Introduction

Inflammatory bowel disease (IBD), along with its two clinical subtypes, Crohn’s disease (CD) and ulcerative colitis (UC), has strong genetic links. DNA underlies nearly every aspect of human health, including function and dysfunction, and the revealing of the entire human genome sequence by the Human Genome Project has made it easier to perform disease-oriented genetic research. Obtaining a detailed picture of how genes and other DNA sequences work together and interact with environmental factors may ultimately lead to the discovery of pathways involved in normal processes and disease pathogenesis. Recent advances in genotyping technology, statistical methodologies, and molecular biology have played key roles in a number of studies of IBD genetics; these studies stand on three pillars; familial aggregation, candidate genes, and genome-wide association (GWA) (Fig. 1), as briefly described in this review. In addition to familial aggregation studies and candidate gene assays, the new technique of GWA scanning has restored momentum to a process that had seemed to stall for 5 years following the identification of the nucleotide-binding oligomerization domain containing 2 (NOD2) as the first CD susceptible gene in 2001 [1]. This restored interest has led to the identification of several IBD susceptibility genes and has opened the door of a new era in IBD research. It is difficult to briefly describe IBD genetics from NOD2 to the autophagy-related gene 16L1 (ATG16L1) [2], as numerous findings are rapidly being presented. In the present article, we describe susceptible genes with respect to epidemiological resources and how this genetic phenomenon has rewritten the story of IBD pathogenesis, with an emphasis on innate immunity.

Fig. 1
figure 1

Trends in inflammatory bowel disease (IBD) initiation. In a genetically susceptible host with or without familial aggregation, environmental triggers (stress, luminal microbes, diet) cause an abnormal immune phenotype and dysregulated barrier function, ultimately leading to IBD. GWA genome-wide association

Full size image

Genetic associations: familial and ethnic groups

Recent molecular advances have aided understanding of the genetics of IBD from epidemiological studies that have examined IBD occurrence within different familial aggregations and ethnic populations. A certain type of IBD that occurs in families and siblings is associated with a phenotype that has an increased risk of developing the disease [3]. In a population-based survey, a significant proportion of affected individuals with IBD had a positive family history, with other members having the disease [4, 5]. The risk of IBD is highest in first-degree relatives of the CD proband, though more distant relatives are also at increased risk [6]. Twins form the highest risk group, especially for CD, as monozygotic twin concordance for CD is reported to range from 42 to 58%, whereas the dizygotic twin concordance is not significantly different from that for all siblings. In contrast, monozygotic and dizygotic concordances for UC range from 6 to 17% and 0 to 5%, respectively [7]. Along with familial history, the influences of age of onset and sex on IBD incidence cannot be ignored. Both CD and UC are primarily diseases of young adulthood, with peak incidence occurring between 15 and 30 years of age, and the average age of onset of familial CD at approximately age 22 years, as compared to age 27 years for sporadic cases [8]. Similarly, in a study conducted in the US, the average ages at diagnosis for familial and sporadic UC were 23.3 and 28.6 years, respectively [9]. An intriguing finding obtained when comparing familial and sporadic cases of IBD is the relative preponderance of female cases among familial IBD, as the female-to-male ratio ranged from 1.23 to 1.68:1 in familial CD, whereas pediatric-onset CD tends to have a slight male preponderance. As opposed to CD, there is a slight male preponderance in cases of UC, representing epigenetic contributions to IBD pathogenesis [10].

Epidemiological studies have shown that IBD incidence and prevalence vary significantly depending on geographic location, and racial or ethnic background, with increases in the incidence rates of CD and UC over the past three decades in nearly all Western countries [11]. In general, there is an increased risk for developing IBD in urban as compared to rural areas, in cohorts with a higher socioeconomic class, and in developed rather than less-developed countries [12, 13]. Incidence rates have also been reported to increase when populations migrate from low-risk geographic areas to higher-risk areas [14]. The prevalence rate for IBD in non-Caucasians in the US has consistently been reported as lower than that in Caucasians. However, diverse rates of IBD preponderance have been seen in different racial (Caucasians, African Americans, Hispanics, and Asians) and ethnic groups (Jewish, and non-Jewish) in the US [9, 13], which have effects on disease occurrence.

