English

• Diterpenoids with unique skeletons and various bioactivities have been constantly found in plants [1,2], especially Isodon species [3], indicating that this genus is a valuable plant source to identify lead molecules for drug discovery. Our research group has been dedicated to search for new diterpenes from the Isodon plants in Guizhou Province [4–8]. In our recent studies, several diterpenoids with unique skeletons have been reported. For example, amethinol A, possesses a six/five/seven-membered tricyclic system [9]; fladin A, a novel diterpenoid, contains an unprecedented cyclic ether group formed between C-4 and C-9 [10]; rubesanolides A and C, two novel abietanes, were identified with rare γ-lactone units formed between C-9 and C-20 for the former [11], and between C-8 and C-20 for the latter [12].

  As a continuous part of the project to discover novel active compounds from I. flavidus, two novel diterpenoids (1 and 2) have been isolated from the twigs and leaves of this plant (Fig. 1). Compound 1 represents an unprecedented diterpenoid with a novel six/seven/five-membered carbon skeleton, and compound 2 is a heterodimeric diterpenoid belonging to a seco-isopimarane coupled with a norabietane unit featuring an unique ester linkage between C-9 and C-3′. The compounds have been evaluated for their antiinflammatory activities based on the inhibitory assays of nitric oxide (NO) production and tumor necrosis factor-α (TNF-α) expression in lipopolysaccharide (LPS)-induced RAW264.7 cells.

  Figure 1

  

  Figure 1.  Structures of 1 and 2.

  DownLoad: Full-Size Img PowerPoint

  Compound 1, colorless crystals, was deduced to have the molecular formula of C₂₀H₃₀O₅ from its high resolution electrospray ionization mass spectrometry (HR-ESI-MS) (m/z 373.1985 [M+Na]⁺, calcd. for 373.1991), which is calculated to have 6 unsaturated degrees. The infrared (IR) spectrum displayed absorption bands at 3474 cm⁻¹ belonging to hydroxy groups and 1730 and 1702 cm⁻¹ belonging to carbonyl groups. The ¹³C nuclear magnetic resonance (NMR), distortionless enhancement by polarization transfer (DEPT)-135° and DEPT-90° spectra of 1 (Table 1) exhibited 20 carbon signals, characterizing as four methyl carbons, six aliphatic methylene carbons, one oxymethine carbon, three non-oxymethine carbons, one ketone carbonyl carbon, one carboxylic carbon, and four aliphatic quaternary carbons including two oxygenated ones.

  Table 1

  

  Table 1.  ¹H (400 MHz) and ¹³C (100 MHz) NMR data for 1 and 2 in CD₃OD (δ in ppm, J in Hz).

  DownLoad: CSV

  Starting with the two singlet methyl signals at δ_H 0.83 and 0.88, other key substructures of 1 could be established from the analysis of the 2D spectral data (Fig. 2A). In the ¹H–¹H correlation spectroscopy (COSY) spectrum, the correlation signals among the protons at δ_H 1.62/1.49, 2.10/1.87 and 1.45/1.24, the correlations among δ_H 1.80, 2.52/1.99 and 4.81, the correlations among δ_H 3.38 and 2.26/1.53, and the correlations among δ_H 1.73, 0.94 and 0.95 indicated the presences of the segments of a -CH₂CH₂CH₂-, a -CHCH₂CH-, a -CHCH₂-, and a -CH(CH₃)₂, respectively. Further analysis of the heteronuclear multiple bond correlation (HMBC) spectral data connected these segments together to form the diterpene skeleton. The -CH₂CH₂CH₂- segment is shown to have HMBC to the quaternary carbon at δ_C 35.5 (C-4), which is then linked to the -CHCH₂CH- segment. The -CHCH₂CH- segment is found to have HMBC with the oxytertiary carbon at δ_C 82.4 (C-8), which is connected to the -CHCH₂- segment. The presence of HMBC of the -CHCH₂- segment to the oxy-tertiary carbon at δ_C 83.7 (C-13), and C-13 to the isopropyl group [-CH(CH₃)₂] further determined the connections of these segments as -CH₂CH₂CH₂-C(CH₃)₂-CH-CH_2-CH(O)-C(O)-CH-CH₂-C(O)-CH-(CH₃)₂. The segment of -CH₂CH₂CH₂- also showed HMBC with the quaternary carbon at δ_C 62.0 (C-10) and the ketone carbon at δ_C 204.9 (C-9). The carbonyl carbon C-9 is linked to the -CHCH₂- segment by observing HMBC between the segment and C-9. The presence of HMBC between the -CHCH₂- segment and the methylene group (-CH₂-) of C-14, between the -CH₂- and C-8, and between the -CHCH₂CH- segment and the carboxylic carbon at δ_C 177.7 (C-20) finally made all the connections of the carbons present in the diterpenoid. C-20 is connected to C-7 to form a lactone ring through the observation of HMBC of H-7 (δ_H 4.81). Taken together, the planar structure of 1 was assigned. In the rotating frame Overhauser effect spectroscopy (ROESY) spectrum (Fig. 2A), the presence of the correlation signals of H-11 with H-14β, and H-15 with H-12α/H-14α, indicated the same orientation of H-11 and 13-OH.

