Isolation, Structure Elucidation, and First Total Synthesis of Quinomycins K and L, Two New Octadepsipeptides from the Maowei Sea Mangrove-Derived Streptomyces sp. B475

Mangrove actinomycetia have been proven to be one of the promising sources for discovering novel bioactive natural products. Quinomycins K (1) and L (2), two rare quinomycin-type octadepsipeptides without intra-peptide disulfide or thioacetal bridges, were investigated from the Maowei Sea mangrove-derived Streptomyces sp. B475. Their chemical structures, including the absolute configurations of their amino acids, were elucidated by a combination of NMR and tandem MS analysis, electronic circular dichroism (ECD) calculation, advanced Marfey’s method, and further unequivocally confirmed by the first total synthesis. The two compounds displayed no potent antibacterial activity against 37 bacterial pathogens and had no significant cytotoxic activity against H460 lung cancer cells.


Introduction
As a unique ecosystem with extreme conditions and high biodiversity, mangrove is becoming a rich source for discovering new actinomycetia and novel pharmaceutical compounds. A multitude of bioactive compounds, including the promising compounds salinosporamide A, xiamycins, and indolocarbazoles, have been isolated from mangrove actinomycetia [1][2][3][4]. The Beibu Gulf is located in the northwestern continental shelf area of the South China Sea, from the Leizhou Peninsula, Qiongzhou Strait, and Hainan Island to Vietnam and northward to Guangxi. The mangrove species in this region are very rich, and mangrove in the region is ranked as the second-largest area and accounts for more than 37% of the total mangrove area in China [5]. Recently, our group has discovered many new actinomycetia taxa along with novel bioactive compounds from mangroves in the Beibu Gulf, such as the Maowei Sea [6,7], Beilun Estuary [8][9][10], and Leizhou Peninsula [11].
In our previous study, five putative new quinomycin analogues were found by MS/MS-based molecular networking analysis [7]. However, due to a limited amount of sample, only two compounds, quinomycins K (1) and L (2), were obtained (1, 3.2 mg; 2, the Beibu Gulf, such as the Maowei Sea [6,7], Beilun Estuary [8][9][10], and Leizhou Peninsula [11]. In our previous study, five putative new quinomycin analogues were found by MS/MS-based molecular networking analysis [7]. However, due to a limited amount of sample, only two compounds, quinomycins K (1) and L (2), were obtained (1, 3.2 mg; 2, 1.8 mg) (Figure 1). Quinomycins are cyclic octadepsipeptides that belong to the quinoxaline family of antibiotics, exhibiting significant antibacterial and antitumor activities due to the bisintercalation of the two quinoxaline rings into DNA [12]. The isolation and synthesis of quinomycins, such as echinomycin and triostin A, have attracted tremendous attention due to their fascinating skeletons and a wide range of biological activities [13][14][15]. In this paper, we present the isolation and complete structural elucidation of quinomycins K (1) and L (2) based on the combination of spectroscopic analyses, ECD calculations, Marfey's methods, and the first total synthesis.
The 1 H NMR data (Table 1) of 1 showed the presence of two exchangeable NH protons (δH 8.70, 8.24), the characteristic signal of quinoxaline-2-carboxylic acid (δH 9.53, 7.97-8.02, 8.19-8.22), two methylamino proton signals (δH 2.92 and 2.85). The SCH3 (δH≈2.10) [18] and SCH2 (δH≈3.60) [14] signals were absent in the 1 H NMR spectrum. The 13 31.2, 29.5). The number of carbons in the molecular formula has twice as much as those in the 13 C NMR, indicating that 1 should be a symmetrical dimer with eight amino acid residues. Comparison of NMR spectra of 1 and triostin A [14,19], showed they were similar except for the loss of the Cys residue signal in 1, suggesting that this Cys residue was replaced by another amino acid group without the intra-peptide disulfide bridge. After examining the NMR data of amino acid residues in 1, it was inferred that the new amino acid residue was N-Me-2-aminobutyric acid (NMe-Abu) [20], which was further confirmed by the 1 H-1 H COSY correlations of H-5 (δH 5.36)/H2-17 (δH 1.66-1.82) and H2-17/ H3-18 (δH 0.80), and the HMBC correlations