Genetic association: candidate genes

In general, two different genetic research strategies have been applied in studies of IBD. In one, candidate genes are the focus of studies combined with fine-mapping efforts in regions of genetic linkage, while in the other, unbiased methods of genome-wide scanning are used in large panels of cohorts [15, 16]. Candidate gene studies of attribute-selected genes have implicated their roles in disease pathogenesis, as revealed by expression or functional analysis, as well as evidence from sources such as linkage studies, animal models, or other related diseases. The candidate gene approach resulted in the identification of the first IBD susceptible gene, NOD2, previously known as caspase activated recruitment domain protein 15 (CARD15), by using a strategy of positional cloning within a region of genetic linkage. That approach revealed two missense single-nucleotide polymorphisms (SNPs), Arg702Trp and Gly908Arg, and one frameshift mutation, Leu1007 (3020insC), which cause a premature stop codon that independently increases the risk for ileal and ileocolonic CD, but not colonic CD or UC [17]. Although researchers in China and Japan have not found associations between NOD2 variants and CD in their populations [18, 19], recent meta-analysis studies of European cohorts have estimated that homozygous or compound heterozygous carriage of NOD2 risk alleles confers a 17.1-fold increased risk of CD [20]. Other than NOD2 variants, a recent attempt to identify IBD susceptibility loci in 158 nuclear families provided evidence for a gene (IBD5) that confers susceptibility to CD in chromosome 5q31 [21]. In addition, another study identified a number of immuno-active candidates, including interferon regulatory factor1 (IRF1) and important cytokine genes such as interleukin (IL)-4, IL-13, IL-5, and IL-3, located within the 250-kb region risk haplotype [22]. Furthermore, organic cation transporter genes (OCTN1/2, also known as SLC22A4/5) are located within a single haplotype block of the IBD5 locus. The presence of a two-locus risk haplotype within OCTN1 (1672 C/T) and OCTN2 (207 G/C) accounts for increased susceptibility to CD [23]. An overwhelming finding in studies of candidate gene strategies is the identification of an association between the major histocompatibility complex (MHC) region and IBD susceptibility. Replicated HLA class II associations in IBD including HLA-DRB1*1502 (serologic marker HLA-DR2) for UC [24], and HLA-DRB1*0103 for both UC and colonic CD have been firmly established [25]. Another recent candidate gene approach to identify a set of SNPs located in a predicted regulatory region on chromosome 1q44 downstream of the Nod-like-receptor family NLRP3 (previously known as CIAS1 and NALP3) associated with CD has been used by Villani et al. [26] to study four sample sets from individuals of European descent. These results suggest that the NLRP3 region is also related to susceptibility for CD. Villani et al. also examined NLRP3 expression in colon tissues isolated from mice with trinitrobenzene sulfonic acid (TNBS)-induced colitis and in biopsy samples from individuals with CD, and noted that the expression was significantly higher in both acute and chronic TNBS-induced colitis models as compared to colon tissues from control mice. NLRP3 expression was also significantly higher in ulcerated intestinal mucosa from human CD than in that from healthy controls. Taken together, these results indicate that the NLRP3 locus can be added to the list of several newly uncovered CD loci in which the common allele has been reported to be a risk allele.

