Cebulactam A₃, a Macrolactam from Marine-Derived Saccharopolyspora sp. PG10, and Its Antibacterial Activity

Abstract

Chemical analysis of the marine-derived Saccharopolyspora sp. PG10 led to the isolation of a novel macrolactam, cebulactam A₃ (1), along with four known congeners, cebulactams A₁ and A₂ (2 and 3) and shengliangmycins B and D (4 and 5). The structure of 1 was established by high-resolution mass spectrometry (HRMS) and comprehensive nuclear magnetic resonance (NMR) analyses, and its absolute configuration was determined using Mosher’s method. Genome analysis identified a putative biosynthetic gene cluster consistent with a hybrid polyketide pathway. Antimicrobial evaluation revealed that shengliangmycin B exhibited the strongest activity, whereas cebulactam analogs exhibited weaker effects. These findings expand the structural diversity of cebulactam-type macrolactams and provide insights into their stereochemical variation.

1. Introduction

Marine-derived microorganisms have emerged as an important source of structurally diverse secondary metabolites, many of which exhibit biological activities not readily observed in compounds from terrestrial origins [1,2]. The physicochemical constraints of marine environments, including osmotic stress, nutrient limitation, and intense ecological competition, are thought to drive the evolution of specialized metabolic pathways [3]. Consequently, marine-derived actinomycetes frequently produce polyketides and peptide-derived metabolites with unusual scaffolds, oxidative patterns, and stereochemical complexity [4,5,6]. Although substantial progress has been made over the past decades, the chemical space represented by marine actinomycetes remains incompletely explored, particularly at the level of strain-specific metabolite diversity [7,8].

Among marine actinomycetes, members of the genus Saccharopolyspora have been recognized as prolific producers of bioactive natural products. These species produce a wide range of secondary metabolites, including macrolides, polyketides, and hybrid structures, many of which exhibit antibacterial or cytotoxic activities [9,10,11]. Notably, Saccharopolyspora-derived compounds are often derived from modular polyketide synthase (PKS) systems and exhibit extensive post-assembly tailoring, resulting in structurally complex frameworks [12,13]. However, compared with other well-studied actinomycete genera, such as Streptomyces, the metabolic output and biosynthetic potential of marine-derived Saccharopolyspora strains remain relatively unknown [11]. Hence, targeted exploration of this genus may yield novel chemical scaffolds or structural variants of known metabolite families.

Macrolactams constitute a distinct class of nitrogen-containing macrocyclic natural products that are frequently assembled through hybrid polyketide synthase–nonribosomal peptide synthetase (PKS–NRPS) or PKS-like biosynthetic pathways. These compounds are characterized by large ring systems incorporating amide functionalities, often appended with aromatic moieties and multiple oxygenated stereocenters [14,15]. They exhibit a range of biological activities, including antibacterial and antiproliferative effects, and have attracted interest as potential leads for drug discovery [16]. Structurally, macrolactams exhibit significant diversity in ring size, substitution pattern, and stereochemistry, reflecting the flexibility of their biosynthetic machinery. In particular, subtle changes in stereochemistry or oxidation state can give rise to distinct analogs within a single metabolite family.

The cebulactam family represents a group of macrolactam natural products derived from Saccharopolyspora species [17]. These compounds are biosynthetically related to ansamycin-type scaffolds and are thought to originate from 3-amino-5-hydroxybenzoic acid (AHBA) starter units extended by polyketide elongation [18,19]. Prior studies have identified several cebulactam analogs with closely related core structures but different substitution patterns and stereochemical features. However, the structural diversity within this family remains limited. Moreover, the relationship between stereochemical variation and biological activity has not been fully explored. In particular, epimeric variation at specific stereocenters has not been extensively documented, leaving open questions about the origin and functional effects of such diversity.