The 1 H NMR data (Table 1) [18] and SCH 2 (δ H ≈3.60) [14] signals were absent in the 1 H NMR spectrum. The 13 C NMR spectra, with the aid of 2D NMR, showed the presence of 26 carbons, including four carbonyls (δ C 171.4, 170.0, 169.3, 167.9), a quinoxaline-2-carboxylic acid moiety carbon (δ C 162.6, 129.0-143.4), one oxygenated methylene (δ C 64.4), four aliphatic α-amino carbons (δ C 62.1, 54.4, 50.9, 47.2), two methylamino carbons (δ C 31.2, 29.5). The number of carbons in the molecular formula has twice as much as those in the 13 C NMR, indicating that 1 should be a symmetrical dimer with eight amino acid residues. Comparison of NMR spectra of 1 and triostin A [14,19], showed they were similar except for the loss of the Cys residue signal in 1, suggesting that this Cys residue was replaced by another amino acid group without the intra-peptide disulfide bridge. After examining the NMR data of amino acid residues in 1, it was inferred that the new amino acid residue was N-Me-2-aminobutyric acid (NMe-Abu) [20], which was further confirmed by the 1 H-1 H COSY correlations of H-5 (δ H 5.36)/H 2 -17 (δ H 1.66-1.82) and H 2 -17/ H 3 -18 (δ H 0.80), and the HMBC correlations from H 3 -18 to C-17 (δ C 21.1) and C-5 (δ C 54.4), from H 3 -19 (δ H 2.85) to C-7 (δ C 171.4) and C-5, and from H 3 -16 (δ H 2.92) to C-4 (δ C 170.0) and C-2 (δ C 62.1) ( Figure 2). The analysis of MS/MS fragmentation patterns provided additional data to support this assignment ( Table 2, Figures 3 and 4). Therefore, the planar structure of 1 was established, as shown in Figure 1.    The relative configuration of 1 was determined by the ROESY spectrum. The key ROESY cross-peak between H 3 -20 (δ H 1.29) and H-5, H-13 (δ H 2.28-2.35), H 3 -14 (δ H 0.95) and H-21 (δ H 8.70) indicated that they were in the same spatial orientation. Thus, two possible isomers (2S,5S,8S,11R,2 S,5 S,8 S,11 R)-1 (1a) and its enantiomer (2R,5R,8R,11S,2 R,5 R,8 R, 11 S)-1 (1b) were concluded. To clarify the absolute configurations of 1, ECD calculation based on the simplified time-dependent density functional theory approach (sTD-DFT) was used, which allowed fast computation of electronic ultraviolet (UV) or circular dichroism (CD) spectra of very large molecules with up to 1000 atoms, such as peptides and proteins [21]. As shown in Figure 5, the calculated ECD spectrum of 1a was in good agreement with the experimental ECD of 1, which established the assignment of the absolute configuration of 1 as 2S,5S,8S,11R,2 S,5 S,8 S,11 R. In addition, the advanced Marfey's analysis of 1 was performed to determine the absolute configurations of amino acid residues [22]. The presence of D-Ser, L-Ala, L-NMe-Abu, and L-NMe-Val was unambiguously confirmed by comparison with authentic standards (Table 3, Figure 6 and Figure S11 in Supplementary Materials), which was consistent with the structure of 1a.      The relative configuration of 1 was determined by the ROESY spectrum. The key ROESY cross-peak between H3-20 (δH 1.29) and H-5, H-13 (δH 2.28-2.35), H3-14 (δH 0.95) and H-21′ (δH 8.70) indicated that they were in the same spatial orientation. Thus, two possible isomers (2S,5S,8S,11R,2′S,5′S,8′S,11′R)-1 (1a) and its enantiomer (2R,5R,8R,11S,2′R,5′R,8′R,11′S)-1 (1b) were concluded. To clarify the absolute configurations of 1, ECD calculation based on the simplified time-dependent density functional theory approach (sTD-DFT) was used, which allowed fast computation of electronic ultraviolet (UV) or circular dichroism (CD) spectra of very large molecules with up to 1000 atoms, such as peptides and proteins [21]. As shown in Figure 5, the calculated ECD spectrum of 1a was in good agreement with the experimental ECD of 1, which established the assignment of the absolute configuration of 1 as 2S,5S,8S,11R,2′S,5′S,8′S,11′R. In addition, the advanced Marfey's analysis of 1 was performed to determine the absolute configurations of amino acid residues [22]. The presence of D-Ser, L-Ala, L-NMe-Abu, and L-NMe-Val was unambiguously confirmed by comparison with authentic standards (Table 3 [23,24] and our previous study [7], compounds 1 and 2 were neighbor nodes in one cluster, indicating that they had the identical or similar structural fragments in their structure. Therefore, the planar structure of 2 was deduced by analyzing the relationship of MS/MS fragmentation ions between 1 and 2 ( Table [23,24] and our previous study [7], compounds 1 and 2 were neighbor nodes in one cluster, indicating that they had the identical or similar structural fragments in their structure. Therefore, the planar structure of 2 was deduced by analyzing the relationship of MS/MS fragmentation ions between 1 and 2 ( Table 2 The absolute configuration of compound 2 was determined by amino acid analysis using the advanced Marfey's method. Comparison of FDLA derivatives of the hydrolysates of 2 with those of appropriate standard amino acids using UPLC-MS techniques indicated L from Ala, NMe-Abu, NMe-Ser, and NMe-Val, and D from Ser in compound 2 (Table 3, Figure 9).  