated in Fig. 5.

Open image in new window

Based on the above analysis, the macrocyclic stereochemical framework of 1–3 can be assigned to the structure shown in Fig. 6. However, the chiral centers located on the complex side chains of 2 and 3 cannot be unambiguously determined solely through NMR comparison or ECD calculations, primarily because C-38 resides in a locally symmetric structural environment, which complicates its stereochemical assignment. Therefore, future work will employ the Mosher's method to experimentally determine the absolute configuration of this chiral center.

Open image in new window

2.3 The proposed biosynthetic pathway of 1

In compound 1, the original naphthoquinone core undergoes cleavage and rearrangement, resulting in the disruption of the C–C bond between the aromatic ketone and the benzene ring and its conversion into a terminal carboxylic acid substituent, thereby constructing a new naphthomycin scaffold. Given the difficulty of cleaving the C–C bond within the naphthoquinone ring, we proposed a plausible biosynthetic pathway.

Genome sequencing of Streptomyces sp. HKIB0008 revealed a biosynthetic gene cluster of 116.7 kb. Comparative analysis with the reported [17] Streptomyces sp. CS biosynthetic gene clusters showed a sequence similarity of 71%, with 14 homologous genes shared between the two clusters, the detailed similarity comparison is provided in Table S5. Notably, Streptomyces sp. HKIB0008 biosynthetic gene cluster contains an additional Baeyer–Villiger monooxygenase gene, natQ (FAD monooxygenase) (Fig. 7). According to the literature [18], Baeyer–Villiger monooxygenases are a class of flavin-dependent monooxygenases that catalyze the conversion of ketones to esters via the Baeyer–Villiger (BV) oxidation, producing lactone intermediates. Based on this, we propose that compound 1 is derived from naphthomycin E through monooxygenase-catalyzed formation of a lactone intermediate, which subsequently undergoes hydrolysis. The proposed biosynthetic pathway is illustrated in Fig. 8.

Open image in new window

Open image in new window

2.4 Biological activities of 1–3

The antibacterial activity (Escherichia coli ATCC 25722, Bacillus subtilis ATCC 23857 and Staphylococcus aureus ATCC 25923) and antifungal activity (Saccharomyces cerevisiae AS 2.399) of the crude extract were detected by filter paper disc method. It was found that the crude extract had antibacterial and antifungal activity (Fig. S36). The crude extracts were obtained from the fermentation broths of Streptomyces sp. HKIB0008, which was cultured for 5 days in five different media (M1, M2, M3, M4, and ISP2) [21], followed by extraction with ethyl acetate, only the secondary metabolites derived from the M3 medium showed antimicrobial activity. When the minimum inhibitory concentration test was evaluated. Based on the initial results, which indicated the strongest inhibitory activity against Staphylococcus aureus, this bacterium was selected as the bacterial indicator for subsequent minimum inhibitory concentration (MIC) assays, with Saccharomyces cerevisiae serving as the fungal indicator.

The minimum inhibitory concentration of 1, 2 and 3 was tested. The minimum inhibitory concentration (MIC) results are presented in Table 3. Among them, compound 2 exhibited activity comparable to that of naphthomycin I, with both being more potent than compound 3 and naphthomycin J. The activity of compound 3 was similar to that of naphthomycin J reported in the literature (MIC ≈ 250 μg/mL), whereas compound 1 showed the weakest activity.

3 Experimental

3.1 General

Streptomyces sp. HKIB0008 is an actinomycete strain isolated from soil, fungi and insects collected from the primitive forest of Yexianggu in Jinghong City, Xishuangbanna Dai Autonomous Prefecture, Yunnan Province (100°53′42″E, 22°5′30″N). 60—80 mesh and 200—300 mesh silica gel column chromatography, Qingdao Marine Chemical Plant Branch. Dichloromethane, methanol, Yunnan Liyan Technology Co., Ltd. Chromatographic grade methanol, chromatographic grade acetonitrile, Shanghai Xingke Gaochun Solvent Co., Ltd. Chloroform-d and Pyridine-d₅, CIL deuterated reagent company. LDZF-75L Vertical High Pressure Steam Sterilizer, Shanghai Shen'an Medical Device Factory. Triple TOF 6600 + Nanoliter Liquid-Ion Mobility-Tandem Quadrupole Time-of-Flight Mass Spectrometry, AB Sciex, USA. SepaBean^® machine U rapid liquid chromatography system, Santai Technology (Changzhou) Co., Ltd. LC-20A High Performance Liquid Chromatograph, Shimadzu, Japan. Avance Ⅲ 600 MHz nuclear magnetic resonance spectrometer (Brucker, Rheinstetten), avance Ⅲ 800 MHz nuclear magnetic resonance spectrometer (Brucker, Germany).

