Inflammatory Bowel Disease: Mechanisms,Redox Considerations,and Therapeutic Targets

A. Alteration of the apical junctional complex

The cytokines interferon γ (IFNγ) and tumor necrosis factor α (TNFα), which are central mediators of intestinal inflammation, are able to damage epithelial barrier functions independently of their pro-apoptotic action. These cytokines have been shown to disassemble tight junctions (TJs) of apical junctional complex and to enhance paracellular permeability, by promoting internalization and cellular redistribution of their junctional adhesion molecule (JAM)-A, occludin, and claudins1, but not of zonula occludens-1 or adherens junction proteins in T-84 cells (32).

A dramatic loss of JAM-A expression has been found not only in both CD and UC, but also in dextran sodium sulfate (DSS)-induced experimental colitis in mice. Consistently, JAM-A null mice, in which colitis was induced by DSS treatment, showed a marked increase in epithelial permeability and inflammatory cytokine production (272).

Redistribution of claudins has also been observed in IBD; however, claudins-3 and −4 seem to be reduced, or almost entirely redistributed to the basolateral surface of the epithelium, the expression of claudin-2 increased predominantly in cells along inflamed crypts of both UC and CD patients. Interestingly, claudin-2 was up-regulated in T-84 cells by interleukin (IL)-13, which is a key effector of the T-helper (Th)2 response and is produced in large amounts in the lamina propria of UC patients (205). In these individuals, IL-13 has been found to induce apoptosis of colonic epithelial cells and to lower transepithelial resistance by increasing paracellular permeability (92).

It is still unclear whether damage to the transmembrane proteins that are components of junctions should be considered a cause or a consequence of intestinal inflammation. Under inflammatory conditions, as in IBD patients, endocytosis of junctional molecules increases, and intracellular redistribution may cause a breakdown of the protective barrier. This process may enhance production of inflammatory infiltrate and of cytokines, which further contribute to amplifying epithelial barrier damage.

B. Impairment of intestinal antimicrobial peptides

The intestinal mucosa is covered by a thick layer of secreted mucus containing antimicrobial molecules, which protect the mucosal surface against microbial invasion. It has been suggested that a possible cause of IBD is a defect in mucus secretion which can expose the mucosa not only to pathogenic microbes, but also to assault by commensal bacteria that become pathogens, and contribute to chronic inflammatory response. Several mucin (MUC) glycoproteins and antimicrobial peptides, such as defensins, cathelicidins, lactotransferrin, and lysozymes, are highly effective in the host defense.

MUCs, trefoil factors (TFFs), and resistin-like molecule β (RELMβ) are produced by goblet cells (GCs) in varying proportions throughout the human gastrointestinal tract. Up to 20 different MUC genes have been identified: MUC2 was the first human secretory MUC to be identified and widely studied. Mice with missense mutations in the MUC2 gene or MUC2-deficient animals spontaneously develop increased susceptibility to colitis (90, 270).

A significant down-regulation of MUC2 in CD and MUC12 gene expression has been found in colon tissue in both CD and UC (171). MUC1, MUC3, and MUC19 have also been identified as possible susceptibility genes for IBD (165, 231).

TFF3 interacts with MUC2 by enhancing the stability of the MUC layer, thus protecting epithelial cells from intestinal stressors (124). TFF3-deficient mice show a higher susceptibility to DSS-induced colitis. Conversely, TFF3 overexpression increases resistance to intestinal damage in mice (126). RELMβ is also strongly expressed in intestinal GCs by colonization with normal enteric bacteria, and in mice with gastrointestinal infections or IBD. It is involved in the induction of CD4⁺ T cell-dependent adaptive immune response (179).

Paneth cells (PCs) control bacterial attack of the host mucosa, regulating bacterial numbers by secreting various defensins, as well as lysozyme and secretory phospholipase A₂ (Plase A₂). Defensins provide an important antimicrobial function, because of their ability to regulate the adherence of specific bacteria to the intestinal epithelium. In humans, two α-defensins secreted under normal conditions by PCs have been identified, namely human α-defensins (HD)-5 and HD-6 (130). Reduced expression of HD-5 and HD-6 by PCs has been observed in ileal mucosa from CD patients, with no changes in lysozyme or secretory Plase A₂ expression (283).

The importance of human β-defensins (HBDs) in IBD pathogenesis is still unclear. HBD-1 is constitutively expressed in small amounts in normal colonic epithelial cells. HBD-1 has been found to be decreased in the inflamed mucosa of IBD patients. In contrast, HBD-2 was strongly induced in the inflamed mucosa of UC, but not CD, patients (210, 284). Defensin production is under the control of the molecules involved in commensal and pathogenic bacteria recognition, namely nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs) and Toll-like receptors (TLRs) (see section “Impairment of receptor-mediated microbial recognition”). Alterations to HD and HBD expression are reported to be associated with caspase activation and recruitment domain 15 (CARD15)/NOD2 mutations, which, in turn, are thought to be responsible for an increased susceptibility to CD (278, 284).

Cathelicidins are antimicrobial peptides, synthesized by neutrophils, macrophages, and intestinal epithelial cells (IEC), that modulate microbial growth, wound healing, and inflammation; they counteract gastrointestinal pathogens, including several Helicobacter strains, Shigella, Salmonella, and Candida albicans (102). It is unknown whether they are associated with IBD; however, recent experimental studies have revealed a new role for these peptides in UC therapy. Administration of a synthetic mouse cathelicidin significantly prevented the development of DSS-induced UC in mice (249).

The response to commensal and pathogens bacteria and viruses of this antimicrobial peptide is mainly mediated by selective activation of the TLR-nuclear factor-κB (NF-κB) pathway (124). In fact, MUC2 has been shown to have NF-κB binding sites in the promoter. A number of Lactobacillus strains up-regulate HBD-2 via induction of NF-κB, activator protein 1 (AP-1) and mitogen-activated protein kinase (MAPK) signaling pathways (224). Recently, due to TFF3 ability to inhibit TNFα production, TLR4 and NF-κB in a well-established model of trinitrobenzene sulfonic acid (TNBS)-induced colitis in mice, TFF3 has been hypothesized to be a potent agent in controlling IBD (258).

C. Impairment of receptor-mediated microbial recognition

The different dietary antigens are recognized by pathogen recognition receptors (PRRs), which discriminate among pathogenic bacteria, commensal bacteria, and nutrients. PRRs recognize conserved molecular motifs known as pathogen-associated molecular patterns (PAMPs) or microbe-associated molecular patterns. PRRs include membrane-bound TLRs, the RNA helicase family of receptors, that is, retinoic acid inducible gene I (RIG-I)-like receptors (RLRs) and cytoplasmic NLRs. TLRs recognize bacteria-derived lipopolysaccharide (LPS), flagellin, or unmethylated cytosine nucleotide next to guanine separated by a phosphate (CpG); RIG-I and the melanoma differentiation-associated gene 5 (MDA5) belong to the RNA helicase family of receptors, and are primary sensors of RNA viruses; and NLRs recognize bacteria-derived peptoglycans (285).

3. NLRs and their modulation by ROS

NLRs are emerging as crucial regulators of the inflammatory response to commensal microflora in the gut. NLRs are cytoplasmic proteins that are characterized by the presence of N-terminal protein–protein interaction domains (such as CARD, pyrin, and baculovirus inhibitor of apoptosis protein repeat domains), a central conserved NOD domain, and the carboxy-terminal leucine-rich repeats (LRRs), which are involved in microbial sensing. Among the different NLRs (23 members in humans), NOD1 and NOD2 have been most widely studied in the intestine. NOD1 is ubiquitously expressed in intestinal cells, including IEC, whereas high levels of NOD2 are constitutively expressed in phagocytes and dendritic cells, and are induced in PCs of the small intestine by inflammatory stimuli. These NLRs recognize distinct PAMPs: NOD1 recognizes γ-glutamyl-meso-diaminopimelic acid, while NOD2 recognizes muramyl dipeptide. They are actively involved in the recognition and activation of inflammatory and immune responses against intestinal pathogenic bacteria, such as Enterobacteriacee, Helicobacter pylori, and Listeria monocytogenes. On activation, NOD1 and NOD2 form cytoplasm platforms called “NOD signalosomes,” which activate MAPKs and NF-κB, by recruiting the receptor-interacting protein 2 (RIP2), and producing pro-inflammatory cytokines (e.g., IL-8 and IL-1β) (74). NOD2-dependent NF-κB activity appears to be closely related to the perturbation of actin dynamics in cells (145). Through NF-κB activation, NOD2 induces the release of antibacterial molecules produced in PCs, such as HD.

Another set of NLRs responds to different stimuli to form protein complexes known as inflammasomes, which are involved in the activation of pro-caspase-1. Caspase-1 is considered pro-inflammatory, because it activates immature forms of the IL-1 family (IL-1β, IL-18, and IL-33) and induces IL-1 secretion. There are different groups of inflammasomes, the main ones being NOD, LRR and pyrin domain containing 1 and 3 (NOD leucine rich repeat and pyrin domain [NLRP]1, NLRP3), and IL-1β converting enzyme-protease activating factor.

Of interest, ROS are crucial in up-regulating NLRP3 inflammasome activity. Most of the danger-associated molecular patterns (DAMPs) listed earlier can activate phagocytic cells, with a consequent overproduction of ROS. For example, enhanced generation of ROS has been observed after macrophage-like THP1 cell exposition to asbestos and silica; NLRP3 activation is triggered by ROS produced by up-regulated NADPH oxidase (61) (See Sections “The IL-1 family and inflammasomes as ROS-dependent targets” and “NADPH oxidase as target of the pathogen recognition pathway”).