Figure 9).  To unequivocally confirm their structure and solve the supply problem for further bioactivity studies, the first total syntheses of compounds 1 and 2 were performed. The solution-phase synthetic procedure described herein will contribute to the development of the synthesis of compounds 1 and 2. We retrosynthetically disconnected the macrocycle 1 and 2 at the amide bonds linking D-Ser and L-Ala residues (Scheme 1). The intermediate tetrapeptide compound 4 could be accomplished via a sequential peptide coupling approach.    To unequivocally confirm their structure and solve the supply problem for further bioactivity studies, the first total syntheses of compounds 1 and 2 were performed. The solution-phase synthetic procedure described herein will contribute to the development of the synthesis of compounds 1 and 2. We retrosynthetically disconnected the macrocycle As shown in Scheme 2, The total synthesis started with the synthesis of Cbz-D-Ser(Boc-MeVal)-OAll (5) according to an optimized procedure previously reported by Nagaswa et al. [14,15]. Compound 5 was subsequently condensed with N-Boc-methyl-Abu-OH affording the tripeptide 6. The next step involved the conjugation of N-Boc-L-Ala-OH with 6 to give tetradepsipeptide 7 in 60% yield over four steps. Half of the obtained compound 7 was then deallylated to give 8. Subsequently, N-Boc deprotection of 7 was followed by condensation with compound 8 using HATU to afford a linear octadepsipeptide 9. Compound 9 obtained by deallylation of 8 was subjected to intramolecular amide bond formation to produce cyclic peptide 10. N-Cbz deprotection of 10 was condensed with quinoxaline-2-carboxylic acid to obtain the target compound 1 in 55% yield. The synthetic procedure for compound 2 was similar to compound 1, except for the O-Bn-N-Boc-methyl -Ser-OH instead of N-Boc-methyl-Abu-OH (Scheme 3). All the spectroscopic data ( 1 H and 13 C NMR, HRESI-MS, UV, IR, and specific rotation) and UPLC-UV-MS retention times (two different eluting conditions) of the synthetic-1 and -2 matched those of the isolated natural products, respectively (Tables 1 and 4, Figures S31-S38 and Tables S1).
As shown in Scheme 2, The total synthesis started with the synthesis of Cbz-D-Ser(Boc-MeVal)-OAll (5) according to an optimized procedure previously reported by Nagaswa et al. [14,15]. Compound 5 was subsequently condensed with N-Boc-methyl-Abu-OH affording the tripeptide 6. The next step involved the conjugation of N-Boc-L-Ala-OH with 6 to give tetradepsipeptide 7 in 60% yield over four steps. Half of the obtained compound 7 was then deallylated to give 8. Subsequently, N-Boc deprotection of 7 was followed by condensation with compound 8 using HATU to afford a linear octadepsipeptide 9. Compound 9 obtained by deallylation of 8 was subjected to intramolecular amide bond formation to produce cyclic peptide 10. N-Cbz deprotection of 10 was condensed with quinoxaline-2-carboxylic acid to obtain the target compound 1 in 55% yield. The synthetic procedure for compound 2 was similar to compound 1, except for the O-Bn-N-Boc-methyl -Ser-OH instead of N-Boc-methyl-Abu-OH (Scheme 3). All the spectroscopic data ( 1 H and 13 C NMR, HRESI-MS, UV, IR, and specific rotation) and UPLC-UV-MS retention times (two different eluting conditions) of the synthetic-1 and -2 matched those of the isolated natural products, respectively (Tables 1 and 4, Figures S31-S38 and Table S1).   Compounds 1-2 were evaluated for antibacterial against 37 different bacterial pathogens and cytotoxicity against the H460 lung cancer cells. Compared with the positive control echinomycin, no prominent antibacterial (MIC > 32 µg/mL) and cytotoxic (IC 50 > 1000 nM) activities were observed in compounds 1-2 (Tables S2 and S3), indicating that the cross-linking through a bridge bond may play a key role in the biological effects of quinomycin-type depsipeptides. This result was consistent with the findings from the previous report [15].

