Racemic trinorsesquiterpenoids from the Beihai sponge Spongia officinalis: structure and biomimetic total synthesis

Dong-Yu Sunab, Guan-Ying Hanbc, Na-Na Yanga, Le-Fu Lana, Xu-Wen Li*a and Yue-Wei Guo*ad
aState Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zu Chong Zhi Road, Zhangjiang Hi-Tech Park, Shanghai 201203, China. E-mail: xwli@simm.ac.cn; ywguo@simm.ac.cn
bSchool of Pharmaceutical Sciences, Jinzhou Medical University, Jinzhou 121001, China
cThe First Affiliated Hospital of Jinzhou Medical University, 2 Ren Min Road, Jinzhou 121001, China
dOpen studio for druggability research of marine lead compounds, Qingdao National Laboratory for Marine Science and Technology, 1 Wenhai Road, Aoshanwei, Jimo, Qingdao, Shandong 266237, China

Received 5th December 2017 , Accepted 7th January 2018

First published on 8th January 2018


Two rare new furan butanolides, namely sponalisolides A (1) and B (2), characterized by an unprecedented furan-bearing trinorsesquiterpene alkyl chain connecting either a butanolide (1) or an N-acyl homoserine lactone moiety (2), were isolated in racemic forms from the Beihai sponge Spongia officinalis, and were further separated, by chiral-phase HPLC, to their corresponding enantiomers 1a/1b and 2a/2b, respectively. The structures, including the absolute stereochemistry, of the two pairs of enantiomeric compounds, were unambiguously established by extensive spectroscopic analysis and biomimetic total synthesis, involving a key Johnson-Claisen rearrangement and a lactone cyclization after epoxidation or dihydroxylation. All the new compounds exhibited the Pseudomonas aeruginosa quorum sensing inhibitory activity.


Introduction

The marine sponge Spongia officinalis (order Dictyoceratida, family Spongiidae) is a prolific source of diverse terpenoids, including spongian diterpenoids, C-21 furanoterpenoids, acyclic furanosesterterpenoids, and scalarane sesterterpenoids.1 Interestingly, these metabolites exhibit broad biological properties, such as cytotoxic, antibacterial, antifungal, analgesic and muscle relaxant effects, etc.1,2 A literature survey revealed that most of the chemically investigated sponge specimens of S. officinalis were collected from the Mediterranean Sea,1 and until now, only one report discusses the chemical constituents of title sponge from the South China Sea, with only several known sterols being isolated.3 As part of our ongoing projects searching for bioactive natural products from Chinese benthic invertebrates,4 the title animal was encountered off the coast of Weizhou island, Beihai, Guangxi Zhuang Autonomous Region, China. Chemical investigation of the n-butanol-soluble portion of the acetone extract of this sponge has led to the isolation of two novel furano-trinorsesquiterpenoids, named sponalisolides A (1) and B (2) (Fig. 1). The racemic nature of both 1 and 2 was recognized by optical rotation measurements and successive chiral resolution via chiral-phase HPLC, affording two pairs of enantiomers 1a/1b and 2a/2b, respectively. Herein, we report the isolation, chiral separation and structural characterization of all the new isolates, and the biomimetic total synthesis of 1a, 1b, 2a, 2b and the unnatural epimers 1c and 1d, as well as their biological evaluation.
image file: c7qo01091e-f1.tif
Fig. 1 Structures of compounds 1–3.

Results and discussion

The frozen sponge specimens (dry weight 510 g) were cut into pieces and exhaustively extracted with acetone. The n-butanol-soluble portion of the acetone extract was repeatedly chromatographed over silica gel columns to yield compounds 1 (3.2 mg) and 2 (5.0 mg), respectively.

