Heterologous Expression of Type II PKS Gene Cluster Leads to Diversified Angucyclines in Streptomyces albus J1074

Abstract

Heterologous expression has emerged as an effective strategy in activating Streptomyces cryptic gene clusters or improving yield. Eight compounds were successfully obtained by heterologous expression of the type II PKS gene cluster spi derived from marine Streptomyces sp. HDN155000 in the chassis host Streptomyces albus J1074. The structures with absolute configurations were elucidated using extensive MS and NMR spectroscopic methods, as well as theoretical NMR calculations and electronic circular dichroism (ECD) calculations. Interestingly, compound WS009 Z (2) contains a rare thiomethyl group, angumycinone T (4) has a novel oxo-bridge formed between C12a and C4, and angumycinone X (3) showed cytotoxicity toward K562 and NCI-H446/EP cell lines.

1. Introduction

Angucyclines are a significant class of aromatic polyketide compounds biosynthesized by type II polyketide synthases (PKS) via an iterative Claisen condensation to generate Poly-β-ketone backbones [1]. So far, over 400 members of angucyclines have been reported [2] and nearly half of them showed a wide range of biological activities, such as antimicrobial and/or anticancer, glutamate receptor agonist, gastric mucosal protectant, and platelet aggregation-inhibiting actions [3], which continually enhanced the importance of expanding the chemical space and obtaining novel angucycline derivatives for drug discovery and development purposes [4].

Streptomyces, a type of actinobacteria, is a Gram-positive bacterium with high GC genomes, primarily isolated from soil and marine sediments and other habitat samples [5,6,7,8], also known for its ability to produce a complex variety of secondary metabolites [9,10,11]. As genome sequencing technology advances, an increasing number of gene clusters (BGC) for natural products are being discovered [12]. However, only a limited number of natural products have been identified to date [13], making it essential to convert genomic information into valuable compounds for drug discovery and also commercial production. Heterologous expression of gene clusters is an effective strategy to address this problem by enabling the activation of cryptic gene clusters in heterologous host strains and increasing the yield of gene clusters expressed at low levels under laboratory conditions, as well as obtaining new analogues other than wild-type [14,15,16]. However, the biosynthetic capacity of an exogenous gene cluster is greatly affected by the metabolic characteristics of the chassis cells; for example, the competitive consuming of precursors in the native biosynthetic pathways will stop or drastically decrease the production of the desired compound. For this reason, it is critical to select a proper surrogate host, because different hosts influence the yield and abundance of compounds, even possibly producing entirely distinct compounds [17,18]. Thus far, various Streptomyces species including S. coelicolor, S. lividans, S. venezuelae, S. albus, S. fradiae, S. roseosporus, and S. toyocaensis have been specifically engineered and frequently employed for heterologous expression [19].

During our ongoing research into bioactive chemicals from marine-derived microorganisms, six angucycline derivatives, including one novel oxaspirocyclic structure, were obtained by heterologous expression of a type II polyketide biosynthesis gene cluster spi in Streptomyces coelicolor A3(2) [20]. In the current study, to further explore the metabolic capacity of the BGC of spi and acquire more angucycline analogues, this cluster was expressed in another feasible heterologous host, S. albus J1074. A total of six angucycline derivatives and two by-products were obtained from the heterologous strains. Here, we will describe the construction of the recombinant strain, and the characterization of the structures obtained from the culture.

2. Results

2.1. Heterologous Expression of spi BGC in S. albus J1074

A type II PKS gene cluster spi (GenBank accession number OP009365, Figure 1a), which is responsible for angucyclines synthesis, has been previously identified and confirmed from the genome of a marine-derived actinobacteria strain of Streptomyces sp. HDN155000 [20]. To develop the metabolic potential of the spi BGC and obtain more angucycline analogues, we decided to have spi expressed in S. albus J1074, which has been demonstrated as a friendly surrogate host for heterologously expressing actinobacteria sourced genes due to its faster growth, small genome size, easy genetic manipulation procedure, and clear metabolic background [21,22,23]. The empty vector p15A was transferred into S. albus J1074 as the negative control (Figure 1b, trace i), and a series of new peaks were detected by HPLC analysis (Figure 1b, trace ii), demonstrating that the spi gene cluster was successfully activated and that changing heterologous host would greatly develop the metabolic capacity of the spi BGC (Figure S2).

