Cyclohexylethanoid derivative and rearranged abietane diterpenoids with anti-inflammatory activities from Clerodendrum bungei and C. inerme

2.1 General experimental procedures

The material and instruments for the isolation and identification of compounds 1–13 were the same as reported (Ding et al., 2020).

2.2 Plant material

The branches and leaves of C. bungei and C. inerme were collected in December 2019 from Neijiang city of Sichuan Province and in November 2020 from Kunming city of Yunnan Province, China, respectively. The botanical identification was conducted by one of the authors (C.X.), and the two voucher specimens (No. 201912CB and 202011CI) were deposited at College of Pharmacy, Nankai University, China.

2.3 Extraction and isolation

The branches and leaves of C. bungei (5 kg) were extracted by refluxing with MeOH three times, followed by methanol elimination under reduced pressure to obtain a residue (775 g). The residue was suspended in water and then partitioned with EtOAc to yield a crude fraction (120 g), which was subjected to a silica gel column eluted with petroleum ether/acetone (100:0–100:30) to obtain five fractions (F1–F5). Fraction F3 (2.7 g) was isolated by medium-pressure liquid chromatography (MPLC) over ODS eluted with 75% and 85% MeOH in H₂O, resulting in five subfractions F3a–F3e. Compounds 3 (t_R = 23 min, 6.7 mg) and 4 (t_R = 29 min, 10.8 mg) were isolated from the subfraction F3b by preparative HPLC (80% MeOH in H₂O). Fraction F4 (4.5 g) gave six subfractions F4a– F4f on the basis of the same MPLC (75–85% MeOH in H₂O). Compound 1 (t_R = 10 min, 161.8 mg) was purified from the subfraction F4a using preparative HPLC (55% MeOH in H₂O), while compounds 5 (t_R = 28 min, 16.9 mg) and 6 (t_R = 35 min, 7.1 mg) were obtained from the subfraction F4c based on HPLC (80% MeOH in H₂O). Furthermore, the subfraction F4e was isolated by preparative HPLC (75% MeOH in H₂O) to give compound 7 (t_R = 82 min, 7.4 mg).

The branches and leaves of C. inerme (7.5 kg) were subjected to methanol extraction by refluxing three times and then methanol was removed under reduced pressure to obtain the crude extract (1100 g), which was subsequently suspended in H₂O and partitioned with EtOAc. The resulting EtOAc-soluble part (155 g) was subjected to a column of silica gel eluted with petroleum ether/acetone (100:0–100:30) to obtain seven fractions (F1–F7). Fraction F2 (8.4 g) was isolated by medium-pressure liquid chromatography (MPLC) over ODS eluted with 75% to 85% MeOH in H₂O to give five subfractions F2a–F2e. F2c was purified by preparative HPLC (80% MeOH in H₂O) to afford compounds 8 (t_R = 61 min, 4.7 mg), 2 (t_R = 65 min, 3.3 mg), 9 (t_R = 66 min, 9.4 mg) and 10 (t_R = 81 min, 10.9 mg). Fraction F3 (13.1 g) provided seven subfractions F3a − F3g by the same MPLC (75–85% MeOH in H₂O). F3b yielded compound 11 (t_R = 53 min, 15.6 mg) through HPLC (75% MeOH in H₂O). Fraction F4 (10.2 g) was purified on the MPLC (75–85% MeOH in H₂O) to afford seven subfractions F4a − F4g. The subsequent purification of F4b using preparative HPLC (71% MeOH in H₂O) gave compounds 12 (t_R = 95 min, 38.9 mg) and 13 (t_R = 197 min, 30.8 mg).

2.4 ECD calculation

ECD spectrum was determined on a MOS-450CD spectrometer (BioLogic, France) and ECD calculation was carried out as previously reported (Ma et al., 2018; Yan et al., 2020).

2.5 X-ray crystal structure analysis

The crystallographic data were measured on a Rigaku XtalAB PRO MM007 DW diffractometer using Cu Kα radiation. Tetragonal, space group P4₃, a = 8.43293(17) Å, b = 8.43293(17) Å, c = 23.1641(9) Å, α = 90°, β = 90°, γ = 90°, Z = 4, V = 1647.30(9) ų, density (calculated) = 1.324 mg/m³, F(0 0 0) = 704, μ(Cu Kα) = 1.54184 mm⁻¹, Flack parameter =  − 0.12 (16), crystal size = 0.11 × 0.09 × 0.09 mm³, reflections collected: 7984, Independent reflections: 3248 [R(int) = 0.0432], R₁ = 0.0423, wR₂ = 0.1077, Final R₁ = 0.0423, wR₂ = 0.1077 (I > 2sigma(I)).