Compound 1 was isolated as an optically inactive colorless oil. Its molecular formula was established as C12H16O4 by HREI-MS (m/z [M+] 224.1048, calcd for 224.1049). The IR absorptions at νmax 3448 and 1763 cm−1 indicated the presence of hydroxyl and ester carbonyl groups in the molecule. The aforementioned functionalities were confirmed by the 13C-NMR spectrum, showing diagnostic signals at δC 75.1 (–CH–OH), and δC 177.1 (C), 88.9 (C) (lactone). Moreover, the 1H- and 13C-NMR data, in combination with HSQC spectroscopic analysis, suggested the presence of a β-substituted furan ring by the characteristic signals at δH 7.37 (t, 1H, J = 1.6 Hz), 7.26 (brs, 1H), 6.29 (brs, 1H), and δC 143.2 (CH), 139.2 (CH), 124.2 (C), 111.0 (CH) (Table 1).1 Other carbon signals were ascribable to one methyl at δC 23.1 and four methylenes at δC 31.4, 29.5, 27.6, and 21.5, respectively (Table 1). Further detailed analysis of the 1H–1H COSY spectrum readily identified three partial structures ac by the clear direct proton-proton correlations of H-1/H-2 (a) on the furan ring; H2-5 (2H, δH 2.73, 2.55)/H2-6 (2H, δH 1.69, 1.61)/H-7 (1H, δH 3.71) connecting to the hydroxyl (b); and H-10 (2H, δH 2.43, 1.81)/H-11 (2H, δH 2.64, 2.62) (c), respectively (Fig. 2). The position of these subunits, bearing in mind three unassigned quaternary carbons, one tertiary and one primary carbons, was connected as shown in Fig. 2 by the well resolved HMBC correlations from H2-5 and H2-6 to C-3 (δC 124.2), from H-7 to C-5 (δC 21.5) and C-8 (δC 88.9), from H3-9 (δH 1.35) and H2-10 (δH 2.43, 1.81) to C-7 (δC 75.1), from H3-9 to C-10 (δC 27.6), and from H2-10 to C-12 (δC 177.1), to complete the planar structure of 1 as depicted.


image file: c7qo01091e-f2.tif
Fig. 2 1H–1H COSY and key HMBC correlations of compounds 1 and 2.
Table 1 1H (400 MHz) and 13C (125 MHz) NMR Data (δ in ppm, J in Hz) for compounds 1a–1d, 2a and 2b in CDCl3
No. 1a and 1b 1c and 1d 2a and 2b
δH, mult (J, Hz) δC δH, mult (J, Hz) δC δH, mult (J, Hz) δC
1 7.37 t (1.6) 143.2 (CH) 7.37 t (1.6) 143.2 (CH) 7.33 t (1.6) 142.7 (CH)
2 6.29 brs 111.0 (CH) 6.28 brs 110.9 (CH) 6.26 brs 111.1 (CH)
3 124.2 (C) 124.1 (C) 124.9 (C)
4 7.26 brs 139.2 (CH) 7.26 brs 139.3 (CH) 7.20 brs 139.0 (CH)
5 2.73 ddd (14.2, 9.2, 4.8); 2.55 m 21.5 (CH2) 2.72 m; 2.54 m 21.0 (CH2) 2.44 dd (7.7, 7.3) 24.9 (CH2)
6 1.69 m; 1.61 m 31.4 (CH2) 1.68 m 31.4 (CH2) 2.24 ddd (14.6, 7.3, 7.0) 28.5 (CH2)
7 3.71 dd (10.6, 2.2) 75.1 (CH) 3.54 dd (7.5, 5.1) 76.2 (CH) 5.22 t (7.0) 125.2 (CH)
8 88.9 (C) 88.9 (C) 134.2 (C)
9 1.35 s 23.1 (CH3) 1.36 s 21.4 (CH3) 1.60 s 16.1 (CH3)
10 2.43 ddd (12.8, 10.1, 8.6); 1.81 ddd (12.8, 8.6, 6.6) 27.6 (CH2) 2.15 dt (12.8, 9.0); 1.92 ddd (12.8, 9.6, 5.5) 29.2 (CH2) 2.33 m 35.1 (CH2)
11 2.64 m; 2.62 m 29.5 (CH2) 2.62 m 30.7 (CH2) 2.33 m 34.9 (CH2)
12 177.1 (C) 176.6 (C) 173.5 (C)
1′ 175.6 (C)
2′ 4.52 ddd (11.7, 8.6, 5.8) 49.4 (CH)
3′ 2.82 ddd (12.2, 8.6, 5.8); 2.08 qd (11.7, 9.1) 30.7 (CH2)
4′ 4.45 t (9.5); 4.27 ddd (11.3, 9.5, 5.8) 66.2 (CH2)
NH 6.16 brs