2.2. Isolation and Purification of Compounds from Recombinant Strain

To specifically identify the compounds, we processed a larger-scale fermentation (20 L) of the strain J1074::spi, and ethyl acetate extraction of the fermentation product yielded 11.3 g crude extract. Subsequent stepwise separation by column chromatography (CC) with SiliaSphere C18 (ODS), medium pressure liquid chromatography (MPLC) and HPLC of the crude extract yielded a total of eight compounds, including four undescribed angucyclinone analogues: angumycinone Z (1) (7.1 mg), WS009 Z (2) (5.6 mg), angumycinone X (3) (6.4 mg), and angumycinone T (4) (4.7 mg). Compounds 5–8 were identified by comparing their 1D NMR and MS data to previously published literature (Figure 2).

Compound 1 was separated as a pale brown powder. Its molecular formula, C₁₉H₁₈O₈, was identified by HR-ESIMS m/z 373.0916 [M − H]⁻, indicating eleven degrees of unsaturation. The signals indicated in the ¹H NMR data illustrate that there are one methyl (δ_H 1.93, s), three methylenes (δ_H 2.89, 2.06; δ_H 1.98, 1.60; δ_H 2.28, 1.74), one vinyl methine (δ_H 5.71, s), three aromatic methines (δ_H 7.34, d, J = 8.2 Hz; δ_H 7.76, t, J = 8.0 Hz; δ_H 7.60, d, J = 7.5 Hz), and five exchangeable protons (δ_H 11.10, s; δ_H 6.32, s; δ_H 6.03, s; δ_H 7.01, s; δ_H 5.66, s) (

Table 1. ¹H Spectroscopic Data for 1–4 in DMSO-d₆ (500 MH_Z).

No.	1	2	3	4
	δH (J in Hz)	δH (J in Hz)	δH (J in Hz)	δH (J in Hz)
2	5.71, s	5.70, s	5.97, s	5.83, s
4	2.89, d (18.0)	2.94, d (18.0)	3.06, d (18.0)	3.98, d (1.3)
	2.06, d (18.0)	2.13, d (18.0)	2.02, d (18.0)	-
5	1.98, m	2.07, m	1.76, m	1.98, m
	1.60, m	1.63, m	1.35, m	-
6	2.28, m	2.36, m	2.53, m	2.24, m
	1.74, m	1.84, m	2.38, m	2.14, m
9	7.34, d (8.0)	7.35, d (8.0)	7.31, m	7.35, dd (8.0, 1.0)
10	7.76, t (8.0)	7.73, t (8.0)	7.69, dd (8.2, 7.6)	7.76, dd (8.0, 7.6)
11	7.60, d (8.0)	7.58, d (8.0)	7.34, m	7.34, dd (7.6, 1.0)
12b	-	-	-	3.14, s
4a-OH	6.32, s	-	2.54, s	5.59, s
6a-OH	6.03, s	-	-	7.05, s
12a-OH	7.01, s	-	-	-
12b-OH	5.66, s	-	-	-
8-OH	11.10, s	11.17, s	11.06, s	11.72, s
13	1.93, s	1.96, s	1.81, s	2.33, d (1.2)
S-CH3	-	1.67, s	-	-

). The ¹³C NMR spectrum combined with the HSQC spectrum reveals the existence of one methyl, three sp3 methylenes, four protonated sp2 carbons, and eleven quaternary carbons (

Table 2. ¹³C Spectroscopic Data for 1–4 in DMSO-d₆ (125 MH_Z).

