Caffeic acid phenethyl ester is protective in experimental ulcerative colitis via reduction in levels of pro-inflammatory mediators and enhancement of epithelial barrier function

Inflammatory bowel disease (IBD) is an idiopathic disorder, generally categorised as either Crohn’s disease (CD) or ulcerative colitis (UC) (Neurath 2014; Ford et al. 2011). There is no therapeutic cure for IBD and the current disease management strategies possess several drawbacks. For example, immunomodulatory agents, such as azathioprine and 6-mercaptopurine, cause bone marrow depletion and damage to both white blood cell and hepatic cell populations. Furthermore, results from recent clinical trials confirm that azathioprine is ineffective in UC (Ardizzone et al. 2006; Kamath et al. 2016; O’Connor et al. 2010) and sulfasalazine causes ruptures in liver tissue and decreases platelets count in blood (de Abajo et al. 2004; Rubin 1994). Furthermore, pulmonary disorders are reported in IBD patients treated with chimeric monoclonal antibodies, such as infliximab (Patel et al. 2016).

The clinical symptoms of IBD range from episodes of relapse and remission with mild inflammation and discomfort to a chronic ulcerative disease requiring surgical removal of the inflamed gut. The current therapeutic strategies for IBD are generally limited, but recent clinical advancement has occurred in immunotherapy using monoclonal antibodies. This approach is directed against inflammatory mediators, such as TNF-α (Targan 2006; Subramanian et al. 2017; Gecse and Lakatos 2017; Chan and Ng 2017). However, these biological agents are expensive and result in severe side effects and life threatening complications (Cote-Daigneault et al. 2015; Blonski and Lichtenstein 2006; Clarke and Regueiro 2012; Cohen and Thomas 2006). Hence, there is a need in the field of IBD therapy to develop new therapeutics, which are effective, safe, and economical. One way to achieve this is investigation into the anti-inflammatory effect of natural compounds and understanding their mechanism of action. Lack of specificity and the encumbrance of severe side effects necessitate further investigation into effective and safer options for treating IBD (Pichai and Ferguson 2012), which is the aim of this current study.

Colonic specimens from UC patients display overexpression of transcription factor nuclear factor kappa beta (NF-κβ) (Atreya et al. 2008). NF-κβ is up regulated by TNF-α, interleukin (IL), interferon, chemokines, and DNA damaging agents during the inflammatory phase. Similar effects are observed following exposure to lipopolysaccharide derived from bacterial cell wall components (Lawrence 2009b; Xavier and Podolsky 2007). In UC, levels of inflammatory mediators, such as TNF-α, interleukins, and interferons, increase due to the over stimulation of NF-κβ during inflammation (Schreiber et al. 1998). It is feasible, therefore, that inhibition of NF-κβ may be of therapeutic benefit in UC, which forms the hypothesis of our current work.

Novel pharmacological inhibitors of NF-κβ are currently available, but these compounds inflect toxicity and severe side effects in humans. Hence, we have selected caffeic acid phenethyl ester (CAPE), a phenolic constituent derived from honeybee propolis and shown in Fig. 1, for further study. It possesses no known adverse side effects (Tolba et al. 2014; Liao et al. 2003). CAPE possesses potent anti-inflammatory properties, which are attributed to its selective inhibition of NF-κβ. Recently, CAPE has been reported to inhibit other relevant pathways, such as MAPK and PI3K (Natarajan et al. 1996; Ozturk et al. 2012; Lin et al. 2013; Pramanik et al. 2013; Cho et al. 2014). CAPE represses translocation of NF-κβ, either by inhibition of Iκβ degradation or by blocking of NF-κβ and DNA binding (Wang et al. 2010; Bezerra et al. 2012). It has been reported that inflammatory markers, such as INF-ϒ, IL-6, IL-β, TNF-α, and IL-10, cause degradation of Iκβ, which results in induced overexpression of NF-κβ (Lang et al. 2004). CAPE inhibits this overexpression of NF-κβ via prevention of degradation of Iκβ (Wang et al. 2010; Lee et al. 2008).

The most recent evaluation of the anti-inflammatory activity of CAPE has been limited to either cell culture models or rat models of 2,4,6-trinitrobenzenesulfonic acid-peptidoglycan polysaccharide-induced UC (Armutcu et al. 2015; Ek et al. 2008; Fitzpatrick et al. 2001; Kim et al. 2013). However, the pathophysiological mechanism, such as macroscopic, microscopic changes in the colon and effect on pro-inflammatory cytokine levels, and mucosal barrier function, by which CAPE exerts its anti-inflammatory activity, have not been fully explored (Fitzpatrick et al. 2001; Michaluart et al. 1999; Ek et al. 2008; Cooper et al. 1993). Hence, the activity of CAPE has not been studied in relevant in vivo models, which are physiologically more representative of the human disease (Chassaing et al. 2014). Thus, the main aspect of this study, which differentiates it from the previous work, is assessment of the activity of CAPE on the colon at macroscopic and microscopic levels, its effect on MPO and pro-inflammatory cytokine levels and altered mucosal permeability in a mouse model of colitis, which is physiologically relevant to human disease (Tambuwala et al. 2015). The findings of this study will provide an insight into the anti-inflammatory efficacy of CAPE during colitis in terms of changes in the levels of the disease activity index (DAI) score, colon MPO, pro-inflammatory cytokines, and epithelial barrier function.

