Anti-inflammatory and analgesic activity of carnosol and carnosic acid in vivo and in vitro and in silico analysis of their target interactions

Abstract

Background and purpose: The diterpenoids carnosol (CS) and carnosic acid (CA) from Salvia spp. exert prominent anti-inflammatory activities but their molecular mechanisms remained unclear. Here we investigated the effectiveness of CS and CA in inflammatory pain and the cellular interference with their putative molecular targets.

Experimental approach: The effects of CS and CA in different models of inflammatory pain were investigated. The inhibition of key enzymes in eicosanoid biosynthesis, namely microsomal prostaglandin E₂ synthase-1 (mPGES-1) and 5-lipoxygenase (5-LO) was confirmed by CS and CA, and we determined the consequence on the eicosanoid network in activated human primary monocytes and neutrophils. Molecular interactions and binding modes of CS and CA to target enzymes were analyzed by docking studies.

Key results: CS and CA displayed significant and dose-dependent anti-inflammatory and anti-nociceptive effects in carrageenan-induced mouse hyperalgesia 4 h post injection of the stimuli, and also inhibited the analgesic response in the late phase of the formalin test. Moreover, both compounds potently inhibited cell-free mPGES-1 and 5-LO activity and preferentially suppressed the formation of mPGES-1 and 5-LO-derived products in cellular studies. Our in silico analysis for mPGES-1 and 5-LO supports that CS and CA are dual 5-LO/mPGES-1 inhibitors.

Conclusion and implications: In summary, we propose that the combined inhibition of mPGES-1 and 5-LO by CS and CA essentially contributes to the bioactivity of these diterpenoids. Our findings pave the way for a rational use of Salvia spp., traditionally used as anti-inflammatory remedy, in the continuous expanding context of nutraceuticals.

Linked articles: This article is part of a themed section on Principles of Pharmacological Research of Nutraceuticals. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v174.11/issuetoc.

Figure 1

Molecular structures of CS (1) and CA (2). Also shown, as compound 3, is the structure of the inhibitor of mPGES‐1, 2‐[[2,6‐bis(chloranyl)‐3‐[(2,2‐dimethylpropanoylamino)methyl]phenyl]amino]‐ 1‐methyl‐6‐(2‐methyl‐2‐oxidanyl‐propoxy)‐ N‐[2,2,2‐tris(fluoranyl)ethyl]benzimidazole‐ 5‐carboxamide.

Figure 2

Effect of CS and CA acid in the carrageenan‐induced hyperalgesia model. CS and CA were injected s.c. (1, 30 or 100 μg in 20 μL), 30 min before injection of 1% carrageenan (50 μL; s.c.) into the same hind paw of mice. Paw withdrawal was recorded 4 h after carrageenan administration. The results obtained are expressed as the mean ± SEM; n = 6. *P < 0.05; significantly different from vehicle control (V); one‐way ANOVA with Dunnett's post hoc test.

Figure 3

Effect of CS and CA in the formalin‐induced pain model. CS and CA were injected s.c. at a single dose (100 μg in 20 μL), 30 min before formalin injection (20 μL; s.c.). Early, licking activity recorded from 0 to 10 min after formalin administration; Late, licking activity recorded from 15 to 40 min after formalin administration. The results obtained are expressed as the mean ± SEM; n = 6. **P < 0.05; significantly different from vehicle control (V); one‐way ANOVA with Dunnett's post hoc test.

Figure 4

Effect of CS and CA on 5‐LOX (A) and mPGES‐1 activity (B). Residual activities (% of control) are shown as mean ± SEM of single determinations obtained in five (A) and three (B) independent experiments. A, *P < 0.05; significantly different from vehicle control; one way ANOVA with Tukey HSD post hoc tests.

Figure 5

Effect of CS and CA on eicosanoid formation in activated human neutrophils (A) and monocytes (B). A, Neutrophils were pre‐incubated with vehicle (DMSO) or test compounds for 10 min before eicosanoid formation was initiated by A23187 (2.5 μM). B, Monocytes were pre‐incubated with vehicle (DMSO) or test items and then stimulated with LPS for 24 h. Heatmaps were prepared using Gene‐E 3.0 (Broad Institute) and show residual activities (percentage of vehicle control) as mean of single determinations obtained in three (A) and five (B) independent experiments. Red indicates a relative increase and blue a decrease of eicosanoid levels. 5‐LOX products: LTB₄, trans/epi‐trans‐LTB₄, 5‐oxo‐HETE, 5,12‐DiHETE, 5,15‐DiHETE, LXA₄/isomers, 5‐HETrE, 5‐HEPE; 12/15‐LOX products: 12‐HETE, 12‐HEPE, 15‐HETE, 15‐HETrE, 5,12‐DiHETE, 5,15‐DiHETE, LXA₄/isomers; COX products: PGE₂, PGE₁, TxB₂, TxB₁, PGF_2α, 11‐HETE, 12‐HHT; mPGES‐1 products: PGE₂, PGE₁.

Figure 6

Superimposition of the inhibitor, compound 3, co‐crystallized (blue) and calculated (light pink) with mPGES‐1 (A). 2D panels of calculated interactions of compound 3 in mPGES‐1 binding site (B). Positive charged residues are coloured violet, negative charged residues are coloured red, polar residues are coloured light blue, and hydrophobic residues are coloured green. The π–π stacking interactions are indicated as green lines, and H‐bond (side chain) are reported as dotted pink arrows. Neutral histidine with hydrogen on the δ nitrogen is shown as HID, and neutral histidine with hydrogen on the ε nitrogen is shown as HIE.

Figure 7

3D models of CS (A) (coloured by atom types: C orange, O red, polar H white) and CA (B) (coloured by atom types: C yellow, O red, polar H white) in the binding site of mPGES‐1 with GSH (coloured by atom types: C green, O red, polar H white). Residues in the active site are represented in tubes (coloured by atom types: C grey, N blue, O red, H white). 2D panels represent the interactions between CS (C), CA (D) and the residues of mPGES‐1 binding site. Positive charged residues are coloured violet, negative charged residues are coloured red, polar residues are coloured light blue, hydrophobic residues are coloured green. The π–π stacking interactions are indicated as green lines, and H‐bonds (side chain) are reported as dotted pink arrows. Neutral histidine with hydrogen on the δ nitrogen is shown as HID, and neutral histidine with hydrogen on the ε nitrogen is shown as HIE.

Figure 8

3D models of CS (A) (coloured by atom types: C orange, O red, polar H white) and CA (B) (coloured by atom types: C yellow, O red, polar H white) in the binding site of 5‐LOX. Residues in the active site are represented in tubes (coloured by atom types: C grey, N blue, O red, polar H white), and Fe ion is depicted as yellow cpk. 2D panels represent the interactions between CS (C), CA (D) and residues of the 5‐LOX binding site. Negative charged residues are coloured in red, polar residues are coloured in light blue, and hydrophobic residues are coloured in green. The H‐bond (side chain), metal coordination and salt bridge are indicated as dotted pink arrows, grey lines and blue/red lines respectively. Neutral histidine with hydrogen on the δ nitrogen is shown as HID, and neutral histidine with hydrogen on the ε nitrogen is shown as HIE.