No.	1	2	3	4
	δC, Type	δC, Type	δC, Type	δC, Type
1	192.8, C	193.5, C	196.7, C	195.6, C
2	122.9, CH	122.7, CH	122.6, CH	124.6, CH
3	158.2, C	159.4, C	156.1, C	162.4, C
4	41.8, CH2	41.6, CH2	38.2, CH2	85.1, CH
4a	75.5, C	75.7, C	65.1, C	81.6, C
5	28.2, CH2	28.8, CH2	25.8, CH2	32.6, CH2
6	21.6, CH2	17.9, CH2	20.1, CH2	26.8, CH2
6a	77.6, C	61.2, C	74.2, C	78.0, C
7	200.5, C	195.5, C	194.5, C	201.6, C
7a	116.3, C	115.4, C	114.7, C	114.9, C
8	159.9, C	160.0, C	159.5, C	161.1, C
9	123.6, CH	124.4, CH	123.9, CH	123.7, CH
10	136.2, CH	136.0, CH	136.4, CH	137.7, CH
11	118.9, CH	119.0, CH	118.9, CH	118.3, CH
11a	133.0, C	131.5, C	132.7, C	135.3, C
12	194.1, C	191.0, C	189.6, C	192.1, C
12a	76.4, C	78.4, C	67.5, C	92.5, C
12b	76.2, C	75.7, C	73.7, C	52.7, CH
13	23.6, CH3	23.7, CH3	23.6, CH3	26.3, CH3
S-CH3	-	13.3, CH3	-	-

). These data imply that compound 1 has a tetracyclic benz[a]anthracene skeleton belonging to the angucycline analogue [24,25], which possibly resulted from the expression of the spi BGC. The planar structure of 1 was subsequently identified through a comprehensive analysis of 2D NMR data (Figure 3). Firstly, the ¹H-¹H COSY correlations from H-9 (δ_H 7.34, d, J = 8.2 Hz) via H-10 (δ_H 7.76, t, J = 8.0 Hz) through H-11 (δ_H 7.60, d, J = 7.5 Hz) indicate the presence of a 1,2,3-trisubstituted benzene ring (ring D) (Figure 3). The HMBC spectrum correlations from 8-OH (δ_H 11.10) to C-8 (δ_C 159.9, s), C-9 (δ_C 123.6, d), and C-7a (δ_C 116.3, s) showed a phenolic OH at C-8. The HMBC correlations from Me-13 to C-4 (δ_C 41.8, t), C-3 (δ_C 158.2, s), C-1 (δ_C 192.8, s), and C-2 (δ_C 122.9 d), from H-2 (δ_H 5.71, s) to C-4 and C-12b (δ_C 76.2, s), as well as from H-4 (δ_H 2.06, d, J = 17.7 Hz) to C-4a (δ_C 75.5, s) and C-12b (δ_C 76.2, s) identified the presence of the A ring. The B ring was determined via the HMBC correlations from the active exchangeable protons 12a-OH (δ_H 7.01) to C-12a (δ_C 76.4, s), C-12 (δ_C 194.1, s), and C-12b (δ_C 76.2, s), from 12b-OH (δ_H 5.66) to C-4a (δ_C 75.5, s), C-12a (δ_C 76.4, s), and C-12b, from 4a-OH (δ_H 6.32) to C-4a, and C-5 (δ_C 28.2, t), and from 6a-OH (δ_H 6.03) to C-6a (δ_C 77.6, s) and C-12a (δ_C 76.4, s). The HMBC correlations from H-11 to C-12 and C-7 (δ_C 200.5, s) confirmed the connection of rings B and D via ring C. Thus, the planar structure of 1 was identified (Figure 3).

The absolute configuration of 1 was determined by NOESY spectrum analysis and ECD calculations. Analyzing the NOESY spectrum data revealed a correlation between 4a-OH (δ_H 6.32) and 12a-OH (δ_H 7.01), as well as a correlation between 4a-OH and 12b-OH (δ_H 5.66), while there was no correlation with 6a-OH (δ_H 6.03), which indicated that 4a-OH, 12a-OH, and 12b-OH were all in the same plane and 6a-OH was on the opposite side. Thus, the relative configuration of 1 was confirmed to be (6aS*, 4aR*, 12aR*, 12bR*)-1 or (6aR*, 4aS*, 12aS*, 12bS*)-1. ECD calculation demonstrated that the latter was more compatible with the measured values, based on which the absolute configuration of compound 1 was confirmed as (6aR, 4aS, 12aS, 12bS)-1 (Figure 4).

