Abstract
Utilizing a binding mode-based physicochemical screening method using d-Ala-d-Ala silica gel, two new macrolactams, named banglactams A (1) and B (2), were discovered from the culture broth of Nonomuraea bangladeshensis K18–0086. In the course of our investigation, we found that d-Ala-d-Ala silica gel precisely differentiated the chemical structures of banglactams and separated them. However, we were not able to obtain enough of 1 to elucidate the structure due to its instability and insolubility. To overcome this challenge, we chemically modified 1 to improve solubility, enabling us to obtain a sufficient material supply for the indirect determination of the structure. Antibacterial activity evaluation of banglactams revealed that 1 binding to d-Ala-d-Ala silica gel exhibited antibacterial activity against Staphylococcus aureus; however, this was not the case with 2. This research indicates the utility of our original binding mode-based PC screening method, and the combination strategy of PC and chemical modifications led us to discover novel antibacterial compounds.
Introduction
Thousands of natural products have been discovered from microbial sources so far. Natural products have facilitated the development of novel drugs including agrochemicals such as pyripyropene and avermectin.1−3 As a general approach for screening of drug candidates, bioassay-guided fractionation has been employed for decades4 and several bioactive compounds have been identified based on their potent biological activities. However, the discovery of new natural products has progressively decreased due to the repurification of known bioactive compounds.5 In contrast to this situation, the cutting edge of whole-genome sequencing technology has shown that diverse and unexplored biosynthetic gene clusters are encoded in many microorganisms.6,7 They are now being recognized as a potential microbial source of new agrochemicals and drugs. To address this issue, we have worked on physicochemical (PC) screening8 to discover new natural compounds guided by physicochemical properties instead of their biological activities. This strategy is opposite to bioassay-guided fractionation and is a superior method for efficient discovery of new compounds. Our research group has discovered several new compounds, such as hatsusamides,9 pochoniolides,10 and shikinefragalides,11 based on this concept. After their initial discovery, many of them were found to show useful biological activities, and staurosporine12 was demonstrated to be a pan protein kinase inhibitor,13 leading to the innovation of an antineoplastic molecular target drug, midostaurin.14 These successes encouraged us to create a new PC screening method. Vancomycin is a broad-spectrum antibiotic, used very commonly in the hospital to treat methicillin-resistant Staphylococcus aureus (MRSA) infections and empirically in severe infections of unknown cause, that binds to the d-Ala-d-Ala moiety of the bacterial cell wall.15 This exceptional binding model inspired us to design and synthesize d-Ala-d-Ala silica gels as a new PC screening method.16 We established the method which enabled us to purify vancomycin from the culture broth of a vancomycin-producing strain, Amycolatopsis orientalis.16 In this paper, we describe our original PC screening, which led us to discover two new macrolactams, named banglactams A (1) and B (2) (Figure 1), from the culture broth of Nonomuraea bangladeshensis K18–0086. Throughout our investigation, we came across challenges in the isolation and characterization of 1 due to its instability and extremely low solubility. A chemical modification approach enabled us to isolate acetyl-banglactam A (3), facilitating the determination of the structure of 1.
Figure 1.
Structure of banglactams.
Results and Discussion
Screening of Our In-House Microbial Secondary Metabolites Library Using a d-Ala-d-Ala Silica Gel and Isolation of 1
We screened our microbial library using d-Ala-d-Ala silica gel and found a hit culture broth of N. bangladeshensis K18–0086 (Figure S2). According to the procedure reported by us,16 we detected a compound (Peak A) as a major component, which bound to d-Ala-d-Ala silica gel and was eluted with a solution of MeOH containing 50 mM TFA (Scheme 1). Based on liquid chromatography–mass spectrometry (LC–MS) analysis of peak A, the molecular formula of the target compound was proposed as m/z 473.3005, and we attempted to isolate the compound to characterize the structure from the 12-day-old culture broth (6 L) of the strain guided by LC-ESI-DAD-MS. The MeOH extract of the mycelial cake was separated by an HP20 column and purified by high-performance liquid chromatography (HPLC) (Scheme S1 and Figure S3–1) to afford the target compound, banglactam A (1, 0.9 mg), as a yellow powder. The molecular formula of 1 was determined as C27H40N2O5 by high-resolution-ESI-MS (HR-ESI-MS) analysis (m/z 473.3005, calculated for C27H41N2O5 [M + H]+ 473.3010), requiring nine degrees of unsaturation (Figure S4–1–1). During our investigations, the compound was found to be unstable, especially under light, and demonstrated extremely low solubility. Attributing to these physicochemical properties, we could not obtain a sufficient amount of the compound to fully elucidate the structure. However, it seemed that 1 was still contained in the mycelial cake; therefore, we decided to modify the compound to improve the solubility.
Scheme 1. Separation Scheme of Peak A and the HPLC Analysis.