Integration of genomic information has become an important approach in natural product research, particularly for linking chemical structures to their biosynthetic origins [20]. In macrolactam-producing actinomycetes, biosynthetic gene clusters (BGCs) typically encode modular PKS systems and enzymes responsible for the formation of AHBA-derived starter units. Such information can support biosynthetic proposals and help rationalize structural variations within related compound families.

As part of our efforts to explore metabolite diversity in marine-derived actinomycetes, we investigated a Saccharopolyspora strain (PG10) isolated from marine sediment in Busan, Republic of Korea. Chemical analysis led to the isolation of a previously unreported macrolactam, cebulactam A₃ (1), together with four known congeners (Figure 1, Table S1). Structural and stereochemical analyses revealed that 1 is a C-7 epimer of a previously reported analog, and genome analysis identified a putative BGC consistent with its assembly.

2. Results and Discussion

2.1. Isolation of Strains Using Diverse Culture Media

In total, 21 strains were isolated from marine sediment using five different media under two antibiotic conditions (Table S2). Among these, strains obtained under antibiotic set A were more abundant, with TWYE medium yielding the highest number of isolates. Of the isolated strains, only strains PG1–PG11 were successfully cultivated in liquid medium and preserved for further analysis. LC/MS-based screening of these cultures identified strain PG10 as the most promising candidate for chemical analysis. Phylogenetic analysis based on 16S rRNA gene sequencing subsequently identified this strain as a member of the genus Saccharopolyspora.

2.2. Structural Elucidation

Cebulactam A₃ (1) was isolated as a brown powder. Its molecular formula was established as C₁₉H₂₃NO₅ from HR-ESI-MS m/z 346.1644 [M + H]⁺ (calculated for C₁₉H₂₄NO₅, m/z 346.1649), indicating nine degrees of unsaturation. The structure elucidation was carried out using 1D and 2D NMR techniques, including correlation spectroscopy (COSY), heteronuclear single quantum correlation (HSQC), heteronuclear multiple bond correlation (HMBC), and rotating-frame Overhauser effect spectroscopy (ROESY), along with infrared spectroscopy (IR) and mass spectrometry (MS).

The ¹H and ¹³C NMR spectra (

Table 1. ¹H and ¹³C NMR spectroscopic data of cebulactam A₃ (1) in DMSO-d₆.

Position	Cebulactam A3 (1)
	δC, Type	δH, Mult (J in Hz)
NH	-	8.87, s
1	169.8, C	-
2	48.4, CH	3.75, q (6.5)
2-Me	17.1, CH3	1.10, d (6.5)
3	209.2, C	-
4	45.5, CH	3.79, m
4-Me	19.5, CH3	1.06, d (6.5)
5	130.0, CH	5.33, dd (9.0, 1.0)
6	135.1, C	-
6-Me	24.7, CH3	1.77, d (1.0)
7	80.9, CH	4.16, d (11.0)
8	38.7, CH	2.05, m
8-Me	13.9, CH3	1.05, d (6.5)
9	69.1, CH	4.20, m
10	129.1, C	-
11	140.1, C	-
12	125.2, C	-
13	112.8, CH	6.36, d (3.0)
14	150.3, C	-
15	111.7, CH	6.82, d (3.0)

¹H and ¹³C data were recorded at 700 and 175 MHz, respectively.

) indicated the presence of a highly functionalized macrolactam framework bearing an aromatic moiety. The ¹³C NMR spectrum exhibited signals for two carbonyl carbons, namely an amide carbonyl (δ_C 169.8, C-1) and a ketone carbonyl (δ_C 209.2, C-3). The aromatic region exhibited signals characteristic of a 1,3,4,5-tetrasubstituted benzene ring, supported by two proton signals at δ_H 6.36 (d, J = 3.0 Hz, H-13) and 6.82 (d, J = 3.0 Hz, H-15). In the olefinic region, one olefinic proton and two olefinic carbons (δ_C 130.0 and 135.1) were observed, indicating the presence of a double bond. In addition, four methyl groups and two oxygenated methines suggested a highly substituted and branched carbon skeleton.