Compound 2 was also obtained as an optically inactive colorless oil, and its molecular formula was determined to be C16H21NO4 by HRESI-MS (m/z [M + Na+] 314.1362, calcd for 314.1363). The presence of a terminal β-substituted furan ring, like 1, was immediately recognized by directly comparing its NMR data with those of compound 1 (Table 1). The presence of a trisubstituted olefin in 2 is also obvious as evidenced by diagnostic 1H- and 13C-NMR resonances (δH 5.22, t, 1H, J = 6.9 Hz; δC 134.2, C; 125.2, CH). Other 1H- and 13C-NMR signals, assisted with DEPT and HSQC experiments, were assigned to two carbonyls (δC 175.6, 173.5), one oxygenated methylene (δH 4.45, t, 1H, J = 9.1 Hz; 4.27, ddd, 1H, J = 11.3, 9.1, 5.9 Hz; δC 66.2), one nitrogenated methine (δH 4.52, ddd, 1H, J = 11.7, 8.6, 5.8 Hz; δC 49.4), and five normal methylenes (Table 1), indicating the presence of an amide and an α-substituted γ-lactone moiety. All above identified fragments were connected, as depicted in 2, by 2D NMR experiments with characteristic 1H–1H COSY and HMBC correlations shown in Fig. 2. The E configuration of Δ7,8 in 2 was deduced by the lack of NOE correlation between H-7 and H3-9. Furthermore, the presence of a butanolide moiety connecting to an amide group at the C-2′ position was confirmed by direct comparison of the corresponding 1H and 13C NMR data with those of N-acyl homoserine lactones (N-AHLs).5 Thus, the planar structure of 2 was also identified.

To complete the entire structure, the next step was to determine the stereochemistry at C-7, C-8 of 1 and C-2′ of 2. Since both 1 and 2, as mentioned above, are respective racemic mixtures, chiral-phase HPLC was then applied to separate each of them. As expected, (+)-sponalisolide A (1a, 1.4 mg) and (−)-sponalisolide A (1b, 1.2 mg) were afforded from 1 with [α]D values of +7.0 (c 0.05, CHCl3) and −5.3 (c 0.1, CHCl3), respectively, whereas (+)-sponalisolide B (2a, 1.1 mg) and (−)-sponalisolide B (2b, 1.2 mg) were obtained from 2 with [α]D values of +6.0 (c 0.05, CHCl3) and −1.3 (c 0.05, CHCl3), respectively. Although two pairs of optically pure enantiomers of 1 and 2 were obtained, it was still a challengeable task to determine their relative configuration (RC). As depicted in structures 1 and 2, the flexibility of the molecules prevented the determination of the RC for the chiral centers at C-7 and C-8 in each enantiomer of 1, and C-2′ in 2 by the NOE experiment. Moreover, the efforts to crystallize 1 and 2 were also unsuccessful. Based on these facts, it appears that the best solution was to synthesize each of them so as to conclude the structure assignments of these new compounds.

The retrosynthetic analysis of 1 and 2 disclosed that both compounds are structurally reminiscent of the model compound 3 (dendrolasin), a furanosesquiterpenoid previously isolated from the Fiji sponge Lasius fuliginosus (Fig. 1).6 Actually, 3 might be the common precursor of 1 and 2. Biogenetically, as outlined in Scheme 1, an oxidative cleavage of the olefin at Δ12,13 of 3 would generate the key carboxylic acid intermediate (4), which could then either be amidated by reacting with D or L-homoserine lactone (5) towards 2a/2b, or underwent the Δ7,8 epoxidation to afford the intermediate 6, of which, lactonization between its C-12 and C-8, accompanying the epoxide ring opening at C-8, would yield compounds 1a/1b.


image file: c7qo01091e-s1.tif
Scheme 1 Plausible biosynthetic pathway of 1 and 2.

Inspired by this biogenetic analysis, the synthetic efforts towards all the four new compounds were carried out as shown in Scheme 2. Starting from the commercially available furfural (7), a Wittig reaction proceeded in the presence of ethyl (triphenylphosphoranylidene)acetate in dichloromethane (DCM) under reflux over 16 h, affording ethyl (E)-3-(furan-3-yl)acrylate (8) with 81% yield. It was then treated with Pd-CaCO3 in MeOH at 40 °C to produce nearly quantitative ethyl 3-(furan-3-yl)propanoate (9), which further underwent a DIBAL-H reduction towards the furan aldehyde (10) in 80% yield.7 The secondary alcohol 11 was obtained in 55% yield by the Grignard reaction between 10 and isopropenylmagnesium bromide in Et2O at −10 °C.7 Afterwards, a key Johnson-Claisen orthoester rearrangement was carried out by treating 11 with triethyl orthoacetate in 10% propanoic acid at 150 °C to generate the intermediate 12 in 70% yield.8


image file: c7qo01091e-s2.tif
Scheme 2 Biomimetic synthetic route of 1 and 2.

