Ulcerative Colitis Impairs the Acylethanolamide-Based Anti-Inflammatory System Reversal by 5-Aminosalicylic Acid and Glucocorticoids

Studies in animal models and humans suggest anti-inflammatory roles on the N-acylethanolamide (NAE)-peroxisome proliferators activated receptor alpha (PPARα) system in inflammatory bowel diseases. However, the presence and function of NAE-PPARα signaling system in the ulcerative colitis (UC) of humans remain unknown as well as its response to active anti-inflammatory therapies such as 5-aminosalicylic acid (5-ASA) and glucocorticoids. Expression of PPARα receptor and PPARα ligands-biosynthetic (NAPE-PLD) and -degrading (FAAH and NAAA) enzymes were analyzed in untreated active and 5-ASA/glucocorticoids/immunomodulators-treated quiescent UC patients compared to healthy human colonic tissue by RT-PCR and immunohistochemical analyses. PPARα, NAAA, NAPE-PLD and FAAH showed differential distributions in the colonic epithelium, lamina propria, smooth muscle and enteric plexus. Gene expression analysis indicated a decrease of PPARα, PPARγ and NAAA, and an increase of FAAH and iNOS in the active colitis mucosa. Immunohistochemical expression in active colitis epithelium confirmed a PPARα decrease, but showed a sharp NAAA increase and a NAPE-PLD decrease, which were partially restored to control levels after treatment. We also characterized the immune cells of the UC mucosa infiltrate. We detected a decreased number of NAAA-positive and an increased number of FAAH-positive immune cells in active UC, which were partially restored to control levels after treatment. NAE-PPARα signaling system is impaired during active UC and 5-ASA/glucocorticoids treatment restored its normal expression. Since 5-ASA actions may work through PPARα and glucocorticoids through NAE-producing/degrading enzymes, the use of PPARα agonists or FAAH/NAAA blockers that increases endogenous PPARα ligands may yield similar therapeutics advantages.


Introduction
Ulcerative colitis (UC) is a chronic relapsing inflammation of the colonic tissue caused by the influence of complex genetic and environmental interactions [1]. Different pro-inflammatory factors, including reactive oxygen and nitrogen metabolites, eicosanoids, platelet-activating factors and cytokines, are upregulated and actively contribute to an exacerbated intestinal immune response to otherwise innocuous stimuli [2,3]. For the management of human UC, several drugs are currently used as 5aminosalicylic acid (5-ASA), glucocorticoids, anti-TNFa and immunomodulators as thiopurines, which interfere with proinflammatory cascades and effectively down-regulate the overstated inflammatory response [4][5][6]. Evidences suggested that the anti-inflammatory effect of 5-ASA can be mediated by the activation of peroxisome proliferator-activated receptors (PPARs), such as PPARa and PPARc [7][8][9][10][11], which are highly expressed in the intestinal and colonic mucosa by both epithelial cells and macrophages [12][13][14]. UC patients seem to have reduced levels of PPARc in their colonic epithelium and similar deficiencies were observed in colitis mouse models, but only in macrophages of the lamina propria [15,16]; confirming the beneficial effects of PPARc agonists on the attenuation of colon inflammation [17,18]. Less information is available on PPARa and b/d receptors.