Compound 2 is an even brown amorphous powder with a molecular formula of C₂₀H₂₀O₇S given based on HRESIMS data. One-dimensional NMR and HSQC spectra showed that there were two methyl (δ_C 23.7 and 13.3), three sp3 methylenes (δ_C 17.9, 28.8, and 41.6), four protonated sp2 carbons (δ_C 119.0, 122.7, 124.4, and 136.0), and eleven quaternary carbons including four non-protonated sp3 carbons (δ_C 61.2, 75.7, 75.7, and 78.4), four non-protonated sp2 carbons (δ_C 115.4, 131.5, 159.4, and 160.0), and three carbonyls (δ_C 191.0, 193.5, and 195.5) (Table 2). These signals all indicate that compound 2 has a backbone highly similar to 1. The difference lies in the high-field chemical shift of C-6 (δ_C 17.9, t), C-6a (δ_C 61.2, s), and C-7 (δ_C 195.5, s), recombined with the HMBC correlation from -SCH₃ (δ_H 1.67) to C-6a (Figure 3) and the HR-ESIMS data, which determined that the hydroxyl signal (6a-OH) in compound 1 was replaced by the thiomethyl group (δ_C 13.3, q) in 2. However, the NOESY data could not determine the relative configuration of the thiomethyl group and three hydroxyl groups. Compounds 2 and 1 are proposed to have originated from the same epoxide precursor (discussed later in Scheme 1), which went through a ring opening process probably promoted by hydrolysis or methane thiol addition, hence the absolute configuration of 2 was tentatively assigned as 6aR, 4aS, 12aS, 12bS.

Compound 3 is a yellowish solid powder. HR-ESIMS analysis established the chemical formula as C₁₉H₁₆O₇, with twelve degrees of unsaturation. The ¹H and ¹³C NMR spectra revealed that 3 contains one methyl, three methylene, four methenyl, and 11 quaternary carbon signals (including three carbonyl and six aromatic quaternary carbons). The ¹H and ¹³C NMR spectra showed that 3 was highly similar to angumycinone A [26], with the only difference being the low-field chemical shift at C-12a (δ_C 67.5), from which it can be deduced that there is hydroxylation at this position, thus determining the planar structure of compound 3. Nevertheless, the relative configuration was unable to be defined by the NOESY data. The stereo-structure was determined based on NMR calculations, DP4+ analysis, and ECD calculations, primarily NMR calculations and DP4+ probability analyses of the four possible configurations (6aS*, 12aR*, 12bS*, 4aR*)-3a, (6aS*, 12aR*, 12bR*, 4aS*)-3b, (6aS*, 12aR*, 12bS*, 4aS*)-3c, and (6aS*, 12aR*, 12bR*, 4aR*)-3d at the B3LYP/6-311+G(d,p) level using PCM model in DMSO. The calculated results for 3d (R² = 0.9958) were more consistent with the experimental data than others (Figure S3). Furthermore, the DP4+ probability calculations identified 3d with a 99.90% probability (Table S7). The absolute configuration of 3 was then determined to be 6aS,12aR,12bR,4aR through ECD calculation at the B3LYP/6-31+G(d,p)//B3LYP/6-31G(d) level was compared with the experimental ECD curve (Figure 5).