  Figure 2

  

  Figure 2.  2D NMR correlations and X-ray data of 1. (A) Key 2D NMR correlations of 1. (B) Plot of X-ray crystallographic data for 1 (Displacement ellipsoids are drawn at the 30% probability level).

  DownLoad: Full-Size Img PowerPoint

  To completely determine its absolute configuration, 1 was crystallized in methanol to afford a crystal of the monoclinic space group P2₁2₁2₁, which was analyzed by X-ray crystallography. The final refinement on the Cu Kα data resulted in a Flack parameter of 0.02 (5) [13], allowing an unambiguous assignment of the absolute structure of 1 (Fig. 2B). The six chiral centers, C-5, C-7, C-8, C-10, C-11 and C-13, were thus determined as S, S, S, R, R and R, respectively. Accordingly, compound 1 was identified as an unprecedented diterpenoid bearing a novel six/seven/five-membered carbon skeleton that has not been reported before. We thus designate this type of diterpene structure as 'flavidane' and report 1 as the first example of flavidanes, which was given the trivial name flavidanolide A.

  Compound 2 was obtained as colorless crystals. Its molecular formula of C₃₉H₆₀O₄ was deduced from the HR-ESI-MS (m/z 615.4379 [M+Na]⁺, calcd for C₃₉H₆₀O₄Na, 615.4384). The IR spectral data indicated the presence of hydroxyl (3324 cm⁻¹) and ester carbonyl (1745 cm⁻¹) groups in 2. The ¹H NMR spectrum (Table 1) coupled with the analysis of the ¹H–¹H COSY spectral data (Fig. 3A) exhibited the characteristic proton signals for five singlet methyls (δ_H 0.81, 0.87, 0.88, 0.92, 1.82), two olefinic protons from their corresponding trisubstituted double bond counterparts [(δ_H 5.37 (d), 5.77 (t)], three olefinic protons from a vinyl group [(δ_H 4.88 (d), 4.94 (d), 5.81 (dd)], two olefinic protons from a methylene group [(δ_H 4.80 (d), 4.87 (d)], and seven aliphatic protons from an isopropyl group [(δ_H 0.92 (dd, 3H); 0.93 (dd, 3H); 1.58 (m, 1H)]. The ¹³C NMR and DEPT spectra (Table 1) revealed that 2 contains 39 carbon signals, which consist of one ester carbonyl carbon (δ_C 171.6), eight olefinic carbons (δ_C 114.1, 109.5, 121.5, 123.6, 136.1, 138.7, 147.5, 150.2), three aliphatic oxyquaternary carbons (δ_C 73.8, 74.8, 84.5), seven methyl carbons (δ_C 16.9 × 2, 17.1, 21.6, 23.7, 27.2, 28.1), thirteen methylene carbons (δ_C 20.7, 21.0, 23.1, 25.1, 29.3, 29.8, 30.3, 31.0, 32.9, 34.8, 36.3, 38.9, 46.5), four methine carbons (δ_C 38.8, 43.2, 44.4, 49.7), and three aliphatic quaternary carbons (δ_C 30.9, 37.1, 37.6). The aforementioned structure information indicated 2 as a diterpenoid dimer, which could comprise an isopimarane unit (part a, Fig. 3A), and a norabietane unit (part b, Fig. 3A) due to the presence of the characteristics vinyl and isopropyl groups.