Detailed analysis of the COSY spectrum established three substructures: fragment I (C-2/2-Me), fragment II (4-Me/C-4/C-5), and fragment III (C-7/C-8/C-9/8-Me) (Figure 2A). These fragments were subsequently assembled through key HMBCs. Correlations from 2-Me (δ_H 1.10) to C-1 (δ_C 169.8) and C-3 (δ_C 209.2) established a methylmalonyl moiety, which was extended to C-5 (δ_C 130.0) based on HMBC correlations from both 2-Me and 4-Me (δ_H 1.06) to C-3. Further extension of the carbon framework to C-9 (δ_C 69.1) was achieved by incorporation of fragment III, supported by HMBC correlations from 6-Me (δ_H 1.77) to C-5 and C-7 (δ_C 80.9) and from H-7 (δ_H 4.16) to C-5. The side chain was connected to a tetrasubstituted aromatic ring at C-10 (δ_C 129.1), as evidenced by HMBC correlations from H-8 (δ_H 2.05) to C-10, H-9 (δ_H 4.20) to C-11 (δ_C 140.1), and H-15 (δ_H 6.82) to C-9 (δ_C 69.1). Closure of the 13-membered macrolactam ring was established by HMBC correlations from the amide proton (δ_H 8.87) to C-1, C-12 (δ_C 125.2), and C-13 (δ_C 112.8). Eight of the nine degrees of unsaturation were accounted for by two carbonyl groups, one benzene ring, one olefinic bond, and the macrolactam ring. The remaining degree of unsaturation was attributed to an additional ring formed via an ether bridge between C-7 and C-11, supported by key HMBCs from H-7 to C-11. Accordingly, 1 was identified as a tricyclic macrolactam.

The geometry of the double bond was assigned as Z, supported by the ROESY correlations between 6-Me and H-5 (δ_H 5.33) and between H-7 and H-4 (δ_H 3.79). The planar structure of 1 was identical to that of cebulactam A₂ (3). However, noticeable differences in the chemical shifts around C-7 suggested that 1 is a diastereomer of 3. The relative configuration of 1 was established by comprehensive analysis of the ROESY spectrum (Figure 2B). In particular, the ROESY correlation between H-9 and 8-Me (δ_H 1.05) indicated an anti relationship between 8-Me and 9-OH. Additionally, the ROESY correlation between H-7 and H-8 suggested that these protons are syn-oriented. Furthermore, the ROESY correlation between H-8 and H-4 indicated that 8-Me and 4-Me are located on the same face of the molecule. The configuration at C-2 was tentatively assigned as S based on biosynthetic considerations. This assignment was supported by the co-produced congeners cebulactams A₁ and A₂ and shengliangmycin B, which possess the same S configuration at the corresponding C-2 position. Moreover, previous reports of structurally related congeners bearing the same C-2 configuration further support this assignment [19,21]. Collectively, the ROESY data and biosynthetic considerations supported the assignment of the relative configuration of 1 as 2S*, 4R*, 7R*, 8S*, 9S*.

The absolute configuration at C-9 was determined using the modified Mosher’s method [22,23,24]. The secondary hydroxy group was derivatized with (R)- and (S)-MTPA chloride to afford the corresponding diastereomeric Mosher esters, which were then analyzed by ¹H NMR spectroscopy. Interpretation of the Δδ_S−R values led to the assignment of the absolute configuration at C-9 as S. Combined with the aforementioned relative stereochemical analysis, the absolute configuration of 1 was assigned as 2S, 4R, 7R, 8S, 9S (Figure 3). Thus, 1 was identified as the C-7 epimer of cebulactam A₂ (3) and was designated as cebulactam A₃ (1). The distinct CD profiles of cebulactam A₃ (1) and cebulactam A₂ (3), which showed different Cotton effect patterns below 250 nm, were consistent with this epimeric relationship (Figure S58).

By comparison of their NMR spectroscopic data with those reported in the literature, compounds 2–5 were identified as cebulactam A₁ [17], cebulactam A₂ [17], shengliangmycin B [19], and shengliangmycin D [19], respectively.