Actinomycetia Material
The strain Streptomyces sp. B475 was isolated from soil collected from the Maowei Sea Mangrove Reserve, Qinzhou City, Guangxi Zhuang Autonomous Region, China, in July 2017. The 16s rRNA gene sequence data were submitted to GenBank with accession NO. MN199475. Its sequence was similar to that of Streptomyces seoulensis NRRL B-24310 T (identity: 99.93%). Accordingly, this strain was identified at the genus level as Streptomyces sp. The strain was deposited at the Department of Microbial Chemistry, Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences and Peking Union Medical College.

Fermentation, Extraction, and Isolation
The strain was grown and maintained on an agar plate with ISP2 medium (glucose 4 g/L, yeast extract 4 g/L, malt extract 10 g/L, distilled water 1 L, pH = 7.2) at 28 • C for 7-10 days. The spores of the strain were inoculated into 500 mL Erlenmeyer flasks containing 100 mL of ISP2 medium at 28 • C for three days with shaking at 180 rpm. The 6 L (60 × 100 mL) of fermentation broth was centrifuged at 4300 rpm for 20 min, and the supernatant was extracted three times with ethyl acetate (6 L/time) to give an organic extract. After 20 fermentations, the combined organic extract (8.

ECD Calculation of Compound 1
Conformational search and geometry optimizations for 1a were performed on Molclus 1.9.9 program [26] by invoking xtb 6.3.3 program [27], Gaussian 16 packages [28], and ORCA 4.2.1 program [29] as described previously [30][31][32]. Firstly, the original structure was used to perform molecular dynamics (MD) simulations in the xtb 6.3.3 program with GFN0-xTB [33], the thermostat temperature was set at 400 K, and the total run time of simulation was 150 ps. These conformers were subjected to semi-empirical geometry optimization using the GFN0-xTB and GFN2-xTB [34] with the GBSA model in the MeOH method successively, the xTB geometries with a difference of distance geometries and energies within 0.5 were clustered by the isostat module in the Molclus program. Then, the clustered geometries within an energy window of 3 kcal/mol were subjected to a DFT geometry optimization and frequency analyses at B3LYP-D3(BJ)/6-31G level of theory with DFT-D3 dispersion correction [35] using Gaussian 16 program, and subsequently subject to ORCA program for calculating high precision single point energy at RI-PWPB95-D3(BJ)/def2-TZVPP level with SMD solvent model in MeOH. Conformer's clustering, their relative Gibbs Free Energy, and Boltzmann distribution in room temperature (298.15 K) were obtained from the isostat module in the Molclus program. Those conformers with Boltzmann distribution over 2% population were subjected to subsequent calculations. ECD calculation was performed using the simplified time-dependent density functional theory (sTD-DFT) approach [21] at the wB97X-D3/def2-SV(P) level of theory in MeOH with the SMD model in the ORCA program. The calculated ECD curves of each conformer and their Boltzmann-weighed ECD curves were generated using Multiwfn 3.7 software [36]. Finally, the ECD of its enantiomer 1b was generated and plotted together with the experimental ECD of 1, calculated ECD of 1a using Origin 2018 software (OriginLab, Northampton, MA, USA).