  Figure 3

  

  Figure 3.  2D NMR correlations and X-ray data of 2. (A) Key 2D NMR correlations of 2. (B) Plot of X-ray crystallographic data for 2 (Displacement ellipsoids are drawn at the 30% probability level).

  DownLoad: Full-Size Img PowerPoint

  After careful comparison of the spectral data of 2 to those of the reported isopimaranes, the similar NMR pattern of 3, 4-seco-isopimara-4(18), 7, 15-triene-3-oic acid [14] except for the obvious upfield shift of the carboxylic carbon (δ_C 171.6 vs. 181.0) is observed for 2, indicating that the 3, 4-seco-isopimara-4(18), 7, 15-triene-3-oic acid is the isopimarane unit of 2, and the 3-oic acid is the linkage site to connect with the other abietane diterpenoid unit. The proposed isopimarane substructure was then verified by analyses of the 2D NMR correlation signals. Based on the ¹H–¹H COSY spectrum (Fig. 3A), the spin systems of H-5′/H-6′/H-7′ and H-9′/H-11′/H-12′ were established, which, in combination with the methyl and vinyl signals mentioned before, could be employed as the starting points for the structure elucidation. The presence of the key HMBC of H-1′ with C-3′, H-7′ with C-9′/C-14′, H-11′ with C-8′, H-15′ with C-12′/C-14′/C17′, H-16′ with C-13′, H-18′ with C-5′/C-19′, and H3–20′ with C-1′/C-5′/C-9′ (Fig. 3A) assigned the planar structure of the isopimarane unit as 3, 4-seco-isopimara-4(18), 7, 15-triene-3-oic acid.

  The planar structure of the abietane diterpenoid unit could also be determined by the 2D NMR spectral data. In addition to the characteristic isopropyl unit, the segments of =CHCH₂CH₂-, -CHCH₂CH₂- and –CH₂CH₂- were established based on the observation of the ¹H–¹H COSY correlation signals (Fig. 3A) among H-1 (δ_H 5.77), H-2 (δ_H 1.39, 1.16) and H-3 (δ_H 2.13, 2.11), among H-5 (δ_H 1.55), H-6 (δ_H 1.67, 1.26) and H-7 (δ_H 1.81, 1.55), and among H-11 (δ_H 2.37, 2.21) with H-12 (δ_H 1.58, 1.51), respectively. Ring A (Fig. 1) with two methyls substituted at C-4 was determined according to the presence of the key HMBC signals of H-1 with C-5 (δ_C 43.2), H-2 with C-4 (δ_C 30.9) and C-10 (δ_C 138.7), and H-18 with C-3 (δ_C 31.0), C-5 and C-19 (δ_C 28.1). The presence of HMBC signals of H-1/H-5/H-7 with C-9 (δ_C 84.5), and H-6 with C-8 (δ_C 74.8) suggested the ring B fused with ring A via C-5 and C-10. Based on the observation of HMBC signals of H-11 with C-8, H-1/H-12 with C-9, H-15 with C-12 (δ_C 29.8) and C-14 (δ_C 38.9), and H-16 with C-13 (δ_C 73.8), ring C was elucidated with an isopropyl group substituted at an oxyquaternary carbon C-13 and it was determined to fuse with ring B through C-8 and C-9. The planar structure of part b was subsequently established to be a 20-norabietane diterpenoid skeleton. Compared to the chemical shifts observed for the two oxy-tertiary carbons of C-8 (δ_C 74.8) and C-13 (δ_C 73.8), the ¹³C NMR chemical signal of C-9 (δ_C 84.5) displayed an obvious downfield shift, indicating it was the connection position with the isopimarane unit. The planar structure of 2 was thus proposed as a diterpenoid dimer with parts a and b connected through an ester bond between C-9 with C-3′.