3.2 Fermentation

HKIB0008 spores were grown on solid ISP2 media (glucose 4 g, yeast extract powder 4 g, malt extract powder 10 g, agar powder 15 g, distilled water 1 L, pH 7.2–7.4, liquid medium without agar powder). Using an agar plate as a growing medium, the strain HKIB0008 was inoculated into a 250 mL baffled Erlenmeyer flask that held 50 mL of ISP2 liquid media. For the purpose of seed culture, the flask was shaken for 24 h at 28 ℃ at 220 rpm on a rotary shaker. The seed culture was transferred into 1000 mL baffled Erlenmeyer flasks containing 300 mL of M3 media (glycerol 50 mL, corn flour 25 g, yeast extract 5 g, distilled water 1 L, pH 7.2–7.4). The flasks were then incubated for 7 days at 220 rpm and 28 ℃ on a rotary shaker, yielding 34.32 L of fermentation broth.

3.3 Extraction

The mycelium and supernatant of the culture broth were separated by centrifugation. For the supernatant, an equal volume of EtOAc was used to extract it three times and the final crude materials were dissolved in MeOH. For the mycelium, MeOH extraction was adopted for three times to get crude materials, which were then combined with crude materials from supernatant. After extraction and concentration, 35 g of crude extract was obtained.

3.4 Isolation

The crude extract was dissolved in methanol and subjected to normal-phase silica gel column chromatography (200–300 mesh), eluted with a stepwise gradient of methanol–chloroform (0, 1, 2, 3, 4, 5, 8, 10, 15, and 100% MeOH). The fractions were collected and concentrated under reduced pressure at 37 ℃, then dissolved in 20 mL of methanol and stored in 50 mL centrifuge tubes. UPLC analysis revealed that compound 3 was enriched in the 4% MeOH fraction (Fr. E), compound 2 in the 5% MeOH fraction (Fr. F), and compound 1 in the 10% MeOH fraction (Fr. H).

3.4.1 Purification

Compounds 1–3 were obtained from the MeOH eluates of fractions H (10%), F (5%), and E (4%), respectively. Each fraction was concentrated to dryness and subjected to stepwise octadecylsilyl (ODS) column chromatography with increasing concentrations of CH₃CN, affording subfractions (Fr. H2, Fr. F2, and Fr. E2) enriched in the target compounds. These subfractions were further purified by Sephadex LH-20 column chromatography using MeOH as the eluent, yielding crude extracts of 83–130 mg. The crude materials (Fr. H3, Fr. F3, and Fr. E3) were analyzed by ultra-performance liquid chromatography (UPLC) under gradient elution conditions specific to each compound, which afforded single peaks at retention times of 23.4 min for compound 1 (36–39% CH₃CN, UV 282/364 nm), 18.5 min for compound 2 (36–46% CH₃CN, UV 235/284 nm), and 27.4 min for compound 3 (30–31.5% CH₃CN, UV 235/309 nm), respectively. Final purification of these fractions was achieved by preparative HPLC, providing pure compounds 1 (25 mg), 2 (75 mg), and 3 (55 mg).

3.5 Compound characterization

Naphthomycin R (1): Yellow oily substance. [α]_D²⁵ –10.20 (c 0.511, MeOH). IR (KBr): υ_max 3428, 2967, 1671, 1458, 1382 and 1206 cm⁻¹. UV (CH₃OH) λ_max (log ε) 247 (0.8) nm. ¹H NMR (600 MHz in CDCl₃) and ¹³C NMR (150 MHz in CDCl₃) data, see Table 1. HRESIMS m/z 720.3377 [M + H]⁺, (calcd. for C₄₀H₅₀NO₁₁⁺, 720.3384).

Naphthomycin S (2): Yellow oily substance. [α]_D²⁵ + 323.60 (c 0.511, MeOH). IR (KBr): υ_max 3401,2929, 1653, 1417, 1336 and 1042 cm⁻¹. UV (CH₃OH) λ_max (log ε) 247 (0.8) nm. ¹H NMR (800 MHz in C₃D₃N) and ¹³C NMR (200 MHz in C₃D₃N) data, see Table 2. HRESIMS m/z 962.4468 [M + H]⁺, (calcd. for C₅₁H₆₈N₃O₁₃S⁺, 962.4472).

Naphthomycin T (3): Yellow oily substance. [α]_D²⁵ + 519.20 (c 0.511, MeOH). IR (KBr): υ_max 3400, 2929, 1654, 1449, 1337 and 1047 cm⁻¹. UV (CH₃OH) λ_max (log ε) 247 (0.8) nm. ¹H NMR (800 MHz in C₃D₃N) and ¹³C NMR (200 MHz in C₃D₃N) data, see Table 2. HRESIMS m/z 948.4315 [M + H]⁺, (calcd. for C₅₁H₆₈N₃O₁₃S⁺, 948.4316).