Advanced Marfey's Analysis of Compounds 1-2
Analysis was carried out following the published method [22]. Samples of 1-2 (each 0.12 mg) were dissolved in 6M HCl (0.2 mL) and heated to 110 • C in a sealed vial for 16 h. The hydrolysates were concentrated under dry N 2 at 40 • C, dissolved in 0.1 mL H 2 O, and divided into two portions (each 50 µL). Each portion was treated with 20 µL of 1M NaHCO 3 , and then we added 100 µL of 1% L-FDLA and 1% D-FDLA in acetone, respectively. The mixtures were vortexed and incubated at 37 • C for 1 h. After cooling to room temperature, the reaction mixture was quenched by adding 20 µL of 1M HCl and diluted with 200 µL of CH 3 CN. A total of 20 µL of the resulting solutions were diluted again with 800 µL of CH 3 CN and then centrifuged at 12,000 rpm for 10 min. The supernatants (1 µL) were analyzed by UPLC-HRESI-MS using an ACQUITY UPLC BEH C18 column (2.1 × 100 mm, 1.7 µm, 0.3 mL/min) with gradient elution (solvent A: water with 0.1% acetic acid; solvent B: CH 3  The N-Cbz-D-Ser(N-Boc-N-Me-L-Val)-OAll 5 (1.18 g, 2.4 mmol, 1.1 equiv) was dissolved in 4 M HCl/dioxane (10 mL, 60 mmol) at 0 • C. Then the resulting solution was warmed to room temperature and stirred at that temperature for 2 h. The mixture was diluted with EtOAc (200 mL) and sat. NaHCO 3 aq. (100 mL). The two layers were separated, and the organic layer was washed with brine (100 mL) and dried over Na 2 SO 4 , and the solvent was removed under reduced pressure to obtain the residue as a white solid. 1  Then, to a solution of the residue (4 g, 10.19 mmol), N-Boc-N-Me-L-Abu -OH (2.21 g, 10.19 mmol, 1 equiv), and DIEA (1.86 g, 11.21 mmol) in DMF (25 mL) was added HATU (0.91 g, 3.3 mmol, 1.5 equiv) at 0 • C and stirred at room temperature overnight. The mixture was diluted with EtOAc (150 mL) and water (150 mL). The two layers were separated, and the organic layer was washed with HCl (1 mol/L, 100 mL) and brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with n-hexane: EtOAc (80:20, v/v) to afford the target compound 6 as a colorless oil (5 g, 82.9% yield). 1 (7) To a solution of N-Cbz-D-Ser[N-Boc-N-Me-L-Abu-N-Me-L-Val]-OAll 6 (4 g, 6.76 mmol) in DCM (9 mL) was added TFA (3 mL) at 0 • C. Then the resulting solution was warmed to room temperature and stirred for 1 h. The solvent was removed under reduced pressure, then the residue was diluted with EtOAc (200 mL) and sat. NaHCO 3 aq. (100 mL). The two layers were separated, and the organic layer was washed with brine (100 mL) and dried over Na 2 SO 4 ; the solvent was removed under reduced pressure to get the residue as a white solid. 1 4, 34.8, 29.9, 29.6, 28.4, 26.8, 25.8, 20.2, 18.4, 17.5, 16.3, 8.4.
Then, to a solution of the residue (2.5 g, 5.09 mmol), N-Boc-N-L-Ala-OH (1.06 g, 5.09 mmol, 1 equiv), and DIEA (1.01mL, 6.01 mmol) in DMF (15 mL) was added HATU (2.12 g, 5.59 mmol) at 0 • C and stirred at room temperature overnight. The mixture was diluted with EtOAc (150 mL) and water (150 mL). The two layers were separated, and the organic layer was washed with HCl (1 mol/L, 100 mL) and brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with n-hexane: EtOAc (50:50, v/v) to afford the target compound 7 as a white solid (2.8 g, 83.1% yield). (1 g, 1.51 mmol), Pd(PPh 3 ) 4 (3.5 mg) in DCM (10 mL) was added morpholine (0.46 ml, 5.28 mmol) and stirred at room temperature for 1 h. Then the mixture was diluted with EtOAc (100 mL) and sat. NH 4 Cl aq. (100 mL). The two layers were separated, and the water layer was extracted with AcOEt (100 mL). The combined organic layers were washed with brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with DCM: MeOH (100:1, v/v) to afford the target compound 8 as a white solid (800 mg, 85.2% yield). The two layers were separated, and the water layer was extracted with EtOAc (100 mL). The combined organic layers were washed with brine (100 mL) and dried over Na 2 SO 4 ; the solvent was removed under reduced pressure. The residue was dissolved in TFA: DCM (1:3, 12 mL) at room temperature and stirred for 1 h. The solvent was removed under reduced pressure. The residue was dissolved in DMF: DCM (1:9, 100 mL) and adjusted to pH 7 with DIEA. The HOAt (60.8 mg, 0.446 mmol) and EDCI (85.7 mg, 0.446 mmol) were added to the mixture at 0 • C and then warmed to room temperature and stirred for 24 h. The mixture was washed with HCl (1 mol/L, 100 mL), sat. NaHCO 3 (100 mL) and brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel using DCM: MeOH (100:1, v/v) as the mobile phase to afford the target compound 10 as a colorless oil (200 mg, 40.6% yield).