  In order to confirm the elucidated chemical structure of 2 and to determine its absolute configuration, 2 was crystallized from methanol to afford crystals suitable for X-ray analysis. The Flack number of −0.01 resulted from the final refinement on the Cu Kα data of the crystal of 2 allows an unambiguous assignment of the absolute structure of 2 (Fig. 3B). The eight chiral centers of C-5, C-5′, C-8, C-9, C-9′, C-10′, C-13, C-13′ were thus determined as S, S, R, S, S, R, R and R, respectively. Finally, the structure of 2 was established as 8β, 13β-dihydyoxyl-20-norabieta-1(10)-ene-9-yl 3, 4-seco-isopimara-4(18), 7, 15-triene-3-carboxylate, and given the trivial name flavidanolide B.

  The plausible biosynthetic pathways of 1 and 2 were proposed in Scheme 1. For 1, a similar biosynthetic pathway to that of amethinol A was proposed with lophanic acid (A) as the same precursor [9]. The difference between the two pathways is that rings B and C in 1 are formed as 5- and 7-membered rings instead of 7- and 5-membered rings for amethinol A, respectively. The cause of the difference is due to the aldol condensation with C-11 as a carbon anion is preferred for 1 instead of C-7 for amethinol A, because a lactone ring is formed between C-20 and C-7, which significantly weakens the nucleophilic strength of the C-7 anion.

  Scheme 1

  

  Scheme 1.  Plausible pathway for the biogenesis of 1 and 2.

  DownLoad: Full-Size Img PowerPoint

  For 2, the abietane diterpenoid unit could also be derived from lophanic acid (A). A undergoes an oxidation to introduce a hydroxyl group of C-1 to produce E, which could be converted to F by epoxidation of the Δ^{8, 9} double bond. The C-20 carboxylic acid could be cleaved off via a decarboxylation mechanism to afford G, which loses a H₂O molecule to form H. The hydrolysis of the exposide could thus produce the norditerpene unit of 2. The seco-isopimarane diterpenoid unit of 2 was proposed to be derived from glutinosin, a constituent richly found in I. flavidus [15]. The dehydration of the two hydroxyl groups in glutinosin (J) forms compound K, which is added a hydroxyl on C-3 to produce L [8(9), 15-isopimaradien-3β-ol from Platycladus orientalis] [16] by an oxidation reaction. Another oxidation further transforms L to M, which undertakes a Baeyer-Villiger reaction to yield N with a 7-membered lactone ring [17]. The hydrolysis of N opens the lactone ring to provide O, which loses the hydroxyl group to form a double bond between C-4′ and C-18′ for P by the dehydration. The other diterpene monomer unit Q [3, 4-secoisopimara-4(18), 7, 15-triene-3-oic acid from Salvia cinnabarina] [18] is thus produced from P via an allylic isomerization to shift the Δ^{8, 9} double bond to Δ^{7, 8}. An esterification reaction then occurs between the two diterpene monomers of I and Q to accomplish the final step of the biosynthetic pathway of compound 2.

  The antiinflammatory effects of 1 and 2 were evaluated based on the LPS induced RAW264.7 cell model [19]. As demonstrated in Table S3 (Supporting information), 2 exhibited the inhibitory effects on NO production in the concentration range of 1–10 µmol/L, comparable with those of the positive control pyrrolidinedithiocarbamate ammonium (PDTC) in the concentration range of 5–10 µmol/L. However, 2 only showed 18.98% inhibitory rate on TNF-α production in the LPS-induced cell model (Table S4 in Supporting information).

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (No. 82204605), the National Natural Science Foundation of China (No. 81560709), the Technology Fund of Guizhou Administration of Traditional Chinese Medicine (No. QZYY-2022-019), Science and Technology Tip-top Talent Foundation of Universities in Guizhou Province (No. Qian jiao he KY (2021) 034), the Research Grant Council of the Hong Kong Special Administrative Region, China (No. HKBU 12102219), the University Grants Committee of the Hong Kong Special Administrative Region, China (UGC Research Matching Grant Scheme, No. RMG2019_1_19).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108621.