Compound 4 is a brown amorphous powder with molecular formula C₁₉H₁₆O₇ deduced from HR-ESIMS, indicating 12 degrees of unsaturation. The ¹H NMR spectrum of 4 displayed one methyl (δ_H 2.33, d, J = 1.2 Hz), two methylenes (δ_H 1.98; δ_H 2.24, 2.14), five methines (δ_H 3.98, d, J = 1.3 Hz; δ_H 3.14, s; δ_H 7.35, dd, J = 8.6, 1.0 Hz; δ_H 7.76, dd, J = 8.3, 7.6 Hz; δ_H 7.34, d, J = 7.5, 1.0 Hz), a vinyl methine (δ_H 5.83, s), and three exchangeable protons (δ_H 5.59, s; δ_H 7.05, s; δ_H 11.72, s) (Table 1). The ¹³C NMR and HSQC spectrum showed that 4 contains one methyl (δ_C 26.3, q), two methylene (δ_C 26.8 and 32.6), six methines (δ_C 52.7, 85.1, 118.3, 123.7, 124.6, and 137.7), and ten quaternary carbons, including three non-protonated sp3 carbons (δ_C 78.0, 81.6 and 92.5), four non-protonated sp2 carbons (δ_C 114.9, 135.3, 161.1, and 162.4), and three carbonyls (δ_C 192.1, 195.6, and 201.6) (Table 2). The planar structure of 4 was then identified via a detailed analysis of 2D NMR (Figure 3). The A ring was determined by the HMBC correlations from H-2 (δ_H 5.83, s) to C-1 (δ_C 195.6), C-4 (δ_C 85.1), and C-12b (δ_C 52.7), from 13-CH₃ (δ_H 2.33, d) to C-2 (δ_C 124.6), C-3 (δ_C 162.4), C-4, and from H-4 (δ_H 3.98, d, J = 1.3 Hz) to C-4a (δ_C 81.6) and C-12b. The ¹H-¹H COSY correlations from H-9 (δ_H 7.35, dd, J = 8.6, 1.0 Hz) via H-10 (δ_H 7.76, dd, J = 8.3, 7.6 Hz) through H-11(δ_H 7.34, dd, J = 7.5, 1.0 Hz) suggested the presence of the 1,2,3-trisubstituted benzene ring (ring D). HMBC correlation analyses from H-9 to C-7a (δ_C 114.9) and C-7 (δ_C 201.6), from H-10 to C-11a (δ_C 135.3) and C-8 (δ_C 161.1), and from H-11 to C-12 (δ_C 192.1) extend ring D to ring C. Further analysis of the HMBC correlations from H-5 (δ_H 1.98, m) to C-6a (δ_C 78.0) and C-12b and from H-6 (δ_H 2.14) to C-12a (δ_C 92.5) determined the presence of ring B, closing the angular skeleton of compound 4. HMBC correlations from the active exchangeable protons 4a-OH (δ_H 5.59) to C-4a, C-4 and C-5 (δ_C 32.6), from 6a-OH (δ_H 7.05, s) to C-6a, C-7 and C-6, together with the fact that there are 12 degrees of unsaturation, suggests that another unsaturation should be the presence of an epoxide between C-12a and C-4 (δ_C 85.1). Thus, the planar structure of 4 was determined.

The stereo configuration of compound 4 was determined by NOESY spectral analysis and ECD calculations. The NOESY spectrum analysis revealed that H-12b (δ_H 3.14) correlates with 4a-OH and 6a-OH, indicating that H-12b, 4a-OH, and 6a-OH are in the same plane, resulting in only two possible structures: (4R*, 4aR*, 6aS*, 12aS*, 12bS*)-4 and (4S*, 4aS*, 6aR*, 12aR*, 12bR*)-4. Finally, the absolute configuration of 4 was determined to be 4S,4aS,6aR,12aR based on the results of ECD calculations (Figure 6). Angucyclines carrying epoxide modifications on the skeletons were frequently discovered in nature [27], however, this is the first example of angucycline derivative with the epoxidation-bridged C-12a and C-4.

Based on the comparison of NMR and MS data with those reported in the literature, the six known compounds isolated in this study were identified as WP 3688-2 (5) [28], gephyromycin (6) [29,30], SEK43 (7) [31], and SEK15 (8) [32].

All new angucyclines were evaluated for cytotoxicity against L-02, MDA-MB-231, K562, ASPC-1, and NCI-H446/EP cell lines in vitro. Adriamycin was used as the positive control. Compound 3 showed cytotoxicity toward five cancer cell lines and especially strongly inhibited K562 and NCI-H446/EP cell lines with IC50 values of 11.72 and 8.92 μM, respectively. It is worth mentioning that both compounds 3 and 4 possess an oxygen bridge and two hydroxyl groups in the structure; however, compound 4 did not display obvious cytotoxicity, which indicated that the oxidative modifications on rings A and B are very important for their mode of actions.