Genetic association: GWA screening

Hypothesis-free methods of genome scanning have provided unbiased surveys of the whole genome for IBD-associated loci, as well as profound insights regarding disease mechanisms. Such methods have evolved from low-resolution linkage analyses based on multiple affected families. GWA studies have identified a number of well-replicated loci and have thus advanced our understanding of IBD genetic associations. The first GWA study among Japanese case–control groups identified a 280-kilobase region on chromosome 9q32, with the most significant markers clustered within the tumor necrosis factor superfamily 15 (TNFSF15) gene [27]. That study also revealed highly significant associations of SNPs and haplotypes within the TNFSF15 genes in Japanese and Caucasian family study populations composed of CD patients. TNFSF15, a potent enhancer of interferon (IFN)-γ production in human T-cells and natural killer cells, exerts its effects most profoundly in the gut, and is upregulated in macrophages and CD4+/CD8+ lymphocytes of the intestinal lamina propria in patients with CD [28]. Another GWA study, involving a North American ileal CD case–control cohort, identified the most significant associations in the IL-23 receptor (IL-23R) gene region on chromosome 1q31 [29]. Within the IL-23R gene region, the most significant association was observed for the amino acid polymorphism, Arg381Gln, with the less common glutamine allele conferring significant protection against CD, similar to the protection against UC conferred by this allele in non-Jewish cohorts. The gene immediately centromeric to IL-23R is its closely related homologue IL-12B2, which is also significantly related to UC [29]. The gene encoding the IL-12B subunit has also recently been reported to be associated with CD [30]. The association of both IL-23R and IL-12B indicates that the interaction between IL-23R and its ligand triggers a signaling cascade central to inflammation leading to CD. In further support of a central role for this pathway in disease susceptibility are the confirmed associations of the janus kinase 2 gene (JAK2, a proximal kinase in the IL-23R pathway) in CD, and the signal transducer and activator of transcription 3 gene (STAT3, found immediately downstream of JAK2) in CD and UC [31, 32]. Apart from these genes, several other noteworthy genes within related loci have recently been implicated in CD pathogenesis, such as chemokine receptor 6 (CCR6) and its ligand macrophage inflammatory protein 3α (MIP-3α), due to their selective expression by IL-17-producing cells and IFNγ-producing TH17/TH1 cells [31]. Within the IBD susceptibility locus resides the secretory and membrane-bound mucin gene family, which protects intestinal epithelial cells from injury and is vital for IBD genetic studies. Considerable evidence has also implicated a critical role for the mucin gene family, as its deficiency potentiates intestinal inflammation in mouse models of colitis [33, 34]. An exciting and unanticipated finding from GWA studies is the important role of autophagy in CD pathogenesis. Two genes involved in this process, ATG16L1 and IRGM, were found to be significantly associated in multiple GWA studies, and a third autophagy-associated gene, leucine-rich repeat kinase 2 (LRRK2), was shown to be one of two CD candidate genes on chromosome 12q12 [2, 31, 35]. For ATG16L1, the SNP showing the strongest association is a non-synonymous amino acid change at position 300 (alanine to threonine) [36]. The finding of an association of the ATG16L1 gene with CD has directed study of the autophagy pathway in the context of inflammatory diseases. These genetic and functional studies, based on the implication of ATG16LI and IRGM involvement in CD shown by GWA analysis, indicate that autophagy is a key pathway in the pathogenesis of CD. Common genetic variations in these and other autophagy genes are likely to have a major impact on the response of the innate immune system to intestinal microbial flora and susceptibility to IBD. In addition to these genes in IBD susceptible loci, recent GWA studies have also pinpointed several genes involved in a number of homeostatic mechanisms, including inducible T-cell co-stimulator ligand (ICOSLG), protein tyrosine phosphatase, non-receptor types 2 and 22 (PTPN2 and PTPN22), CDKAL1, and intelectin 1 (ITLN1) [31]. In Table 1, we have summarized some IBD susceptible genes and their disease contribution based on recent results. The IBD designations are cited from National Center for Biotechnology Information (NCBI) and online Mendelian inheritance in man (OMIM) resources.

Table 1 Inflammatory bowel disease (IBD) susceptible genes/loci identified by linkage analysis and genome-wide association (GWA) studies
Full size table