  
  
• 1. [1]

     Y.Y. Feng, S.Q. Zha, H.Q. Zhang, et al., Chin. Chem. Lett. 34 (2023) 107742. doi: 10.1016/j.cclet.2022.107742

  2. [2]

     J.J. Zhao, S.Z. Du, K. Hu, et al., Chin. Chem. Lett. 34 (2023) 107737. doi: 10.1016/j.cclet.2022.08.017

  3. [3]

     W.G. Wang, X. Du, X.N. Li, et al., Org. Lett. 14 (2012) 302–305. doi: 10.1021/ol203061z

  4. [4]

     J. Zou, X. Du, G. Pang, et al., Org. Lett. 14 (2012) 3210–3213. doi: 10.1021/ol3013205

  5. [5]

     G.Y. Luo, R. Deng, J.J. Zhang, et al., Asian Nat. Prod. Res. 20 (2018) 227–233. doi: 10.1080/10286020.2017.1317754

  6. [6]

     X. Yu, X. Song, Y. Zhang, et al., Molecules 27 (2022) 6836. doi: 10.3390/molecules27206836

  7. [7]

     K. He, J. Zou, Y.X. Wang, et al., Molecules 26 (2021) 3865. doi: 10.3390/molecules26133865

  8. [8]

     W.F. Li, Z.M. Liang, C.L. Zhao, et al., Molecules 27 (2021) 3098.

  9. [9]

     C.L. Zhao, M.S. Sarwar, J.H. Ye, et al., Acta Cryst. C 74 (2018) 635–640. doi: 10.1107/S2053229618005740

  10. [10]

      J.X. Li, Q.J. Li, Y.F. Guan, et al., J. Etnopharmacol. 191 (2016) 372–378. doi: 10.1016/j.jep.2016.06.046

  11. [11]

      J. Zou, L.T. Pan, Q.J. Li, et al., Org. Lett. 13 (2011) 1406–1409. doi: 10.1021/ol200086k

  12. [12]

      J. Zou, L.T. Pan, Q.J. Li, et al., Org. Biomol. Chem. 10 (2012) 5039–5044. doi: 10.1039/c2ob25192b

  13. [13]

      S. Parsons, H.D. Flack, T. Wagner, Acta Cryst. B 69 (2013) 249–259. doi: 10.1107/S2052519213010014

  14. [14]

      A. Bisio, B. Pagano, A. Romussi, et al., Molecules 12 (2007) 2279–2287. doi: 10.3390/12102279

  15. [15]

      Q.S. Zhao, J. Tian, J.M. Yue, et al., Phytochemistry 48 (1998) 1025–1029. doi: 10.1016/S0031-9422(97)00608-0

  16. [16]

      J. Asili, M. Lambert, H.L. Ziegler, et al., J. Nat. Prod. 67 (2004) 631–637. doi: 10.1021/np034033e

  17. [17]

      O. Abril, C.C. Ryerson, C. Walsh, et al., Bioorg. Chem. 17 (1989) 41–52. doi: 10.1016/0045-2068(89)90006-0

  18. [18]

      G. Romussi, G. Ciarallo, A. Bisio, et al., Planta. Med. 67 (2001) 153–155. doi: 10.1055/s-2001-11511

  19. [19]

      Y.Y. Liu, Y.M. Yan, D.W. Wang, et al., Org. Lett. 23 (2021) 8657–8661. doi: 10.1021/acs.orglett.1c02378

• 

  Figure 1  Structures of 1 and 2.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  2D NMR correlations and X-ray data of 1. (A) Key 2D NMR correlations of 1. (B) Plot of X-ray crystallographic data for 1 (Displacement ellipsoids are drawn at the 30% probability level).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  2D NMR correlations and X-ray data of 2. (A) Key 2D NMR correlations of 2. (B) Plot of X-ray crystallographic data for 2 (Displacement ellipsoids are drawn at the 30% probability level).

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Plausible pathway for the biogenesis of 1 and 2.

  下载: 全尺寸图片 幻灯片

  Table 1.  ¹H (400 MHz) and ¹³C (100 MHz) NMR data for 1 and 2 in CD₃OD (δ in ppm, J in Hz).

  下载: 导出CSV
•