Recent studies in experimental colitis suggested that PPARa ligands also have anti-inflammatory properties, which was enhanced after glucocorticoid treatment, but was weakened in PPARa-null mice [19][20][21][22][23]. Moreover, 5-ASA is able to induce PPARa expression and promote its translocation to the nucleus in  an animal model of irradiation-induced intestinal inflammation [11]. PPARa is specifically expressed in the more differentiated colonic epithelial cells facing the intestinal lumen of the small intestine and colon [12][13][14]. Thus, PPARa has been proposed to participate in the intestinal epithelial barrier system; absence of PPARa expression resulted in an increase of tight junction permeability associated with apoptosis in an animal model of experimental colitis [20]. PPARa signaling system is an antiinflammatory system composed of the PPARa receptor and its endogenous ligands, the N-acylethanolamides oleoylethanolamide (OEA) and palmitoylethanolamide (PEA). It also includes the enzymes involved in their biosynthesis and release, such as N-acyl phosphatidylethanolamide-specific phospholipase D (NAPE-PLD), as well as mechanisms for cellular uptake and hydrolysis, such as fatty acid amide hydrolase (FAAH) and N-acylethanolamidehydrolyzing acid amidase (NAAA) [24][25][26]. Increased PPARa expression or enhanced PPARa ligand production can attenuate inflammatory process observed in current animal models of experimental colitis. For instance, treatment with FAAH antagonists or genetic ablation of FAAH protected against colitis inflammation [27][28][29]. Thus, PPARa system is positioned to exert a putative role in many of the points where homeostasis breaks in UC, although the anti-inflammatory role of PPARa remains to be determined in humans.
The aim of the present study is to analyze the expression and distribution of components of the acylethanolamide-PPARa antiinflammatory system such as PPARa receptor and the enzymes involved in endogenous ligand degradation (FAAH and NAAA) and biosynthesis (NAPE-PLD) in the normal human colonic tissue compared to untreated active UC at disease onset and after achieving remission, according to clinical and endoscopic criteria, and depending on treatment received (5-ASA, glucocorticoids and/or immunomodulators).

Ethics statement
Biopsies and colonic resection samples used for the present study were obtained after a written inform consent from all the patients, as requested by the clinical guides of Hospital del Mar. Research procedures were approved by the Hospital del Mar and Hospital Carlos Haya Clinical Research and Ethics Committee and were conducted according to the principles expressed in the Declaration of Helsinki.

Subjects
We selected retrospectively 24 consecutive patients diagnosed from January 2006 to December 2007 of a first flare of UC, with extensive or left-side extension according to the Montreal classification (E2 and E3) [30]. UC was defined by the criteria of Lennard-Jones [31]. All patients had to achieve clinical and endoscopic remission after medical treatment according to Truelove-Witts index (,6 points) [32] and endoscopic Mayo score (score = 0) [33] during 12 months of diagnosis. We excluded patients with distal UC according to Montreal classification (E1) and patients without clinical and endoscopic remission criteria after treatment. Thus, we obtained several endoscopic samples of UC mucosa from each patient collected before any treatment (active group), and after medical treatment and endoscopic remission (quiescent group).
Colonic samples were retrieved from tissue bank of Pathology Service at the Hospital del Mar. Data from each patients were collected retrospectively from medical records including age, sex, smoking habits and alcohol history, body mass index and For the control group, we selected retrospectively 22 patients who were pathohistologically given a definite diagnosis of colorectal cancer and who had not received preoperative radiotherapy or chemotherapy treatment and underwent colonic resections for colorectal cancer. Several colonic resections were obtained from each patient at least 10 cm from the tumour (control group). We confirmed histopathologically the absence of microscopic alterations (Table 2).
Colonic samples were frozen at 280uC for molecular analysis (N = 7-8) or fixated with 4% paraformadehyde in 0.1 M phosphate buffered saline (PBS) by immersion and included in paraffin until immunohistochemical analysis (N = 22-24). The analysis of the immunostaining patterns was carried out at transmural planes of the normal and pathological colonic tissue by comparing it with hematoxylin-eosin staining.