2.3. Plausible Biosynthetic Pathways for Compounds 1–8 Produced by the Mutant Strain

The plausible biogenetic pathway of compounds 1–8 was determined based on the bioinformatics analyses (Table S2) and literature surveys. First, the polyketide chain was generated by an iterative Claisen condensation reaction of a starting unit malonyl-CoA and nine extension unit malonyl-CoA under the action of the minimal PKS SpiA, SpiB and SpiC, and was subsequently folded by cyclase SpiE and SpiD to form UWM6 [33,34], a characteristic intermediate of angucyclines; however, if it is spontaneously folded, the by-products 7 and 8 will be produced. Thereafter, UWM6 can achieve the oxidation of C-12 catalyzed by monooxygenase SpiH2 [35] and subsequently undergo reduction at C-6, dehydration and enoyl reduction at C-5 and C-6, and oxidation at C-12b to produce 5, which could be converted into compounds 1 and 2 via a 6a, 12a-epoxide intermediate. The methanethiol residue of 2 was probably derived from methionine which was incorporated via methane thiol addition to the epoxide. In alternative routes, UWM6 may undergo multi-step catalysis including oxidation, dehydration and oxo-bridge generation to give 3, 4 and 6 in the presence of relevant endogenous enzymes of the heterologous host.

3. Materials and Methods

3.1. General Experimental Procedures

High-quality actinobacteria genomes were obtained using methods previously described in the literature [36,37]. Polymerase chain reaction (PCR) was performed using 2 × Phanta Flash Master Mix (P510, Vazyme, Nanjing, China). PCR analyses were conducted with a 2 × Hieff^® PCR Master Mix (with dye, Yeasen, Shanghai, China). Shanghai Sangon DNA Technologies (Shanghai, China) offered custom oligonucleotide synthesis and fragment sequencing services.

NMR spectra were measured on an Agilent 500 MHz DD2 spectrometer with tetramethylsilane (TMS) as an internal standard. Optical rotations in MeOH were recorded on a JASCO-1020 digital polarimeter. A Thermo Scientific LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) was used to collect HRESIMS data. UV-vis spectra were acquired on the UFLC system (Shimadzu, Tokyo, Japan). The JASCO J-715 spectropolarimeter was used to record the ECD spectra. A UFLC system (Shimadzu, Tokyo, Japan) with a C18 column (Shimadzu, 4.6 mm × 150 mm, 5 μm, 1 mL/min) coupled to an LCMS-2020 mass spectrometer (Shimadzu, Tokyo, Japan) was used to perform LC-MS analyses. Column chromatography (CC) was performed with SiliaSphere C18 (Octadecylsilyl, ODS) monomeric (SiliCycle Inc., Quebec, QC, Canada, 50 μm, 120 A), and Sephadex LH-20 (GE Healthcare, Uppsala, Sweden). MPLC was performed using a C18 column (Welch Materials Inc., Ultimate^® XB-C18, 21.2 mm × 250 mm, 5 μm, 10 mL/min).

Prediction and analysis of gene clusters using the online tool antiSMASH. DNA and protein sequence homology analysis using the online tool NCBI.

3.2. Materials and Culture Conditions

Streptomyces sp. HDN155000 was obtained from a marine sediment sample collected from the South China Sea at coordinates 125°28.550′ E and 29°1.618′ N. The strain was identified via genome sequencing and submitted to GenBank (accession number MN822699). S. albus J1074 was grown for 6 days at 28 °C on MS plates with 2% mannitol, 2% soya bean flour, and 1.5% agar. For genome extraction, the strain was inoculated into 250mL Erlenmeyer flasks containing 50 mL of TSB liquid medium (1.7% tryptone, 0.3% peptone, 0.25% glucose, 0.25% KH₂PO₄, and 0.3% NaCl) for 4 days.

Escherichia coli strains DH10B as the general host for cloning and E. coli ET12567/pUZ8002 as the donor in intergeneric conjugation were cultured on Luria–Bertani plates or liquid medium (1% tryptone, 0.5% yeast extract, and 1% NaCl) at 37 °C.