With the pivotal furan ester 12 on hand, the initial biomimetic synthetic plan was to install the five-membered lactone by an asymmetric dihydroxylation of the olefin Δ7,8, followed by an intramolecular transesterification towards 1a and 1b (Scheme 2). To realize this goal, 12 was treated with AD-mix β and AD-mix α, respectively, by following the Sharpless asymmetric dihydroxylation procedure,9 giving the corresponding intermediates 14a/14b (Scheme 3), which underwent an immediate cascade transesterification on the basic condition in one pot, yielding two end products (Scheme 2).10 These two synthesized compounds, identified as the 7,8-erythro geometric products 1c [7R,8R, [α]D +7.5 (c 0.5, CHCl3)] and 1d [7S,8S, [α]D −13.3 (c 0.3, CHCl3)], respectively, were proved not to be the expected 1a and 1b, due to their obviously different NMR signals at C-7 from those of 1a/1b. In fact, the chemical shifts of H-7 and C-7 were significantly shifted from δH 3.71 (J = 10.6, 2.2 Hz) and δC 75.1 in 1a/1b to δH 3.54 (J = 7.5, 5.1 Hz) and δC 76.2 in 1c/1d (Table 1), suggesting that the RCs at C-7 and C-8 of 1a/1b should be threo (7R,8S/7S,8R) geometry. In light of this observation, we had to explore another solution so as to accomplish the total synthesis of target compounds 1a and 1b. Thus, 12 was first treated with m-CPBA to give a racemic mixture of 7R,8R (13a) and 7S,8S (13b) epoxides in 90% yield, which was then intramolecularly transesterified, in the presence of hydrochloric acid in MeOH, to the racemic mixture of target compounds 1a and 1b (Scheme 2). This mixture was further separated by the chiral-phase HPLC with the same condition as the natural products, showing the same retention time in the HPLC column, the identical NMR data, and the similar [α]D values [+7.4 (c 0.3, CHCl3) for synthetic 1a (7R,8S); −6.3 (c 0.1, CHCl3) for synthetic 1b (7S,8R)] as the corresponding natural products 1a and 1b, respectively (see below for stereochemical assignments).


image file: c7qo01091e-s3.tif
Scheme 3 Mechanism of the oxidation and cyclization from the furan ester 12 to all the isomers 1a–1d (compound 12 was drawn according to the mechanism model of Sharpless dihydroxylation).

As shown in Scheme 3, the mechanisms of the non-stereoselective epoxidation followed by the intramolecular transesterification towards 1a/1b, and those of the two cascade reactions comprising the Sharpless asymmetric dihydroxylation towards 1c/1d, reasonably explained the established absolute stereochemistry of all the four isomers of 1. Moreover, to secure the above defined absolute configuration (AC) of 1a/1b and 1c/1d, the modified Mosher's method was applied to confirm the AC of 7-OH on the synthetic 1a and 1c, respectively.11 As displayed in Fig. 3, the R configuration at the C-7 position was demonstrated for both diastereoisomers 1a and 1c.


image file: c7qo01091e-f3.tif
Fig. 3 Result of modified Mosher's method for compounds 1a and 1c.

Comparatively, the synthesis of compound 2 was relatively simple. The hydrolysis of 12 gave the key intermediate 4, which then reacted with L- and D-homoserine lactone hydrochloride, respectively, in the presence of DCC, DMAP in DCM, yielding 86% of 2a [2′S, [α]D +7.5 (c 0.5, CHCl3)] and 79% of 2b [2′R, [α]D −10.9 (c 0.5, CHCl3)], respectively (Scheme 2).12 The identity of synthetic and natural products was proven by the same retention time in the chiral HPLC column, the identical NMR data, and similar [α]D values.