mRNA isolation and quantitative RT-PCR analysis
In order to evaluate the mRNA expression we collected prospectively 7 colonic endoscopic biopsies from patients with a first flare of active UC and 8 colonic resections, at least 10 cm from the tumour, of patients with colorectal cancer (control group). Colonic resections were divided into mucosa, containing both epithelium and lamina propria, and submucosa layers, containing smooth muscle and enteric plexi. Reverse transcript reaction was carried out from 4 mg of mRNA using the Transcriptor Reverse Transcriptase kit and random hexamer primers (Transcriptor RT, Roche Diagnostic GmbH, Manheim, Germany). Quantitative real-time reverse transcription polymerase chain reaction (quantitative RT-PCR) was performed using a CFX96TM Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA), and the SYBR Green detection format (FastStart Universal Master Kit, Roche, Mannheim, Germany). Each reaction was run in duplicate and contained 5 ml of cDNA. Quantification was carried out with the classic standard curve method run at the same time. We analyzed the housekeeping genes SP1 transcription factor and bACTIN, selecting the most suitable according to their homogeneity ( Figure  S1). Absolute values from each sample were normalized with regard to the housekeeping gene SP1. Primers for PCR reaction were designed based on NCBI database sequences of human reference mRNA (Table 3), checked for specificity with BLAST software from NCBI website (http://blast.ncbi.nlm.nih.gov/Blast. cgi) and synthesized by Invitrogen.

Quantification of mucosa immunostaining
For epithelium, we carried out a densitometrical quantification for each protein. For lamina propria, we evaluated the type and the number of immunostained immune cells per area (mm 2 ) analyzed. In addition, quantification was segregated depending on UC severity and treatment received: 5-ASA, glucocorticoids, and/ tissue showed prominent immunoreactive bands of the expected size for PPARa (52 kDa), NAAA (31 kDa), NAPE-PLD (46 kDa) and FAAH (62 kDa

Statistical analysis
Data were analyzed using SPSS 15.0 software (Statistical Package for the Social Sciences Inc., Chicago, Illinois, USA). Results are expressed as mean 6 S.E.M. Differences between groups were evaluated using Student t test for parametric observation and Mann-Whitney U and Wilcoxon tests for non parametric observations. A P value of P,0.05 was considered statistically significant.

Results
Presence and distribution of PPARa, NAAA, NAPE-PLD and FAAH in the normal human colonic tissue The normal colonic tissue showed gene expression of PPARa, NAAA, NAPE-PLD and FAAH in the mucosa, including epithelium and lamina propria, and the submucosa layers, containing smooth muscle and enteric plexi (Figs. 1A-D). Protein extracts from normal colonic tissue confirmed the presence of protein levels of PPARa, NAAA, NAPE-PLD and FAAH. They appeared as prominent immunoreactive bands of expected molecular masses at ,52 kDa for PPARa, ,31 kDa for NAAA, ,46 kDa for NAPE-PLD and ,62 kDa for FAAH (Figs. 1E-H).
Results of the immunohistochemical distribution were summarized in a rating scale (Table 4). PPARa immunoreactivity was observed in the colonic epithelium of both absorptive and goblet  Figure 1K). Moderate to low intensity of NAAA immunostaining was observed in the colonic epithelium (Figs. 1L, M). Interestingly, we detected numerous NAAA immune cells in the lamina propria, which showed a variety of shapes and sizes (Fig. 1M). Muscularis mucosae, muscularis externa, plexi and serosa showed very weak staining for NAAA (Fig. 1N). Intense NAPE-PLD immunoreactivity was widely distributed in the colonic epithelium, being prominent in the perinuclear portion of the absorptive cells (Figs. 1O, P, inset). Some positive plasma cells were also observed in the lamina propria (Fig. 1P). Strong NAPE-PLD immunostaining defined both layers of muscularis externa, but low immunostaining was detected in fibers of the myenteric plexi (Fig. 1Q). FAAH immunoreactivity was mainly detected in the colonic epithelium, which shows higher expression in the apical portion of the epithelial cells (Figs. 1R, S). A low number of immunoreactive immune plasma cells were observed in the lamina propria and no staining was detected in the muscularis mucosae, the muscularis externa and the serosa. However, we can observe a specific FAAH immunoreactivity in nervous cells of the myenteric plexi (Fig. 1T).