2.6 RAW 264.7 cell culture and nitrite determination

RAW 264.7 cells were purchased from the Procell Life Science&Technology Co., Ltd. The experimental procedures were the same as previously reported (Cheng et al., 2023).

2.7 Cell viability assay

The method for assessing the cell viability was similar as previously described (Cheng et al., 2023). The detailed procedure was provided in the Supporting Information.

2.8 Western blot assays

The western blot assays were carried out as reported (Cheng et al., 2023). The following antibodies were used: IκB-α (AF5002, Affinity), p-IκB-α (AF2002, Affinity), NF-κB p65 (AF5006, Affinity), NF-κB p-p65 (AF2006, Affinity) and β-tubulin (T0023, Affinity).

2.9 RNA extraction and RT-qPCR

The methods for the total RNA extraction and RT-qPCR analysis were the same as previously described (Cheng et al., 2023). The primers used for SYBR Green real-time RT-PCR were as follows:: IL-1β, 5′-AGCTTCAGGC AGGCAGTATC-3′ (Forward) and 5′-AAGGTCCACGGGAAAGACAC-3′ (Reverse); IFN-γ, 5′-TGA ACGCTACACACTGCATCT-3′ (Forward) and 5′-TGTTGCTGATGGCCTGATTGT-3′ (Reverse); iNOS, 5′-GGGCTGACCTGTTTCCT ACT-3′ (Forward) and 5′-GGAGGTTGAGACCCAATGGA-3′ (Reverse); IL-6, 5′-CAACGATGATGCACTTGCAGA-3′ (Forward) and 5′-TGTGACTCCAGCTTA TCTCTTGG-3′ (Reverse); TNF-α, 5′-CC CACGTCGTAGCAAACCA-3′ (Forward) and 5′-ACAAGGTACAACCCATCGGC-3′ (Reverse); IL-4, 5′-GGCAGGAGAACATGGAGAGT-3′ (Forward) and 5′-CTTTCCTTCCTGCCTCTCCA-3′ (Reverse); IL-10, 5′-TAACTGCACCCACTTCCCAG-3′ (Forward) and 5′-AAGGCTTGGCAACCCAAGTA-3′ (Reverse); Arg-1, 5′-CTCCAAGCCAAAGTCCTTAGAG-3′ (Forward) and 5′-GGAGCTGTCATTAGGGACATCA-3′ (Reverse); β-actin, 5′-TGAGCTGCGTT TTACACCCT-3′ (Forward) and 5′-GCCTTCACCGTTCCAGTTTT-3′ (Reverse); GAPDH, 5′-GGACTTA CAGAGGTCCGCTT-3′ (Forward) and 5′-CTATAGGGCCTGGGTCAGTG-3′ (Reverse). Mean Ct of each gene was determined from triplicate measurements and normalized with the mean Ct of GAPDH or β-actin, a control gene.

2.10 Animals

Male C57BL/6 mice (weight 19–21 g) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. All animal experiments have been approved by the Animal Protection and Use Committee of Nankai University (No. 2022-SYDWLL-000586) and followed Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. Mice were classified into five groups (n = 8 for each): Sham group, CLP group, CS group (CLP + intraperitoneal injection of Cefpirome sulfat, 200 mg/kg each time), low-dose group (CLP + intraperitoneal injection of compound 1, 5 mg/kg each time) and high-dose group (CLP + intraperitoneal injection of compound 1, 15 mg/kg each time). After mice were acclimatized to the laboratory conditions for 7 days, multibacillary sepsis was stimulated by CLP method. The mice were administered at 2 h before and 24 h after CLP. The H&E staining was carried out according to the previous procedure (Cheng et al., 2023).

2.11 ELISA assay

Blood samples were collected from five groups at 48 h post-surgery and then centrifuged at 2,000 g for 20 min at 4 °C. Serum was collected and stored at − 20 °C before subjecting to further analysis. The levels of TNF-α, IL-6 and IFN-γ were determined using ELISA kits (Jianglai Biological Technology, Shanghai, China) by reading the absorbance at 450 nm.

2.12 Statistical analysis

Prism 8.0 software (GraphPad) for statistical comparison, two-tailed student T test to compare two data sets, and ANOVA for multiple comparison were performed, respectively. The statistical significance was considered based on P value < 0.05.

