3.1 Separation and Identification of compounds from different fractions
Compounds 1-11 were obtained from C1 by extraction, silica gel column, D101 macroporous resin, and semi-preparative high-performance liquid chromatography (Fig.1). Their structures were determined via NMR and mass spectrometry (MS) techniques. Furthermore, compound 1 was chosen as an example to indicate the structural identification of other compounds in the next section.
Compound 1 was obtained as an amorphous brown powder. Mg-HCl and Molisch reactions were positive, which indicated that compound 1 flavonoid. The molecular formula was determined as C27H30O15 by HR-ESI-MS at m/z 595.16. Its structure was determined by 1D (1H, 13C) and 2D (HSQC and HMBC) NMR experiments. The 1H-NMR spectrum of compound 1 was obtained with a carefully dried sample of 1 dissolved in DMSO-d6. which showed the shift at 13.50 (s), indicating a skeleton of a flavonol with a free hydroxyl group at C-5 position. Ring A and ring C had three protons for H-3, H-6 and H-8 confirmed by their downfield shift at 6.98 (s, 1H), 6.88 (s, 1H)and 6.56 (s, 1H). The data suggested that C-7 and C-5 were both substituted. The aromatic protons of B ring appeared as a doublet (8.09, 2H, J=8.5 Hz) assigned to H-2’, 6’ and a doublet (δ 7.21, 2H, J=8.5Hz) for H-3’, 5’ with a downfield shift consisting of AA’BB’ coupling system, which indicated that C-4’ was substituted. Moreover, the spectrum exhibited two anomeric proton signals resonated at 5.04(1H, d, J=7.2Hz),4.61(1H, d, J=9.4Hz), corresponding to two-configuration glucose units and the shift at δ: 5.04~3.20 assigned to sugar protons overlapped with –OH proton signals. The coupling constant of anomeric proton showed the β-configurations of the glucose.
The 13C-NMR spectrum confirmed the presence of 13 carbon signals assigned to the flavonol and 12 carbon signals for the sugar units. The compound can be determined as flavonoids by five SP2 hybrid ether oxygen carbon signals at δ 182.8(C=O), 164.7, 160.9, 164.3 and 157.2 confirming the above deduction. Moreover, the group of signals at δ 102.8, 74.0, 79.2, 70.5, 77.0, 61.5 and δ 100.4, 76.6, 81.3, 73.5, 71.3, 65.5 corresponded to β-D-glucose. In the HSQC spectrum, proton signals at δ6.98, 6.88 and 6.56 were correlated with δ103.8, 90.8 and 94.2 respectively, which proved that δ 6.98, 6.88 and 6.56 were assigned to H-3, H-8 and H-6.
In the HMBC, δ6.56 showed a long-range correlation with 160.9 and 164.3. Moreover, δ 6.88 showed a long-range correlation with 164.3 and 157.2. C-5 at 160.9, indicated that C-7 at 164.3, C-9 at 157.2 and C-7 were oxygenated. Furthermore, δ 8.09 (H-2’, 6’) showed a long-range correlation with 160.8, and 164.7. Similarly, δ 7.21 (H-3’, 5’) showed a long-range correlation with 124.3 and 160.8, which confirmed C-2 at 164.7, C-1’at 124.4, C-4’ at 160. 8 and that C-4’ was oxygenated.
The position of the glucose unit was assigned to C-4’ on the basis of HMBC correlation between the anomeric proton of glucose at 5.04 and C-4’ at 160.8. The anomeric proton signals of sugar 4.62 (H-1’’’) remote related with 65.5(C-6”) on the basis of HMBC, which proved that the two β-glucopyranoside moieties were linked to C-4” and C-6’’’, respectively. The results were found to be in agreement with the proposed structure as 5,7-dihydroxy-flavone-4’-(6”-O-β-D-glucopyranoside)-O-β-D-glucopyranoside.
3.2 Anti-inflammatory evaluation of Xylene-induced ear edema and Carrageenan-induced paw edema
After statistical processing, the inhibition rate of the blank control group and the test group on the degree of auricle swelling and paw edema of mice were obtained. The inhibition rate(%)of the five fractions (C1,C11,C12,C13 and C120)in the xylene-induced mice ear edema and carrageenan-induced mice paw edema compared to the standard drugs dexamethasone are shown in Fig.3, respectively. Compared to the blank group, extracts C11 increased the degree of auricle swelling and paw edema in mice. The inhibition rate(%)of C13 is not significant in the xylene-induced mice ear edema. Treatment with C1, C12 and C120 significantly suppressed ear edema and paw edema.
Therefore, based on the above experimental results of ear swelling and paw edema, fractions(C1,C12,C120)were selected for further mechanistic investigation.
