Mollebenzylanols A and B, Highly Modified and Functionalized

27 mins ago - Two highly modified and functionalized diterpenoids, mollebenzylanols A (1) and B (2), and a known grayanane diterpenoid rhodojaponin II...
2 downloads 9 Views 1MB Size
Download PDF
Letter Cite This: Org. Lett. XXXX, XXX, XXX−XXX

pubs.acs.org/OrgLett

Mollebenzylanols A and B, Highly Modified and Functionalized Diterpenoids with a 9‑Benzyl-8,10-dioxatricyclo[5.2.1.01,5]decane Core from Rhododendron molle Junfei Zhou, Junjun Liu, Ting Dang, Haofeng Zhou, Hanqi Zhang, and Guangmin Yao* Hubei Key Laboratory of Natural Medicinal Chemistry and Resource Evaluation, School of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China S Supporting Information *

ABSTRACT: Two highly modified and functionalized diterpenoids, mollebenzylanols A (1) and B (2), and a known grayanane diterpenoid rhodojaponin III (3) were isolated from Rhododendron molle. Their structures were determined by spectroscopic data analysis, an electronic circular dichroism (ECD) exciton chirality method, ECD calculations, and X-ray diffraction analysis of the p-bromobenzoate ester of 1 (1a). Compounds 1 and 2 possess an unprecedented diterpene carbon skeleton featuring a unique 9-benzyl-8,10-dioxatricyclo[5.2.1.01,5]decane core, and their plausible biosynthetic pathways are proposed. Their PTP1B inhibitory activity and modes of action were investigated.

N

benzyl-8,10-dioxatricyclo[5.2.1.01,5]decane core. In the reported aromatic diterpenoids, the phenyl unit is usually fused with the carbon rings;2−7 however, there is no report of the free benzyl unit in the diterpene backbone. Compounds 1 and 2 represent the first examples of diterpenoids bearing a free benzyl unit in the diterpene backbone. Herein, the isolation, structure elucidation, biogenetic pathway, PTP1B inhibitory activity, and molecular dockings of 1 and 2 are described. Mollebenzylanol A (1) was obtained as a white amorphous powder, [α]20D + 26.5 (c 0.1, MeOH). The molecular formula of 1 was determined as C20H28O4 by a [M + H]+ ion at m/z 333.2049 (calcd for C20H29O4, 333.2066) in the high-resolution electrospray ionization mass spectrometry (HRESIMS) and 13C nuclear magnetic resonance (NMR) data, suggesting seven degrees of unsaturation. The 1H NMR spectrum of 1 (Table S1) exhibited resonances for four methyl singlets at δH 1.36 (s, H316), 2.31 (s, H3-17), 1.07 (s, H3-18), and 1.20 (s, H3-19), a methyl doublet at δH 1.24 (d, J = 7.0 Hz, H3-20), three oxygenated methines at δH 3.72 (dd, J = 7.5, 6.6 Hz, H-2), 3.50 (d, J = 6.6 Hz, H-3), and 3.93 (dd, J = 10.1, 1.7 Hz, H-6), and a 1,3-disubstituted benzene ring at δH 7.15 (t, J = 7.6 Hz, H-12), 7.07 (br. s, H-9), 7.01 (br. d, J = 7.6 Hz, H-11), and 7.04 (br. d, J = 7.6 Hz, H-13). The 13C NMR spectrum showed a total of 20 carbon resonances, assignable by DEPT and HSQC spectra as five methyls, one methylene, five methines (three oxygenated), and two quaternary carbons (an oxygenated), a ketal, a disubstituted benzene ring. A benzene ring accounts for four degrees of unsaturation, and the rest of three degrees of unsaturation required the existence of a tricyclic system in 1. The planar structure of 1 was deduced from 1H−1H correlation (COSY), heteronuclear single quantum correlation