Derivatization of the Target Compound for the Structure Elucidation
Based on HR-ESI-MS and infrared (IR) analyses, the presence of hydroxyl and amine groups in the compound was anticipated. As a first trial, the introduction of the tert-butoxycarbonyl group, which is a well-known protecting group for amines, was investigated. As the reaction time passed, the substrate gradually decomposed, and the corresponding product was not obtained. Due to the unexpectedly low reactivity of the nucleophilic functional groups, we chose a more reactive reagent, acetic anhydride, which successfully progressed to finally furnish acetylated product 3 (Scheme 2). To our delight, this chemical modification enabled us to obtain 20.8 mg of 3 by simple preparative TLC (Figure S3–2). Interestingly, this transformation also provided a new analogue of 3, Ac-banglactam B (4) as a minor product (<1 mg). Acetylation dramatically improved the solubility of 1 as we expected, and we advanced to the elucidation of the structure of 3. The molecular formula of compounds 3 and 4 was determined as C33H46N2O8 by HR-ESI-MS analysis (m/z 599.3302, calculated for C33H47N2O8 [M + H]+ 599.3327) and C33H45NO9 by HR-ESI-MS analysis (m/z 600.3178, calculated for C33H46NO9 [M + H]+ 600.3167), requiring 12 degrees of unsaturation (Figure S4–3–1 and S4–4–1), which indicated compounds 3 and 4 should be triacetylated from compounds 1 and 2. The 1H NMR spectrum of compound 3 in CDCl3 (Table 1 and Figure S4–3–4) revealed one amide NH proton signal (δH 7.65), 12 olefinic proton signals (δH 7.05, 6.40, 6.34, 6.32, 6.23, 6.10, 6.06, 6.01, 5.80, 5.68, 5.61, and 5.03), three doublet methyl signals (δH 1.13, 1.01, and 0.97), and three singlet methyl signals (δH 2.17, 2.16, and 1.96). The 13C NMR and HSQC spectra of compound 3 (Figure S4–3–5 and S4–3–7) indicated the presence of four amide or ester carbonyl carbons (δC 170.7, 170.6, 169.5, and 167.3), 12 olefinic carbons (δC 139.1, 138.4, 134.5, 134.3, 133.1, 132.4, 132.0, 131.9, 130.5, 127.3, 126.9, and 121.3), one anomeric sp3 methine carbon (δC 99.4), four oxygen-bearing sp3 methine carbons (δC 85.1, 70.1, 69.4, and 65.4), one nitrogen-bearing sp3 methine carbon (δC 44.9), one nitrogen-bearing sp3 methylene carbon (δC 46.2), two sp3 methylene carbons (δC 36.3 and 36.0), two sp3 methine carbons (δC 37.6 and 31.5), and six methyl carbons (δC 23.2, 21.2, 20.8, 18.6, 16.9, and 16.4).
Scheme 2. Chemical Modification of the Mycelial Cake.
Table 1. 1H and 13C NMR Data of 2′O, 3′N, and 4′O-Triacetylbanglactam A (3) Measured in CDCl3.
| 2′O, 3′N, 4′O-triacetylbanglactam A (3) |
||||
|---|---|---|---|---|
| position | δCa, type | δH (mult., J in Hz)b | HMBC | COSY |
| 1 | 167.3, C | – | ||
| 2 | 121.3, CH | 5.61 (d, 11.4) | 1, 3, 4 | 3 |
| 3 | 139.1, CH | 6.32 (dd, 11.5, 11.4) | 1, 2, 4, 5 | 2, 4 |
| 4 | 127.3, CH | 7.05 (dd, 14.7, 11.5) | 2, 3, 5, 6 | 3, 5 |
| 5 | 138.4, CH | 6.34 (m) | 3, 4, 6, 7 | 4, 6 |
| 6 | 134.3, CH | 6.23 (dd, 15.2, 10.8) | 4, 5, 7, 8 | 5, 7 |
| 7 | 131.9, CH | 5.80 (m) | 5, 6, 8, 9 | 6, 8 |
| 8 | 36.3, CH2 | 2.55 (m) | 6, 7, 9, 10 | 7, 9 |
| 9 | 85.1, CH | 3.40 (m) | 7, 8, 10, 11, 20, 1′ | 8, 10 |
| 10 | 37.6, CH | 1.47 (m) | 8, 9 11, 12, 20 | 9, 11, 20 |
| 11 | 36.0, CH2 | 1.69 (m) | 9, 10, 12, 13, 20 | 10, 12 |
| 2.07 (m) | 10, 12 | |||
| 12 | 132.0, CH | 5.68 (ddd, 14.0, 12.2, 3.9) | 10, 11, 13, 14 | 11, 13 |