3.6.6. Cyclic Peptide (1) A solution of Cbz-cyclic peptide 10 (50 mg, 0.05 mmol) and 20% Pd(OH) 2 /C (5 mg, 10 wt %) in MeOH (5 mL) was stirred at room temperature under an H 2 atmosphere (1 atm) for 12 h. Then, the mixture was filtered, and the filtrate was concentrated under reduced pressure. To a solution of 2-quinoxalinecarboxylic acid (17 mg, 0.1 mmol, 2 equiv) in DMF (2 mL) was added the residue dissolved in DMF (1 mL). After cooling to 0 • C, The HATU (36 mg, 0.1 mmol, 2 equiv) and DIEA (16 µL, 0.1 mmol, 2 equiv) were added to the mixture, and the solution was stirred at room temperature for 2 h. The mixture was diluted with CH 2 Cl 2 (50 mL) and washed with water (50 mL). The organic layer was separated, washed with brine (50 mL), dried over Na 2 SO 4 , and concentrated under reduced pressure. The residue was purified by pre-HPLC eluted with ACN-H 2 O (50:50 to 60:40, v/v, 0-30 min, 2.0 mL/min) to afford the target compound 1 (10 mg, 20% yield) as a white solid. 1 H NMR (DMSO-d 6 , 600 MHz) and 13   41 mmol) at 0 • C and stirred at room temperature overnight. The mixture was diluted with EtOAc (150 mL), and the organic layer was washed with HCl (1 mol/L, 100 mL) and brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with n-hexane: EtOAc (70:30, v/v) to afford the target compound 11 as a white solid (3.5 g, 66.9% yield). 52 mmol) in DCM (9 mL) was added TFA (3 mL) at 0 • C. Then the resulting solution was warmed to room temperature and stirred for 1 h. The solvent was removed under reduced pressure, and then the residue was diluted with EtOAc (200 mL) and sat. NaHCO 3 aq. (100 mL). The two layers were separated, and the organic layer was washed with brine (100 mL) and dried over Na 2 SO 4 , and the solvent was removed under reduced pressure to get the residue as a white solid. Then to a solution of the residue (2.2 g), N-Boc-N-L-Ala-OH (942 mg, 4.98 mmol, 1 equiv), and DIEA (0.82 mL, 4.98 mmol) in DMF (15 mL) was added HATU (1.89 g, 4.98 mmol) at 0 • C and stirred at room temperature overnight. The mixture was diluted with EtOAc (150 mL), and the organic layer was washed with HCl (1 mol/L, 100 mL) and brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with n-hexane: EtOAc (50:50, v/v) to afford the target compound 12 as a colorless oil (2.5 g, 72.0% yield).  5 g, 1.51 mmol), Pd(PPh 3 ) 4 (26 mg) in DCM (10 mL) was added morpholine (0.21 mL, 2.31 mmol) and stirred at room temperature for 1 h. Then the mixture was diluted with EtOAc (100 mL) and sat. NH 4 Cl aq. (100 mL). The two layers were separated, and the water layer was extracted with EtOAc (100 mL). The combined organic layers were washed with brine (100 mL) and dried over Na 2 SO 4 . The solvent was removed under reduced pressure, and the residue was purified by flash column chromatography on silica gel eluted with DCM: MeOH (100:1, v/v) to afford the target compound 13 (1.3 g, 91.5% yield).
12022. Levofloxacin and echinomycin were used as the positive controls. The medium of the agar dilution method was Mueller-Hinton agar. Suspensions of each microorganism were prepared to contain approximately 10 6 colony forming units (CFU)/mL and applied to plates with serially diluted compounds to be tested by multipoint inoculator and then incubated at 37 • C overnight. The MIC was considered as the lowest concentration that completely inhibited the growth on agar plates.

Conclusions
In conclusion, chemical investigation of the Beibu Gulf mangrove-derived Streptomyces sp. B475 led to the isolation of two novel quinomycin-type octadepsipeptides, quinomycins K (1) and L (2), which were characterized by the loss of intra-peptide disulfide or thioacetal bridge. Their planar structures and absolute configurations were elucidated by detailed NMR, MS spectroscopic analyses, calculated ECD analyses, advanced Marfey's method, and total syntheses. Compounds 1 and 2 did not show potent antibacterial and antitumor activities but displayed the structure-activity relationship of quinomycins. The presence of cross-linking through a bridge bond should contribute to the potent biological activity of quinomycin-type octadepsipeptides.