GC-MS analysis of PREO
From the GC-MS analysis shown in Table 1, 57 compounds were determined. Mostly are terpenoids and monoterpenoids. The most significant two compounds are Citronellol, and Geraniol. Citronellol is a natural monoterpenoid, geraniol is a both monoterpenoid and alcohol.
The GC-MS chromatogram tracing and quantitative analysis of active constituents is shown as Table 1.
Effects of LPS and PREO individually on HaCaT cells
As shown in Fig. 1(B), LPS up to 2.5 µg/ml do not have any cytotoxicity. However, the expressions of pro-inflammatory cytokines in different incubation period determined the oxidative stress level. After 20hrs of incubation (Fig. 1A), IL-8, IL-1β and IκB-α expressions were almost doubled compared to control (no-LPS). For 6h,12h and 18h, expressions were not very significant to be considered as inflammation model. However, from our result, LPS can induce oxidative stress in HaCaT cells in a dose-time dependent manner. In case of PREO induced MTT assay, cells show decrease in viability with the increase of concentration after 24 h. According to Fig. 1C, the highest concentration was 1%, which is visibly toxic to cells. So, LPS 2.5 µg/ml and 0.001%-0.1% PREO had been used for further explorations as they promote cell growth without affecting cell viability.
Effects of PREO on LPS-induced HaCaT cell viability
After treated with 2.5 ug/ml of LPS for 20 h, the cells were treated with 0.001-0.1% (v/v) PREO for 6–18 h to determine the protective effects of PREO on LPS induced cell viability loss. LPS showed a slight decrease in cell viability after 6h of inductions which was regained after 18h probably because of cell adaptation. The results showed that LPS and PREO had promoted cell growth, and the cell survival rate was 90% and above (Fig. 1D).
Effects of PREO on NO and ROS production in LPS induced HaCaT cells
NO and ROS are one of the main criteria for producing oxidative stress. PREO was evaluated with Griess reagent method to determine its effect on reduction of NO level in LPS-induced HaCaT cells. As shown in Fig. 2A, the levels of NO markedly increased in response to 20h induction of LPS. Treatment with 0.001%, 0.01% and 0.1% PREO showed no effects on the NO level after 6 h, but after 12h of induction, PREO reduced the NO level notably and after 18h 0.001% and 0.1% PREO decreased it further more. To determine the cytoprotective effects of PREO in LPS-stimulated HaCaT cells, ROS levels were measured via fluorescent probe DCFH-DA. LPS increased the ROS level in a high range. Conversely, treatment with 001%, 0.01% and 0.1% PREO for 12h significantly reduced ROS levels but both 6h and 18h incubation showed slight raise compared to control (Fig. 2B). These results indicate that PREO can possibly restore the endogenous antioxidant defense mechanisms impaired by LPS. The above findings indicated that the effects of PREO on ROS level were related to the concentration of PREO and induction time.
Effects of PREO on activity of SOD and production of MDA
SOD and MDA are the key markers of oxidative stress. Respective kits measured the levels of SOD activity and MDA content. As shown in Fig. 3 (A, B), LPS induction reduced the activity of SOD but it was remarkably increased at 0.1% of PREO treatment for 6h and showed decrease after 18h (compared to 6 h) in LPS-induced HaCaT cells. The MDA levels markedly increased in response to 20 h of LPS stimulation, whereas treatment with PREO significantly inhibited it in time and dose dependent manner. PREO treated at 0.001% concentration for 6h did not reduce the production of MDA, it is possibly because the treatment time was too short and the concentration was too low. These results showed that PREO could alleviate the oxidative stress to adjust the effect of inflammation.
