Diacerein inhibits the pro-atherogenic & pro-inflammatory effects of IL-1 on human keratinocytes & endothelial cells

We investigated IL-1-induced regulation of genes related to inflammation and atherogenesis in human keratinocytes and endothelial cells, and if ‘diacerein’, an oral IL-1 inhibiting drug currently approved for use in osteoarthritis, would reverse IL-1’s effects on these cells. Primary human keratinocytes and coronary artery endothelial cells were treated with either IL-1α or IL-1β, with and without diacerein. Using PCR-array, we assessed differential gene-expression regulated by IL-1 and diacerein. We identified 34 pro-atherogenic genes in endothelial cells and 68 pro-inflammatory genes in keratinocytes significantly (p<0.05) regulated at least 2-fold by IL-1, in comparison to control. Diacerein completely or partially reversed this regulation on almost all genes. Using ELISA, we confirmed diacerein’s ability to reverse IL-1-driven gene-regulation of 11 selected factors, at the protein level. The results support a novel idea that diacerein acts as an inhibitor of the pro-atherogenic and pro-inflammatory effects of IL-1. Diacerein may have therapeutic applications to diminish IL-1-induced skin inflammation in psoriasis and attenuate IL-1-induced development of atherosclerosis. Further investigation into diacerein’s effect on skin inflammation, atherogenesis and cardiovascular risk in animal models or humans is warranted.


Introduction
Psoriasis, a chronic inflammatory skin disease, affects up to 8.5% of adult populations [1]. Psoriasis was previously considered a solely cutaneous entity but recent studies, including a metaanalysis, have shown that moderate-to-severe psoriasis patients have increased cardiovascular risk, with more frequent atherosclerotic events (ex. myocardial infarction) and life expectancy reduction by up to 4 years [2][3][4]. Although one study showed no link between psoriasis and cardiovascular morbidity [5], data in the literature predominantly demonstrates that an association exists.
Interleukin-1alpha (IL-1α) [6, 7] and interleukin-1beta (IL-1β) [8], the two major subunits of IL-1, are implicated in psoriasis pathogenesis. Increased expression of IL-1α and IL-1β have a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 been found in psoriatic lesional skin in mouse models and in human subjects, as these cytokines directly contribute to the inflammation present in the skin [6,8,9]. However, inflammation in psoriasis is not confined to skin; evidence of chronic systemic inflammation exists [10]. The systemic inflammatory milieu in psoriasis likely contributes to the atherosclerosis (considered an inflammatory disease) present in this disease [11]. Specifically, increased levels of IL-1 have been found in this systemic inflammatory milieu [12]. Since IL-1α and IL-1β have been shown to contribute to vascular inflammation and atherosclerosis [13][14][15], the increased levels of serum IL-1 noted in psoriasis may be responsible for promoting atherosclerosis.
Diacerein is an IL-1 inhibitor, approved in Europe as an oral anti-inflammatory treatment for osteoarthritis. The aim of this study is to investigate IL-1-driven regulation of genes related to inflammation and atherogenesis in human coronary artery endothelial cells (EC) and keratinocytes (KC), and to study the effects of diacerein on this gene regulation. ECs and KCs were utilized because they are directly involved in atherogenesis and psoriasis pathogenesis, respectively [8,16].

Discussion
Our results demonstrate that IL-1 acts as a pro-atherogenic and pro-inflammatory mediator in ECs and KCs. Addition of diacerein yielded complete or partial reversal of IL-1-induced gene regulation.