| 13 | 132.4, CH | 6.01 (dd, 12.2, 11.2) | 11, 12, 14, 15 | 12, 14 |
| 14 | 133.1, CH | 6.10 (m) | 12, 13, 15, 16 | 13, 15 |
| 15 | 126.9, CH | 6.40 (dd, 14.8, 11.2) | 13, 14, 16, 17 | 14, 16 |
| 16 | 130.5, CH | 6.06 (dd, 11.2, 10.8) | 14, 15, 17, 18 | 15, 17 |
| 17 | 134.5, CH | 5.03 (dd, 10.8, 10.7) | 15, 16, 18, 19, 21 | 16, 18 |
| 18 | 31.5, CH | 3.35 (m) | 16, 17, 19, 21 | 17, 19, 21 |
| 19 | 46.2, CH2 | 2.49 (m) | 1, 17, 18, 21 | 18, NH-1 |
| 3.69 (ddd, 13.6, 7.1, 3.9) | 18, NH-1 | |||
| 20 | 16.9, CH3 | 1.01 (d, 6.6) | 9, 10, 11 | 10 |
| 21 | 18.6, CH3 | 0.97 (d, 6.6) | 17, 18, 19 | 18 |
| NH-1 | – | 5.65 (m) | 1, 2, 18, 19 | 19 |
| 1′ | 99.4, CH | 4.90 (br. s) | 9, 2′, 3′, 5′ | |
| 2′ | 69.4, CH | 4.78 (br. d, 3.7) | 1′, 3′, 4′ | |
| 3′ | 44.9, CH | 4.65 (ddd, 8.6, 3.7, 3.7) | 1′, 2′, 4′, 5′ | |
| NH-3′ | – | 5.76 (br. d, 8.6) | 2‴ | |
| 4′ | 70.1, CH | 5.11 (br. d, 3.7) | 2′, 3′, 5′, 6′ | |
| 5′ | 65.4, CH | 4.23 (br. q, 6.5) | 1′, 3′, 4′, 6′ | 6′ |
| 6′ | 16.4, CH3 | 1.13 (d, 6.5) | 4′, 5′ | 5′ |
| 1″ | 21.2, CH3 | 2.16 (s)c | 2″ | 3″, NH-2″ |
| 2″ | 170.6, C | – | 3″, NH-2″ | |
| 1‴ | 23.2, CH3 | 1.96 (s) | 2‴ | |
| 2‴ | 169.5, C | – | 3‴ | |
| 1‴′ | 20.8, CH3 | 2.17 (s)c | 2‴′ | 2‴, 4‴ |
| 2‴′ | 170.7, C | – | ||
Measured at 100 MHz.
Measured at 400 MHz.
Overlapped.
Analysis of the 1H–1H COSY and HMBC spectra (Figures 2, S4–3–6, and S4–3–8) revealed the presence of a 20-membered macrolactam ring and a sugar moiety as shown in Figure 1.
Figure 2.
Structure elucidation of 2′O, 3′N, and 4′O-triacetylbanglactam A (3).
The macrolactam ring was elucidated by COSY cross-peaks of H-2 (δH 5.61)/H-3 (δH 6.32)/H-4 (δH 7.05)/H-5 (δH 6.34), H-6 (δH 6.23)/H-7 (δH 5.80)/H2-8 (δH 2.55)/H-9 (δH 3.40)/H-10 (δH 1.47)/ H3-20 (δH 1.01), H2-11 (δH 2.07 and 1.69)/H-12 (δH 5.68)/H-13 (δH 6.01), H-16 (δH 6.06)/H-17 (δH 5.03), H-18 (δH 3.38)/H3-21 (δH 0.97), and H2-19 (δH 3.69 and 2.49)/NH-1 (δH 5.65), as well as cross-peaks in HMBC from NH-1 to C-1 (δC 167.3), from H-2 to C-1, from H-4 to C-6 (δC 134.3), from H-5 to C-7 (δC 131.9), from H-6 to C-4 (δC 127.3), from H-7 to C-5 (δC 138.4), from H3-20 to C-11 (δC 36.0), from H-14 (δH 6.10) to C-16 (δC 130.5), from H-15 (δH 6.40) to C-13 (δC 132.4) and C-17 (δC 134.5), from H-16 to C-14 (δC 133.1), and from H3-21 to C-17 and C-19 (δC 46.2). The geometry of the six double bonds in the macrolactam ring was determined to be 2Z, 4E, 6E, 12Z, 14E, and 16Z by 1H–1H coupling constants (JH-2/H-3 = 11.4 Hz, JH-4/H-5 = 14.7 Hz, JH-6/H-7 = 15.2 Hz, JH-12/H-13 = 12.2 Hz, JH-14/H-15 = 14.8 Hz, and JH-16/H-17 = 10.8 Hz).
The sugar moiety was established to be a 2′O, 3′N, 4′O-triacetylated 3′-amino-6-deoxypyranose by COSY cross-peaks of H-1′ (δH 4.90)/H-2′ (δH 4.78)/H-3′ (δH 4.65)/H-4′ (δH 5.11)/H-5′ (δH 4.23)/H3-6′ (δH 1.13) and H-3′/NH-3′ (δH 5.76) combined with HMBC cross-peaks from H-1′ to C-5′ (δC 65.4), from H-2′ to C-2″ (δC 170.6), from H3-1″ (δH 2.16) to C-2″, from NH-3′ to C-2‴ (δC 169.5), from H3-1‴ (δH 1.96) to C-2‴, from H-4′ to C-2‴′ (δC 170.7), from H3-1‴′ (δH 2.17) to C-2‴′, and from H3-6′ to C-4′ (δC 70.1). The downfield-shifted 13C chemical shift (δC 99.6) of C-1′ suggested that this sugar moiety has a β-glycosidic bond. The relative configuration of this sugar moiety was established to be 1′R*, 2′R*, 3′S*, 4′S*, and 5′S* by 1H-1H coupling constants between 1′/2′ (0 Hz), 2′/3′ (3.7 Hz), 3′/4′ (3.7 Hz), and 4′/5′ (0 Hz) (Table 1).