Effects of PREO on inflammatory cytokines production in mRNA level
One of the most obvious initial changes due to inflammation is the increase of inflammatory cytokines expression levels. IL-1β, IL-6, IL-8 and TNF-α are the most common cytokines over expressed due to inflammation. As shown in Fig. 4, the mRNA production of IL-1β, IL-6, IL-8 and TNF-α in LPS-induced HaCaT cells markedly increased in response to the LPS (p < 0.05) compared to control group both after 6h and 18h of induction. Treatment with 0.001%, 0.01% and 0.1% PREO significantly reduced LPS-induced mRNA level activation of IL-1β, IL-6, IL-8 and TNF-α in a time and dose-dependent manner compared to only LPS exposed HaCaT cells (Fig. 4A-D). After 6h of treatment, PREO significantly decreased the transcription level of the cytokines, while after 12 h and 18h of inductions caused more reductions with time. 0.001% PREO showed same results as control after 18h of incubation following further decrease with 0.01% and 0.1% PREO. Together, the above results indicated that PREO is a potent inhibitor of LPS-induced inflammatory cytokines expressions.
Effects of PREO on the NF-κB production and phosphorylation of related proteins
The production of inflammatory mediators is strongly affected by NF-κB pathways in the HaCaT cells. p65 and IκB-α are the major component of NF-κB activated by LPS in the HaCaT cells. As shown in Fig. 5 (A-D), LPS remarkably increased that expression of p65 and IκB-α and promoted their phosphorylation. Furthermore, concentration of 0.01% PREO can significantly reduce the expression of p65 and IκB-α and inhibit the phosphorylation of p65 and IκB-α in HaCaT cells after 18h of incubation (Fig. 5B, D). These experimental findings hint that PREO can mediate the inflammation in the HaCaT cells induced by LPS through NF-κB pathways.
Effects of PREO on TLR4 pathway in LPS-induced HaCaT cells
As LPS is a well-known ligand of TLR4, we evaluated the level of mRNA expression involved in this pathway. MyD88, TBK1, Cas-8, TRAK-4, TAK1, IKKβ, IKKΣ, p38 and TRIF mRNA expression levels were measured using RT-PCR analysis. We chose to treat the cell with 0.01% PREO for 18 h following 20 h of LPS pre-induction. We also exposed the cell to only 0.01% PREO to check its individual effects (Fig. 6)
LPS significantly raise the expression level for all mRNA compared to control. This refers to LPS boosting their expression. However, 18 h of exposure with 0.01% of PREO on LPS induced cells relatively lowers the inflammation level. When cells were treated with 0.01% PREO without LPS, MKK5, TAK1, IKKβ, IKKΣ, p38 and TRIF mRNA expression levels were increased compared to normal cell. MyD88, TBK1, MKK5 and TRAK-4 expression had slight changes while Cas9 showed decreased expression in LPS free PREO induced cells.
Binding affinity of ligands with OR2AT4
The physiochemical properties of OR2AT4 were represented in Table 3. The 7TM region was represented in Fig. 7B. Depending on the results from Saves v.6 (Table 4). the model from I-Tasser was selected for modeling. The evaluation results were shown in Fig. 7B, C. The Ramachandran plot showed 82.8% of amino acids in favored region while ERRAT value was 94.15 and verify3D value was 57.19%. This result revealed that the predicted model had very little local errors, compatible with amino acid sequence and conformation of the chain was acceptable for docking. The binding sites for OR2AT4 were 1, 6, 11, 16, 18, 19, 21, 23, 24, 29, 36, 78, 82, 83, 85, 86, 89, 90, 93, 94, 105, 106, 109, 110, 112, 113, 114, 164, 165, 167, 168, 180, 182, 183, 184, 185, 186, 188, 189, 190, 200, 203, 204, 207, 208, 211, 212, 256, 257, 260, 261, 263, 264, 267, 274, 277, 278, 280, 281, 284, 285, 288 residues. Geraniol (compound ID = 637566 ) and Citronellol (compound ID = 8842) were used as ligands. The binding affinity for the best predicted model geraniol was − 5.6 kcal and for citronellol its − 5.5 kcal. The binding pocket of Geraniol was Pro63, Phe67, Val122, Tyr125, Met141, Leu149, Asp126, Tyr 137, Asn146 and the binding pocket of citronellol was Val122, Tyr125, Val129, Met141, Asn146, Leu149, Pro63, Phe67, Asp126, Tyr137, Leu 140. All the docking results were presented in Table 5. Protein structure (ribbon) and ligands was presented in Fig. 7A, F-G. The interactions were presented in 2D format for individual proteins in Fig. 7H, I. The docked model with protein ligand interaction was presented in Fig. 7.