Luteolin suppresses UVB-induced photoageing by targeting JNK1 and p90 RSK2

Abstract

Multiple lines of evidence suggest that natural compounds can prevent skin ageing induced by ultraviolet light. Luteolin, a bioactive compound found in chilli, onion, broccoli, celery and carrot, has been reported to exhibit anti-photoageing effects in vitro. However, the molecular targets and mechanisms of luteolin are still poorly understood. In this study, we sought to investigate the effects of luteolin on UVB-induced photoageing and the molecular mechanisms involved, using HaCaT human keratinocytes and SKH-1 hairless mice. Luteolin was found to inhibit UVB-induced MMP-1 expression in HaCaT cells, as well as UVB-induced activation of AP-1, a well-known transcription factor targeting the MMP-1 promoter region, as well as c-Fos and c-Jun, which comprise the AP-1 complex. In contrast, Western blot data showed that UVB-induced phosphorylation of JNK, ERK and p90RSK was not inhibited by luteolin. In vitro kinase assay data revealed that luteolin significantly suppressed JNK1 and p90RSK activity, but not that of JNK2 and ERK2. Pull-down assays showed that luteolin binds JNK1 in an ATP-competitive manner and p90RSK2 in an ATP-independent manner. Luteolin also inhibited UVB-induced wrinkle formation and MMP-13 expression, a rodent interstitial collagenase in mouse skin, in vivo. Taken together, our observations suggest that luteolin exhibits anti-photoageing effects in vitro and in vivo and may have potential as a treatment for the prevention of skin ageing.

Fig. 1

Effect of luteolin on UVB-induced MMP-1 expression in HaCaT human keratinocytes. (A) Structure of luteolin. (B) UVB induces MMP-1 expression in HaCaT cells. HaCaT cells were starved for 24 hrs in serum-free DMEM, washed with PBS and then irradiated with UVB (0.01 J/cm²) in a small volume of PBS. The cells were incubated for the indicated times and the level of MMP-1 expression in the culture media was determined by Western blot analysis using specific antibodies against the MMP-1 protein as described in Materials and Methods. Data are representative of three independent experiments that gave similar results. (C) Luteolin inhibits UVB-induced MMP-1 expression in HaCaT cells. HaCaT cells were starved for 24 hrs in serum-free DMEM and pretreated with luteolin at the indicated concentrations (0, 5, 10 μM). After 1 hr, the cells were washed with PBS and then irradiated with UVB (0.01 J/cm²) in a small volume of PBS. The cells were incubated for 48 hrs and the level of MMP-1 expression in the culture media was determined by Western blot analysis using specific antibodies against the MMP-1 protein as described in Materials and Methods. Data are representative of three independent experiments that gave similar results.

Fig. 2

Effect of luteolin on UVB-induced AP-1 transactivation and c-Fos promoter activation in HaCaT human keratinocytes stably transfected with AP-1 and c-Fos luciferase reporter plasmids. (A and B) Luteolin inhibits both UVB-induced AP-1 transactivation and c-Fos promoter activation. For the luciferase assay, HaCaT cells stably transfected with AP-1 or c-Fos luciferase reporter plasmid were starved for 24 hrs in serum-free DMEM and pretreated with luteolin at the indicated concentrations (0, 5, 10 μM). After 1 hr, the cells were washed with PBS and then irradiated with UVB (0.01 J/cm²) in a small volume of PBS. The cells were then incubated for 5 hrs (A) or 3 hrs (B), and relative luciferase reporter activity was determined by analysing the cell lysates as described in Materials and Methods. AP-1 and c-Fos luciferase activities are expressed as the percent inhibition relative to non-UVB irradiated cells. Data are represented as the mean ±SD of AP-1 and c-Fos luciferase activity from three independent experiments (* and **significant differences at P < 0.05 and P < 0.01 between groups treated with UVB and luteolin and the group treated with UVB alone respectively).

Fig. 3

Effect of luteolin on UVB-induced JNK, c-Jun, ERK and RSK phosphorylation in HaCaT human keratinocytes. (A) Luteolin inhibits UVB-induced phosphorylation of c-Jun, but not JNK. (B) Luteolin does not inhibit UVB-induced phosphorylation of ERK, or p90RSK2. HaCaT cells were starved for 24 hrs in serum-free DMEM and pretreated with luteolin at the indicated concentrations (0, 5, 10 μM). After 1 hr, the cells were washed with PBS and then irradiated with UVB (0.01 J/cm²) in a small volume of PBS. The cells were incubated for 3 hrs and the levels of phosphorylated and total JNK, c-Jun, ERK and p90RSK protein were determined by Western blot analysis using specific antibodies against the corresponding phosphorylated and total proteins as described in Materials and Methods. Data are representative of three independent experiments that gave similar results.