Finally, cross-peaks in HMBC from H-1′ to C-9 (δC 85.1) led us to elucidate the planar structure of compound 3 as a 20-membered macrolactam having a 2′O, 3′N, 4′O-triacetylated 3′-amino-6-deoxypyranose moiety at the C-9 position, and we designated compound 3 as 2′O, 3′N, 4′O-triacetylbanglactam A (Ac-banglactam A).
Isolation and the Structure Elucidation of 1
Based on this information, we were able to indirectly determine the structure of banglactam A (1) through the structure elucidation of 3, and the discovery of 4 by the chemical modification made us realize the existence of banglactam B (2), which made us to isolate 2.
Comparing the structural difference between 3 and 4 based on each mass value, we envisioned the proposed structure of 2, where the 3′ position is substituted with a OH group compared to 1. We also assumed that the OH group in 2 might improve its solubility and aimed to isolate 2 and directly determine the structure. In fact, the MeOH extract of the mycelial cake was separated by an HP20 column (Scheme S1) and purified by HPLC (Figure S3–1) to afford compound 2 (4.5 mg). As our assumption, 2 is soluble in DMSO, which enabled us to elucidate the structure. Banglactam B (2) was isolated as a yellow powder. The molecular formula of compound 2 was determined as C27H39NO6 by HR-ESI-MS analysis (m/z 474.2861, calculated for C27H40NO6 [M+H]+ 474.2850), requiring nine degrees of unsaturation (Figure S4–2–1).
The 1H NMR spectrum in DMSO-d6 (Table 2 and Figure S4–2–2) revealed one amide NH proton signal (δH 7.94), 12 olefinic proton signals (δH 7.18, 6.34, 6.31, 6.28, 6.23, 6.03, 6.00, 5.95, 5.74, 5.66, 5.61, and 5.02), and three doublet methyl signals (δH 1.10, 0.90 and 0.86). The 13C NMR and HSQC spectra (Figure S4–2–3 and S4–2–5) indicated the presence of one amide carbonyl carbon (δC 166.4), 12 olefinic carbons (δC 138.3, 137.5, 135.5, 134.3, 132.5, 132.4, 132.0, 131.8, 129.8, 127.9, 127.2, and 122.2), one anomeric sp3 methine carbon (δC 103.4), five oxygen-bearing sp3 methine carbons (δC 83.6, 72.5, 70.7, 67.1, and 65.5), one nitrogen-bearing sp3 methylene carbon (δC 45.8), two sp3 methylene carbons (δC 35.8 and 34.9), two sp3 methine carbons (δC 37.7 and 30.4), and three methyl carbons (δC 18.4, 16.7, and 16.4).
Table 2. 1H and 13C NMR Data of Banglactam B (2) Measured in DMSO-d6.
| banglactam B (2) |
COSY | |||
|---|---|---|---|---|
| position | δCa, type | δH (mult., J in Hz)b | HMBC | TOCSY |
| 1 | 166.4, C | – | ||
| 2 | 122.2, CH | 5.61 (d, 11.2) | 1, 3, 4 | 3 |
| 3 | 138.3, CH | 6.23 (dd, 11.2, 11.5) | 1, 2, 4, 5 | 2, 4 |
| 4 | 127.9, CH | 7.18 (dd, 11.5, 11.5) | 2, 3, 5, 6 | 3, 5 |
| 5 | 137.5, CH | 6.31 (dd, 11.5, 11.1) | 3, 4, 6, 7 | 4, 6 |
| 6 | 134.3, CH | 6.34 (dd, 11.2, 11.1) | 4, 5, 7, 8 | 5, 7 |
| 7 | 132.0, CH | 5.74 (ddd, 14.8, 11.2, 4.0) | 5, 6, 8, 9 | 6, 8 |
| 8 | 35.8, CH2 | 2.49 (m) | 6, 7, 9, 10 | 7, 9 |
| 2.57 (ddd, 14.8, 14.0, 1.8) | 7, 9 | |||
| 9 | 83.6, CH | 3.31 (m) | 7, 8, 10, 11, 20, 1′ | 8, 10 |
| 10 | 37.7, CH | 1.19 (m) | 8, 9 11, 12, 20 | 9, 11, 20 |
| 11 | 34.9, CH2 | 1.53 (ddd, 12.6, 12.6, 2.9) | 9, 10, 12, 13, 20 | 10, 12 |
| 2.16 (dd, 12.6, 11.4) | 10, 12 | |||
| 12 | 131.8, CH | 5.66 (ddd, 13.0, 11.4, 2.9) | 10, 11, 13, 14 | 11, 13 |
| 13 | 132.4, CHc | 6.03 (d, 13.0, 10.3)c | 11, 12, 14, 15 | 12, 14 |
| 14 | 127.2, CH | 6.28 (dd, 12.6, 10.3) | 12, 13, 15, 16 | 13, 15 |
| 15 | 132.5, CHc | 6.00 (d, 12.6, 10.5)c | 13, 14, 16, 17 | 14, 16 |
| 16 | 129.8, CH | 5.95 (dd, 10.8, 10.5) | 14, 15, 17, 18 | 15, 17 |
| 17 | 135.5, CH | 5.02 (dd, 10.8, 10.7) | 15, 16, 18, 19, 21 | 16, 18 |
| 18 | 30.4, CH | 3.38 (m) | 16, 17, 19, 21 | 17, 19, 21 |
| 19 | 45.8, CH2 | 2.37 (ddd, 12.4, 11.0, 5.2) | 1, 17, 18, 21 | 18, NH-1 |
| 3.30 (m) | 18, NH-1 | |||
| 20 | 16.4, CH3 | 0.90 (d, 6.0) | 9, 10, 11 | 10 |
| 21 | 18.4, CH3 | 0.86 (d, 6.6) | 17, 18, 19 | 18 |
| NH-1 | – | 7.94 (br. dd, 5.2, 5.2) | 1, 2, 18, 19 | 19 |
| 1′ | 103.4, CH | 4.74 (br. s) | 9, 2′, 3′, 5′ | |
| 2′ | 70.7, CH | 3.53 (br. s) | 1′, 3′, 4′ | |
| 3′ | 65.5, CH | 3.54 (br. d, 2.4) | 1′, 2′, 4′, 5′ | |
| 4′ | 72.5, CH | 3.45 (br. s) | 2′, 3′, 5′, 6′ | |
| 5′ | 67.1, CH | 3.85 (br. q, 6.5) | 1′, 3′, 4′, 6′ | 6′ |
| 6′ | 16.7, CH3 | 1.10 (d, 6.5) | 4′, 5′ | 5′ |
Measured at 150 MHz.