Endothelial cells
Diacerein reverses IL-1α-induced regulation of atherosclerosis-related genes in ECs (Table 1). IL-1α was found to be pro-atherogenic by regulating several genes in ECs including up-regulation of ACE, ICAM1and other genes; these regulations were reversed by diacerein, thus acting in an anti-atherogenic manner. ACE (angiotensin-1 converting enzyme) increases angiotensin-II, which elevates blood pressure. Elevated blood pressure accelerates atherogenesis [16] while angiotensin-II may independently promote vascular inflammation by increasing oxidative stress in vessel walls [17] ICAM1 (inter-cellular adhesion molecule-1) was also up-regulated by IL-1α in other studies [18,19]. ICAM1, involved with inflammatory cell recruitment, was increased in atherosclerotic lesions [20]. Another study showed that reduced expression of cellular adhesion molecules in mice decreased atherogenicity [21]. Our IL-1α up-regulation of MMP1 (matrix metalloproteinase-1), was consistent with the findings of Hanemaaijer et al. [22] and was reversed by diacerein. MMP1 contributes to inflammatory cell intimal infiltration, plaque instability and rupture [23] and is correlated with increased total plaque burden [24]. Metalloproteinases are also implicated in multiple phases of atherosclerosis [25]. PLIN2 (perilipin-2), promoting foam-cell formation, was reported as a safe target for anti-atherogenic therapy [26]. SELE (E-selectin; facilitates vascular inflammatory cell infiltration) was up-regulated by IL-1α, consistent with Etter et al. [27]. We found diacerein significantly reduced EC protein expression of E-selectin, confirming functional down-regulation of the gene (Fig 1). E-selectin's importance to atherosclerosis was demonstrated in studies showing increased E-selectin on arterial plaque surfaces [28] and reduced development of atherosclerotic lesions in mice lacking SELE [29]. Diacerein also inhibited up-regulation of SELPLG (P-selectin ligand), which helps to recruit leukocytes to the endothelium [30]. Two genes were regulated in unique ways. TNFAIP3 (TNF-alpha-induced protein-3) was down-regulated by IL-1α, in contrast to genes discussed above. However IL-1α's effect was still pro-atherogenic because TNFAIP3 is reported to decrease inflammation and atherosclerosis in murine models [31]. Regarding LIF (leukemia inhibitory factor), two studies showed that its expression is inversely correlated with coronary atherosclerosis [32, 33]. We found IL-1α increased LIF expression, but diacerein further up-regulated the gene, enhancing LIF's anti-atherogenic correlation. (Table 2). IL-1β was found to be pro-atherogenic by regulating several genes in ECs including up-regulating CCL2, CCL5and other genes; these were reversed by diacerein, thus acting in an anti-atherogenic manner. CCL2 and CCL5 (chemokines) promote vascular inflammation by attracting leukocytes to vessel walls. Additionally, CCL2 is absent in endothelium in normal conditions but is increased in the setting of atherosclerosis and associated with elevated risk of myocardial infarction [34]. CD44 and VCAM1 were up-regulated by IL-1β, in accordance with another report of IL-1β up-regulation of VCAM1 in ECs [35]. Both CD44 and VCAM1 recruit inflammatory cells to the endothelium, while CD44 also interacts with hyaluronan, a glycosaminoglycan increased in atherosclerotic lesions. CD44 and VCAM1 were increased in an atherosclerosis mouse-model too [36]. Interestingly, we found IL-1β upregulated IL-1α in ECs; IL-1α is reported to be a strong facilitator of atherogenesis and a prospective therapeutic target to reduce vascular inflammation [14]. Thus, diacerein may reverse IL-1α's effects on atherosclerosis and also reduce its expression by ECs via suppressing IL-1β activity. TNC (tenascin-C), an extra-cellular matrix protein, is involved with induction of proinflammatory cytokines and metalloproteinases [37]. TNF (tumor necrosis factor-alpha), upregulated by IL-1β in our study and another, [38] is thought to be pro-atherogenic by disrupting endothelial barrier function, inducing metalloproteinases and promoting vascular inflammation [39]. Diacerein reduced TNC and TNF protein expression by ECs (Fig 2). Other genes regulated in ECs with unclear links to atherosclerosis. CSF2 (granulocytemacrophage colony-stimulating factor; recruits inflammatory cells) was down-regulated by IL-1α but highly up-regulated by IL-1β in ECs (including increased protein expression, Fig 2) and up-regulated by IL-1α in KCs. Given conflicting regulation of this gene, net effects that diacerein would exert on GM-CSF activity are unclear.