3.6 Filter paper disc diffusion method

The antibacterial and antifungal activity of the crude extract of HKIB0008 strain was determined by filter paper disk diffusion method [19]. LB agar plates (composition: tryptone 10 g, yeast extract 5 g, NaCl 10 g, agar 15 g, distilled water 1 L) were inoculated with Escherichia coli ATCC 25722, Bacillus subtilis ATCC 23857, and Staphylococcus aureus ATCC 25923. Similarly, YPD agar plates (tryptone 20 g, yeast extract 10 g, glucose 20 g, agar 15 g, distilled water 1 L) were inoculated with Saccharomyces cerevisiae AS 2.399. A total of 20 μL of each crude extract was applied dropwise to standard 8 mm sterile paper discs. After the solvent evaporated, the discs were aseptically placed onto the surface of the inoculated agar plates. All plates were incubated at 28 ℃. The incubation periods were 8–12 h for bacterial strains and 15–18 h for yeast. Following incubation, the diameters of the inhibition zones were measured, and these data were used to select the optimal culture medium and conditions for large-scale fermentation.

3.7 96-Well plate method

The MIC values of naphthomycins R-T were determined by the microbroth dilution method (96-well plate method) [20]. The minimum inhibitory concentration of the known compound naphthomycin J [15] was 250 µg/mL, and the gradient dilution was set based on this. Naphthomycin I, naphthomycin J reported in the literature and naphthomycins R-T were added to DMSO solution at a concentration of 1 mg/mL. Staphylococcus aureus and yeast cultures were diluted 1000 times. The assay was performed as follows: 5 µL of each compound stock solution was added to 195 µL of the standardized inoculum in the first well of a row, resulting in a 2 × final concentration. Then, a two-fold serial dilution was performed across the plate. Specifically, 100 µL was transferred from the first well to the second well containing 100 µL of fresh inoculum, mixed thoroughly, and this process was repeated sequentially through to the seventh well. From the seventh well, 100 µL was discarded, leaving the eighth well as a negative control (inoculum only). A well containing only medium served as the sterility control. The 96-well plates were incubated in an incubator at 28 ℃. Staphylococcus aureus was incubated for 8–12 h, and yeast was incubated for 15–18 h. The MIC value was determined by colorimetry.

4 Conclusion and discussion

Three new members of the naphthomycin family secondary metabolites were isolated and identified from Streptomyces sp. HKIB0008, including a novel skeleton compound, naphthomycin R, and two new analogs, naphthomycin S and naphthomycin T. Genome sequencing of Streptomyces sp. HKIB0008 revealed a 116.7 kb biosynthetic gene cluster. Comparative analysis with previously reported naphthomycin gene clusters identified a possible Baeyer–Villiger monooxygenase gene, natQ´ (FAD-dependent monooxygenase). Based on the known biosynthetic pathway of naphthomycin E, a putative biosynthetic route for naphthomycin R was proposed. Finally, the minimum inhibitory concentrations (MIC) of compounds 1–3 were determined. The decrease antimicrobial activity of 1 indicated that the ring-opening might be a detoxification result, since the production of 1 is only observed in the late fermentation stage. Compared to the known cysteine modifications on the naphthalenoid core observed in naphthomycins I and J, compounds 2 and 3 feature more complex side chains. Meanwhile, the side chain decoration on 2 and 3 are still unknown from the perspective of biofunction and biosynthesis pathway.

Notes

Acknowledgements

The authors thank Wu-Lai Jihuo, Cui-Xuan Mei, and Pei-Dong Shi for their assistance in sample collection and strain purification.

Author contributions

Yuan Chen conducted the experiments, processed the data, and wrote the original draft. Mingming Cao supervised and guided the experiments. Fei Zhang reviewed the manuscript for errors. Qian Zhao, Zi-Kang Zhao, and Lin-Fang Tang verified the spectral data of the compounds. Yu-Xin Sun, Mi-Mi Tian, and Xing-Tao Li assisted with the experimental procedures.

Funding

This research was supported financially by grants from the Key Research and Development Project of Yunnan Province (202003AD150012, China), the National Science Foundation of China (22177050), the Xingdian Talent Support Program of Yunnan Province (XDYC-QNRC-2022-0774), Yunnan Provincial Science and Technology Department (202301AT070323), and Youth Innovation Promotion Association CAS to (2022395).

Data availability

The data supporting the findings (mass spectrometry, infrared, ultraviolet, electronic circular dichroism ECD, and NMR spectra) were provided in the Supplementary Information.

Declarations

Competing interests

The authors declare no competing financial interests.