Fig. 4

Effect of luteolin on JNK1, JNK2, p90RSK2 and ERK2 activities. (A) Luteolin inhibits JNK1 kinase activity. JNK1 kinase assays were performed as described in Materials and Methods. (B) Luteolin does not inhibit JNK2 kinase activity in vitro. (C) Luteolin inhibits p90RSK2 kinase activity in vitro. (D) Luteolin does not inhibit ERK2 kinase activity in vitro. Kinase assays were performed as described in Materials and Methods. Data are representative of three independent experiments that gave similar results. (**significant differences at P < 0.01, between groups treated with active kinase (JNK1, JNK2, p90RSK2 or ERK2) and luteolin, and the groups treated with active kinase alone).

Fig. 5

Binding affinity of luteolin with JNK1, JNK2 and p90RSK2. (A) Luteolin exhibits higher binding affinity to JNK1 than JNK2. Upper panel: in vitro luteolin binding was determined by immunoblotting with specific antibodies against JNK1 and JNK2 as described in Materials and Methods. Lane 1, active JNK1 (0.1 μg); lane 2, negative control, JNK1 does not bind with Sepharose 4B beads; lane 3, JNK1 binds with luteolin–Sepharose 4B beads; Lane 4, active JNK2 (0.1 μg); lane 5, negative control, JNK2 does not bind with Sepharose 4B beads; lane 6, JNK2 does not bind with luteolin–Sepharose 4B beads. Middle panel: lane 1, HaCaT cell lysate; lane 2, negative control, JNK1 does not bind with Sepharose 4B beads; lane 3, HaCaT cell lysate precipitated with Sepharose 4B beads. Lower panel: luteolin binds JNK1 in an ATP-dependent manner. Lane 1, active JNK1 (0.1 μg); lane 2, negative control, JNK1 does not bind with Sepharose 4B; lane 3, positive control, JNK1 binds with luteolin–Sepharose 4B; lanes 4 and 5, increasing concentrations of ATP suppress binding of luteolin to JNK1. (B) Luteolin directly binds with p90RSK2. Upper panel: lane 1, active p90RSK2 (0.1 μg); lane 2, negative control, p90RSK2 does not bind with Sepharose 4B beads; lane 3, p90RSK2 binds with luteolin–Sepharose 4B beads. Middle panel: lane 1, HaCaT cell lysates (0.1 μg); lane 2, negative control, p90RSK2 does not bind with Sepharose 4B beads; lane 3, p90RSK2 binds with luteolin–Sepharose 4B beads. Lower panel: luteolin binds with p90RSK2 in an ATP-independent manner. Lane 1, active p90RSK2 (0.1 μg); lane 2, negative control, p90RSK2 does not bind with Sepharose 4B beads; lane 3, positive control, p90RSK2 binds with luteolin–Sepharose 4B; lanes 4 and 5, increasing concentrations of ATP do not suppress luteolin binding with p90RSK2. Data are representative of three independent experiments that gave similar results.

Fig. 6

Luteolin inhibits UVB-induced wrinkle formation and MMP-13 expression in SKH-1 hairless mice. (A) Representative photographs showing anti-photoageing effects of luteolin. Mice were divided into four groups; controls treated with 200 μl acetone and non-irradiated; UVB group, treated with 200 μl acetone and UVB (0.18 J/cm²) irradiated; low dose sample group, treated with 10 μmol luteolin in 200 μl acetone and UVB irradiated; high dose sample group, treated with 40 μmol luteolin in 200 μl acetone and UVB irradiated. (B) Luteolin reduces UVB-induced wrinkle formation. The mice were treated as described for (A), and the extent of wrinkling was graded using a modification of the method proposed by Bissett et al., . Data are represented as the mean ±SD (n = 3) (**significant differences at P < 0.01, between the UVB group and sample treated groups). (C) Luteolin inhibits UVB-induced MMP-13 expression. Protein was extracted from mouse dorsal skin as described in the Materials and Methods, and the level of MMP-13 expression was determined by Western blot analysis using specific antibodies against the MMP-13 protein.

Fig. 7

Simplified depiction of the proposed anti-photoageing mechanism induced by luteolin.