Measured at 600 MHz.
Exchangeable.
Analysis of the 1H-1H COSY and HMBC spectra (Figures 3, S4–2–4, and S4–2–6) revealed the presence of a 20-membered macrolactam ring and a sugar moiety as shown in Figure 1.
Figure 3.
Structure elucidation of banglactam B (2).
The macrolactam ring was deduced by COSY cross-peaks of H-2 (δH 5.61)/H-3 (δH 6.23)/H-4 (δH 7.18)/H-5 (δH 6.31)/H-6 (δH 6.34)/H-7 (δH 5.74)/H2-8 (δH 2.57 and 2.49)/H-9 (δH 3.31)/H-10 (δH 1.19)/H2-11 (δH 2.16 and 1.53)/H-12 (δH 5.66)/H-13 (δH 6.03)/H-14 (δH 6.28)/H-15 (δH 6.00) /H-16 (δH 5.95)/H-17 (δH 5.02)/H-18 (δH 3.38)/H2-19 (δH 3.30 and 2.37)/NH-1 (δH 7.94), H-10/H3-20 (δH 0.90), and H-18/H3-21 (δH 0.86), as well as cross-peaks in HMBC from H-2 to C-1 (δC 166.4) and from NH-1 to C-1. The geometry of six double bonds in the macrolactam ring was determined to be all Z by small 1H–1H coupling constants (JH-2/H-3 = 11.2 Hz, JH-4/H-5 = 11.5 Hz, JH-6/H-7 = 11.2 Hz, JH-12/H-13 = 13.0 Hz, JH-14/H-15 = 12.6 Hz, and JH-16/H-17 = 10.8 Hz).
The sugar moiety was established to be a 6-deoxypyranose by COSY cross-peaks of H-1′ (δH 4.74)/H-2′ (δH 3.53) and H-5′ (δH 3.85)/H3-6′ (δH 1.10) combined with HMBC cross-peaks from H-1′ to C-5′ (δC 67.1), from H-2′ to C-4′ (δC 72.5), from H-3′ (δH 3.54) to C-4′, from H-4′ (δH 3.45) to C-2′ (δC 70.7), from H-5′ (δH 3.85) to C-1′ (δC 103.4), and from H3-6′ to C-4′. The downfield-shifted 13C chemical shift (δC 103.4) of C-1′ suggested that this sugar moiety has a β-glycosidic bond. The relative configuration of this sugar moiety was established to be 1′S*, 2′R*, 3′S*, 4′S*, and 5′S* by 1H–1H coupling constants between 1′/2′ (0 Hz), 2′/3′ (0 Hz), 3′/4′ (2.4 Hz), and 4′/5′ (0 Hz) (Table 2).
Finally, cross-peaks in HMBC from H-1′ to C-9 (δC 83.6) and H-9 to C-1′ led us to elucidate the planar structure of compound 2 as a 20-membered macrolactam having a 6-deoxypyranose moiety at the C-9 position, and we designated compound 2 as banglactam B.