Inflammasomes may play an important role in IBD, especially because of their ability to activate pro-inflammatory caspases. Recent reports have hypothesized a relationship between NLRPs dysregulation and IBD. Activation of NLRP1 has been observed in metaplastic PCs in the colon of IBD patients (237). Furthermore, NLRP3 inflammasome has been found crucial in inducing experimental colitis in mice: NLRP3 ^−/− mice developed a markedly less severe clinical picture after DSS treatment (19). NLRP3 was found to be overexpressed both in the experimental model of TNBS-induced colitis in mice, and in the damaged mucosa of human CD patients. However, the precise role of NLRP3 inflammasome in IBD is still under debate. Recent reports have shown that NLRP3 inflammasome-induced cytokine production might, under specific conditions, confer protection against experimental colitis (95, 296).

In conclusion, consistently impaired signaling by TLRs and NLRs takes place during the development of IBD, with consequent enhancement and endorsement of inflammatory reactions within the gut wall (see the schematic representation in Fig. 1).

OPEN IN VIEWER

FIG. 1.

Inappropriate response of the mucosa-associated immune system to commensal and pathogenic microbiota in IBD. IBD, inflammatory bowel disease; NLRs, nucleotide-binding oligomerization domain-like receptors recognizing bacteria-derived peptidoglycan; TLRs, Toll-like receptors recognizing bacteria-derived lipopolysaccharide, flagellin, or unmethylated cytosine nucleotide next to guanine separated by a phosphate.

D. The role of the endothelium in IBD progression

IBD development comprises alternate phases of tissue damage and tissue repair. The latter process implies neoangiogenesis, then the synthesis of pro-angiogenic factors, such as vascular endothelial growth factor and fibroblast growth factor, by macrophages and lymphocytes recruited in the inflammatory areas. Endothelial cells of the inflamed capillaries also contribute to the production of angiogenic factors.

Intestinal endothelial cells express CD40 and CD40L proteins, which belong to the TNF receptor superfamily, and that have co-stimulatory activity for immune cells, such as T cells, and for non-immune cells. CD40 pathway activation leads to the amplification of immune and inflammatory responses. In both UC and CD, CD40 and CD40L are overexpressed, especially in severely inflamed mucosa (54).

Bacterial antigens that can penetrate the epithelial barrier may reach the endothelium, where they activate PRR signaling: Various TLRs and both NOD1 and NOD2 have also been found expressed on the surface of endothelial cells, and are up-regulated by specific bacterial fragments. The possibility of hyper-activation of PRRs in inflammatory conditions suggests further function of endothelial cells as a second barrier against antigens (91).

Expansion of the microvasculature in inflamed intestinal tissue stimulates the recruitment of pro-inflammatory mediators, thus further amplifying inflammation and tissue damage.

A. Cytokines produced by Th cells

Lymphocytes Th1 are the main mediators of type1 immunity. They act against intracellular pathogens by means of intense phagocyte activity and produce IL-2, IFNγ, and TNFα; the same cytokines amplify this response by inducing macrophages to release further amounts of TNFα, IL-6, and IL-1β. Th1 cells and related cytokines are predominant in active CD. The histopathological features of CD are shown in Figure 4. The expressions of IL-1β, IL-6, and TNFα are increased in the affected mucosa of patients with active IBD. Conversely, the same patients showed decreased levels of IL-2 and IFNγ in the peripheral blood cells (209).

OPEN IN VIEWER

FIG. 4.

Histological evidence of CD in colonic biopsy. Diffuse inflammation involves the greater part of the tissue section. Giant cells, histiocytes, and a large number of intraepithelial lymphocytes are present. Glands appear normal but hyperplastic. The specimen was stained with Hematoylin and Eosin (20×magnification) and viewed using a microscope equipped with a LEICA Photocamera. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Among the various cytokines produced during the Th1 response, the overexpression of TNFα undoubtedly plays a key role in the induction of intestinal damage, especially in CD patients. This molecule can trigger pro-inflammatory and survival pathways, as well as the apoptotic process. Its effect probably depends on the amount and time of cell exposure to the cytokine, the type of receptor that binds TNFα, and the sequence of events that take place on ligand binding. Regulation of NF-κB activity appears to be fundamental in determining the specific cellular response to TNFα (211). Based on the extensive research that has been done into the biological activity of this molecule, the use of TNFα blockers in treating CD has recently generated considerable interest.

Cytokines produced by Th2 cells are selectively increased in the inflamed mucosa of UC. These cells synthesize IL-4, IL-5, IL-10, and IL-13, and are required for humoral response (185). Th2 lymphocytes support the synthesis of all types of immunoglobulins, by secreting IL-4 and IL-5, and inhibit Th1-activated macrophages by secreting IL-4, IL-10, and IL-13. IL-4 and IL-10 are considered anti-inflammatory cytokines. IL-4 secretion by dendritic cells is also a fundamental up-stream effector of Th2 differentiation. Conversely, IL-13 is considered a pro-inflammatory molecule that is mostly implicated in allergic inflammation. Mononuclear cells from intestinal lamina propria of UC patients show high levels of Th2-related IL-13 (92).

The function of specific lymphocytes is more complicated. For example, both Th1 and Th2 cells can increase transcription of MUC2, through NF-κB activation. Th2 cells utilize IL-4 and IL-13, which mediate NF-κB-dependent MUC2 transcription through MAPK signaling. On the other hand, Th1 cells up-regulate NF-κB-dependent MUC2 transcription through the TNFα-phosphatidylinositol 3 kinase (PI3K)/Akt cell pathway (124).

Lymphocytes Th17 secrete pro-inflammatory cytokines IL-17, IL-22, TNFα, and IL-6, and have been identified as potent inducers of tissue inflammation, because they recruit neutrophils and macrophages when Th1-type and Th2-type immunity fails to clear pathogens (156). The cytokine IL-23 is essential for maintaining Th17 response, and is characteristically associated with Th17 cell lineage differentiation and expansion. Furthermore, IL-23 counteracts anti-inflammatory pathways, by inhibiting the generation of inducible T regulatory cells, and reducing IL-10 secretion by other CD4⁺ T cells (175).

B. Cytokines produced by T-regulatory cells

Treg cell subsets are activated by the two anti-inflammatory cytokines IL-10 and TGFβ, mainly those produced by dendritic cells (46). They are characterized by the expression of the transcription factor forkhead box protein 3 (FoxP3), and show the so-called CD4⁺CD25(high) phenotype in humans. They are, thus, also termed CD4⁺CD25(high)FoxP3⁺ cells. During the active phase of the disease, both CD and UC patients showed a decreased number of peripheral suppressive Tregs, whereas an expansion of the Treg population has been found in mucosal lymphoid tissues of these patients (20, 295).

Tregs may indirectly control the production of pro-inflammatory TNFα, through the T-box transcription factor T-bet. T-bet-deficient mice developing severe colitis showed a selective increase of TNFα production by colon dendritic cells. This finding suggests that colonic T-bet deficiency in the dendritic cells may confer genetic susceptibility to aberrant TNFα-related inflammatory response, by triggering strong apoptotic processes in the intestinal epithelium, which becomes more sensitive to pathogenic microbiota (79).

IL-10 and TGFβ have long been known as the two main cytokines involved in the negative regulation of inflammation and immune response, thus contributing to the maintenance of gut homeostasis.

The protective role of IL-10 in IBD had already been demonstrated in an IL-10 knock-out (KO) mouse model of colitis, which required gut microbial intervention to develop inflammation. This gave rise to the idea that IL-10 could be involved in restricting the mucosal immune response to the enteric flora (138). IL-10 inhibits pro-inflammatory cytokine expression and nitric oxide (NO) generation in macrophages. It has been observed that IL-10 activates different cell signaling pathways in order to exert its anti-inflammatory action. In particular, it promotes proliferation in selected immune cells through JAK/STAT3 or PI3K/Akt pathways. IL-10 can also inhibit TNFα production and hemeoxygenase-1 induction in macrophages through the activation of p38 MAPK pathway (201). Impairment of IL-10 signaling has been reported in a pediatric group of IBD patients with granuloma-positive colitis (20). These findings suggest that IL-10 is a potential therapeutic target for IBD, although no apparent beneficial effect of IL-10 therapy has yet been observed in a number of human clinical trials, probably due to IL-10 instability and its short half-life (114).

The TGFβ cytokine family is recognized as a key regulator of cell proliferation, differentiation, and apoptosis in the intestinal mucosa, as well as of the control of immunological self-tolerance, inducing Treg activation by up-regulating FoxP3; high levels of the TGFβ1 isoform have been found to inhibit induction of Th17 cells (297). TGFβ1 is very probably implicated in organ fibrosis, and may be overexpressed by CD stricture fibroblasts. Tissue levels of this cytokine are transiently diminished in the early phases of IBD, whereas they increase at later disease stages (33, 110). It has been hypothesized that the TGFβ1 cell signaling pathway is defective in IBD: Up-regulation of Smad7, the intracellular inhibitory molecule of the TGFβ1 signal, followed by a block of phosphorylation of the principal transduction molecule Smad3, have been reported to occur in murine IBD models (173). Finally, TGFβ1 has long been recognized to possess marked pro-angiogenic activity, so that the pleiotropic action of this cytokine appears to play a primary role in the chronicization of IBD (Fig. 5).

OPEN IN VIEWER

FIG. 5.