3.3. S. albus J1074 Intergeneric Conjugation

The intergeneric conjugation between E. coli and S. albus J1074 was performed as previously described with some modifications [38]. The culture of the donor E. coli ET12567/pUZ8002 containing the heterologous vector p15A::spi was incubated with appropriate antibiotics LB liquid culture medium (containing 50 mg/L kanamycin, 15 mg/L chloramphenicol, and 50 mg/L apramycin,) until the optical density at 600 nm (OD600) of 0.6–0.8. Bacteriophages were collected at 9000 rpm, rinsed three times with 1 mL 2× YT medium [39] to remove antibiotics, and then resuspended in 0.2 mL of 2× YT (containing 10 mM MgCl₂) as the donor cells. S. albus J1074 spores were washed three times and suspended in 2× YT broth at a concentration of 10⁹ per mL. Subsequently, spores obtained in the previous step were heated at 50 °C for 10 min to serve as recipients. Donor and recipient cells were mixed evenly on MS plates (containing 30 mM MgCl₂ and 30 mM CaCl₂), and cultivated for 18 h at 30 °C. After incubation, plates were covered with 1 mL of water with apramycin (50 μg/mL) and nalidixic acid (50 μg/mL) and then kept at 30 °C for 4–6 days until exconjugants appeared. The conjugation frequency reached 7.8 × 10⁻⁵.

3.4. Fermentation and LC/LC-MS Analysis

Heterologous expression strains and negative control strains were cultured for 7 days at 28 °C in 500 mL Erlenmeyer flasks with 100 mL of M1 medium (2 g/L peptone, 4 g/L yeast extract, 10 g/L starch), after which the culture products were collected and extracted three times with ethyl acetate. The organic phase was evaporated, and the residue was re-dissolved in 150 μL MeOH, which was analyzed using HPLC (C18 column, Shimadzu, 4.6 mm × 150 mm, 5 μm, 1 mL/min), revealing a considerable change in metabolite synthesis in the heterologous expression strain J1074::spi (Figure 1).

For the isolation of secondary metabolites, we scaled up cultures with the 500 mL Erlenmeyer flask (total 20 L) according to the conditions used for small-scale analyses. The fermentation supernatant was extracted with ethyl acetate three times and then evaporated to dryness, obtaining a total of 11.3 g of crude product.

3.5. Extraction, Isolation, and Purification

The crude extract was eluted using a step gradient of MeOH-H₂O using metabolite detection on a C18 column, yielding seven subfractions (Fr.1-Fr.7, 30% to 100%). Fr.2 was purified by semi-preparative HPLC (33:67 MeOH-H₂O, 3 mL/min) to generate compounds 1 (7.1 mg, t_R = 15.1 min) and 6 (3.1 mg, t_R = 16.7 min). Fr.3 was purified by a semi-preparative C18 HPLC column (47:53 MeOH-H₂O), producing compound 3 (6.4 mg, t_R = 14.7 min). Fr.4 was separated using preparative C18 HPLC column (58:42 MeOH-H₂O) to yield two sub-fractions (Fr.4.1 and Fr.4.2). Then, Fr.4.1 and Fr.4.2 were purified by semi-preparative HPLC (53:47, 60:40, MeOH-H₂O, 3 mL/min) to provide compounds 4 (4.7 mg, t_R = 18.1 min), 5 (3.4 mg, t_R = 20.7 min), 7 (3.8 mg, t_R = 16.9 min), and 8 (3.0 mg, t_R = 19.4 min), respectively. Fr.5 was applied to a Sephadex LH-20 column and eluted with methanol, obtaining compound 2 (5.6 mg).

Angumycinone Z (1): pale brown powder, ${\left[\alpha \right]}_D^{25}$ + 19.5 (c 0.03, CH₃OH); UV (DAD) λ_max 208 nm, 250 nm, 327 nm, 386 nm; CD (MeOH) λ_max (∆ε) 210 (−19.06), 243 (+33.91), 283 (−4.88), 302 (+3.13), 336 (−3.39), 376 (+5.21); ¹H and ¹³C NMR data, see Table 1 and Table 2; negative ion HRESIMS m/z 373.0916 [M − H]⁻ (calcd. for C₁₉H₁₇O₇⁻, 373.0928).

WS009 Z (2): even brown amorphous powder, ${\left[\alpha \right]}_D^{25}$ + 55.8 (c 0.03, CH₃OH); UV (DAD) λ_max 249, 302, 352; ¹H and ¹³C NMR data, see Table 1 and Table 2; positive ion HRESIMS m/z 405.0996 [M + H]⁺ (calcd. for C₂₀H₂₁O₇S⁺, 405.1003).