Antibacterial Activity Evaluation of Banglactams
The discovery of banglactams prompted us to evaluate their antibacterial activity against S. aureus, which has the d-Ala-d-Ala moiety in the growing cell wall. In this investigation, we sought the relevance of the binding ability of each compound and their antibacterial activity. Compounds 1–3 were tested for antibacterial activity against S. aureus on a paper disk method (Figure S5). Compound 1 displayed antibacterial activity with inhibition zones of 8.0 mm/30 μg and 7.5 mm/10 μg, respectively, while compounds 2 and 3 showed no antibacterial activity even at 30 μg/disk. This result is consistent with their binding ability to d-Ala-d-Ala silica gel. Considering the structure difference between 1 and 2, either or both of the 3′-NH2 moiety and the olefin geometries in the 20-membered macrolactam ring would influence the binding ability to d-Ala-d-Ala silica gel resulting in the exhibition of the antibacterial activity of 1 but not that of 2. It was also suggested that d-Ala-d-Ala silica gel would be able to precisely recognize the chemical structures of binding compounds, which also suggests the reliability of our previous binding experiments using antibiotics such as daptomycin.16
In conclusion, we were fascinated by PC screening as a powerful method for the discovery of novel natural compounds and have established a new binding mode-based PC screening method using d-Ala-d-Ala silica gel. In this report, we displayed the utility of our method based on the discovery of new natural products, banglactams A (1) and B (2). Throughout their structure analysis, we encountered the issue attributed to the instability and insolubility of 1 and conducted the chemical modification to change the physicochemical property. Acetylation of the mycelial cake containing banglactams dramatically improved the solubility of 1 and made us realize the existence of 2 as a natural product, enabling the determination of the chemical structure of 1. We believe that the combination strategy of the binding mode-based PC screening method using d-Ala-d-Ala silica gel and chemical modifications pioneers a new approach in the natural product screening research area. We are currently synthesizing new banglactam A derivatives to improve the stability and solubility, and the result will be reported in due course.
Experimental Section
General Experimental Procedures
Reagents and solvents were commercially available and used without further purification. For thin-layer chromatography (TLC) analysis, precoated silica gel plates with a fluorescent indicator (Merck 60 F254, Merck KgaA, Darmstadt, Germany) were used. The Diaion HP20 resin (Mitsubishi Chemical Co., Ltd., Tokyo, Japan) was used for open column chromatography. An HPLC column, Pegasil ODS SP100 (5 μm, 20 i.d. × 250 mm), was purchased for isolation from Senshu Science Co., Ltd. (Tokyo, Japan). Liquid chromatography–ultraviolet (LC-UV) data were taken using system (Agilent Technologies, Santa Clara, CA) with an HPLC column, Symmetry C18 3.5 μm (2.1 i.d. × 150 mm, Waters Co., Ltd., Milford, MA). LC-ESI-MS data were obtained using an AB Sciex Triple TOF 5600+ LC-MS/MS Systems (AB Sciex, Framingham, MA) with a reversed-phase HPLC column, Capcell Core C18 (2.7 μm, 3.0 i.d. × 100 mm, Osaka Soda Co., Ltd., Osaka, Japan). NMR spectra were measured in DMSO-d6 or CDCl3 using an XL-400 spectrometer (Agilent Technologies) or a JNM-ECA-500 spectrometer (JEOL Ltd., Tokyo, Japan) or an AVANCE III HD-600 (Bruker, Billerica, MA) with 1H NMR at 400 or 500 or 600 MHz and 13C NMR at 100 or 125 or 150 MHz, respectively. The chemical shifts are expressed in ppm and are referred to as DMSO-d6 (2.48 ppm) or CDCl3 (7.26 ppm) in the 1H NMR spectra and DMSO-d6 (39.5 ppm) or CDCl3 (77.0 ppm) in the 13C NMR spectra, respectively. Infrared (IR) spectra were recorded using a JASCO FT-/IR-4600 spectrometer. UV spectra were measured with a U-2800 spectrophotometer (Hitachi Ltd., Tokyo, Japan). Optical rotation was measured with a JASCO P-2200 polarimeter (JASCO Corporation, Tokyo, Japan).
Taxonomic Studies of Strain K18–0086
Strain K18–0086 was isolated from a cinnamon root collected in Ome, Tokyo, Japan. The strain grew well in International Streptomyces Project (ISP) medium 2, forming orange colonies with white aerial mycelia (Figure S2). The 16S rRNA gene sequence analysis was performed according to our method.17 The strain was identified as N. bangladeshensis based on a high similarity value of 99.9% to N. bangladeshensis 5–10–10T (AB274966).
Fermentation and Isolation
We cultured and maintained strain K18–0086 on agar medium (1.0% starch, 0.3% N-Zamine, 0.1% yeast extract, 0.1% meat extract, 0.3% CaCO3, 1.2% agar, and distilled water at pH 7.0). As seed culture, we inoculated a loop of strain K18–0086 into 100 mL of seed medium (2.4% starch, 0.1% glucose, 0.3% peptone, 0.3% of 35% Ehrlich bonito extract, 0.5% yeast extract, and 0.4% CaCO3) (adjusted to pH 7.0 before sterilization) in a 500 mL Erlenmeyer flask. The flask was shaken at 210 rpm for 3 days at 27 °C. We inoculated the seed culture (1 mL) into 100 mL of production medium (2.0% soluble starch, 1.0% defatted wheat germ, 0.5% glycerol, 0.3% dry yeast, 0.3% of 35% Ehrlich bonito extract, and 0.3% CaCO3) in a 500 mL Erlenmeyer flask (total 60 flasks, 6 L), which was incubated on a rotary shaker at 210 rpm for 12 days at 27 °C.