Bioinformatic analysis of the draft genome of strain PG10 identified a putative cebulactam biosynthetic gene cluster (BGC) spanning approximately 60 kb, containing 40 predicted coding sequences (CDSs) (Table S3). This cluster was found to encode a hybrid type I PKS system along with enzymes potentially involved in precursor supply, tailoring, and regulation, consistent with the biosynthesis of macrolactam-type polyketides.

The PKS modules (ctg1_2182–2184) were proposed to utilize AHBA as the starter unit and methylmalonyl-CoA and malonyl-CoA as extender units to construct the macrolactam backbone 6. The presence of CDSs associated with AHBA biosynthesis further supported this assignment. Subsequently, 6 undergoes cyclization between C-2 and the 11-OH group, followed by hydroxylation, to afford 5. In parallel, the p-dihydroxybenzene moiety is oxidized to a 1,4-benzoquinone, yielding 4. Further cyclization between C-7 and C-11 via an ether bridge generates 1 and 3, which exist as a pair of epimers. Finally, 3 undergoes isomerization of the double bond to form 2 (Figure 4).

The formation of the C-7 epimeric pair can be alternatively explained by the intramolecular 1,4-addition via an allylic carbocation intermediate. First, activation of the conjugated diene generates an allylic carbocation, with the positive charge delocalized over the C-4/C-7 framework. Subsequently, the 11-OH can act as an intramolecular nucleophile and attack the planar allylic cation at C-7 from either the re- or si-face. This non-stereoselective intramolecular capture leads to the formation of two epimeric products, cebulactams A₃ (1) and A₂ (3), thereby accounting for the observed stereochemical variation at C-7.

2.3. Antimicrobial Activity Assessment

The antimicrobial activities of the isolated compounds were evaluated against five microbial strains, including the Gram-positive bacterium Bacillus subtilis, the Gram-negative bacteria Pseudomonas aeruginosa, Escherichia coli, and Erwinia rhapontici, and the fungus Candida albicans, using a broth microdilution assay. Among the tested strains, inhibitory activity was only detected against B. subtilis and P. aeruginosa, whereas no activity was detected against E. coli, Er. rhapontici, or C. albicans under the tested conditions. Among the tested compounds, shengliangmycin B exhibited the most potent activity, with IC₅₀ values of 3.37 µM against P. aeruginosa and 14.10 µM against B. subtilis, whereas shengliangmycin D exhibited no detectable activity. In contrast, cebulactam derivatives exhibited only weak to moderate activity against P. aeruginosa; the IC₅₀ values for cebulactams A₁ (2), A₂ (3), and A₃ (1) were 200.50, 218.98, and 203.07 µM, respectively (Table S4). Notably, cebulactam A₂ was the only analog that exhibited measurable activity against B. subtilis (IC₅₀ = 109.67 µM). These results indicate that shengliangmycin-type compounds are more potent than cebulactam analogs and that variation at C-7 has a limited impact on antibacterial activity within this scaffold.