atural products are potent sources for new drug discovery, and diterpenoids, such as taxol and ingenol mebutate, are the representative examples.1 Diterpenoids are widespread in nature, while aromatic diterpenoids bearing a multisubstituted benzene ring are relatively rare and mainly belong to abietane,2 icetexane,3 cassane,4 podocarpane,5 totarane,6 taiwaniaquinoid,7 and cleistanthane types.8 Due to their intriguing bioactivities, the total synthesis of aromatic diterpenoids has attracted interest from the synthetic community.2−7 Rhododendron molle G. Don (Ericaceae) is a well-known medicinal plant for treating pain, rheumatism, and asthma, as well as bruises,9 and its flowers, fruits, and roots have been proven to be a rich source of grayanane and related diterpenoids.10 However, phytochemical investigations on the leaves of R. molle are rare, and only a rhodomollane,11a three 2,3:5,6-di-secograyanane,11b and related grayanane diterpenoids were recently reported by our group.11 In a continuing search for novel tyrosine phosphatase 1B (PTP1B) inhibitors,11 the leaves of R. molle were investigated, leading to the isolation of two highly modified and oxygenated diterpenoids, mollebenzylanols A (1) and B (2) (Figure 1), as well as their plausible related biogenetic precursor rhodojaponin III (3).12 Compounds 1 and 2 possess an unprecedented diterpene carbon skeleton featuring a unique 9-

Figure 1. Structures of mollebenzylanols A (1) and B (2), and the 2,3bis-p-bromobenzoate ester of 1 (1a). © XXXX American Chemical Society

Received: February 19, 2018

A

DOI: 10.1021/acs.orglett.8b00606 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

The UV spectrum of 1a (Figure S1) exhibited a strong absorption at 245 nm (logε 3.64) attributable to the two bromobenzoyl groups. Consistent with this UV maximum absorption, the ECD spectrum (Figure 3) of 1a showed a

(HSQC), and heteronuclear multiple bond correlation (HMBC) spectra analyses (Figure 2). 1H−1H COSY spectrum revealed the

Figure 2. 1H−1H COSY, HMBC, and NOESY correlations of 1.

presence of three partial structures: (a) C(20)H3−C(15)H− C(1)H−C(2)H−C(3)H, (b) C(6)H−C(7)H 2 , and (c) C(11)H−C(12)H−C(13)H. In the HMBC spectrum of 1, the cross peaks from H3-18/H3-19 to C-3, C-4, and C-5 indicated the direct connections of C-3, C-5, C-18, and C-19 toward the quaternary carbon C-4. The connection of C-1 and C-6 through C-5 was established by the HMBC correlations from H-1 to C-5/ C-6 and from H-6 to C-1/C-4/C-5. HMBC correlations from H3-16 to C-14/C-15 and from H3-20 to C-14/C-15 established the connections of C-15 and C-16 to C-14. The connection of C7 and C-8 was established by HMBC correlations from H2-7 to C-8/C-9/C-13. The HMBC correlations from H3-17 to C-9/C10/C-11 combined with 1D NMR data confirmed the presence of a 3-methyl-benzyl unit in 1. The large chemical shifts of C-5 (δC 99.2), C-6 (δC 82.0), and C-14 (δC 111.1) suggested the presence of two oxygen bridges between C-5 and C-14, and between C-6 and C-14, respectively, which were supported by the molecular formula and the resulted degree of saturation of 1. Thus, mollebenzylanol A (1) was established to be 3,4dihydroxyl-2,2,6,7-tetramethyl-9-(3-methyl-benzyl)-8,10dioxatricyclo[5.2.1.01,5]decane (Figure 2). The relative configuration of compound 1 was determined by nuclear Overhauser effect spectroscopy (NOESY) analysis (Figure 2). H-1 was randomly assigned as α-orientation. NOESY correlations between H-1α and H-3 and between H-3 and H3-19 indicated that H-3 and CH3-19 were α-oriented. The β-orientation of H-2 was deduced from the NOESY correlation between H-2 and H3-18. The NOESY correlations between H2β and H-15 suggested the β-orientation of H-15. NOESY correlations between H-1α and H2-7 indicated that H-6 is βoriented. The α-orientation of H3-16 was assigned by its NOESY correlations with H3-20α. Due to the presence of two adjacent hydroxyl groups in 1, an electronic circular dichroism (ECD) exciton chirality method could be used to determine its absolute configuration by introducing two benzoate ester chromophore groups.13 Thus, compound 1 was treated with p-bromobenzoyl chloride in anhydrous pyridine to yield the 2,3-bis-p-bromobenzoate ester of 1 (1a) (Scheme 1).