Quantification of PPARa, PPARc, NAAA, iNOS, NAPE-PLD and FAAH gene expression in the mucosa of UC patients
In order to evaluate any changes on the expression of PPARa signaling system in the colonic mucosa (epithelium and lamina propria) of UC patients, we analyzed the relative differences in the mRNA levels of selected genes such as PPARa, PPARc, NAAA, iNOS, NAPE-PLD and FAAH in the UC mucosa, containing epithelium and lamina propria, by quantitative RT-PCR. We detected significantly lower levels of PPARa (P,0.05), PPARc (P,0.01) and NAAA (P,0.05) mRNA in the mucosa of UC patients compared to that of control ones ( Figs. 2A-C). In contrast, iNOS and FAAH gene expression was significantly higher in the mucosa of UC patients (P,0.05) (Figs. 2D, F). We observed no change in the levels of NAPE-PLD mRNA between both groups (Fig. 2E).
Densitometrical quantification of PPARa, NAAA, NAPE-PLD and FAAH immunoreactivity in the epithelium of UC patients depending on treatment  Figures 3D, H, L, P respectively. We detected a decrease of PPARa and NAPE-PLD immunoreactivity in the epithelium of UC patients compared to that of control ones (P,0.01 and P,0.001 respectively) (Figs. 3D, L). In contrast, NAAA immunoreactivity was more prominent in the epithelium of active UC patients (P,0.01) (Fig. 3H). No change was detected in FAAH immunoreactivity in the epithelium between active UC and control groups (Fig. 3P). In order to address the disease severity, we analyzed the NAE-PPARa signaling system depending on the clinical score (mild, moderate and severe) in active UC patients ( Figure S2). UC patients with moderate clinical score showed a significant reduction (P,0.05) of PPARa immunohistochemical expression ( Figure S2A). When NAPE-PLD immunoreactivity was analyzed, we detected significant decreases in UC patients with mild (P,0.01), moderate (P,0.01) and severe (P,0.001) clinical score ( Figure S2B). However, FAAH immunoreactivity was no affected ( Figure S2C). Finally, UC patients with moderate (P,0.05) and severe (P,0.01) clinical score showed significant increases in NAAA immunoreactivity ( Figure S2D). We also analyzed the possible effect of gender and smoking habits on the NAE-PPARa signaling system in active UC patients. We did not detect differences between females and males or between smokers and non-smokers in the immunohistochemical expression in the epithelium of active UC patients (Figures S3 and S4).
We also quantified the immunoreactivity in the colonic epithelium of quiescent UC patients depending on the treatment received: 5-ASA (3 cases), 5-ASA and glucocorticoids (15 cases), or 5-ASA, glucocorticoids and immunomodulators (6 cases). 5-ASA treatment produced an increase of PPARa immunoreactivity in the colonic epithelium of active UC patients (P,0.05), but not when UC patients were treated with 5-ASA in combination with other drugs (Fig. 3D). Thus, no difference in PPARa immunoreactivity was observed in UC patients treated with 5-ASA plus glucocorticoids only or 5-ASA, glucocorticoids and immunomodulators with respect untreated active UC ones. Interestingly, the decrease in PPARa immunoreactivity observed in the epithelium of UC patients treated with 5-ASA, glucocorticoids and immunomodulators was significant compared to that of UC patients treated with 5-ASA only (P,0.05). However, we cannot detect significant changes in NAAA immunoreactivity in the epithelium of quiescent UC patients treated with any of the drugs, being similar to that of active UC patients (Fig. 3H). Regarding NAPE-PLD immunoreactive levels, there was a significant increase to control levels in UC patients treated with 5-ASA (P,0.05) or 5-ASA and glucocorticoids (P,0.01) compared to the active UC patients (Fig. 3L). However, a wide variability in the intensity of NAPE-PLD immunoreactivity was detected in UC patients treated with 5-ASA, glucocorticoids and immunomodulators. Finally, we cannot observe any difference in FAAH immunoreactivity in the UC epithelium after treatment (Fig. 3P).