Angumycinone X (3): yellowish, solid powder, ${\left[\alpha \right]}_D^{25}$ + 79.1 (c 0.03, CH₃OH); UV (DAD) λ_max 206, 238, 309, 364; CD (MeOH) λ_max (∆ε) 211 (−12.77), 242 (+25.31), 281 (−3.28), 300 (+2.63), 334 (−2.96), 372 (+4.21); ¹H and ¹³C NMR data, see Table 1 and Table 2; negative ion HRESIMS m/z 355.0821 [M − H]⁻ (calcd. for C₁₉H₁₅O₇⁻, 355.0823).

Angumycinone T (4): brown amorphous powder, ${\left[\alpha \right]}_D^{25}$ + 31.0 (c 0.03, CH₃OH); UV (DAD) λ_max 206, 236, 298, 350; CD (MeOH) λ_max (∆ε) 207 (+3.30), 227 (+19.67), 243 (−11.30), 274 (+14.31), 332 (−2.67), 370 (+2.94); ¹H and ¹³C NMR data, see Table 1 and Table 2; negative ion HRESIMS m/z 355.0815[M − H]⁻ (calcd. for C₁₉H₁₅O₇⁻, 355.0823)

3.6. NMR and ECD Calculations

Conformation searches based on molecular mechanics with MMFF force fields were performed for stereoisomers to obtain stable conformers. The Gaussian 16 programme package [40] used the density functional theory (DFT) approach at the B3LYP/6-31G(d) level to optimize all conformers. Gauge Independent Atomic Orbital (GIAO) calculations of their ¹H and ¹³C NMR chemical shifts were performed utilizing density functional theory (DFT) at the mPW1PW91/6-311+G(d,p) level with the PCM model in DMSO [41]. The calculated NMR data of these conformers were averaged using the Boltzmann distribution theory and their respective Gibbs free energy. The ¹H and ¹³C NMR chemical shifts for TMS were computed using the same methodology and served as a reference. Following calculation, the experimental and calculated data were compared using linear correlation coefficients (R²) and the improved probability DP4+ technique [42].

All stable conformers were further optimized at the B3LYP/6-31G(d)-GD3BJ level with the Gaussian 16 programme package. The ECD was computed using time-dependent density functional theory (TDDFT) at a B3LYP/6-311+G(d,p) level with the IEFPCM model. The calculated ECD curves were all generated using the SpecDis 1.71 computer package and the calculated ECD data of all conformers were Boltzmann averaged using Gibbs free energy [43].

3.7. Cytotoxicity Assay

The cytotoxic assay involved five human cancer cell lines: K562 (using the MTT method), ASPC-1, NCI-H446/EP, MDA-MB-231, and L-02 (using the SRB method). Adriamycin (ADM) was employed as a positive control. The detailed procedures for biological testing were conducted as previously stated [20,44]. The cancer cell lines are purchased from the National Collection of Authenticated Cell Cultures of China (Shanghai).

4. Conclusions

Secondary metabolites from actinobacteria have long been a valuable source of natural medicines [45]. However, the success rate for identifying actinobacteria-derived novel compounds is limited by gene cluster silencing or low expression under laboratory conditions, especially for micro-organisms from special habitats such as marine actinobacteria. In this study, the type II PKS gene cluster spi derived from Streptomyces sp. HDN155000 was successfully heterologously expressed in S. albus J1074, and a total of eight compounds were obtained through further fermentation, isolation, and purification. Based on the literature and our previous study, we proposed a plausible biosynthetic pathway leading to those new compounds [4,34]. Generally, compounds 3–6 were directly produced from UWM6 by a multi-step enzymatic reaction involving oxidation, dehydration, enoyl reduction, etc. Compounds 1 and 2 were derived from compound 5 via dehydrogenation, hydrolysis, methane thiol addition, and other processes. Angucyclines are a chemical family that has been extensively explored in actinobacteria since the first structure was described in 1965 in Streptomyces rimosus [2]. More than 400 cases have been recorded, with a range of skeleton modifications; however, compound 4 is the first example of oxidized angucyclines with an oxo-bridge between C-12a and C-4 [3]. Compound 3 showed cytotoxic to the K562 and NCI-H446/EP cell lines, providing an alternative lead compound for the study of human chronic myeloid leukemia and lung cancer. The above data enrich our understanding of the type II PKS pathways and related aromatic products, and also indicate heterologous expression as a promising strategy for maximizing the metabolic capability of Streptomyces.