Figure 3. Experimental ECD spectrum of 1a and the calculated ECD spectra for 1a and its enantiomer in MeOH. The most stable conformer of 1a calculated by Gaussian is shown.

positive Cotton effect at 233.6 nm (Δε +17.0) and a negative Cotton effect at 252.0 nm (Δε −37.2) due to the transition interaction between two identical bromobenzoyl moieties. The negative chirality suggested that the transition dipole moments of the two bromobenzoyl segments were oriented in a counterclockwise manner (Figure 3) and established the configuration of C-2 and C-3 both as R. In order to further confirm the absolute configuration, the time-dependent density functional theory (TDDFT) ECD calculation for 1a and its enantiomer was carried out.14 Results (Figure 3) showed that the Boltzmann-averaged ECD spectrum of (1R,2R,3R,5R,6R,14R,15S)-1a matched the experimental one, whereas the enantiomer of 1a showed completely opposite curves. However, there was a slight difference between the calculated and experimental ECD spectra at 201 nm. Therefore, more solid evidence, such as single-crystal X-ray diffraction, was required to confirm the absolute configuration of 1a. After many attempts, a high-quality single crystal of 1a was obtained from MeOH/ THF/H2O (1:1:0.1) and subjected to X-ray diffraction experiment with Cu Kα radiation. Single crystal X-ray diffraction of 1a (Figure 4) not only confirmed its structure but also unambiguously assigned its absolute configuration as 1R,2R,3R,5R,6R,14R,15S by a Flack parameter of −0.002(5).15 Mollebenzylanol B (2) was obtained as an amorphous powder and possessed the same molecular formula as 1 by the HRESIMS ion at m/z 333.2052 [M + H]+ (calcd for C20H29O4, 333.2066). The NMR data of 2 were similar to those of 1, and the major differences were that H-15 (δH 2.41, dq, J = 8.5, 7.3 Hz) in 2 shifted downfield, compared to that (δH 2.10, qd, J = 7.0, 3.6 Hz) in 1, while C-15 (δC 46.7) in 2 shifted upfield, compared to that (δC 52.2) in 1. Notably, the coupling constant of H-1 and H-15 (JH‑1/H‑15 = 3.6 Hz) in 2 was much smaller than that (JH‑1/H‑15 = 8.5 Hz) in 1. Thus, 2 should be a 15-epimer of 1. The strong NOESY correlation between H-2 and H3-20 supported this deduction. Two-dimensional NMR data of 2, including HSQC, 1 H−1H COSY, HMBC, and NOESY (Figure S2), further

Scheme 1. Preparation of the p-Bromobenzoate Ester of 1

B

DOI: 10.1021/acs.orglett.8b00606 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

neutralized by H2O to generate 10, which is further oxidized to a diketone 11. The carbon bond linkage of C-11 and C-12 in 11 is cleaved by a retro-aldo reaction16 to yield a triketone 12. After reduction, 8,9-carbon bond cleavage by a retro-aldo reaction,16 and aromatization, an aromatized diterpenoid 15 is formed. Finally, compounds 1 and 2 are generated by the successive ketal formation and addition of 15. Since grayanane diterpenoids were reported to have PTP1B inhibitory activities,17 compounds 1−3 and 1a were evaluated for their PTP1B inhibitory activities in vitro. As shown in Table 1 Table 1. PTP1B Inhibitory Activities of 1−3 and 1a Figure 4. ORTER drawing of 1a.

confirmed this structure. The ECD spectrum (Figure S3) of 2 showed a positive Cotton effect at 214.4 nm (Δε +6.1), which fits well with that of 1 (212.4 nm, Δε +6.2). Thus, the absolute configuration of 2 was defined to be 1R,2R,3R,5R,6R,14R,15R. Mollebenzylanols A (1) and B (2) possess an unprecedented diterpene carbon skeleton featuring a unique 9-benzyl-8,10dioxatricyclo[5.2.1.01,5]decane core, and this new diterpene skeleton is named mollebenzylane. Their biosynthetic precursor could be traced back to the coisolated grayanane diterpenoid rhodojaponin III (3).12 As shown in Scheme 2, grayanane

compd

inhibition (%)a

IC50 (μM)b

1 2 3 1a oleanolic acidc

75.4 70.5 3.4 90.6 99.2

22.99 ± 0.43 32.24 ± 0.74 >200 11.56 ± 1.93 4.71 ± 0.16

a Preliminary screening concentration was 200 μM. bValues are expressed as the means ± SD, n = 3. cOleanolic acid was used as the positive control.