Diacerein reverses IL-1β-induced regulation of atherosclerosis-related genes in ECs
Although TGFB1 (transforming growth factor-beta1) and TGFB2 (transforming growth factor-beta2) were amplified within atherosclerotic plaques in one study [40], more convincing evidence indicates these proteins actually have an athero-protective effect [41]. Diacerein decreased TGFB1 and TGFB2 mRNA and protein expression in ECs, but only to levels that remained higher than those in untreated ECs (Fig 1). Thus, diacerein may not have great impact on atherogenesis through TGF-beta regulation.

Keratinocytes
KCs are clearly involved in psoriasis pathogenesis but unlike ECs, are not directly implicated in atherogenesis. However, gene regulation observed in KCs may still have implications for atherosclerosis in the following way. It has been suggested that the skin is a source of mediators that exert not only local inflammatory effects but also enter the circulation and cause systemic inflammation [42]; as discussed previously, systemic inflammation is associated with atherosclerosis. This may be one way skin inflammation leads to systemic and vascular inflammation, ultimately contributing to atherogenesis (Fig 4). Also, there may be a pro-inflammatory positive feedback-loop between the epidermis and endothelium which perpetuates this effect [6]. We found various inflammatory genes regulated by IL-1 in KCs, several of which may produce mediators that have atherogenic effects downstream in the vasculature. Diacerein reversed IL-  1-induced regulation of all genes in KCs, thus acting in an anti-inflammatory and potentially anti-atherogenic manner. Diacerein reverses IL-1α-induced regulation of genes linked to atherosclerosis in KCs (Table 3). IL-1α is potentially pro-atherogenic by regulating several genes in KCs; these regulations were reversed by diacerein. IL17C, a cytokine typically produced by Th17 cells, is the most plentiful IL-17 isoform in psoriatic lesions [6] and may stimulate cytokine production involved in vascular inflammation [43]. IL-23A is a cytokine that helps transform naïve CD4+ T-cells into Th17 cells, which as reported, may be pro-atherogenic [44,45]. Serum IL-23 was also significantly elevated in patients with peripheral arterial disease, a manifestation of atherosclerosis [46]. Previous reports also showed IL-1 stimulates IL-17 and IL-23 production in Tcells [47,48] Taken together, our data shows that IL-1α may increase Th17-mediated vascular inflammation.

Diacerein reverses IL-1β regulation of genes linked to atherosclerosis in KCs
Diacerein and apoptosis. Diacerein was reported to induce or inhibitor apoptosis in different cell types [64][65][66]. Interestingly, there are reports that diacerein does not affect cell apoptosis [67,68]. In our experimental setting, we found that diacerein inhibits apoptosis in keratinocytes while it has no role in endothelial cell apoptosis (Fig 5). The molecular mechanisms for diacerein's role in apoptosis of keratinocytes and endothelial cells need further investigation.

Conclusions
Our results show that diacerein significantly opposes pro-atherogenic and pro-inflammatory effects of IL-1 on multiple genes in ECs and KCs. Psoriasis has increased IL-1 activity and increased risk of atherosclerosis and cardiovascular morbidity [3,12]. Because our EC data and previous studies have demonstrated that IL-1α and IL-1β contribute to vascular inflammation/ atherogenesis [13][14][15], the elevated activity of IL-1 in psoriasis may contribute to this increased risk of atherosclerosis. Additionally, our KC data show that diacerein may diminish skin inflammation but also may have potential implications for atherogenesis in psoriasis: diacerein inhibits IL-1-induced regulation of several inflammatory genes in KCs reported to have pro-atherogenic properties, downstream in the vasculature. We therefore propose a novel idea that diacerein may act to diminish IL-1-induced atherogenesis both indirectly via skin and directly on ECs in the vasculature. Interestingly, there is a recombinant, non-glycosylated antagonist of the human IL-1 receptor, anakinra. Although limited clinical data exists, one pilot study showed anakinra Diacerein blocks IL-1 in skin & endothelial cells to have modest benefit in patients with psoriasis and psoriatic arthritis [69]. In conclusion, this study provides evidence that diacerein reverses the pro-atherogenic and pro-inflammatory gene regulation caused by IL-1 in ECs and KCs, potentially preventing progression of skin inflammation and inflammation-induced atherosclerosis. Diacerein may have an advantage over biologic therapies. Biologics like TNF-alpha inhibitors are proteins with a partial animal component and have the potential to induce an antibody response in patients, leading to the drug's ineffectiveness. Diacerein, being a small-molecule drug administered orally, has a substantially reduced tendency to induce such antibodies. Future investigation into clinical use of diacerein to benefit both the skin and vasculature by diminishing atherogenesis and inflammation, is warranted.