3.1 Structural elucidation

Compound 1 was isolated as rufous oil. Its HR-ESIMS showed an [M + Na] ⁺ peak at m/z 333.1312, indicating the molecular formula of C₁₆H₂₂O₆ (calcd for C₁₆H₂₂O₆Na 333.1314) and six degrees of unsaturation. The IR data revealed hydroxyl (3376 cm⁻¹), carbonyl (1717 cm⁻¹) and double bond (1682 cm⁻¹) functional groups. The ¹H NMR spectrum (Table 1) displayed the presence of two olefinic methines at δ_H 5.96 (d, J = 10.2 Hz, 1H) and 6.79 (dd, J = 1.6, 10.2 Hz, 1H), two oxygenated methines at δ_H 4.14 (td, J = 1.6, 4.5 Hz) and 3.89 (m, 1H), and two oxygenated methylenes at δ_H 3.84 (m, 2H), 3.94 (m, 1H), 4.02 (m, 1H). The ¹³C NMR spectrum (Table 1) of 1 showed 16 carbon signals consisting of eight methylenes [two oxymethylenes (δ_C 65.7, 65.5)], four methines [two oxymethines (δ_C 80.8 and 83.4) and two olefinic methines (δ_C 127.4 and 149.3)], and four quaternary carbons [two ketone carbonyl (δ_C 197.8, 211.4) and two oxygenated tertiary carbon (δ_C 74.1, 76.5)] on the basis of the DEPT and HMQC spectra. Since three degrees of unsaturation were assignable to two ketone carbonyl and one double bond groups, the remaining ones were thus attributed to three saturated rings. The NMR data of 1 indicated a resemblance with those of the known compound clerodenone A also occurring in C. bungei (Liu et al., 2009) except for the appearance of two aliphatic methylenes (δ_C 32.7, 34.5) instead of the double bond carbons (δ_C 148.8, 128.1), as evidenced by ¹H–¹H COSY correlations and HMBC correlation (Fig. 2). Thus, the planar structure of 1 was determined to be that depicted.

Close

Fig. 2

¹H–¹H COSY and key HMBC correlations of compounds 1 and 2.

Export to PPT

The NOESY correlations of HO-8/H-13 and HO-16/H-5 (in DMSO‑d₆) (Fig. 3), as well as the small coupling, implied a relative cis orientation (Liu et al., 2009). Furthermore, the NOESY correlation of HO-8 and HO-16 revealed that HO-8, H-13, HO-16 and H-5 were all the same orientation and assigned as a β-orientation. The calculated ECD spectrum matched the experimental data closely (Figure S10), demonstrating a (5R, 8S, 13R, 16R) absolute configuration for 1. Compound 1 is thus established and named clerodenone B.

Close

Fig. 3

Selected NOESY correlations of compounds 1 and 2.

Export to PPT

The molecular formula of compound 2 was deduced to be C₂₀H₂₄O₄ according to the negative ion peak at m/z 327.1601 [M − H]^– (calcd for C₂₀H₂₃O₄ 327.1596) in HR-ESIMS spectrum with the 9 degrees of unsaturation. Four methyl signals at δ_H 1.21 (s, 3H), 1.51 (d, J = 6.3 Hz, 3H), 1.63 (s, 3H), and 1.66 (s, 3H), one oxygenated methine at δ_H 5.12 (m, 1H), and one characteristic signal of a hydrogen-bonded hydroxyl group at δ_H 13.44 (s, 1H) were observed in the ¹H NMR spectrum (Table 1). The ¹³C NMR spectrum displayed 20 carbon signals, which were divided into one carbonyl carbon at δ_C 204.1, eight sp² quaternary carbons, one sp³ quaternary carbon, four sp³ methylenes, two sp³ methines and four methyl groups with the aid of DEPT and HMQC spectra. This suggested that 2 might be a 17(15 → 16), 18(4 → 3)-diabeo-abietane diterpenoid (Fan et al., 2000), and was structurally similar to the known compound uncinatone (11) (Tian et al., 1993) except for the shielded resonances at C-5 (Δδ_C − 125.3 ppm) and C-6 (Δδ_C − 85.9 ppm). Combined with the ¹H–¹H COSY correlations of H-5/H₂-6 and the HMBC interactions of H-6/C-7 and C-10, compound 2 was inferred to be a 5, 6-dihydro derivative of uncinatone. The complete planar structure of 2 was established by the detailed interpretation of ¹H–¹H COSY and HMBC spectra (Fig. 2). The NOESY correlations of H-5/H-1β, H-1α/H₃-20, and H-6α/H₃-20 suggested that the A/B rings are trans-fused (Fig. 3). X-ray diffraction analysis (CCDC 2160720) of 2 determined the absolute configuration as 5S,10R,16S (Fig. 4). Thus, the structure of 2 was identified as (5S,10R,16S)-12,16-epoxy-11,14-dihydroxy-17(15 → 16), 18(4 → 3)- diabeo-abieta-3,8,11,13-tetraen-7-one and named 15,16-dihydrouncinatone.