and Figure S4, new mollebenzylane diterpenoids 1 and 2 exhibited moderate PTP1B inhibitory activities with IC50 values of 22.99 ± 0.43 and 32.24 ± 0.74 μM, respectively, while grayanane diterpenoid 3 did not show significant PTP1B inhibitory activity (IC50 > 200 μM). Interestingly, the 2,3-bisp-bromobenzoate ester of 1 (1a) showed more potent PTP1B inhibitory activity (IC50 = 11.56 ± 1.93 μM) than 1 and 2. Thus, the 9-benzyl-8,10-dioxatricyclo[5.2.1.01,5]decane core of this new diterpene skeleton may be essential for the PTP1B inhibitory activity. To further investigate their structure−activity relationship and modes of action, the binding modes and energies (Table S2) of compounds 1−3 and 1a with PTP1B were obtained by molecular docking. Results (Figures 5 and S5−S9) show that 2α-OH in 1 has a strong hydrogen bond with the Cys215 residue in the highly catalytic PTP-loop of PTP1B, which is responsible for executing the nucleophilic attack on the substrate phosphate moiety,18 and the 6-oxygen atom in 1 has a strong hydrogen bond with Try46 residue of the pTyr loop, which recognizes common features of

Scheme 2. Proposed Biosynthetic Pathway for 1 and 2

diterpenoid 3 is dehydrated to form 4, from which a 5,9-epoxygrayanane diterpenoid 5 is formed by an addition reaction. A triepoxy grayanane diterpenoid 6 is obtained by an enzymemediated oxidation of 5. A carbocation center is formed at C-12 in 8 by the protonation of the 5,9-epoxy in 6 and then the ring cleavage of the 5,9-epoxy in 7. The migration of the 13,14-bond to C-12 is accomplished by an enzymatic Wagner−Meerwein rearrangement14 in 8, which is the key step to form a new carbon−carbon bond between C-12 and C-14 in 9. The carbocation center created at C-13 in intermediate 9 is

Figure 5. Low-energy binding conformations of 1−3 and 1a (A−D) bound to PTP1B enzyme generated by virtual ligand docking. Dotted green and orange lines indicate hydrogen bonds and π−π interactions, respectively. C

DOI: 10.1021/acs.orglett.8b00606 Org. Lett. XXXX, XXX, XXX−XXX

Letter

Organic Letters

(2) Gonzalez, M. A. Nat. Prod. Rep. 2015, 32, 684. (3) Simmons, E. M.; Sarpong, R. Org. Lett. 2006, 8, 2883. (4) Mahdjour, S.; Harche-Kaid, M.; Haidour, A.; Chahboun, R.; Alvarez-Manzaneda, E. Org. Lett. 2016, 18, 5964. (5) Zhu, G.; Liu, B. Tetrahedron 2017, 73, 4070. (6) Li, F.; Tu, Q.; Chen, S.; Zhu, L.; Lan, Y.; Gong, J.; Yang, Z. Angew. Chem., Int. Ed. 2017, 56, 5844. (7) Majetich, G.; Shimkus, J. M. J. Nat. Prod. 2010, 73, 284. (8) Zheng, X. H.; Yang, J.; Lv, J. J.; Zhu, H. T.; Wang, D.; Xu, M.; Yang, C. R.; Zhang, Y. J. Fitoterapia 2018, 125, 89. (9) State Administration of Traditional Chinese Medicine. Zhong Hua Ben Cao; Shanghai Science and Technology Publishing Company: Shanghai, 1999; Vol. 6, p 5266. (10) (a) Chen, S. N.; Zhang, H. P.; Wang, L. Q.; Bao, G. H.; Qin, G. W. J. Nat. Prod. 2004, 67, 1903. (b) Wang, S. J.; Lin, S.; Zhu, C. G.; Yang, Y. C.; Li, S.; Zhang, J. J.; Chen, X. G.; Shi, J. G. Org. Lett. 2010, 12, 1560. (c) Li, Y.; Liu, Y. B.; Zhang, J. J.; Li, Y. H.; Jiang, J. D.; Yu, S. S.; Ma, S. G.; Qu, J.; Lv, H. N. Org. Lett. 2013, 15, 3074. (d) Zhou, S. Z.; Yao, S.; Tang, C. P.; Ke, C. Q.; Li, L.; Lin, G.; Ye, Y. J. Nat. Prod. 2014, 77, 1185. (e) Li, Y.; Liu, Y. B.; Liu, Y. L.; Wang, C.; Wu, L. Q.; Li, L.; Ma, S. G.; Qu, J.; Yu, S. S. Org. Lett. 2014, 16, 4320. (f) Li, Y.; Liu, Y. B.; Yan, H. M.; Liu, Y. L.; Li, Y. H.; Lv, H. N.; Ma, S. G.; Qu, J.; Yu, S. S. Sci. Rep. 2016, 6, 36752. (11) (a) Zhou, J.; Zhan, G.; Zhang, H.; Zhang, Q.; Li, Y.; Xue, Y.; Yao, G. Org. Lett. 2017, 19, 3935. (b) Zhou, J.; Sun, N.; Zhang, H.; Zheng, G.; Liu, J.; Yao, G. Org. Lett. 2017, 19, 5352. (c) Zhou, J.; Liu, T.; Zhang, H.; Zheng, G.; Qiu, Y.; Deng, M.; Zhang, C.; Yao, G. J. Nat. Prod. 2018, 81, 151. (12) Klocke, J. A.; Hu, M. Y.; Chiu, S.-F.; Kubo, I. Phytochemistry 1991, 30, 1797. (13) Harada, N.; Nakanishi, K. Acc. Chem. Res. 1972, 5, 257. (14) Zhang, M.; Zhu, Y.; Zhan, G.; Shu, P.; Sa, R.; Lei, L.; Xiang, M.; Xue, Y.; Luo, Z.; Wan, Q.; Yao, G.; Zhang, Y. Org. Lett. 2013, 15, 3094. (15) Parsons, S.; Flack, H. D.; Wagner, T. Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2013, 69, 249. (16) Zhou, L.; Tuo, Y.; Hao, Y.; Guo, X.; Tang, W.; Xue, Y.; Zeng, J.; Zhou, Y.; Xiang, M.; Zuo, J.; Yao, G.; Zhang, Y. Org. Lett. 2017, 19, 3029. (17) Liu, C. C.; Lei, C.; Zhong, Y.; Gao, L. X.; Li, J. Y.; Yu, M. H.; Li, J.; Hou, A. J. Tetrahedron 2014, 70, 4317. (18) Feldhammer, M.; Uetani, N.; Miranda-Saavedra, D.; Tremblay, M. L. Crit. Rev. Biochem. Mol. Biol. 2013, 48, 430. (19) Sarmiento, M.; Zhao, Y.; Gordon, S. J.; Zhang, Z. Y. J. Biol. Chem. 1998, 273, 26368.