The previous studies have indicated that CAPE is an effective inhibitor of NF-κβ and related cytokines in vitro, and also has the ability induce apoptosis in inflammatory cells (Fitzpatrick et al. 2001). In our current study, we have shown for the first time that CAPE significantly ameliorates the severity of the disease in a mouse model of UC.

One of the initial events that occur during the onset of IBD is disruption of the intestinal epithelial barrier function. This dysfunction leads to unwanted movement of luminal antigenic material into the lamina propria. This is followed by activation of mucosal immune cells and triggering of an inflammatory response. It has been suggested that one of the critical events in the development of inflammation in the intestine could be the regulation of intestinal epithelial cell apoptosis (Cummins et al. 2008; Tambuwala et al. 2010). In cases of chronic inflammation during IBD, constant intestinal epithelial cell apoptosis could lead the loss of the epithelial barrier, which will result in spread of inflammation, resulting in increased severity of the disease. This creates an imbalance in the innate and adaptive immunity of the gut (Nenci et al. 2007; Zaph et al. 2007).

In the present work, we have shown for the first time that CAPE, a potent inhibitor NF-κβ, is profoundly protective in an in vivo mouse model of acute colonic inflammation. Although we hypothesise that the protective effects of CAPE are mediated through the inhibition of the over activation of the NF-κβ pathway, we cannot exclude the possibility of NF-κβ independent mechanisms of action, such as inhibition of hydroxylases and activation of hypoxia inducible pathways (Cummins et al. 2008). However, several researchers have indicated that the NF-κβ pathway plays an important role during intestinal inflammation (Wei and Feng 2010; Buhrmann et al. 2011; Lawrence 2009a; Fitzpatrick et al. 2001) and inhibition of this pathway targets pro-inflammatory cytokines, such as interferons and tumour necrosis factor alpha. These are known to play key role during the development and progression of UC (Baird et al. 2016; Bishop et al. 2014; Ferrari et al. 2016).

CAPE treatment significantly ameliorated the severity of disease after acute DSS exposure in all parameters studied, including weight loss (Fig. 2a), clinical DAI score (Fig. 2b), reduction of colon length, and appearance of blood in faeces (Fig. 3a, b). A marked improvement in colon histology was observed (Fig. 4a), together with improved blinded inflammation scores (Fig. 4b). In the murine model of DSS-induced colitis, the increase in MPO and pro-inflammatory cytokines occurred after the disruption to the intestinal barrier, indicating that compromised barrier function results in progression inflammation. CAPE-treated mice did not have increased MPO (Fig. 5a) and only small increase in other pro-inflammatory cytokines (Fig. 5b–f), suggesting that CAPE treatment prevented the damage to colon epithelial cells caused by DSS and helps in maintaining the epithelial barrier function, which is evident by reduced permeability of FITC in mice treated with CAPE (Fig. 6). However, whether improved epithelial barrier function or indeed lowered cytokine expression is the cause or consequence of the protective effects of CAPE remains to be elucidated. This critical question will be the topic of further investigations.

In mice treated with CAPE, in the absence of DSS exposure, there were no alterations in MPO or cytokines in the colon and no alterations in colon histology or length. This confirms that in the acute 6-day treatment regimen used in this study, CAPE did not alter physiologic inflammation in normal tissue, but instead suppressed inflammation in the colon when occurred due to DSS-induced disruption of the barrier function.

In this study, we have observed that there was an increase in NF-κβ activity in the colon of mice treated with DSS and that CAPE downregulates this increase, thereby exerting a protective event in a mouse model of UC. Since CAPE is a natural compound, with no known side effects (Tambuwala 2016), its therapeutic benefits are obvious and desirable when measured up against over other novel compounds with pro-tumorigenic effects; such as DMOG, which have also shown to be protective in experimental colitis (Cummins et al. 2008). The findings of this work indicate that CAPE can be used an effective first-line treatment for patients with UC, improving intestinal barrier function and halting the progression of disease, whilst promoting mucosal healing. The next stage of our work will focus on two elements, namely (1) development of nanoparticle-based colonic drug delivery of CAPE, which could allow for local delivery of the drug to inflamed tissue to ensure effective therapeutic outcomes using a lower dose and (2) identification of the NF-κβ subunit most affected by CAPE.