Close

Fig. 4

X-ray crystallographic structure of 2.

Export to PPT

The known compounds were identified as teuvincenone E (3) (Rodríguez et al., 2002), teuvincenone F (4) (Cuadrado et al., 1992), 6-methoxyvillosin C (5) (Wang et al., 2013), villosin B (6) (Li et al., 2014), villosin C (7) (Li et al., 2014), teuvincenone D (8) (Rodríguez et al., 2002), villosin A (9) (Chen et al., 2017), teuvincenone B (10) (Ulubelen et al., 1994), uncinatone (11) (Tian et al., 1993), 6α,11,12,16-Tetrahydroxy-7-oxo-abieta-8,11,13-triene (12) (Yadav et al., 2010), and 6,11,12,16-tetrahydroxy-5,8,11,13- abitetetraen-7-one (13) (Han et al., 2008), respectively, be means of spectral data.

3.2 Anti-inflammatory effect of compounds 1–13 on LPS-activated NO production

The cytotoxicity of all the isolates 1–13 on RAW 264.7 cells was firstly assessed by Cell Counting Kit 8 (CCK8) assay. As exhibited in Fig. 5A, all compounds except 13 have no significant toxicity on RAW 264.7 macrophages in vitro. Next, the inhibitory effects of 1–12 on NO generation in LPS (1 µg/ml)-mediated RAW 264.7 cells were examined, where BAY 11–7082 (BAY, 4 µg/ml) was used as a positive control. The anti-inflammatory result showed that compounds 1 and 12 significantly restrained the LPS-stimulated production of pro-inflammatory factor NO (Fig. 5B).

Close

Fig. 5

(A) Cell viability of RAW 264.7 cells treated with compounds 1–13 (40 μM). (B) NO production in LPS-mediated RAW 264.7 macrophages after compounds 1–12 treatment (20 μM). (C-D) Different concentrations treatment and the IC₅₀ value of compound 1 against LPS-mediated NO production in RAW 264.7 macrophages. Data are expressed as mean ± SD; *p < 0.05, ****p < 0.0001, compared with the LPS group.

Export to PPT

Due to low cytotoxicity (Fig. S41) and the greatest NO inhibitory activity in vitro, compound 1 was used to further investigate the anti-inflammatory activity. Cells were treated with different concentrations of 1 for 30 min followed by a 24 h treatment with LPS (1 µg/ml). The elevated NO levels stimulated by LPS were reduced significantly by pre-treatment with compound 1 in a dose-dependent fashion with the IC₅₀ value of 6.33 μM (Fig. 5C-5D).

3.3 Effect of compound 1 on the mRNA levels of inflammatory cytokines in LPS-exposed RAW 264.7 cells

In order to further validate the anti-inflammatory effect of compound 1, we investigated whether compound 1 regulated the mRNA expression levels of the pro-inflammatory cytokines iNOS, IL-1β, IFN-γ, TNF-α, IL-6 and the anti-inflammatory cytokines IL-4, IL-10 and Arg-1 in LPS-induced RAW 264.7 cells. As shown in Fig. 6A-H, compound 1 inhibited the LPS-stimulated mRNA expression of iNOS, IL-1β, IFN-γ, TNF-α, IL-6, while increased the mRNA expression of IL-4, IL-10 and Arg-1 in RAW 264.7 cells. These results showed that compound 1 exerted an inhibitory effect on the LPS-induced immune response.

Close

Fig. 6

Effects of compound 1 on LPS-induced iNOS (A), IL-1β (B), IFN-γ (C), TNF-α (D), IL-6 (E) IL-4 (F), IL-10 (G) and Arg-1 (H) mRNA level alteration in RAW 264.7 cells. mRNA levels were normalized to GADPH or β-actin mRNA and expressed as fold change relative to the control group. The data presented are the means ± SD of three independent experiments; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, compared with the LPS group.

Export to PPT

3.4 Compound 1 significantly inhibited NF-κB signaling pathway in LPS-induced RAW 264.7 cells

Since the activation of NF-κB is the central event leading to the inflammatory process, targeting it represents an attractive approach for anti-inflammatory therapies. Phosphorylated IκBs are polyubiquitinated and afterwards degraded by the proteasome, thereby permitting NF-κBs to translocate to the nucleus and to exert their function (Bollrath et al., 2009). The phosphorylation of NF-κB p65 promotes its own nuclear translocation, which is an important marker of NF-κB pathway activation (Liu et al., 2017).