Primary contribution of the pleiotropic cytokine TGFβ1 to the chronicization of IBD process. ROS, as well as lipid oxidation products such as oxysterols and HNE, contribute toward regulating the activity of TGFβ1. This cytokine is a crucial molecule produced during inflammatory reactions. TGFβ1 is the “crossroads” for cell fate: It is responsible for enterocyte differentiation and aging, with consequent induction of cell death. In parallel, it also activates intestinal wound healing, inducing angiogenesis and fibrosis, this last mainly involved in CD pathogenesis. TGF, transforming growth factor; ROS, reactive oxygen species; L^•, lipid radical; LO^•, alkoxy radical; LOO^•, peroxy radical; H₂O₂, hydrogen peroxide; O₂ ^−•, superoxide; HNE, 4-hydroxynonenal.

A. Antioxidant defenses in IBD

B. Oxidative insult in IBD

A significant mass of data now points to a preponderance of ROS and RNS being the antioxidant defense affected in the inflamed areas of the gut of IBD. Quite recently, experimental and human studies on the steady state of ROS and antioxidants in the inflamed colon mucosa have been nicely put together and analyzed (213). IBD is a classic chronic disease process in which inflammatory cells, cytokines, chemokines, and intestinal flora interact, until they induce a net imbalance of the redox equilibrium of colonic mucosa, with the consequent development of an “oxy-radical overload” (181, 202, 269). ROS levels have been correlated to the clinical and endoscopic assessment of IBD severity (134). Of note, RNS and ROS have been demonstrated to up-regulate a number of different genes involved in adaptive and innate immune responses in the gastrointestinal tract.

2. ROS in IBD progression

The primary activity of ROS in defense against invading microbes is the well-known respiratory burst induced in phagocytic cells. In particular, stimulated macrophages generate high concentrations of ONOO⁻ and H₂O₂, acting as bactericidal agents. ROS comprise many species, that is, superoxide (O₂ ^−.), hydroxyl radical (HO⁻), and H₂O₂. Various studies have demonstrated that ROS are implicated in determining mucosal injury in UC (206). Gene expression profiles in the three different mouse models of colitis revealed alterations in the expression of important genes involved in the regulation of H₂O₂ production in the intestinal tissue (256). Furthermore, the mucosal barrier dysfunction has been related to disorganization of the cytoskeleton in epithelial cells, due to excessive ROS generation (14).

An excess of ROS causes massive DNA damage to accumulate. A reaction of the hydroxyl radical at the C-8 position of deoxyguanosine induces formation of 8-hydroxy-2′-deoxyguanosine (8-OHdG), which is considered a marker of oxidative DNA damage in biological systems (83). Oxidative DNA lesions are relevant to mutagenesis, producing several base modifications such as G:C to T:A transversions, which have been associated to TP53 mutations in UC (100). The colonic mucosa concentration of 8-OHdG has been found to be doubled in rats treated with DSS. The role of inflammation in inducing DNA oxidative damage, in terms of 8-OHdG production, has been verified in colonocytes by using the cyclooxygenase (COX)-2 inhibitor nimesulide (253, 254). It is widely thought that chronic inflammation is actively involved in intestinal tumorigenesis via the induction of DNA mutations by ROS (58, 59, 67, 166).

An important source of oxidative DNA damage is the mitochondria, because of the central role they play in energy oxidative cell metabolism. Mitochondrial DNA is particularly exposed to ROS attack, due to its localization, close to the intracellular electron transport chain in the inner mitochondrial membrane, and to the absence of protective histones. Polymerase chain reaction analyses to detect large deletions in the mitochondrial DNA, and 8-OHdG tissue concentrations, were measured in mucosal specimens from patients affected by UC. The increased gene instability of mitochondrial DNA in UC patients has been related to the high development rate of colorectal cancer (188).

However, owing to their short half-life and high reactivity, direct detection of ROS in vivo is difficult. Oxidative epithelial-cell injury has been indirectly evaluated in the mucosa of IBD patients, in whom the production of oxidants was excessive if compared with their antioxidant defenses (134). The assessment of oxidative and/or nitrative modifications of proteins has long been one of the few methods available to detect ROS/RNS generation in vivo. Inflamed mucosa of UC patients with active disease shows increased cytoskeletal protein oxidation and nitration (116). Direct evidence of in vivo ROS production in IBD has been provided using chemiluminescence methods (192, 247) or fluorescent probes (see Fig. 6) detecting O₂ ^−., H₂O₂, and NO in lymphocytes, monocytes, and granulocytes isolated from the peripheral blood of both CD and UC patients (21).

OPEN IN VIEWER

FIG. 6.

Intracellular production of ROS in IECs. ROS production was visualized in colon intestinal CaCo-2 cells using the fluorescent probe 2′,7′-dichlorofluorescein derived from the cleavage and oxidation of its reduced form, dichlorofluorescin. The fluorescence was monitored directly by laser scanning confocal microscopy (microscope Zeiss LSM 510, Carl Zeiss SpA, Arese, Milan, Italy) that was equipped with a plan neofluar lens 20×/0.75. (A), untreated cells; (B), CaCo-2 cells treated with 1 mM H₂O₂. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Recent studies, aimed at evaluating the therapeutic effect of probiotics, have demonstrated abnormal intestinal bacterial colonization in IBD, which often causes damage by inducing exaggerated oxidative reactions. Certain bacteria, such as Bacteroides, Clostridium, and Enteroccocus, can be considered detrimental; whereas Bifidobacterium and Lactobacillus may be beneficial (65). E. faecalis has been implicated in tissue damage in IBD because of its ability to induce high concentrations of H₂O₂ and of inflammatory cytokines, such as TNFα and IL-12p70 (161). ROS production is critical for efficient autophagic targeting of Salmonella enterica serovar Typhimurium in phagocytes (98). Conversely, Lactobacillus strains may ameliorate IBD severity, by activating antioxidant enzymes and anti-inflammatory cytokines, both in experimental animal models and in children (37, 153).

The importance of ROS has been stressed in terms of their bactericidal effect, as regulators of molecules involved in the recognition and processing of intestinal pathogens. ROS are essential for autophagosomal degradation of proteins: In particular, H₂O₂ specifically inactivates ATG4, most likely activating PI3K signaling, thereby promoting ATG8 lipidation, and resulting in enhanced autophagy (223). Conversely, another in vitro study has shown that ATG16L1 dysfunction, which has been observed in CD patients, results in an overproduction of ROS, which diverts cells away from autophagy, to activating NLRP3-related IL-1β production (219).

ROS generation is closely related to TLR and NLR recognition that is aimed at killing the pathogen directly. ROS are reported to be involved in the NF-κB pathway activation mediated by TLR4 in neutrophils (12). Autophagy induced by the LPS is mediated by TLR4-induced signaling. Generation of ROS with consequent recruitment of autophagy light chain peptide LC3 to phagosomes during phagocytosis of TLRs has been shown in the murine macrophage Raw 264.7 cell line (289). Various stressors, such as asbestos and silicas, activated NLRP3 to secrete IL-1β in macrophages through ROS generation mediated by adenosine triphosphate (ATP). ROS can activate the PI3K transduction pathway, which, in turn, is required to activate pro-caspase-1 and to secrete pro-inflammatory IL-1 and IL-18 (50, 61). ROS may either directly activate inflammasomes or indirectly interact with other molecules able to modulate inflammasome activity (162).

3. The role of lipid oxidation end-products in IBD

Most studies determined the presence of a condition of oxidative stress in experimental colitis and/or human IBD in terms of increased lipid oxidation products, measuring thiobarbituric acid reacting substances or aldehydic end-products, such as 4-hydroxynonenal (HNE) and malondialdehyde (213). Dietary lipids, in particular ω6-polyunsaturated fatty acids (PUFAs) and their oxidative modifications, are undoubtedly crucial in promoting chronic inflammation: They provide the main substrates for PGs and leukotrienes (LTs) formation through COXs and lipoxygenases (LOXs) activity and, for reactive aldehyde production, by non-enzymatic breakdown.

As far as COX activity is concerned, phagocytes of the lamina propria produce high levels of ω6-arachidonic acid (AA) metabolites through inducible COX-2. The eicosanoids PGE2 and LTB4 may activate Th cells to synthesize either pro- or anti-inflammatory mediators of the response to dietary antigens, by selectively triggering different signal transduction pathways (87).

A large body of literature is now available on cell-damaging effects of aldehydes deriving from non-enzymatic degradation of ω6-PUFA; some of these carbonyl derivatives appear likely to be involved in chronic inflammatory processes characterized by excessive fibrogenesis, especially in atherosclerosis and neuronal diseases (184).

The aldehyde HNE is considered a crucial molecule in cell signaling activation, because of its strong reactivity with biomolecules containing amino and thiol groups. High concentrations of HNE exert necrogenic and genotoxic effects, whereas at low concentrations, HNE appears to act as second messenger of free radical-mediated reactions. HNE appears to be involved in regulating a number of cell signals, especially in apoptosis- and inflammation-related signaling pathways. There is growing evidence of increased HNE tissue/blood levels in a wide variety of human diseases characterized by inflammation, pointing to the implication of this molecule in their pathogenesis (203).