Acylethanolamide producing/degrading enzyme ratio in the colonic epithelium
In order to analyze whether the differential immunohistochemical expression of either acylethanolamide producing or degrading enzymes may have resulted in an altered PPARa endogenous ligand tone in the untreated active and treated quiescent UC epithelium, we calculated the ratios between NAPE-PLD and NAAA expressions, and between NAPE-PLD and FAAH expressions. These ratios can suggest possible changes of OEA/PEA levels (Fig. 4). The main result of these analysis was that there was a significant decrease of both NAPE-PLD/NAAA (P,0.001) and NAPE-PLD/FAAH (P,0.05) ratios in the epithelium of untreated active UC patients. Interestingly, we detected an increase of NAPE-PLD/NAAA and NAPE-PLD/FAAH ratios (both at P,0.05) only in the epithelium of quiescent UC patients treated with 5-ASA and glucocorticoids, but not with 5-ASA or 5-ASA, glucocorticoids and immunomodulators (Fig. 4).

NAAA, NAPE-PLD and FAAH immunoreactive cells in the lamina propria of UC patients and after treatment
The number of NAAA and FAAH immunoreactive cells in the lamina propria showed significant changes in active UC patients and after treatment (quiescent group). We found a significant low number of NAAA-ir cells (2.75-fold; P,0.001) in the infiltrate of active UC patients, but was completely restored, similar to control level, after treatment (Fig. 5). Performing double immunofluorescence, NAAA expression was found in CD19-positive (+) B   lymphocytes, CD3+ T lymphocytes and CD14+ macrophages (Fig. 6). In contrast, the number of FAAH-ir cells of the lamina propria increased dramatically (10-fold; P,0.001) in active UC patient (Fig. 7). After treatment, the number of FAAH-ir cells decreased significantly in quiescent UC patients (P,0.001), but did not reach control level. We did not find change in the number of NAPE-PLD-ir cells in the lamina propria of active UC patients and after treatment (Fig. 8). Nearly all the FAAH-ir cells in the infiltrate expressed the plasma cell-specific CD38, which include B lymphocytes and natural killer cells (Figs. 9A-C). NAPE-PLD expression was mainly found in CD38+ plasma cells and CD3+ T lymphocytes (Figs. 9 D-I).

Discussion
The key findings of this study are to demonstrate that profound changes in the acylethanolamide-PPARa anti-inflammatory system are produced in human UC. Overall the findings suggest that active UC deactivate this anti-inflammatory system while 5-ASA/ glucocorticoids treatment restores its normal expression ( Table 5). The process involves both receptors and enzymes for acylethanolamides.
Considering PPARa receptor we found that it is mainly expressed in the human colonic epithelium, but not in immune cells of the lamina propria [14]. Interestingly, colonic mucosa (epithelium and lamina propria) in active UC patients at disease onset showed a significant down-regulation of both PPARa and PPARc mRNA expression in colonic mucosa of active UC patients. These data indicate that, not only PPARc, but also PPARa, are implicated in the pathophysiology of the human colonic inflammation. We also detected an over-expression of iNOS mRNA, a pro-inflammatory mediator that produces nitric oxide species and leads oxidative stress and cell death [11,36,37]. This enzyme is under the active control of PPARa receptor since PPARa agonists enhance its degradation [38]. Immunohistochemical results demonstrated that PPARa mRNA down-expression in the UC mucosa correlated with PPARa protein downexpression in the UC epithelium. Moreover, only 5-ASA treatment increased immunohistochemical expression of PPARa to control expression level, but not when UC patients were treated with 5-ASA in combination with glucocorticoids and/or immunomodulators.