peptide substrates and are important for peptide substrate binding and/or E-P formation.19 The benzyl group in 1 and 2 exhibits strong π−π interactions with the Tyr46 residue of the pTyr loop. Compared to 2, compound 1 has more hydrogen bonds with the Ala217, Gly218, Gly220, and Arg221 residues in the PTP-loop, which is why 1 has more potent PTP1B inhibitory activity than 2. However, compound 3 has no obvious interactions with the Try46 and Cys215 residues, although 14OH in 3 has strong hydrogen bonds with the Ser 216, Ala217, and Arg221 residues. That could be used to explain the poor PTP1B inhibitory activity of 3 (IC50 > 200 μM). Interestingly, although the di-p-bromobenzoate derivative of 1 (1a) has no obvious interactions with the Try46 and Cys215 residues, the benzene ring of the 3-p-bromobenzoate in 1a exhibited a strong π−π interaction with the residue Phe182 of the catalytic WPDloop, which is not only important for substrate binding and/or EP formation but also important for the E−P hydrolysis step.19 Thus, the potential PTP1B inhibitors are required to interact with the Tyr46, Phe182, or Cys215 residues to bind at the catalytic binding site, which is consistent with the previous results.11a,b These findings not only enriched the chemical diversity of Ericaceae diterpenoids but also provided useful clues to design novel PTP1B inhibitors.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b00606. Detailed experimental procedures, NMR spectroscopic data, binding energies and docking models, HRESIMS, UV, IR, and NMR spectra for compounds 1−3 and 1a (PDF) Accession Codes

CCDC 1823682 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_ [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Junjun Liu: 0000-0001-9953-8633 Guangmin Yao: 0000-0002-8893-8743 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported financially by the National Natural Science Foundation of China (U1703109, 81001368, and 31170323) and the Fundamental Research Funds for the Central Universities (HUST: 2016YXMS148). We thank the Analysis and Measurement Centre at HUST for the IR and ECD data collection, and the Instrumental Analysis Center of Shanghai Jiao Tong University for single-crystal X-ray diffraction analysis.

■

REFERENCES

(1) Newman, J.; Cragg, G. M. J. Nat. Prod. 2016, 79, 629. D

DOI: 10.1021/acs.orglett.8b00606 Org. Lett. XXXX, XXX, XXX−XXX