Diet may influence tissue levels of HNE, which derives from arachidonic and linoleic ω6-acids, are often abundant especially in Western diets (246). However, HNE levels can be also affected by antioxidant intake, as well as by intestinal microbiota. HNE and its urinary metabolites have been found to be reduced as a response to improved plasma antioxidant concentrations in humans. In diabetic rats, dietary supplementation with the antioxidant α-lipoic acid reduced plasma levels of HNE and prevented the progression of diabetes mellitus complications, by inhibiting oxidative DNA damage of the peripheral lymphocytes. An in vitro model of dietary fiber fermentation decreased HNE-induced DNA damage and increased GSH-S-transferase in IEC (13, 139, 222).

High levels of HNE have been consistently detected in the intestinal mucosa and in the plasma of CD patients (26, 27). An increase in etheno-modified DNA bases, generated by the reaction of DNA with HNE, has been found in inflamed pancreatic tissue of patients with chronic pancreatitis, and in the colon mucosa of IBD patients (178).

In male Sprague-Dawley rats, in which colitis was induced by LPS injection, a transient increase of HNE in colonic mucosa was shown to modify IgA molecules, thus resulting in a reduction of bactericidal activity at the level of the intestinal epithelial layer (125). In other experimental models of colitis, induced by DSS or TNBS in C3H/HeN and C3H/HeJ mice, high levels of HNE and malonaldehyde, along with reduced antioxidant defenses (i.e., GSH content, SOD and CAT activities), have been detected in inflamed colon mucosa (144). The consistent evidence of HNE overproduction during inflammatory disease processes, along with HNE's strong reactivity with important functional biomolecules, point to this molecule being an effector of oxidative damage in IBD.

Besides lipid peroxidation-derived aldehydes, other lipid oxidation products might impact not only IBD's progression and perpetuation, but also its initiation. At least in the case of high dietary consumption of animal fats, several oxidation products of cholesterol, termed oxysterols, have recently been demonstrated as exerting a strong pro-inflammatory effect on human colonic cells (163).

The pathogenetic role of dietary cholesterol and oxysterols in several inflammatory diseases, mainly atherosclerosis, has been established unequivocally (197, 241). Oxysterols belong to a family of 27-carbon cholesterol derivatives, which originate from the diet through non-enzymatic reactions, or endogenously through the activity of specific hydroxylases, cytochrome P450-dependent or -independent enzymes. Dietary oxysterols derive from cholesterol-rich food by autoxidation reactions, promoted either by incorrect food processing and storage, or by increased ROS steady-state levels (30). The addition of oxygen to the cholesterol molecule gives the resulting metabolite hydrophilic properties and much higher mobility than the parent compound within the cell membrane, with consequent changes in structure and function of the phospholipid bilayer. Dietary oxysterols are absorbed completely from the bowel, cleared from plasma into lipoproteins, and taken up by different tissues more rapidly than cholesterol (78). The principal dietary oxysterols formed on non-enzymatic cholesterol oxidation are found in a mixture; they comprise 7α-hydroxycholesterol, 7β-hydroxycholesterol (7β-OH), 7-ketocholesterol (7K), 5α,6α-epoxycholesterol, and 5β,6β-epoxycholesterol. Their concentrations are in the 10–100 μM range in processed and/or stored cholesterol-rich foods (111).

Oxysterols perform important physiological activities that include sterol and bile acid biosynthesis and reverse cholesterol transport. However, the tissue accumulation of oxysterols has been found to be potentially more pro-inflammatory and more toxic than cholesterol accumulation for different cell types, including IEC. The main cell pathways activated by oxysterols are summarized in Figure 7. A mixture of dietary oxysterols has recently been demonstrated as interfering with the homeostasis of the human digestive tract, promoting and sustaining inflammatory processes and inducing apoptosis of enterocyte-like CaCo-2 cells. In these cells, oxysterols not only increased the expression of IL-1β (see Fig. 8), IL-6, IL-8, IL-23, and MCP-1, but also up-regulated mRNA levels of the recognition receptors TLR2 and TLR9. The NADPH oxidase isoform 1 (NOX1) appears to contribute to the pro-inflammatory and pro-apoptotic effects of oxysterols, as these events can be prevented by cell pre-treatment with the NADPH oxidase inhibitor diphenyleneiodonium. Among the oxysterols studied, 7β-OH and 7K displayed the highest cytotoxicity. However, in food, they are mixed with other oxysterols and do not elicit the same strong response, suggesting a partial competition with other oxysterols for the same cell functional sites (25, 163). Table 2 shows the main inflammatory molecules involved in IBD that are induced by lipid oxidation products.

OPEN IN VIEWER

FIG. 7.

Role of oxysterols in regulating cell signaling pathways related to oxidative reactions. Among various signaling pathways induced by oxysterols, the activation of Ca²⁺-dependent Plase A2, followed by AA release and NADPH oxidase up-regulation, has been hypothesized. The consequent ROS generation induces a complex network of cellular signals, which can lead to inflammation through NF-κB or to cell death by activating mitochondrial apoptotic pathway. Plase A₂, phospholipase A₂; AA, arachidonic acid; PYK2, proline-rich tyrosine kinase 2; PKC, protein kinase C; ERK1,2, extracellular signal-regulated kinase1,2; NF-κB, nuclear factor kappaB; Bim, Bcl-2 interacting mediator of cell death; Bad, Bcl-xL/Bcl-2 associated death promoter.

OPEN IN VIEWER

FIG. 8.

IL-1β induction in enterocytes stimulated with oxysterols. Stimulation of CaCo-2 enterocyte-like cells with a mixture of dietary oxysterols comprising: 42.96% 7-ketocholesterol, 32.3% 5α,6α-epoxycholesterol, 5.76% 5β,6β-epoxycholesterol, 4.26% 7α-hydroxycholesterol, 14.71% 7β-hydroxycholesterol. (A) untreated cells; (B) oxysterol-treated cells. IL-1β production was detected by red immunefluorescence, using a confocal microscope Zeiss LSM 510 (Carl Zeiss SpA, Arese, Milan, Italy) with a plan neofluar lens 40×/0.5. IL, interleukin. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

OPEN IN VIEWER

Table 2.

Enhanced Synthesis of Inflammatory Cytokines, Chemokines, and Adhesion Molecules, Induced by Lipid Oxidation Products Relevant to IBD progression

Lipid oxidation product/s	Cell type	Induced molecule/s	Reference
7K	macrophages	IL-1β	157
7K, 7β-OH	macrophages	IL-8	155
7K	endothelial cells	adhesion molecules	233
7K/7α-, 7β-OH	endothelial cells	adhesion molecules/IL-1β	146
7α-OH	endothelial cells	MCP-1	147
7K	smooth muscle cells	IL-6	245
7β-OH	colon epithelial cells	IL-8	163
HNE	macrophages	TGFβ1	148
HNE	macrophages	MCP-1	189
Ox-PLs	macrophages	VEGF/IL-8/COX-2	29
Ox-PLs	macrophages	GM-CSF/IL-1β, IL-6/IL-8/MCP-1/TNFα	275

7K, 7-ketocholesterol; 7α-OH, 7α-hydroxycholesterol; 7β-OH, 7β-hydroxycholesterol; HNE, 4-hydroxynonenal; Ox-PLs, oxidized phospholipids; IL, interleukin; TGFβ1, transforming growth factor beta1; MCP-1, monocyte chemotactic protein-1; VEGF, vascular endothelial growth factor; COX-2, cyclooxygenase-2; GM-CSF, granulocyte-macrophage colony-stimulating factor; TNFα, tumor necrosis factor-alpha.

In the mechanism by which ROS propagate inflammation, a key role is certainly played by the activation of NF-κB, which is involved in the regulation of several immune and inflammatory genes.

A. Nuclear factor-κB

The transcription factor NF-κB is a major regulator of multiple important physiological processes. At the intestinal epithelium, NF-κB is involved in the immune homeostasis of IEC, and in modulating intestinal barrier permeability (200). However, chronic NF-κB activation may play a major role in exacerbating chronic inflammation of intestinal epithelium. This is supported by the aberrant NF-κB activation that is consistently found in IBD (7, 215, 274). Furthermore, chronic NF-κB activation and the resulting inflammation are associated with a general risk for most cancer types (1), the link between IBD and colorectal cancer being paradigmatic (53, 66).

Transcription factor NF-κB is composed of homodimers and heterodimers of the Rel/NF-κB family of proteins (c-Rel, RelB, p65 [RelA], p50/p105, and p52/p100). NF-κB usually resides in the cytosol, forming an inactive complex with the inhibitor of κB (IκB). Its activation via the canonical pathway involves IκB kinase (IKK)-mediated phosphorylation of two IκBα serine residues (S32 and S36). IKK consists of two regulatory subunits, IKK1 (IKKα) and IKK2 (IKKβ), plus the NF-κB essential modulator (NEMO or IKKγ) regulatory subunit. IκBα phosphorylation triggers its subsequent ubiquitination and degradation by the proteasomes, and translocation of the active NF-κB dimer to the nucleus. NF-κB binds to DNA κB sites and regulates the expression of many genes that can be grouped into those which modulate the immune and stress response, cell proliferation, and cell survival (288).

NF-κB activation is a down-stream pathway in microbial recognition by IEC. On activation by different PAMPs, TLRs react with adaptor proteins MyD88 and TRIF, which, in turn, activate NF-κB. Furthermore, recognition of the bacteria-derived peptoglycan by NLRs results in the recruitment of RIP2/RIP-like interacting caspase-like apoptosis regulatory protein kinase and the subsequent activation of NF-κB (3). Altered NF-κB expression led to the dysregulation of antimicrobial peptides, which influenced the proliferation of pathogenic bacteria in the intestinal epithelium of Drosophila melanogaster, leading to the host's death (218).