All the natural products (1a, 1b, 2a and 2b) and their synthetic diastereoisomers (1c and 1d) were subjected to various bioassays, such as cytotoxic, antibacterial and immunological activity tests. Unfortunately, none of them showed obviously positive effects on the abovementioned assays. Nevertheless, in the bacteria quorum sensing (QS) inhibitory activity assay, we do have observed some positive effects. In order to determine the Pseudomonas aeruginosa QS inhibitory potential of these six compounds, a LasR inhibition assay based on a lasA promoter-lux transcriptional fusion (p-lasA-lux) in P. aeruginosa PAO1 was used. lasA codes for the protease involved in proteolysis and elastolysis and is under the transcriptional control of LasR.13 As shown in Fig. 4, we found that compounds 1a, 1b, 2a and 2b significantly decreased the expression of p-lasA-lux fusion, respectively. Additionally, these compounds did not affect the bacterial growth at the indicated concentrations (Fig. 4). These observations suggest that the new natural products may possess inhibitory activity against LasR and functioned as P. aeruginosa QS inhibitors.


image file: c7qo01091e-f4.tif
Fig. 4 Effect of 1a, 1b, 2a, and 2b on lasA promoter activity (left) and growth (right). The PAO1/p-lasA-lux treated with indicated concentrations of compounds (A: 250 μg mL−1, B: 125 μg mL−1) were grown in LB at 37 °C and promoter activities were measured as counts per second (CPS) of light production with a Synergy 2 Multi-Mode Microplate Reader. Values represent means ± standard error of the mean (SEM) and each value was performed with triplicate biological replicates.

Conclusions

In summary, although furanoterpenoids are a class of frequently encountered natural products in marine invertebrates,14 this type of metabolites containing the butanolide motif was rarely reported. In particular, trinorsesquiterpenoids bearing both furan and butanolide moieties are unprecedented. Sponalisolides A (1) and B (2), herein reported, represent the first examples of such terpenoids in nature. Moreover, the successful total synthesis of each enantiomer of 1 and 2 by a biomimetic strategy not only allowed the determination of their AC, but also provided chemical evidence for the possible biosynthetic pathway of such intriguing secondary metabolites. In the bioassay, all the new compounds were found to exhibit inhibitory activity against LasR and functioned as P. aeruginosa QS inhibitors. Further studies should be conducted to understand the ecological roles played by these racemic natural products in the life cycle of the sponges for more targeted pharmacological investigations.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research work was financially supported by the Natural Science Foundation of China (No. 81520108028, 41476063, and 41676073), NSFC-Shandong Joint Fund for Marine Science Research Centers (No. U1606403), STCSM Project (No. 15431901000), and the SKLDR/SIMM Project (SIMM 1705ZZ-01). X.-W. Li acknowledges the financial support of “Youth Innovation Promotion Association” (Grant No. 2016258) from the Chinese Academy of Sciences, “Young Elite Scientists Sponsorship” from the China Association for Science and Technology (2016QNRC001), Shanghai “Pujiang Program” (No. 16PJ1410600), and SA-SIBS Scholarship Program.