3. Materials and Methods

3.1. General Experimental Procedures

Optical rotations were measured using a JASCO P-2000 polarimeter (JASCO, Hachioji, Tokyo, Japan) in a 1.0 cm cell at room temperature. IR spectra were recorded on a PerkinElmer Spectrum 400 FT-IR/FT-NIR spectrometer (PerkinElmer, Waltham, MA, USA). CD spectra were obtained with a 1 mm cell using Chirascan Plus (Applied Photophysics Ltd., Leatherhead, Surrey, UK). ¹H, ¹³C, and 2D NMR spectra were acquired on Bruker Avance II 600 and 700 MHz NMR spectrometers (Bruker, Billerica, MA, USA) at the Korea Basic Science Institute (KBSI), Ochang, Republic of Korea; MTPA ester derivatives were analyzed on the 600 MHz instrument. UV spectrometry and low-resolution electrospray ionization mass spectrometry (LR-ESI-MS) were performed using an Agilent G6125B MSD coupled to an Agilent 1260 Infinity II LC system (Agilent Technologies, Santa Clara, CA, USA) equipped with a Phenomenex Luna reversed-phase C₁₈ column (100 × 4.6 mm, 5 μm). LC/MS analysis was performed using H₂O (Daejung Chemicals & Metals, Siheung, Republic of Korea) and CH₃CN (Daejung Chemicals & Metals, Siheung, Republic of Korea) as mobile phases, each containing 0.1% formic acid (Sigma-Aldrich, St. Louis, MO, USA), at a flow rate of 0.4 mL/min. The gradient was as follows: 10% CH₃CN for 3 min, 10–100% CH₃CN over 20 min, 100% CH₃CN for 5 min. HR-ESI-MS data were recorded on an Agilent 1290 Series HPLC system coupled to an Agilent 6530 iFunnel Q-TOF mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). Microplate absorbance was measured using an Epoch microplate spectrophotometer (BioTek Instruments, Winooski, VT, USA).

3.2. Bacterial Material

3.2.1. Bacterial Isolation and Identification

Strain PG10 was isolated in August 2024 from marine sediment collected from Gwangalli, Busan, Republic of Korea (35°09′19″ N, 129°08′16″ E, depth: 13 m). The sediment samples were dried in a clean bench for 2 days. The dried sediment (2 g) was then mixed with distilled water (12 mL) at a 1:6 ratio and sonicated for 20 min. Subsequently, 100 μL of the supernatant was spread onto prepared SIM, ISP5, SCNA, TWYE, and GS agar plates (for details, see Table S5). All media were prepared using artificial seawater containing 33 g/L sea salt to approximate the salinity of natural seawater. Depending on the medium, antibiotic set A [cycloheximide (100 mg/L; Thermo Scientific, Waltham, MA, USA) and nalidixic acid (20 mg/L; Acros Organics, Geel, Belgium)] or antibiotic set B [cycloheximide (80 mg/L), gentamicin (10 mg/L; Thermo Scientific, Waltham, MA, USA), and novobiocin (10 mg/L; Thermo Scientific, Waltham, MA, USA)] was added. Strain PG10 was then purified by cultivation on TWYE agar. To identify strain PG10 at the molecular level, its 16S rRNA gene was sequenced using two independent primers: 785F and 907R. The resulting forward and reverse reads were assembled based on their overlapping region to generate a single contig of 1458 bp. BLASTN analysis [25] of the assembled sequence revealed the highest similarity to S. cebuensis strain SPE 10-1 (accession no. NR_044047.1), with 1448 identical nucleotides out of 1452 aligned positions, corresponding to 99.72% sequence identity. Thus, strain PG10 was assigned to the genus Saccharopolyspora and was most closely related to S. cebuensis (Figure S59).

3.2.2. Whole-Genome Sequencing and Assembly

Whole-genome sequencing was performed by Macrogen, Inc. (Seoul, Republic of Korea). Genomic DNA of Saccharopolyspora sp. PG10 was subjected to de novo whole-genome sequencing using a hybrid strategy combining Illumina short-read and PacBio SMRT sequencing. The genome was assembled using the Microbial Genome Analysis pipeline (SMRTLINK, v25.1.0.257715) and subsequently polished with Illumina reads. The final assembly comprised two contigs with a total length of 6.29 Mb, an N50 value of 6.24 Mb, and a GC content of 72.7%. The estimated genome size was 6.01 Mb, and the average read depth was 114.37-fold. Assembly validation metrics indicated high quality, including 99.32% Illumina read mapping with 100% coverage, 99.74% PacBio read mapping with 100% coverage, and 100% complete BUSCO recovery, supporting the high completeness and reliability of the assembled genome. Moreover, genome-based phylogenomic analysis using the TYGS platform [26] placed strain PG10 in a clade with S. cebuensis JCM 18116, indicating a close phylogenomic relationship between the two strains (Figure S60). Together with the 16S rRNA gene sequence data, these results support the assignment of strain PG10 as a member of the genus Saccharopolyspora that is most closely related to S. cebuensis. The whole-genome shotgun project has been deposited in the NCBI database under the accession JBWXRK000000000. The associated BioProject and BioSample accession numbers are PRJNA1448979 and SAMN57104874, respectively.