NF-κB is a redox-sensitive transcription factor. Increases in ROS tissue may partly explain the chronic activation of NF-κB in IBD. ROS disrupt the redox interaction between the dynein light chain LC8 peptide and IκBα, which prevents IκBα phosphorylation by IKK (109). TNFα activates NF-κB, which involves NADPH oxidase activation, increased ROS production, release of LC8, and IκBα phosphorylation (107). In support of a role of oxidative stress in IBD-associated NF-κB activation, phenyl-N-tert-butylnitrone, an oxidant-trapping compound known to inhibit NF-κB, ameliorates DSS-induced inflammation by inhibiting the expression of NF-κB-regulated proteins, that is, TNFα, IFNγ, and iNOS (180). Consistently, this model of UC is associated with high levels of colonic mucosa oxidation and luminal nitrite/nitrate content (180).

Although the etiology of IBD is probably multifaceted, NF-κB activation is a common finding in this disease. Administration of the NEMO-binding domain peptide, which blocks the association of NEMO with IKKβ subunit, was found to inhibit IKK activation and to reduce the severity of colon inflammation in different mouse models of colitis (232). High levels of p65 are found in macrophages, endothelial cells, and lamina propria fibroblasts from IBD patients (80, 215). Hyper-activation of NF-κB can cause an increased transcription of various genes that are central to IBD pathogenesis: (i) pro-inflammatory cytokines: IL-1β, IL-6, IL-8, IL-16, and TNFα (198), which are crucial in the IBD sustained cycle of intestinal barrier disruption by furthering the NF-κB-dependent release of pro-inflammatory mediators (43, 255); (ii) genes regulating cell proliferation and survival: An NF-κB-dependent increased expression of the pro-apoptotic p53 up-regulated modulator of apoptosis protein contributes to colitis in humans and mice (208), while increased expression of certain NF-κB-regulated genes may be involved in the proliferation and survival of transformed epithelial cells leading to tumorigenesis (43); (iii) genes regulating the permeability of the intestinal barrier: In this connection, TNFα induces the partial permeabilization of the intestinal barrier through NF-κB-dependent up-regulation of myosin light chain kinase (292); (iv) metalloproteinases, which may degrade the extracellular matrix and mucosal cells (212); and (v) enzymes (e.g., COX-2 and iNOS), which by participating in the metabolism of ROS and RNS may contribute to NF-κB activation and to the disruption of the barrier integrity (7) (Fig. 9).

OPEN IN VIEWER

FIG. 9.

NF-κB target genes likely implicated in the development of IBD. NF-κB, nuclear factor-κB; TNF, tumor necrosis factor; ICAM, intercellular adhesion molecule; VCAM, vascular cell adhesion molecule; MCP-1, monocyte chemotactic protein-1; MIP-1, macrophage inflammatory protein 1; RANTES, regulated upon activation, normal T cell expressed and secreted; PDGF, platelet-derived growth factor; EGF, epidermal growth factor; IAP, inhibitor of apoptosis, FLIP, flice-like inhibitory protein; MLCK, myosin light-chain kinase; VEGF, vascular endothelial growth factor; MMP, matrix metalloproteinase; COX-2, cycloxygenase-2; MnSOD, manganese superoxide dismutase; iNOS, inducible nitric oxide synthase.

The relevance of NF-κB dysregulation in IBD pathogenesis is supported by the evidence that drugs widely used in treating patients with IBD (i.e., sulfasalazine, steroids, TNFα antibodies) are capable of inhibiting NF-κB (193, 265, 282). Furthermore, in support of a role of NF-κB, induced IBD in mice was successfully treated with NF-κB decoy oligonucleotides (69).

In summary, although NF-κB has relevant physiological functions within the intestinal mucosa, its sustained activation accompanies the chronic inflammatory processes in IBD, and may play a major role in both the etiopathogenesis of IBD and the IBD-associated higher risk for colorectal cancer development (193).

B. Nuclear factor-erythroid 2-related factor 2

In recent years, accelerating research has drawn attention to another redox-sensitive transcription factor, namely nuclear factor-erythroid 2-related factor (Nrf2), that is known to play a key role in the regulation of several genes coding for primary antioxidant and/or type 2 detoxifying enzymes (127, 172). Nrf2 may contribute to maintaining the homeostasis of the intestinal mucosa, by preventing and/or counteracting excess production of ROS within the intestinal epithelium and the lamina propria, as occurs in IBD. Further, Nrf2 appears to negatively modulate pro-inflammatory reactions occurring in the gut mucosa, through antioxidant and, possibly, non-antioxidant mechanisms. A number of in vivo experimental studies, most of which focused on lung inflammation and sepsis, have clarified the significant role of this transcription factor in regulating innate immunity and inflammatory processes (106, 266).

Direct evidence of the Nrf2-dependent negative regulation of gut inflammation was obtained in the DSS-induced mouse model of colitis using Nrf2 wild-type and KO animals. Colitis induced in Nrf2^+/+ mice after 6–7 days of 1% DSS oral administration (in the drinking water) caused the loss of about one/third of mucosal crypts. When Nrf2 KO mice underwent the same treatment, the loss of crypts was significantly greater at about two thirds. This increased vulnerability of Nrf2^−/− mice was associated with a net lower expression of antioxidant/phase II detoxifying enzymes, such as hemeoxygenase-1 and GSH-S-transferase Mu-1. However, even more importantly, the expression of genes coding for key pro-inflammatory molecules, such as IL-1, IL-6, TNFα, iNOS, and COX-2, was significantly increased in the colonic mucosa of Nrf2^−/− mice versus the wild-type mice (119, 120).

The in vitro and in vivo experimental findings of Osburn et al. further strengthened the evidence of negative regulation exerted by Nrf2 on the expression of inflammatory molecules and on intensity of inflammatory reactions also in the gut mucosa. The group's main aim was to investigate the contribution of colonic inflammation in inducing colonic pre-neoplastic lesions after a single initiating dose of azoxymethane, using Nrf2^+/+ and Nrf2^−/− mice. Two weeks after an intraperitoneal injection of azoxymethane, mice were given 1% DSS in the drinking water, for 6 or 7 days. Inflammation was only characterized at the end of DSS treatment, while the possible onset of pre-neoplastic lesions was evaluated 3 weeks later. DSS-induced colitis was much more severe in Nrf2 mice than in wild-type mice, with particularly marked damage to the mucosa, which was extensively infiltrated by leukocytes, mainly macrophages and neutrophils (194).

Consistently, at the end of DSS treatment, a significantly higher transcription of the pro-inflammatory genes IL-1β, TNFα and IL-12p40 was detectable in the colonic mucosa of Nrf2 KO mice than in those with active Nrf2. The morphological evidence of colonic wall infiltration by professional phagocytic cells in Nrf2^−/− mice was biochemically confirmed when a net increase of MPO was found in the colon of these animals. Other relevant findings were a net rise of 3-nitrotyrosine-positive macrophages and of lipid peroxidation products in the colonic mucosa of Nrf2^−/− mice versus Nrf2^+/+ mice (194).

Besides confirming and expanding the finding of a more marked inflammatory reaction in the colon from Nrf2 KO mice versus wild-type mice, the same study reported that Nrf2-deficient mice were also more susceptible to inflammation-associated pre-invasive colorectal carcinogenesis versus wild-type mice. In fact, 3 weeks after having stopped DSS treatment, the number of aberrant crypt foci in the colon of azoxymethane plus DSS Nrf2^−/− mice was thrice higher than it was in azoxymethane plus DSS treated wild-type mice (194). Subsequently, adopting the same cancer initiation treatment, that is, an intraperitoneal injection of a single dose of azoxymethane, 2 weeks interval, then 1 week of DSS, but in this case with a follow-up of a further 17 weeks (20 weeks in total), Khor et al. were able to evaluate the incidence of colon tumors in Nrf2 KO and wild-type mice, and found a marked difference in the tumorigenesis rate: Only 53% of wild-type mice versus 92% of KO mice showed a macroscopic tumor. Of note, while in Nrf2^+/+ tumor-bearing mice, 80% were adenomas and 20% were adenocarcinomas, in the Nrf2^−/− mice, incidence of benign versus malignant colon tumors was the converse, that is, 20% were adenomas and 80% were adenocarcinomas (119).

With regard to oxidative stress, a biochemical condition that significantly exacerbates the expression of IBD, NOX is considered one of the main cellular sources of ROS; using a mouse model of LPS-induced sepsis, Nrf2 has recently been shown to be essential for the physiological regulation of this enzyme's activity. According to the researchers involved, the transcription factor down-regulates the activity of protein kinase C, which is a redox-sensitive kinase, as well as reduces the consequent activation of NOX, by increasing intracellular levels of antioxidants, in particular of GSH. This would at least partly explain why Nrf2 deficient mice display increased generation, as well as increased steady-state levels, of ROS (129).

The overall primary function of Nrf2 in moderating and controlling the expression and extent of inflammatory processes has been carefully analyzed in a recent review (122). Disruption of Nrf2 activity was shown to allow an exaggerated production of pro-inflammatory cytokines, such as IL-1β, IL-6, and TNFα, to take place; on the contrary, activation of this transcription factor efficiently down-regulated the expression of key chemokines, such as MCP-1 and IL-8; adhesion molecules, such as VCAM-1, metalloprotease-9, as well as down-regulated COX-2-dependent inflammatory reactions (122).