To clarify whether the inhibitory effect of compound 1 on inflammation in RAW 264.7 cells correlated with the NF-κB signaling pathway, the effects of compound 1 on LPS-triggered phosphorylation of NF-κB p65 and IκBα were investigated using Western blotting. Compound 1 significantly restrained NF-κB p65 and IκBα phosphorylation (Fig. 7A-C), indicating that compound 1 demonstrated the anti-inflammatory effect by impeding the activation of the NF-κB signaling pathway.

Close

Fig. 7

(A-C) Effects of compound 1 on LPS-mediated NF-κB signaling pathway related proteins expression in RAW 264.7 cells. The protein expression levels were determined using specific antibodies for IκB-α, p-IκB-α, NF-κB p65, NF-κB p-p65 and β-tubulin. The data presented are the means ± SD of three independent experiments; **p < 0.01, ****p < 0.0001, compared with the LPS group.

Export to PPT

3.5 Compound 1 attenuated CLP-induced sepsis in mice

Sepsis, severe infections and the unbalanced immune response deleteriously affects physiological homeostasis of vital organ and often evolves into multi-organ failure (Lelubre et al., 2018). Cecal ligation and puncture (CLP) is a standard experimental model of sepsis (Dejager et al., 2011). In the CLP model, a large number of intestinal bacteria enter the abdominal cavity flowed by bacterial peritonitis, and then the bacteria enter the systemic circulation. The infection triggers the excessive generation of inflammatory factors, causing a systemic inflammatory response that eventually progresses to multiple organ failure and collapse of the host defense system.

To further investigate the therapeutic potential of compound 1 for sepsis, we assessed survival rates and body weight at 12, 24, 36, 48 h after CLP in mice relative to controls. After 24 h post-CLP induction, septic mice showed progressive symptoms including decreased respiratory rate, more and more laborious breathing and minimal response to auditory and tactile stimuli. The results showed that the survival rate of mice increased after compound 1 treatment. Compared with CLP group in which the survival rate was 75 % at 48 h, the survival rate of mice in high dose group was 87.5 % (Fig. 8B). Meanwhile, the results showed weight loss in the CLP group. However, both low and high doses of compound 1 as well as CS treatment alleviated this phenomenon (Fig. 8C), which suggested that CS and compound 1 intervention could keep weight of septic mice to a certain extent. The serum TNF-α, IL-6 and IFN-γ levels were also measured through ELISA. Those pro-inflammatory cytokines were elevated in the mice of CLP group, while decreased after compound 1 treatment in a dose-dependent fashion (Fig. 8D-F). High-dose compound 1 (15 mg/kg) has a similar or better effect compared with the positive drug CS. These data indicated that compound 1 was a promising inhibitor against CLP-mediated inflammatory responses in mice models of sepsis.

Close

Fig. 8

Protective effect of compound 1 on CLP-mediated sepsis. (A) Schematic diagram of the process of animal experiment. (B) Compound 1 was administered (5 or 15 mg/kg; intraperitoneal injection), and the survival rate was continuously observed. (C) The weight change of mice after CLP or sham operation. (D-F) Serum levels of TNF-α, IL-6 and IFN-γ were detected through ELISA at 48 h after CLP operation. Data presented are the means ± SD of three independent experiments; *p < 0.05, **p < 0.01, ****p < 0.0001, compared with the CLP group.

Export to PPT

Furthermore, we conducted H&E staining to evaluate pathomorphological changes in lung and kidney tissues. As displayed in Fig. 9, we found that CLP enhanced inflammatory infiltration in lung and kidney tissues, induced morphological changes in renal tubules, which was alleviated in compound 1 and CS groups. In CLP group, the lung tissue of mice showed thickened diaphragm, infiltration of inflammatory cells and edema. As shown in the treatment group, the structure of lung tissue was more clear, and there were fewer edema bands or inflammation (Fig. 9). Renal tubular epithelial cell edema, vacuolar degeneration, brush border disappearance and renal tubular lumen dilatation were observed in the renal tissue of CLP group mice, and compound 1 treatment could significantly attenuated the pathological damage of renal tissue in sepsis model (Fig. 9). Taken together, treatment with compound 1 increased survival and alleviated tissue injury in septic mice, implying that compound 1 exhibited a potent anti-inflammatory effect in model mice with CLP-induced sepsis.

Close

Fig. 9

Effect of compound 1 on CLP-mediated organ injury in mice. Lung and kidney histopathology at post-operation 48 h from five groups; magnification: 200 ×.

Export to PPT