Compounds 1 and 2 were isolated from a 12-day-old cultured material of N. bangladeshensis K18–0086 by guided with UV and MS profiles using LC-DAD-ESI-MS analysis (Scheme S1). They were purified under light-shielded conditions due to their light sensitivity. The fermentation broth (6 L) was centrifuged to separate into the mycelial cake and supernatant. To the obtained mycelial cake was added MeOH (10 l), and the mixture was filtrated in vacuo. To the obtained MeOH extract solution was added H2O (15 l), and the mixture was applied to a Diaion HP20 open column (1 L resin, Mitsubishi Chemical Co., Tokyo, Japan) equilibrated with water. The column was eluted stepwise with 80% MeOH aq (3 L) and 100% MeOH (3 L). A part (2 L) of 80% MeOH aq. fraction stocked at 4 °C was concentrated in vacuo to remove MeOH and extracted with EtOAc (2 L) several times. The obtained EtOAc layer was concentrated in vacuo to remove EtOAc and was added a small amount of MeOH. A part (5 mL) of MeOH aq. solution (20 mL) was subjected to reversed-phase HPLC using a Pegasil ODS SP100 column (20 i.d. × 250 mm, Senshu Scientific Co., Tokyo, Japan) with an isocratic solvent system of CH3CN-H2O-TFA (45:55:0.1) at a flow rate of 7.0 mL min–1 detected by UV 254 nm. The two fractions with retention times of 13 and 26 min were collected (Figure S3–1), evaporated in vacuo, and freeze-dried to afford banglactams A (1, 0.9 mg) and B (2, 4.5 mg), respectively.
Banglactam A (1)
Yellow powder; 1H, see Figure S4–1–4; HR-ESI-MS m/z 473.3005 [M + H]+ (calcd for C27H41N2O5, 474.2850).
Banglactam B (2)
Yellow powder; 1H and 13C NMR data, see Table 2; HR-ESI-MS m/z 474.2861 [M + H]+ (calcd for C27H40NO6, 474.2850).
Preparation and Isolation of 2′O, 3′N, 4′O-Triacetylbanglactam A (3)
To the ODS column eluate of 100% MeOH fraction (1.5 L, Scheme S1) was added H2O, concentrated in vacuo, and then freeze-dried under the light-shielded condition. The obtained powder was dissolved in pyridine (2.5 mL) and were added 4-dimethylaminopyridine (2 mg) and acetic anhydride at 0 °C. After being stirred for 16 h at room temperature, the mixture was added to EtOAc (25 mL) and 1 M HCl (25 mL). The EtOAc extract was purified by preparative TLC developed with EtOAc, and the fraction (Rf = 0.5) was extracted with MeOH/CHCl3 (1:1) and concentrated in vacuo to yield 2′O,3′N,4′O-triacetylbanglactam A (3) with 20.8 mg as a yellow powder.
2′O, 3′N, 4′O-Triacetylbanglactam A (3)
[α]D21 −89.6 (c = 0.05, CHCl3); IR (ATR, diamond prism) νmax 2936, 1660, 1371, 1200, 1129, 1026, 799, 722, 529, 493, 451, 430, 418, 407 cm–1; 1H and 13C NMR see Table 1; HR-ESI-MS m/z 599.3302 [M + H]+ (calcd for C33H47N2O8, 599.3327).
Antimicrobial Assay in Paper Disk Method
Test strain of S. aureus (MSSA FDA-209P), stored at – 80 °C in microbanks (IWAKI&Co., Ltd.), was inoculated into 5 mL of Mueller-Hinton Broth (BD&Co.) and incubated at 37 °C for 20 h with shaking. The test plates were prepared by mixing 1% of the seed culture into 30 mL of Mueller-Hinton Agar (BD&Co.) with 0.05% BSA in a 10–14 cm plate.
Paper disks (6 mm, thin, Advantec Toyo Kaisha, Ltd., Tokyo, Japan) soaked with 10 μL of each sample were placed on the test plates. After incubation at 37 °C overnight, the diameter (mm) of the inhibition zone was measured. Paper disk soaked with 10 μL of 0.3 mg/mL vancomycin aqueous solution was used as a positive control.
Acknowledgments
The authors are grateful to Distinguished Emeritus Professor Satoshi O̅mura (Kitasato University) for his helpful support and valuable guidance and suggestions. They thank Dr. Kenichiro Nagai, Reiko Seki, and Noriko Sato (School of Pharmacy, Kitasato University) for various instrumental analyses. This research was partially supported by Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research [BINDS]) from AMED under Grant Number JP21am0101096 (Phase I) and JP22ama121035 (Phase II). This study was supported by funds from the Institute for Fermentation, Osaka (IFO), Japan.
Supporting Information Available
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c00145.
PC screening using d-Ala-d-Ala silica gel; taxonomy of K18–0086 strain; isolation of banglactams A (1) and B (2) and derivatization from 1 to 3; spectral data of banglactams A (1) and B (2) and triacetylbanglactams A (3) and B (4); and antibacterial activity on paper disk method (PDF)
The authors declare no competing financial interest.