On the basis of the available literature, it might be concluded that NF-κB up-regulation initiates and sustains a variety of pro-inflammatory reactions, whereas Nrf2 orchestrates a number of antioxidant-related biochemical events; together, these might provide a fundamental quenching and negative regulation mechanism of inflammation itself.

A. TNFα-mediated NF-κB signaling

TNFα plays a key role in the induction of intestinal inflammation, through regulating NF-κB. The use of anti-TNFα antibodies in the IBD therapy has recently generated considerable interest. Several TNFα blockers are currently licensed for IBD treatment, namely infliximab, adalimumab, and certolizumab, which have been approved by the U. S. Food and Drug Administration for the treatment of moderate-to-severe CD, with inadequate response to standard medications (23). Infliximab is a chimeric monoclonal IgG1 antibody that binds with high affinity both the soluble and the transmembrane forms of human TNFα in the intestinal mucosa, thereby neutralizing its effect. Adalimumab is a complete human immunoglobulin G1 that specifically blocks interactions with TNFα receptors. Certolizumab is a monoclonal antibody, with the Fab fragment of human anti-TNFα combined with polyethylene glycol.

Infliximab appears to be the most effective of these various TNFα blockers, especially in fistulizing active CD. A CD Clinical study evaluating infliximab in a new long-term treatment regimen (ACCENT I) performed from 1999 to 2000, on 573 CD patients from 55 sites in North America, Europe, and Israel, provided the first significant evidence of the efficacy and safety of repeated infusions of infliximab, in patients who responded after an initial infusion of this compound (86). A second ACCENT II study demonstrated the efficacy of infliximab in sustaining fistula closure, in patients with active fistulizing CD (who had single or multiple draining fistulae) with a higher likelihood of abscess development (220). More recently, a randomized, double-blind Study of Biologic and Immunomodulator Naïve Patients in CD trial, conducted in 92 centers from March 2005 to November 2008, in 508 adult patients with moderate-to-severe CD (who were naïve to both agents) treated with infliximab and/or azathioprine, showed that the combination therapy was more effective at inducing remission and mucosal healing (45).

However, infliximab therapy brings the risk of significant side effects or refractory disease, especially in patients treated with infliximab who also receive concomitant immunomodulators, such as azathioprine or mercaptopurine, as well as corticosteroid therapy (57, 135). Combined therapy has been associated with significant toxicity, especially with an increased risk of serious infections. Thus, patients with latent infections, such as tuberculosis or hepatitis B, should be excluded or treated before starting biological therapy (268). In addition, long-term exposure to infliximab has been per se associated with a loss of response, due to a dysregulation of T-cell subsets' activation (239). Treatment with infliximab for a long period can result in the development of antibodies by the host against this molecule, reducing the duration of response to treatment, and increasing the risk of adverse events (52). A multicenter study on corticosteroid-refractory UC patients, in whom cyclosporin had failed, showed that the use of infliximab avoids colectomy in a relatively high percentage of patients, although 23% developed adverse events (38).

Recently, numerous studies using anti-TNFα agent adalimumab as therapy have provided evidence of clinical improvement in CD patients in whom previous infliximab therapy had failed (47). Adalimumab therapy showed a sustained steroid-free remission and fistula healing in Canadian patients with moderate-to-severe CD (199).

TNFα signals initiate through binding to its surface receptors, TNF-R1 (ubiquitously expressed), and TNF-R2 (mainly found in immune cells). Many cell types co-express the two TNF receptors. TNFα induces their trimerization and results in the activation of apoptosis or inflammation. The induction of different processes depends on the recruitment of signal transducers, which activate three distinct down-stream effectors: Fas-associated death domain protein, which is responsible for apoptosis recruiting and activating caspase 8, RIP1, which appears to be a key effector in the TNF-R1 dependent NF-κB activation, and the TNF-receptor-associated factor 2 (TRAF2), which is activated by both TNF-R1 and TNF-R2, and leads to activation of their common cell signals IKK, JNK, and p38 MAPK. (18).

The key factor in regulating the balance between apoptosis and inflammation is the redox transcription factor NF-κB. On TNFα stimulation, the different components of TNFα signaling undergo post-translational modifications, namely ubiquitination and phosphorylation, which activate and amplify the inflammatory response mediated by NF-κB, which, in turn, suppresses the apoptotic cascade. Conversely, the duration of NF-κB activation is negatively controlled by the induction of different inhibitors associated to the TNFα signal. Among the NF-κB negative regulators identified, a TNF-R2a site located in the carboxyl terminus of TNF-R2 has been found to block TNFα signaling, binding TRAF2 (300). Furthermore, ubiquitin ligase A20, which is also called TNFα-induced protein 3, is said to play an important role in the down-regulation of TNFα-NF-κB axis. A20 appears to block the recruitment of RIP and TNF-R1-associated death domain to TNF-R1 complex; RIP itself may be degraded by A20 (150). Interestingly, A20 has recently been found to be up-regulated by TLR signaling during inflammation, attenuating TLR response in a negative feedback loop (195). These findings suggest the potential clinical importance of A20, as a new target molecule playing a crucial role in the management of intestinal inflammation. The treatment of IBD patients with TNFα antibodies is indeed capable of inhibiting NF-κB (193).

Recently, several attempts to reduce NF-κB and inflammatory cytokines have been made using anti-oxidant compounds such as phenolic acids. For instance, sylimarin was found to reduce TNFα and NF-κB activation in experimental colitis (169), and black tea restored adhesion molecule activation, by blocking NF-κB and the JNK cell pathway in IEC treated with LPS (238). Mitochondrial prohibitin (PHB) (for details see section “The mitochondrial target PHB”) has been shown to decrease TNFα and NF-κB, both in IEC and in colonic mucosa of PHB transgenic mice (with a two-fold increase in PHB expression in colon mucosa), suggesting that this molecule may be protective against intestinal injury in IBD (259).

It has been reported that TNFα potently activates NOX1, inducing transactivation of the NOXO1 gene in human colon epithelial cells; hence, this oxidase is a potential target in treating colon inflammation (140). Furthermore, TNFα was also found to influence NF-κB, by enhancing ROS production mediated by another NADPH oxidase isoform, NOX2 (151).

Future perspectives in IBD therapy involving TNFα signaling should focus on the various molecules that lead to induction of NF-κB activity, rather than using antibodies directed against TNFα itself.

B. The IL-1 family and inflammasomes as ROS-dependent targets

Different members of the IL-1 family, such as IL-1α, IL-1β, IL-1 receptor antagonist (IL-1Ra), IL-18, and IL-33, are known to be actively involved in innate and adaptive immune responses. In particular, IL-1β and IL-18 appear to be of great interest in the pathogenesis of intestinal inflammation. IL-1β has been widely investigated for its strong pro-inflammatory effects, which in many cases overlap those of TNFα, in terms of cell type and signaling molecules involved, despite the fact that there is no homology with any group of proteins activated by TNFα. IL-1β has an anti-microbial function and regulates systemic reactions, that is, fever and septic shock, as well as local inflammatory response. IL-18, similar to IL-1β, should be processed by caspase-1 in order to be activated.

The concept that intestinal inflammation induced by IL-1β and IL-18 is only a contributory factor in tissue damage is now being revisited, based on recent reports on DSS- and TNBS-induced colitis-associated tumorigenesis, which suggest that the NLRP3 inflammasome and IL-1-related signaling are critically involved in protection against colorectal tumor development (4, 296). However, an altered response to luminal agents, with excessive inflammatory reactions, could maintain the inflammasome response harmfully active, which, in turn, might influence the activity of IL-1β and IL-18. High levels of these ILs have, indeed, been detected in the inflamed mucosa of numerous IBD patients. Increased expressions of IL-18 and of its binding protein have recently been observed in the chronically inflamed mucosa of children with CD, and in the serum of adults with active CD, but not in UC patients (143, 177). The levels of IL-1β have been found to be increased in macrophages of DSS colitic mice, in which ROS generation was necessary for p38 MAPK and extracellular signal regulated kinase (ERK)1/2 activation, and consequently for IL-1β production. Furthermore, IL-1β release and caspase-1 activity induced by DSS treatment were significantly inhibited by MAPK kinase inhibitors, and by NAC, a precursor of the antioxidant GSH (141). Extra-intestinal complications, such as thrombosis, have been related to IL-1β enhancement in mice with DSS-induced colitis (293). As discussed at length earlier, the implication of IL-1β in the pathogenesis of IBD gained new importance with the discovery of the NLRP3 inflammasome.

ROS are crucial for the secretion of IL-1β via inflammasome activation. The experimental model of DSS-induced colitis in mice, both wild type and deficient in different components of NLRP3, showed that ROS signaling is indispensable to activate the inflammasome-dependent IL-1β production by macrophages (19). Mitochondria are important cellular sites that are actively involved in ROS generation, which leads to NLRP3 inflammasome activation (240). It has been proposed that exposure to stressors may activate NLRP3-IL-1β, either through NF-κB induction or through the interaction between ROS and a Trxip. ROS can be generated not only by direct exposure to PAMPs or DAMPs, but also in response to ATP-dependent potassium efflux and phagosomal or lysosomal damage. Trxip may directly interact with the redox domain of antioxidant Trx, and might function as a negative regulator of Trx reductase. In the presence of oxidative stress, Trxip dissociates from Trx and binds to NLRP3, which then undergoes conformational changes and activation (298). In UC, the mucosal production of IL-1β, H₂O₂, and NO may contribute to the impaired Ca ²⁺ release, decreasing sigmoid smooth-muscle contractility (36).

