The endothelial cell is one of the central components of various inflammatory diseases such as psoriasis, diabetes, cancer and rheumatoid arthritis [
20]. Recent evidence has shown that chronic inflammation is associated with accelerated atherosclerosis and increased cardiovascular mortality and morbidity [
21]. It is believed that the therapy targeting the inflammatory processes shared in psoriasis and atherosclerosis may reduce cardiovascular risk in psoriatic patients [
22]. In our study, we analyzed the influence of DMF, a potent anti-psoriatic and anti-MS drug, on the constitutive and TNF-α-induced expression of MCP-1 and other important pro-inflammatory and pro-atherogenic cyto- and chemokines. We additionally examined the underlying mechanisms of this expression. We demonstrated that DMF significantly suppressed the constitutive and TNF-α-induced expression of MCP-1, CCL-5, PDGF-BB, GM-CSF and IL-6 in HUVEC. Furthermore, DMF eliminated the TNF-α-induced expression of CCL-5 and PDGF-BB. Recently, Seidel et al. demonstrated that DMF suppresses TNF-α-induced CCL-5 and IL-6 expression in smooth muscle cells via reduced p65 nuclear entry [
23]. Our findings not only confirm these observations in another cell type involved in atherogenesis but also broaden these results regarding the chemokines MCP-1, PDGF-BB and GM-CSF. The TNF-α-induced expression of these pro-inflammatory cyto- and chemokines is known to be mainly transcriptionally regulated by NfκB [
24]. We used MCP-1 as a model chemokine and demonstrated a transcriptional way of regulation. To determine whether this regulation was controlled by the p38/JNK, PI3K/Akt or p42/44 signaling pathways, we performed Western blot analysis. We detected prolonged and superinduced posphorylation of p38 and JNK in response to DMF and TNF-α compared with TNF-α alone. Interestingly, Seidel et al. demonstrated an enhancement of PDGF-BB-induced p38-phosphorylation by DMF in smooth muscle cells [
25]. On the other hand, Gesser et al. found that DMF did not influence IL-1ß-induced p38-phosphorylation in human keratinocytes [
26]. Recently, Xie et al.demonstrated that DMF induces p38, JNK and p42/44 phosphorylation in colon carcinoma cell lines and induces necroptosis [
27]. In our experiments, we did not find any significant influence of DMF on p42/44 phosphorylation. Therefore, this effect might be cytokine- and/or cell-type specific. To determine whether the signaling pathways were related to the inhibition of TNF-α-induced MCP-1 expression, we blocked them using specific inhibitors. Interestingly, blocking the p38, JNK, Akt and p42/44 pathways did not affect the DMF-induced MCP-1 suppression. Therefore, we showed that the prolonged activity of p38 and JNK by DMF treatment was independent of its effect on MCP-1 expression.
It is well known that TNF-α is a potent NfkB inductor and that MCP-1 and the other analyzed cyto- and chemokines are regulated by NfκB nuclear translocation and phosphorylation [
28]. Therefore, we studied the TNF-α-induced nuclear translocation and phosphorylation of p65 to determine whether DMF has the ability to reduce these effects. We found that DMF clearly inhibits the TNF-α-induced nuclear p65 translocation. This phenomenon has also been described by Loewe et al. [
13]. We extended this observation concerning the phosphorylation status of p65 by demonstrating that the TNF-α-induced p65 phosphorylation was almost completely eliminated by DMF. In addition, we found that treatments with DMF resulted in slightly reduced early degradation of IκBα . This observation only explains in part the early decreased p65 nuclear translocation (5-30 min). The later inhibition of p65 translocation by DMF seems to be independent of IκBα degradation, as demonstrated by Peng et al. in dendritic cells treated with DMF [
29]. Sasaki et al. demonstrated that phosphorylation of p65 on Serine 536 defines an IκBα-independent NF-κB pathway [
30]. This group suggests that there are two mutually exclusive populations of p65 that are distinguished in their phosphorylation status at serine 536. They postualte that during activation, the populations translocate independently and induce transcription of different genes. These results are in accordance with our experiments demonstrating a significant reduction of p65 phosphorylation at Serine 536, which now might explain the reduced p65 translocation mainly independent of IκBα in human endothelial cells. Thus, our results not only confirm but also broaden and explain the results demonstrated by Loewe et al. concerning the nuclear entry of p65 [
13].
To prove that the DMF-conveyed reduced nuclear translocation and phosphorylation of p65 is functionally relevant, we performed NFkB luciferase promoter assays demonstrating significant inhibition of constitutive and TNF-α-induced promoter activity by DMF.
In summary, we demonstrated that DMF inhibits constitutive and TNF-α-induced expression of pro-inflammatory and pro-atherogenic cytokines in human endothelial cells. Our findings are consistent with the inhibition of nuclear translocation, phosphorylation and activity of p65, which can be explained in part by the reduced early degradation of IκBα and more important the reduced phosphorylation of p65 at Serine 536. These results, which may help to explain the cardioprotective and anti-inflammatory effects of DMF, highlight its potential importance in the future treatment of vascular diseases.