5-ASA is structurally related to nonsteroidal anti-inflammatory drugs that shares molecular targets including inflammation, proliferation and/or apoptosis [39][40][41]. 5-ASA inhibits inflammation by scavenging free radicals and thus interfering with the arachidonic acid metabolism [42]. Recent studies indicated that the anti-inflammatory effect of 5-ASA is mediated by the activation of PPARc [9][10][11], a nuclear receptor whose agonists can suppress or delay inflammation effectively by inhibiting multiple steps in NF-kB and AP-1 signaling pathways [7,8] and attenuating the production of nitric oxide (iNOS) and macrophage-derived cytokines such as TNFa, IL-1 and IL-6 in mouse models of colitis [16,43,44]. Moreover, Linard et al. [11] showed that 5-ASA is able to induce PPARa, PPARc and RXRa coexpression and promote their translocation to the nucleus in an animal model of irradiation-induced intestinal inflammation. In the present study, we demonstrated that 5-ASA specifically increased the expression of PPARa in the human UC epithelium suggesting that, not only PPARc, but also PPARa can be a key receptor for the potent anti-inflammatory effect of 5-ASA in the human UC [9,10]. At this time, nothing at all is known about the regulation of PPARa expression and much more studies are needed to elucidate the anti-inflammatory mechanisms of 5-ASA.
Others components of the PPARa signaling system, such as NAPE-PLD, FAAH and NAAA, are expressed in the healthy colonic epithelium and immune cells of the colonic lamina propria in humans [35]. NAPE-PLD is one of several N-acylethanolamidebiosynthesis enzymes that catalyze the release of N-acylethanolamide (NAE) from N-acyl-phosphatidylethanolamine (NAPE), converting endogenous lipids into chemical signals like oleoylethanolamine (OEA), palmitoylethanolamine (PEA) and anandamide (AEA) [41,42,45,46]. Some studies showed that the biological activity of PEA, such as anti-inflammatory and analgesic activities [47], and OEA, such as food intake [48][49][50], are mediated by noncannabinoid receptors among which PPARa is probably the most important [51][52][53]. In mammalian tissues, three enzymes responsible for hydrolyses of NAEs to fatty acids and ethanolamine have been identified: FAAH-1, FAAH-2 (human isozyme) and NAAA [24,[54][55][56][57]. Thus, it has been shown that selective FAAH or NAAA inhibitors produced an anti-inflammatory effect [26][27][28][29]. Interestingly, FAAH and NAAA have different catalytic properties and substrate specificity [24]. FAAH is catalytically active at neutral and alkaline pH and shows the highest reactivity with anandamide, followed by OEA and PEA [58]. In contrast, NAAA activity is optimum at pH 4.5-5, being inactive at alkaline pH, and hydrolyzes PEA much faster than others NAEs [24,55]. Therefore, alterations of FAAH and NAAA activity can be as a result of variations of luminal pH in colonic inflammation, and it is conceivable that reduced intracolonic pH in active UC impairs the anti-inflammatory effects of PPAR endogenous agonists [59].
In the present study, we demonstrated that mRNA and protein expression of NAPE-PLD, FAAH and NAAA was partially altered in active colitis, and immunohistochemical expression of these enzymes was partially restored after treatment (quiescent colitis) in a tissue-dependent manner (epithelium and immune cells of the lamina propria). Overall, the present data suggested that both increase of NAAA expression and lack of change in FAAH expression in the UC epithelium agree with a substantial reduction of luminal pH in the colon of UC patients [59]. Therefore, NAPE-PLD down-expression and NAAA over-expression in UC epithelium might let to a net reduction in NAEs turnover (specifically PEA) in the epithelium, leading the attenuation of the antiinflammatory response via the activation of PPAR receptors (Fig. 10). Interestingly, inflammation associated with osteoarthritis and rheumatoid arthritis showed a lower concentration of PEA in the synovial fluid compared to non-inflamed normal volunteers [60]. Changes observed in NAPE-PLD, FAAH and NAAA mRNA expression in the mucosa (epithelium and lamina propria) correlated completely with changes observed in the number of immunoreactive cells in the lamina propria of UC patients, but not with their immunohistochemical expression in the UC epithelium (see Table 5). These discrepancies can be explained by a higher expression of these enzymes in the immune cells during UC  Table 5. Summary of the changes detected in PPARa signaling system (PPARa, NAPE-PLD, FAAH and NAAA) in the colonic epithelium and lamina propria of active UC patients and after treatment (quiescent UC patients) 1 . infiltration, but also the different roles of the NAE-PPARa signaling system in colonic epithelium and lamina propria.