Supplementary Material
References
- Newman D. J.; Cragg G. M. Natural products as sources of new drugs over the nearly four decades from 01/1981 to 09/2019. J. Nat. Prod. 2020, 83 (3), 770–803. 10.1021/acs.jnatprod.9b01285. [DOI] [PubMed] [Google Scholar]
- Kandasamy R.; London D.; Stam L.; von Deyn W.; Zhao X.; Salgado V. L.; Nesterov A. Afidopyropen: New and potent modulator of insect transient receptor potential channels. Insect Biochem. Mol. Biol. 2017, 84, 32–39. 10.1016/j.ibmb.2017.03.005. [DOI] [PubMed] [Google Scholar]
- Burg R. W.; Miller M. B.; Baker E. E.; Birnbaum J.; Currie A. S.; Hartman R.; Kong Y-L.; Monaghan L. R.; Olson G.; Putter I.; Tunac B. J.; Wallick H.; Stapley O. E.; Oiwa R.; O̅mura S. Avermectins, New Family of Potent Anthelmintic Agents: Producing Organism and Fermentation. Antimicrob. Agents Chemother. 1979, 15, 361–367. 10.1128/AAC.15.3.361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Weller M. G. A unifying review of bioassay-guided fractionation, effect-directed analysis and related techniques. Sensors 2012, 12 (7), 9181–9209. 10.3390/s120709181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bérdy J. Thoughts and facts about antibiotics: where we are now and where we are heading. J. Antibiot. 2012, 65 (8), 385–395. 10.1038/ja.2012.27. [DOI] [PubMed] [Google Scholar]
- Keller N. P. Fungal secondary metabolism: regulation, function and drug discovery. Nat. Rev. Microbiol. 2019, 17 (3), 167–180. 10.1038/s41579-018-0121-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kikuchi Y.; Kawashima M.; Iwatsuki M.; Kimishima A.; Tsutsumi H.; Asami Y.; Inahashi Y. Comprehensive analysis of biosynthetic gene clusters in bacteria and discovery of Tumebacillus as a potential producer of natural products. J. Antibiot. 2023, 76 (6), 316–323. 10.1038/s41429-023-00609-y. [DOI] [PubMed] [Google Scholar]
- Nakashima T.; Takahashi Y.; O̅mura S. Search for new compounds from Kitasato microbial library by physicochemical screening. Biochem. Pharmacol. 2017, 134, 42–55. 10.1016/j.bcp.2016.09.026. [DOI] [PubMed] [Google Scholar]
- Miyano R.; Matsuo H.; Nonaka K.; Mokudai T.; Niwano Y.; Shiomi K.; Takahashi Y.; O̅mura S.; Nakashima T. Pochoniolides A and B, new antioxidants from the fungal strain Pochonia chlamydosporia var. spinulospora FKI-7537. J. Biosci. Bioeng. 2018, 126 (5), 661–666. 10.1016/j.jbiosc.2018.05.003. [DOI] [PubMed] [Google Scholar]
- Koike R.; Watanabe Y.; Kato K.; Mikasa Y.; Shimizu K.; Nonaka K.; Honsho M.; Suga T.; Hokari R.; Ishiyama A.; Asami Y.; Iwatsuki M. Shikinefragalides A-D, new tricyclic macrolides produced by Stachybotryaceae sp. FKI-9632. J. Antibiot. 2022, 75 (4), 199–206. 10.1038/s41429-022-00512-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Izuta S.; Kosaka S.; Kawai M.; Miyano R.; Matsuo H.; Matsumoto A.; Nonaka K.; Takahashi Y.; O̅mura S.; Nakashima T. Dipyrimicin A and B, microbial compounds isolated from Amycolatopsis sp. K16–0194. J. Antibiot. 2018, 71 (5), 535–537. 10.1038/s41429-018-0028-0. [DOI] [PubMed] [Google Scholar]
- O̅mura S.; Iwai Y.; Hirano A.; Nakagawa A.; Awaya J.; Tsuchya H.; Takahashi Y.; Masuma R. A new alkaloid AM-2282 of Streptomyces origin Taxonomy, fermentation, isolation and preliminary characterization. J. Antibiot. 1977, 30 (4), 275–282. 10.7164/antibiotics.30.275. [DOI] [PubMed] [Google Scholar]
- Tamaoki T.; Nomoto H.; Takahashi I.; Kato Y.; Morimoto M.; Tomita F. Staurosporine, a potent inhibitor of phospholipid/Ca++dependent protein kinase. Biochem. Biophys. Res. Commun. 1986, 135 (2), 397–402. 10.1016/0006-291X(86)90008-2. [DOI] [PubMed] [Google Scholar]
- Lena A.; Toppino A.; Miele K.; Roletto J.; Paissoni P.. Process for the Preparation of Midostaurin with High Purity. WO Patent, WO2020200945A12023.
- Perkins H. R. Vancomycin and related antibiotics. Pharmacol. Ther. 1982, 16 (2), 181–197. 10.1016/0163-7258(82)90053-5. [DOI] [PubMed] [Google Scholar]
- Kimishima A.; Suzuki Y.; Kosaka S.; Honsho M.; Honma S.; Sakai K.; Tsutsumi H.; Inahashi Y.; Watanabe Y.; Iwatsuki M.; Asami Y.; Nakashima T. Design and synthesis of d-Ala-d-Ala silica gel for a binding mode-based physicochemical screening method. ACS Omega 2023, 8 (42), 39035–39040. 10.1021/acsomega.3c03837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Inahashi Y.; Matsumoto A.; Danbara H.; O̅mura S.; Takahashi Y. Phytohabitans suffuscus gen. nov., sp. nov., an actinomycete of the family Micromonosporaceae isolated from plant roots. Int. J. Syst. Evol. Microbiol. 2010, 60, 2652–2658. 10.1099/ijs.0.016477-0. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.