3.3 Cell viability in RAW264.7
The results of MTS detection showed that, the concentration of C1, C12, and C120 (15.625 ~ 1000 μ g/mL) did not affect the cell viability of RAW264.7 macrophages as compared to the negative control (Fig.4), indicating minor cytotoxic or stimulatory effect of C1, C12, and C120 up to a concentration of 1000 μg/mL on RAW264.7 cells. Therefore, the largest concentration of 500 μ g/mL was used for cell culture.
3.4 NO release quantity
NO is an important inflammatory mediator, which is related to many inflammatory diseases. LPS can significantly promote the release of NO from macrophage RAW264.7. Therefore, the inhibitory effect on the release of NO can be used to evaluate the anti-inflammatory effect of the extract of Thlaspi arvense Linn and compounds isolated from it. Compared to the model group, the extracts of Thlaspi arvense Linn (C1, C12, and C120) could effectively inhibit the production of NO at the concentration of 62.5-1000 μg/mL. The inhibition rate was 50% when the concentration reached 500 μg/mL. The MTS results further showed that the cell viability decreased by 10% at a concentration of 1000 μg/mL. Therefore, concentrations of 125, 250, and 500 μg/mL were selected for further study.
3.5 Modulation of cellular production of IL-6 and TNF-α by extracts from the Thlaspi arvense Linn
In this experiment, lipopolysaccharide stimulates RAW264.7 macrophages to overexpress pro-inflammatory cytokines COX-2 and IL-6. To confirm the anti-inflammatory effect of the extracts of Thlaspi arvense Linn, the release of IL-6 and TNF-α was detected by an ELISA kit. The results showed that after the intervention of C1, C12 and C120, the levels of TNF-α and IL-6 in LPS-stimulated RAW264.7 cells decreased significantly in a dose-dependent manner (Fig.6). To sum up, the extract of Thlaspi arvense Linn may play a good anti-inflammatory role by inhibiting the excessive production of TNF-α and IL-6 in the process of inflammation. The results showed that C12 had the best anti-inflammatory effect and the most significant inhibitory effect against IL-6 and TNF-α produced by LPS-stimulated macrophage 264.7. Therefore, C12 was selected to make an in-depth study on the mechanism.
3.6 Effects of C12 on protein expression of iNOS, COX-2, proteins of NF-κ B pathways in LPS-induced RAW 264.7 cells
The expression of iNOS and COX-2 was significantly up-regulated by LPS. The expression of these two proteins was significantly improved in a dose-dependent manner after different concentrations of C12 were administered (Fig. 7). The expressions of iNOS and COX-2 were not significantly affected by 125 and 250 μg/mL of C12. The NOS expression was significantly inhibited by 500 μg/mL of C12, whereas the LPS-induced COX-2 expression was also significantly decreased by 125-500 μg/mL of C12. To elucidate the anti-inflammatory mechanism of the extract C12 more clearly, the NF-κ B pathway was studied.
In unstimulated RAW264.7 cells, NF-κ B was inhibited by I κ B kinase and remained in the cytoplasm. When RAW264.7 cells were stimulated by LPS, the inflammatory signals lead to I κ B phosphorylation and degradation. NF-κ B is phosphorylated and transferred from the cytoplasm to the nucleus, which promotes the secretion of related inflammatory factors through a series of transcription and translation. C12 can inhibit the overexpression of TLR-4 protein in LPS-induced RAW264.7 macrophages, in addition to the inhibition of the hyperphosphorylation of key proteins downstream of TLR-4 in RAW264.7 macrophages.
To explore the mechanism of the anti-inflammatory effect of C12, we detected the expression of related proteins in the classical TLR-4/NF-κ B signal pathway that mediates the inflammatory response of RAW264.7 macrophages.
As shown in Fig. 7A, stimulation of LPS (1.0 μg/mL) with C12 for 24 hours could significantly induce the overexpression of TLR-4, and different doses of C12 could inhibit the overexpression of TLR-4 to different extents. LPS can significantly induce the phosphorylation of NF-κ B p65, leading to the enhanced production of inflammatory response. At the same time, different doses of C12 could inhibit the phosphorylation level of NF-κ B p65 to different extents, especially at 500 μg/mL (Fig.7B). Besides, the level of I κ B-α increased significantly with the increase of C12 concentration (Fig. 7C), indicating that C12 could inhibit the degradation of I κ B-α. Subsequently, the dissociation of NF-κ B decreased, which prevented the production of pro-inflammatory factors. Inhibition of the phosphorylation level of I κ B-α to different extents can also suppress the depolymerization of I κ B-α polymer and can exert an anti-inflammatory effect. At the same time, the release of inflammatory factor IL-1β secretory protein was also significantly inhibited. Overall, the results confirmed that C12 had an obvious inhibitory effect at 500 μg/mL. The overall inflammatory signaling pathway is shown in Fig. 7.