In the lamina propria of healthy human colon, we found that the number of NAAA-ir immune cells was 50-fold higher than the number of FAAH-ir immune cells. These data can suggest a higher rate of PEA hydrolysis in comparison with AEA hydrolysis. In the lamina propria of active UC, we found that the number of FAAH-ir immune cells increased up to 10-fold, whereas the number of NAAA-ir immune cells decreased up to 2.75-fold, suggesting a concomitant increase of AEA hydrolysis as well as decrease of PEA hydrolysis. These results can be related with the fact that AEA activates cannabinoid (CB1 and CB2) receptors, whereas PEA is inactive on these receptors, but activates PPARa [51], playing different roles in inflammatory activation. Previous biochemical and immunochemical analysis demonstrated NAAA expression in macrophage cells of the rat lung and brain [25].
Here, we showed that NAAA is predominantly expressed in macrophages and B and T lymphocytes in the lamina propria of UC patients. Most FAAH-ir cells in the lamina propria of UC patients expressed CD38, a surface glycoprotein found in plasma B and natural killer cells, and this result agrees with previous studies showing FAAH activity in lymphocytes [61]. NAPE-PLDir cells in the lamina propria of UC patients were CD38+ plasma cells and CD3+ T lymphocytes, but not CD14+ macrophages, contrary to expectation after pro-inflammatory stimuli [62].
UC-specific treatments produced tissue-dependent impairments in the expression of PPARa signaling system. NAPE-PLD, but not NAAA or FAAH, responded to treatment in the epithelium, while NAAA and FAAH, but not NAPE-PLD, responded to treatment in the immune cells of the lamina propria of UC patients. 5-ASA produced an increase of NAPE-PLD immunohistochemical expression (similar to control levels) in the quiescent UC epithelium, which was enhanced after corticosteroid treatment. Interestingly, the analysis of the NAPE-PLD/NAAA and NAPE-PLD/FAAH ratios suggested an increase of NAEs production in the UC epithelium after 5-ASA/corticosteroid treatment, but not when patients were treated exclusively with 5-ASA. It is clear that 5-ASA treatment leads to an increase of NAPE-PLD and PPARa expression, so probably both 5-ASA and the concomitant overproduction of NAEs via glucocorticoids can enhance an antiinflammatory response in the epithelium of UC patients by the activation of PPARa (Fig. 10). This hypothesis agrees with previous data indicating that glucocorticoids generate antiinflammatory regulatory responses by promoting arachidonic acid-containing lipid biosynthesis [63]. Treatment also increases the number of NAAA-ir immune cells, reaching control levels and, probably, normalizing PEA hydrolysis. However, the significant decrease of FAAH-ir immune cells after treatment did not reach control levels, so there may be still an over-degradation of AEA in the lamina propria of UC patients.
We must pay attention on two limitations related with the cohort of patients used in the present study. As a result of prioritizing clinical, endoscopical and histopathological considerations to obtain a homogeneous cohort, control and UC groups were not-age matched. Additionally, smoker patients and patients and controls from both genders were included in the study. However, these factors cannot be included in additional analysis because of the size of the cohort, designed to be a within-subject design (patients were they own control for quiescence status).
In conclusion, our results indicated that PPARa, NAPE-PLD, FAAH and NAAA form part of a key lipid signaling system that regulates UC-activated inflammatory response in human. 5-ASA, through PPARa receptor, and glucocorticoids, through acylethenolamide producing/degrading enzymes, reduces colitis-associated inflammation suggesting PPARa agonists or FAAH/NAAA inhibitors as potential drugs for the treatment of inflammatory bowel diseases in human. Figure S1 Housekeeping gene expressions of SP1 transcription factor (A) and bACTIN (B) represented by the threshold cycles (C(t)). We cannot detect differences in gene expression between control and active UC patients. Student t-test