Notes and references

  1. (a) A. G. Gonzalez, D. M. Estrada, J. D. Martin, V. S. Martin, C. Perez and R. Perez, Tetrahedron, 1984, 40, 4109–4113 CrossRef CAS; (b) G. Cimino, R. Morrone and G. Sodano, Tetrahedron Lett., 1982, 23, 4139–4142 CrossRef CAS; (c) G. Cimino, D. D. Rosa, S. D. Stefano and L. Minale, Tetrahedron, 1974, 30, 645–649 CrossRef CAS; (d) G. Cimino, S. D. Stefano, L. Minale and E. Fattorusso, Tetrahedron, 1971, 27, 4673–4679 CrossRef CAS; (e) G. Cimino, S. D. Stefano and L. Minale, Tetrahedron, 1972, 28, 5983–5991 CrossRef CAS; (f) A. De Giulio, S. De Rosa, G. Di Vincenzo and N. Zavodnik, J. Nat. Prod., 1989, 52, 1258–1262 CrossRef CAS; (g) L. Garrido, E. Zubía, M. J. Ortega and J. Salvá, J. Nat. Prod., 1997, 60, 794–797 CrossRef CAS; (h) E. Manzo, M. L. Ciavatta, G. Villani, M. Varcamonti, S. M. Sayem, R. Soest and M. Gavagnin, J. Nat. Prod., 2011, 74, 1241–1247 CrossRef CAS PubMed.
  2. R. D. Pasquale, C. Circosta, F. Occhiuto, S. D. Rosa and S. D. Stefano, Phytother. Res., 1991, 5, 49–53 CrossRef.
  3. Y. Wan, L. Zeng and J. Su, Chin. J. Mar. Drugs, 1997, 16, 1–3 CAS.
  4. (a) F. Ye, Z.-D. Zhu, J.-S. Chen, J. Li, Y.-C. Gu, W.-L. Zhu, X.-W. Li and Y.-W. Guo, Org. Lett., 2017, 19, 4183–4186 CrossRef CAS PubMed; (b) X.-W. Li, S.-H. Chen, F. Ye, E. Mollo, W.-L. Zhu, H.-L. Liu and Y.-W. Guo, Tetrahedron, 2017, 73, 5239–5243 CrossRef CAS; (c) X.-L. Li, T. Kurtán, J.-C. Hu, A. Mándi, J. Li, X.-W. Li and Y.-W. Guo, J. Agric. Food Chem., 2017, 65, 1550–1555 CrossRef CAS PubMed; (d) H.-L. Liu, D.-Q. Xue, S.-H. Chen, X.-W. Li and Y.-W. Guo, Helv. Chim. Acta, 2016, 99, 650–653 CrossRef CAS; (e) R.-Y. Huang, W.-T. Chen, T. Kurtán, A. Mándi, J. Ding, J. Li, X.-W. Li and Y.-W. Guo, Future Med. Chem., 2016, 8, 17–27 CrossRef CAS PubMed; (f) E. Manzo, M. L. Ciavatta, M. Gavagnin, E. Mollo, Y.-W. Guo and G. Cimino, J. Nat. Prod., 2004, 67, 1701–1704 CrossRef CAS PubMed; (g) Z.-F. Zhou, X. Wang, H. Zhang, J.-Y. Sun, L.-H. Zheng, H.-C. Liu, J.-D. Wang, A.-J. Shen, M.-Y. Geng and Y.-W. Guo, Acta Pharm. Sin. B, 2015, 5, 62–66 CrossRef PubMed.
  5. D. Jakubczyk, C. Merle, G. Brenner-Weiss, B. Luy and S. Bräse, Eur. J. Org. Chem., 2013, 5323–5330 CrossRef CAS.
  6. A. Quilico, F. Piozzi and M. Pavan, Tetrahedron, 1957, 1, 177–185 CrossRef CAS.
  7. C. J. Davis, T. E. Hurst, A. M. Jacob and C. J. Moody, J. Org. Chem., 2005, 36, 4414–4422 CrossRef PubMed.
  8. (a) M. B. Gravestock, W. S. Johnson, B. E. Mccarry, R. J. Parry and B. E. Ratcliffe, J. Am. Chem. Soc., 1978, 100, 4274–4282 CrossRef CAS; (b) S. Lochyński, B. Frąckowiak, T. Olejniczak, Z. Ciunik and C. Wawrzeńczyk, Tetrahedron: Asymmetry, 2002, 13, 1761–1767 CrossRef; (c) E. Redero, C. Sandoval and F. Bermejo, Tetrahedron, 2001, 57, 9597–9605 CrossRef CAS.
  9. I. Paterson, I. Boddy and I. Mason, Tetrahedron Lett., 1987, 28, 5205–5208 CrossRef CAS.
  10. Y. Kobayashi, S. Yoshida and Y. Nakayama, Eur. J. Org. Chem., 2001, 1873–1881 CrossRef CAS.
  11. (a) X.-W. Li, L. Weng, X. Gao, Y. Zhao, F. Pang, J.-H. Liu, H.-W. Zhang and J.-F. Hu, Bioorg. Med. Chem. Lett., 2011, 21, 366–372 CrossRef CAS PubMed; (b) I. Ohtani, T. Kusumi, Y. Kashman and H. Kakisawa, J. Am. Chem. Soc., 1991, 113, 4092–4096 CrossRef CAS.
  12. M. K. Singh, J. Waybright and Q.-S. Zhang, Tetrahedron, 2017, 73, 3677–3683 CrossRef CAS.
  13. (a) D. S. Toder, M. J. Gambello and B. H. Iglewski, Mol. Microbiol., 1991, 5, 2003–2010 CrossRef CAS PubMed; (b) B. A. Cowell, S. S. Twining, J. A. Hobden, M. S. F. Kwong and S. M. J. Fleiszig, Microbiology, 2003, 149, 2291–2299 CrossRef CAS PubMed.
  14. Y. Liu, S. Zhang and P. J. M. Abreu, Nat. Prod. Rep., 2006, 23, 630–651 RSC.

Footnotes

Electronic supplementary information (ESI) available. See DOI: 10.1039/c7qo01091e
Authors contributed equally to this work.

This journal is © the Partner Organisations 2018