3.3. Cultivation and Extraction

Strain PG10 was initially cultivated in 50 mL of TSBY seawater medium (Table S5) in a 100 mL Erlenmeyer flask. After incubation at 27 °C for 3 days on a rotary shaker at 180 rpm, 3.5 mL of the seed culture was transferred into a 500 mL Erlenmeyer flask containing 150 mL of R4 seawater medium and further cultivated for 2 days under the same conditions. Subsequently, 20 mL of this culture was inoculated into 1 L of R4 seawater medium (Table S5) in a 2.5 L Ultra Yield flask and fermented at 27 °C for 5 days with shaking at 180 rpm. The entire culture broth was then extracted with ethyl acetate (EtOAc). The organic layer was separated using a 3 L separatory funnel, dried over anhydrous Na₂SO₄, and concentrated in vacuo to yield 1.7 g of dried extract.

3.4. Purification of Cebulactams and Shengliangmycins

The dried extract of strain PG10 was fractionated by reversed-phase C₁₈ open-column chromatography (YMC ODS-A-C₁₈, 50 μm) using 300 mL each of aqueous MeOH (20%, 40%, 60%, 80%, and 100%), followed by MeOH/DCM (1:1). From the 40% aqueous MeOH fraction, cebulactams A₁–A₃ and shengliangmycin D were isolated by semi-preparative reversed-phase HPLC (YMC-Pack ODS-A-C₁₈ column, 5 μm, 250 × 10 mm) using 20% aqueous CH₃CN containing 0.1% formic acid at a flow rate of 2.0 mL/min, with UV detection at 280 nm. As a result, compounds 1, 2, 3, and 5 were obtained as pure compounds with the following retention times and yields: 1 (t_R = 25 min, 15.0 mg), 2 (t_R = 22 min, 19.6 mg), 3 (t_R = 39 min, 28.3 mg), and 5 (t_R = 18 min, 16.5 mg). From the 60% aqueous MeOH fraction, shengliangmycin B was purified using the same HPLC method with 33% aqueous CH₃CN containing 0.1% formic acid, affording compound 4 as a pure compound (t_R = 30 min, 8.3 mg).

Cebulactam A₃ (1):

Brown powder; ${\left[\mathit{\alpha }\right]}_{\mathit{D}}^{25}$ −125.7 (c 0.10, MeOH); IR ν_max (ATR) 3245, 2974, 2932, 1714, 1661, 1607, 1461, 1376, 1198, 1155, 1005 cm⁻¹; UV (MeOH) λ_max (log ε) 200 (2.24), 305 (1.54) nm; HR-ESI-MS [M + H]⁺ m/z 346.1644 (calculated for C₁₉H₂₄NO₅ 346.1649) (Figures S55–S57); ¹H NMR (DMSO-d₆, 700 MHz) and ¹³C NMR (DMSO-d₆, 175 MHz) (Figures S1–S10).

3.5. MTPA Esterification of Cebulactam A₃

Two samples of cebulactam A₃ (1.5 mg each) were placed in separate 40 mL vials and lyophilized overnight. Each sample was treated with anhydrous pyridine (1.0 mL) and DMAP (1.0 mg) under nitrogen and stirred at room temperature for 5 min. R- or S-MTPA-Cl (30 µL) was added to each vial, and the mixtures were stirred at 35 °C for 4 h. Purification by reversed-phase HPLC on a YMC-Pack C₈ column (5 µm, 250 × 10.0 mm) using a stepwise CH₃CN/H₂O gradient (60–100% CH₃CN over 60 min, followed by 100% CH₃CN for 20 min; 2 mL/min; UV 280 nm) yielded the S-MTPA ester of cebulactam A₃ (1a) and the R-MTPA ester (1b), both of which eluted at 40 min. The Δδ_S−R values were determined based on the ¹H NMR and ¹H − ¹H COSY spectra.