The intestinal epithelium thus appears to be highly sensitive to both decreased and increased expression of the inflammasome effectors, IL-1β and IL-18, which can culminate in severe mucosal injury. In this connection, synthetic antibodies against IL-1β and IL-18 have been developed, in an attempt to reduce the severity of colitis in animal models (118, 235, 293).

The IL-1Ra is considered an important cytokine, whose action comprises competitive inhibition of the binding of IL-1β to its receptor. A low ratio between IL-1Ra and IL-1β has been found in explant cultures of colonic biopsies from both UC and CD (60). Different polymorphisms in gene sequences of IL-1β and IL-1Ra have been identified in IBD populations (48, 291). IL-1Ra has been shown to have beneficial effects in experimental models of colitis (77, 214), and has been considered for use in managing intestinal inflammation. At present, sIL-1R, a non-glycosylated recombinant analogue of secreted IL-1Ra known as Anakinra, has been used in human rheumatoid arthritis and other rheumatic disorders, but solid evidence regarding its application in the therapy of IBD is yet to be obtained (16).

C. NADPH oxidase as target of the pathogen recognition pathway

The recent findings with regard to the involvement of different NADPH oxidase isoenzymes, in inducing ROS in NLRP3, suggest that NOXs may be possible targets of IBD therapy (162, 174).

Efficient bacterial recognition, processing, and killing require a discrete production of bactericidal ROS, via NADPH oxidase enzyme complexes. These are complex enzymes consisting of cytosolic and membrane subunits, which assemble variously to form tissue-specific isoforms of NOX and dual oxidase (DUOX). In particular, NOX1 and DUOX2 are present in epithelial cells, whereas NOX2 is mainly found in phagocytes. Notably, NOX and DUOX have been shown to regulate the number of bacteria in response to commensal and infectious microbes in the intestine of D. melanogaster (112). These enzymes are closely involved in the autophagy mechanisms activated by ROS generation. NOD2 signaling involves ROS generation via DUOX2 and the activation of down-stream inflammasomes and NF-κB: Colon tissue levels of ROS in wild-type mice stimulated with muramyldipeptide were increased; while in muramyldipeptide-stimulated NOD2^–/– mice, ROS generation was abrogated (154).

Hyper-activation of different NOX/DUOX isoforms, highly expressed in mucosal cells, may inflict damage to the host tissue. The association between susceptibility to develop gastrointestinal diseases and altered function of phagocytic and non-phagocytic NADPH oxidase isoforms has recently been hypothesized (40, 163, 216). Lymphocytes present in the mucosal lesions of CD and UC were also found to be strongly positive for NOX1 (248). A significant increase of DUOX2 mRNA expression was found in colonic biopsies from both CD and UC patients (154).

NOX family stands as a promising target for new IBD therapeutic strategies, despite the fact that presently available NOX inhibitors are not fully specific. In relation to this, a recent study has identified a potent, highly selective inhibitor of NOX1, namely the compound ML090 (CID-616479) (31).

D. The mitochondrial target PHB

Mitochondria are another primary site of ROS generation in the immune response against dietary antigens. The recent characterization of a mitochondrial antiviral signaling protein, which mediates NF-κB and IRF3 activation, supports the evidence of active involvement of the mitochondria in microbial defense (229). TLR1, TLR2, and TLR4 can also increase ROS production in phagosome mitochondria (286). Intact mitochondria with intense respiratory activity are necessary for NLRP3 induction, in order to control bacterial recognition and immune response (115, 240). Furthermore, mitochondrial DNA may play an active role by contributing to the secretion of IL-1β and ATP in NLRP3 activation (182, 240).

Mitochondrial derangement may be involved in intestinal inflammatory diseases. Interestingly, a complex regulating network has been hypothesized between mitochondrial machinery and NLRP3: NLRP3 is suggested to efficiently sense high concentrations of ROS produced by malfunctioning mitochondria, thus causing excessive activation of the inflammatory response. This points to an important role of mitochondrial derangement in the development of IBD (299). Marked mitochondrial damage in IEC and mucosal injury in terms of increased protein oxidation have been found in IBD patients (64). In addition to that, increased TNFα expression, with transient enhancement of ROS production leading to mitochondrial derangement, has been observed throughout all the intestinal layers of neonates with necrotizing enterocolitis, and in rat IEC treated with TNFα; the possible role of TNFα-induced ROS in the activation of mitochondrial autophagy was stressed (15). An increase in mitochondrial respiratory chain complex IV, with an enhancement of lipid peroxidation-end products, was shown in the colon of DSS-treated rats (51).

Recently, functional alterations of PHB, which are evolutionarily conserved and ubiquitously expressed membrane proteins mainly localized in the mitochondria, have been suggested in various pathological processes characterized by oxidative modifications (186). PHB levels are decreased in IEC during oxidative stress, and in the IBD inflamed mucosa (97, 260). Low levels of the PHB complex, with changes in PHB mitochondrial expression and localization, resulted in increased ROS generation and sensitivity to free radicals (242, 261).

The eukaryotic mitochondrial PHB complex comprises two subunits, PHB1 and PHB2, which assemble into a ring-like complex and anchor to the inner mitochondrial membrane through N-terminal hydrophobic regions present in both PHB1 and PHB2. PHB1 appears to be present in larger amounts than PHB2 in mitochondria, and is mainly involved in maintaining mitochondrial morphology and distribution. PHB complex contributes to the stabilization of the mitochondrial genome (281); it is implicated in regulating membrane protein degradation by the mitochondrial m-AAA protease (244), and can function as a holdase/unfoldase chaperone that holds and stabilizes unassembled membrane proteins (187). PHB complex plays a role in mitochondrial morphogenesis (167), functioning as a scaffold that recruits membrane proteins to a specific lipid environment; this function ensures the integrity and functionality of the mitochondrial inner membrane (11, 196).

PHBs are also present in other cellular compartments, including nucleus and plasma membrane, the specific location depending on cell type and situation. PHB complex has been identified as a component of the cell-surface-associated molecular complex in human intestinal Caco-2 cells, binding to Vi. The interaction of Vi with this complex makes S. typhi resistant to phagocytosis and to the action of serum complement. PHB complex, thus, inhibits the inflammatory response to S. typhi in the intestine. In addition, this interaction on Caco-2 cells causes a reduction in IL-8 secretion by involving the MAPK pathway (230). This finding suggests that cell-surface-bound PHB complex might be an important target for the modulation of the inflammatory response by various molecules. In this connection, it may be hypothesized that small molecules, peptides, or monoclonal antibodies, which interact with the PHB complex on intestinal cells, might play a therapeutic role in a variety of gastrointestinal diseases, including IBD, by reducing or inhibiting the inflammatory responses (170).

Since IBD is associated with increased oxidative stress and mitochondrial dysfunction, PHB1 could play a role in combating oxidative stress in the gut, in particular by inhibiting GSH depletion. Indeed, IEC overexpressing PHB have been found to show an increase in GSH levels and an induction of GSH-S-transferase π expression (260).

Furthermore, TNFα (96) may contribute to decreasing PHB1 mucosal levels: TNFα has been found to reduce PHB1 levels in cultured IEC and in animals (259). TNFα activates the transcription factor NF-κB, which decreases PHB expression. Overexpression of PHB1 in these cells protects against the deleterious effects of TNFα and NF-κB on the intestinal barrier function by inhibiting TNFα-induced NF-κB (259). This finding suggests that PHB1 may be crucial in modulating intestinal inflammatory processes.

Contrary to TNFα, IL-6 has been shown to up-regulate PHB expression in IEC and in the intestine; cytokine increases PHB mRNA and protein and induces PHB promoter activation. The IL-6 response element site in the PHB promoter is required for maximal basal promoter activity and responsiveness to IL-6 (262). Thus, IL-6 may protect intestinal mucosa from oxidative stress and inflammation by increasing PHB levels and inducing antioxidant defenses. In support to this observation, IL-6^−/− mice show reduced PHB expression in the colon compared with wild-type mice (262), and suffer more severe erosion of the intestinal epithelium during experimental colitis (257).

PHB1 levels in the IEC of transgenic mice overexpressing PHB1 can be restored during inflammation (264); in these mice, it has been observed that PHB acts as an antioxidant, by decreasing oxidative stress and regulating the antioxidant response. Indeed, PHB has been shown to sustain activation of the transcriptional regulator of antioxidant responses Nrf2, and to protect from inflammation-associated oxidative stress (264).

To demonstrate the therapeutic efficacy of PHB restoration for human IBD, adenovirus infection and nanoparticle delivery of biologically active PHB1 have been tested in mice during experimental colitis: Both methods of delivery have resulted in increased PHB1 levels in colon epithelial cells and in reduced severity of induced colitis (261).

Taken together, all these in vivo and in vitro experimental studies consistently indicate that the PHB complex tends to protect IEC from inflammation-associated oxidative stress, thereby counteracting the mucosal barrier disruption induced by increasing steady-state levels of oxidants. On these bases, the restoration of physiological levels of PHB1 in the intestinal mucosa affected by IBD appears to be a very promising and interesting approach (263).

E. Suitable dietary regimens in IBD patients

Over recent years, a small number of clinical trials have been carried out to evaluate the efficacy of special nutritional regimens in preventing relapse in patients with quiescent IBD. The only study thus far performed in a randomized, multicenter, placebo-controlled manner, which employed polyphenols to prevent IBD reactivation, showed that chronic dietary administration of curcumin plus sulfasalazine for 6 months, to patients (n=45) with UC, significantly reduced the percent incidence of disease relapse (5%) versus a similar group of patients (n=44) receiving only the anti-inflammatory drug (20%). The two groups of UC patients were then followed for six further months, during which only the anti-inflammatory therapy was administered, without curcumin. At the end of this second period, the number of relapses was quite similar in the two groups (85).