Cell culture and treatment
Human primary keratinocytes (Invitrogen, Carlsbad, CA, catalog# C-005-25P-A) and human primary coronary artery endothelial cells (Cell Applications Inc., San Diego, CA, catalog# 300K-05a) were cultured under standard conditions (humidified atmosphere, 5% CO2 at 37˚C) in growth medium supplemented with sodium pyruvate, non-essential amino acids and penicillin/streptomycin; media were replaced every 48 hours. KCs were differentiated from basal to mature keratinocytes by placing into a higher-calcium medium for 48-72 hours. At 80% confluence, cells were treated for 24 hours either with IL-1α or IL-1β only (25ng/ml IL-1α or 10ng/ml IL-1β in KCs and 10ng/ml IL-1α or 10ng/ml IL-1β in ECs), IL-1α or IL-1β combined with diacerein (diacerein concentrations: 10μM in KCs and 15μM in ECs with IL-1α; 20μM in KCs and 50μM in ECs with IL-1β-these concentrations were established using a dose-response curve prior to the experiment), or vehicle (DMSO). After 24 hours, cell culture supernatant was collected and stored at -80˚C, and cells were washed with ice-cold phosphatebuffered-saline and frozen at -80˚C until RNA isolation.

RNA isolation and real-time polymerase chain reaction (PCR) analysis
Total RNA from KCs and ECs were isolated using Trizol reagent per manufacturer's instructions. Reverse-transcription using First strand kit formed cDNA. Genes reported in the literature to be regulated by IL-1 and relevant to the pathogenesis of skin or vascular inflammation/ atherosclerosis, were tested by real-time PCR using custom primers (Integrated DNA Technologies, Coralville, IA).

PCR-array
Using KC and EC cDNA, PCR-arrays were used per manufacturer's instructions. 10μL of master mix with cDNA was loaded into each well, and PCR reaction was run (ViiA-7 PCR machine, Applied Biosystems, Waltham, MA). Cycling conditions used: 95˚C (10 minutes), then 40 repeating cycles of 95˚C (15 seconds) followed by 60˚C (60 seconds). Experiments were repeated in triplicate using four biologic replicates. Raw data were analyzed by a webbased Array Analysis Software (Qiagen). Data was normalized using house-keeping genes (ACTB, B2M, GAPDH, HPRT1, RPLP0). We then identified genes that were at least 2-fold up-or down-regulated after addition of IL-1 and/or diacerein in comparison to control, with p-value<0.05.

Pan-caspase assay
Pan-caspase assay was performed using CaspaTag™ Pan-Caspase In Situ Assay Kit (EMD Millipore, Billerica, MA) according to the manufacturer's instructions. Briefly, cells were seeded in 12-well plates and harvested 24 hours after treatment. Cells were trypsinized and centrifuged after trypsin inactivation. Next 90 μl of 1% charcoal-dextran-treated FBS was added into each tube with 3 μl of 30X FLICA to resuspend cells. Then cells were incubated for 1 hour at 37˚C under 5% CO2. After a series of washes, cells were resuspended in PBS and counted. Absorbance using an excitation wavelength of 485 nm and an emission wavelength of 520 nm was read in a Tecan Infinite 200 Pro multimode reader (Tecan US Inc., Morrisville, NC).

Statistical analysis
PCR-array raw data were analyzed as listed above. Other data reported in this study (ELISA, real-time PCR) are the mean of 4 biologic replicates +/-standard error of the mean and were analyzed by ANOVA with p-value <0.05 taken as significant.