S-MTPA ester of cebulactam A₃ (1a): ¹H NMR (600 MHz, DMSO-d₆) δ_H 9.12 (NH, s), 7.46–7.58 (12H, overlapped), 7.06 (1H, d, J = 2.0), 6.79 (1H, d, J = 2.0), 6.23 (1H, d, J = 9.5), 5.30 (1H, d, J = 8.5), 4.61 (1H, d, J = 10.0), 3.85 (2H, overlapped), 3.59 (3H, s), 3.46 (3H, s), 2.59 (1H, m), 1.70 (3H, s), 1.12 (3H, d, J = 6.5), 1.03 (3H, d, J = 6.5), 0.86 (3H, d, J = 6.5); LR-ESI-MS [M + Na]⁺ m/z 800.1.

R-MTPA ester of cebulactam A₃ (1b): ¹H NMR (600 MHz, DMSO-d₆) δ_H 9.11 (NH, s), 7.46–7.58 (16H, overlapped), 7.00 (1H, d, J = 2.0), 6.62 (1H, d, J = 2.0), 6.25 (1H, d, J = 9.5), 5.35 (1H, d, J = 9.0), 4.63 (1H, d, J = 10.0), 3.81 (1H, m), 3.78 (1H, q, J = 6.5), 3.55 (3H, s), 3.53 (3H, s), 2.68 (1H, m), 1.77 (3H, s), 1.09 (3H, d, J = 6.5), 1.01 (3H, d, J = 6.5), 1.00 (3H, d, J = 6.5); LR-ESI-MS [M + Na]⁺ m/z 800.1.

3.6. Antimicrobial Activity Assay

The Gram-positive bacterium B. subtilis (ATCC 6051) and the Gram-negative bacteria P. aeruginosa (KCTC 22073), E. coli (ATCC 11775), and Er. rhapontici (ATCC 29283) were cultured on Luria–Bertani agar (LB agar), while the fungus C. albicans (KCTC 7965) was cultured on yeast extract–peptone–dextrose agar (YPD agar). After overnight incubation at 27 °C, bacterial cells were transferred to LB broth and cultured at 30 °C for 24 h, whereas C. albicans was cultured in YPD broth under the same conditions. The harvested bacterial and fungal cells were inoculated into Mueller–Hinton broth (MHB) and YPD broth, respectively, to an initial optical density (OD) of 0.0008 at 600 nm. Test compounds were dissolved in DMSO and serially diluted two-fold with the corresponding broth to final concentrations of 200–0.4 µg/mL. The plates were then incubated at 30 °C for 18 h, and microbial growth was measured by recording the absorbance at 600 nm. Gentamicin and cycloheximide were used as reference compounds for bacteria and C. albicans, respectively, and were tested under the same conditions at final concentrations of 200–0.4 µg/mL.

4. Conclusions

In this study, a previously undescribed macrolactam, cebulactam A₃ (1), was isolated from the marine-derived Saccharopolyspora sp. PG10, together with four known congeners. The structure of 1 was elucidated by comprehensive spectroscopic analyses, and its absolute configuration was established using Mosher’s method. ROESY data identified 1 to be a C-7 epimer of cebulactam A₂, representing a rare example of stereochemical variation within the cebulactam family. Genome analysis identified a putative BGC consistent with a hybrid PKS pathway, and a plausible mechanism was proposed to account for the formation of the C-7 epimeric pair.

Antibacterial evaluation provided additional biological information on the isolated analogs, including the previously unreported antibacterial activity of shengliangmycin B under the tested conditions. Overall, these findings expand the structural diversity of cebulactam-type macrolactams and provide insights into their stereochemical and biosynthetic variability.