More recently, other clinical trials have used polyphenols as a dietary supplement to counteract or prevent overexpression of inflammatory cytokines and adhesion molecules, not in IBD patients but in individuals with a high risk of developing atherosclerosis. In 67 male high-risk volunteers, who drank a standardized amount either of ethanol or of dealcoholized red wine, the phenolic extract from red wine was shown to down-regulate expression of certain adhesion molecules and cytokines in blood mononuclear cells, in particular intercellular adhesion molecule-1, E selectin, and IL-6 (39).

Another type of so-called “nutritional therapy” employed a comprehensive mixture of aminoacids and vitamins, including the antioxidant vitamins A, C, and E (elemental diet patented as Elental), for home treatment of quiescent IBD, with the aim of delaying disease reactivation. Fifty CD patients in remission were randomly divided into two subgroups, one receiving a free diet and the other receiving a diet supplemented with the aminoacids/vitamins. After 2 years' follow-up, the number of cases of disease reactivation was reduced to half in patients receiving nutritional therapy (34% relapses) versus those on a free diet (64% relapses) (250). Use of the same elemental diet has recently been compared with the frequently used drug 6-mercaptopurine, as maintenance therapy of CD in remission. After 2 years' follow-up, the number of patients (n=32) receiving the elemental diet, rich in vitamins and aminoacids, showed a relapse rate (51%) which was quite similar to that of CD patients (n=30) receiving 6-mercaptopurine instead (40%). The researchers concluded that suitable nutritional therapy could achieve comparable efficacy to purine antimetabolites in preventing reactivation of CD (84).

F. Lipoxins and resolvins: new drug candidates for IBD treatment

There is considerable interest in the relationship between fatty acids and gut inflammation. During acute inflammation, resident cells in the injured tissue produce soluble pro-inflammatory mediators, such as eicosanoids deriving from ω6-AA. This regulates early events in the inflammatory response, promoting the influx of neutrophils from the blood and their phagocytic activity. The resolution of acute inflammation is an active programmed process with a self-limited response of leukocytes, which provides a temporal switch in lipid mediator production within the family of eicosanoids, from pro- to anti-inflammatory PGs and LTs, and also their fatty acid precursors. Indeed, during the latter phase of inflammation, changes occur in the quantity and quality of the leukocyte infiltrate. In this phase, neutrophil numbers decrease, dying by apoptosis, while lymphocytes and monocytes/macrophages predominate, and combine to resolve inflammation by means of the endogenous generation of specific ω3- and ω6-PUFA derivatives with anti-inflammatory and pro-resolving activity, including lipoxins (LXs) and resolvins (Rvs) (228).

Chronic inflammation is now believed to be the evolution of unresolved acute inflammation that can no longer be controlled. To prevent chronicization, it appears to be more important to improve the resolution of the process, rather than inhibiting inflammation by using anti-inflammatory glucocorticoids (226).

Endogenous LXs and Rvs have aroused interest, because of their ability to promote the clearance of apoptotic cells by macrophages, and to limit the infiltration of pro-inflammatory leukocytes, which eventually leave the inflamed site. LXs and Rvs are also generated after acetyl salicylic acid treatment, providing a basis for understanding a novel mechanism of action of both nonsteroidal anti-inflammatory drugs (NSAIDs) and dietary ω3-PUFA supplementation.

Their function in mediating the resolution of inflammation makes them new promising candidates for regulating inflammation-associated disorders such as IBD.

LXs are generated through transcellular biosynthesis, in which cell–cell interactions activate 15-LOX in different cell types at the inflammation site, to produce 15(S)-hydroperoxyeicosatetraenoic acid from AA. This compound is rapidly incorporated in leukocytes, and converted by 5-LOX and hydrolase into Lipoxin (LX)A4 or LXB4.

Endogenous LX analogs are also produced after exogenous administration of aspirin (but not with other NSAIDs), which irreversibly acetylates COX-2 in endothelial cells or leukocytes, switching AA derivatives from pro-inflammatory PGs and LTs to LXs. Acetylated COX-2 is still active, but generates 15(R)-hydroxyeicosatetraenoic acid that, in turn, in endothelial or epithelial cells is converted by 15-LOX into the potent anti-inflammatory 15-LXA4 epimer, known as 15-epi-LXA4 or aspirin-triggered lipoxin (ATL) (226, 243).

LXs limit the number of neutrophils in inflamed areas: They stimulate macrophages to ingest and clear neutrophilic apoptotic bodies, and0 promote only the infiltration of non-inflammatory monocytes, which produce anti-inflammatory and pro-fibrogenic cytokines. In particular, the widely studied LXA4 signals, through a G-protein-coupled receptor, to inhibit chemotaxis and transmigration, and down-regulates pro-inflammatory pathways, such as ERK and p38 MAPK; it limits the activity of AP-1 and NF-κB transcription factors, NOX-dependent ROS production, and inhibits the expression of inflammatory cytokines such as IL-8 and TNFα (9, 108, 137). Synthetic metabolically stable LX analogs have been made to mimic anti-inflammatory LX actions, and prolong its active effect. A stable LXA4 analog has been found to down-regulate IL-8 mRNA expression via NF-κB suppression T84 polarized IEC infected by S. typhimurium, which can be considered an in vitro model of inflammation (81). The therapeutic anti-inflammatory effect of LXA4 analogs has been verified in the DSS-induced murine colitis model, in which oral administration of this active analog led to the attenuation of colitis and the down-regulation of pro-inflammatory gene expression (81). Similar effects are reported for a synthetic analog of ATL examined in TNBS-induced experimental colitis. Oral administration of this analog decreased mucosal mRNA level of COX-2, macrophage inflammatory protein-2, TNFα, IL-2, and IFNγ, showing anti-inflammatory efficacy which was similar to that observed with prednisolone (71).

Studies in animal models of colitis, in which exogenous LX analogs resolve inflammation, support the hypothesis that LX deficiency plays a role in IBD. Mangino et al. found a decrease of LX production in colonic mucosal tissue from UC patients, compared with both organ donors and patients affected by slow-transit-time constipation. They found a decrease in epithelial 15-LOX in UC biopsies, suggesting defective biosynthesis of LXs in these patients (160).

These findings were strengthened by a recent study on humans that found elevated LXA4 biosynthesis in mucosal biopsies from UC patients in medically induced remission, confirming the hypothesis that LXA4 exerts protective and/or reparative action in human intestinal inflammation (276).

Rvs are potent bioactive compounds that exert their anti-inflammatory function by blocking trans-endothelial leukocyte migration and infiltration. They derive from the degradation of eicosapentanoic (EPA) and docosahexaenoic (DHA) ω3-PUFAs through COX-2 and LOX enzymes. RvE-series has been identified in endothelial cells expressing COX-2 treated with aspirin: as in the case of LXs, Rvs are generated by acetylated COX-2, which converts EPA to 18(R)-hydroxyeicosapentanoic acid, and eventually to RvE1 or RvE2, by leukocyte 5-LOX. RvD-series is generated from DHA by 15-LOX or aspirin-derived acetylated COX-2, and 5-LOX, through intermediates 17(S)-hydroxydocosahexaenoic acid or 17(R)-hydroxydocosahexaenoic acid (in the presence of aspirin). Notably, biosynthesis of RvE1 can be also initiated by P450-like enzymes in bacteria (227).

Rvs may provide the molecular means underlying ω3-PUFA's beneficial effects in IBD. Exogenous RvE1 administration protects against the development of TNBS-colitis in mice. In fact, 100 ng of RvE1 reduced leukocyte peritoneal infiltration by 50%, and increased survival and histological scores. Notably, a significant reduction in TNFα, IL-12-p40 subunit, iNOS, and COX-2 was observed in mice that had received RvE1 (10). It has recently been reported that RvE1 treatment in mice with DSS-induced colitis led to less severe histological features of colitis, with a reduction in body-weight loss. In this experimental model, RvE1 administration reduced NF-κB phosphorylation and expression of the NF-κB-dependent pro-inflammatory mediators TNFα, IL-1β, and IL-6. (105). Anti-inflammatory effects of the aspirin-triggered RvD1, of its precursor 17-hydroxy-DHA, and of RvD2 were demonstrated in DSS- or TNBS-induced colitis, reducing colonic cytokine levels as well as mRNA expression of NF-κB and adhesion molecules (22).

Elucidating the molecular basis of ω3-PUFA action in generating local mediator Rvs and their effectiveness in clinical applications as anti-inflammatory therapy in IBD are still major issues; data currently available only concern experimental models. Transgenic fat-1 mice with high tissue levels of ω3-PUFA, because they have been engineered to express C. elegans fat-1 gene encoding for ω3 fatty acid desaturase, showed a significant production of Rvs. Further, the same experimental model was protected against DSS-induced colitis, leading to a manifest reduction in tissue injury inflammation, with significantly low amounts of TNFα, IL-1β cytokines, and reduced NF-κB activity (99).

A diet involving the intake of the recommended ω6/ω3-PUFA ratio could undoubtedly help delay or prevent the progression of IBD. However, clinical and epidemiological studies on the role of PUFA supplementation in dampening inflammation in IBD, through the local production of anti-inflammatory lipid derivatives, still do not afford conclusive results (35, 287).