IBD candidate genes and GWA studies: roles in innate immunity

IBD is characterized by a complex interplay between genetic, bacterial, and immunological factors that induce intestinal inflammation. Therefore, mutations/polymorphisms in any one of a number of different genes in the IBD susceptible loci may disrupt the homeostasis of the mucosal immune system. In this section, we focus on recent advances in understanding of the influence of genetic factors on the innate immune system in IBD. The discovery of NOD2 as the susceptibility gene for CD has shifted the focus of research of the pathogenesis of IBD firmly toward the innate immune response. At the N-terminal part of NOD2 there are two CARDs, which play major roles in apoptosis and nuclear factor (NF)-κB activation pathways [37]. Treatment with either bacterial lipopolysaccharide (LPS) or peptidoglycan (PGN) was shown to induce NF-κB in cells transfected with wild-type NOD2/CARD15 [38]. However, the frameshift variant, Leu1007fsinsC, which truncates the final 3% of the NOD2/CARD15 protein, is associated with marked hyporesponsiveness toward NF-κB activation with LPS treatment. In contrast, the Arg702Trp and Gly908Arg variants respond to LPS to a greater extent than the frameshift variant, though overall they show a significantly diminished ability to activate NF-κB with LPS treatment [39, 40]. Based on these features, it has been clearly shown how NOD2 mutations and impaired NF-κB activation confer susceptibility to CD. NOD2, a member of a family of intracellular proteins that share homology to disease resistance (R) proteins present in plants [37], is the LRR of plant R genes, as well as being a member of the family of toll-like receptors (TLRs), which recognize pathogen-associated molecular patterns (PAMPs) and activate the innate immune system, and ultimately induce the expression of a variety of immune-response genes, including inflammatory cytokines [41]. The recognition of PAMPS by some TLRs leads to activation of the NF-κB signaling pathway. Therefore, it is possible that extracellular TLRs and intracellular NOD2 synergistically participate as pattern-recognition receptors (PRRs) in regulating mucosal innate immune responses to intestinal microbes. Consequently, a decrease in NF-κB activation because of NOD2 LRR mutations in this pathway might be associated with impaired killing of intracellular microbes, resulting in homeostatic imbalance by compensatory increases in other components, such as proinflammatory cytokines, produced by the interactions of mucosal cells involved in the intestinal inflammatory/immune response. It is well established that the activation of NF-κB signaling pathways in response to bacterial components mediates protection of the host against invading pathogens [42, 43]. Since CD is characterized by a strong induction of NF-κB [44], controversy has arisen regarding how the mutant NOD2 protein potentiates the activation or inhibition of NF-κB. Ogura et al. [39] noted that NOD2 mutants seemed to induce only a weak activation of NF-κB, while the controversy gained momentum when Maeda et al. [45] performed a study on mutant mice expressing the homologue of the c3020insC human mutation, and reported that NF-κB activity and IL-1β processing were induced. Although NF-κB is thought to be one of the major effectors for NOD2, it should be noted that NF-κB is more effectively activated by bacterial products via TLRs. Thus, NF-κB activation is not solely reliant on NOD2 and its loss may not compromise NF-κB signaling in response to bacterial infection.

Next, we describe a member of the CATERPILLER family of genes [46], NOD-like receptor protein 3 (NLRP3), which is comprised of a nucleotide-binding domain and an LRR domain. As reported by Mariathasan and Monack [47], NLRP3 encodes cryopyrin, which regulates the activation of caspase-1 and the processing of IL-1β via an inflammasome-mediated pathway; their study noted the importance of cryopyrin in inflammation by highlighting its hyper-responsiveness, due to gain-of-function mutations within its NOD domain, in some hereditary periodic fever syndromes. NLRP serves as an intracellular sensor of microbial motifs and ‘danger signals,’ which have emerged as crucial components of innate immune responses and inflammation. The importance of the NALP3 inflammasome is emphasized by the identification of mutations in the NALP3 gene that are associated with a susceptibility to inflammatory disorders. In an IBD cohort, two functionally distinct SNPs within the predicted NLRP3 locus were identified [26]. One type exhibited gain-of-function mutations leading to hyperproduction of IL-1β, while the other was identified in the regulatory region downstream of NLRP3 that is associated with hypoproduction of IL-1β as a result of decreased NLRP3 expression. Due to a dysregulated innate immune response, both these forms contribute to CD pathogenesis.

Consistent with the implications of NOD2 and NLRP3, studies of PRRs have significantly advanced our understanding of innate immunity. Pasare and Medzhitov [48] presented a breakthrough discovery that TLRs activate NF-κB, and suggested that these receptors may link innate and adaptive immunity. LPS triggers signaling through TLR4 after associating with the LPS-binding proteins CD14 and MD2. CD14 is located in the IBD susceptibility locus on chromosome 5q13 [49]. In addition, the CD14 promoter polymorphism 159C/T has been associated with CD in a Greek study, while this polymorphism was associated with UC in a Japanese IBD cohort [50, 51]. DNA sequence variants of TLR genes in IBD have also received substantial research interest. TLR2 (4q31.3), TLR3 (4q35.1), TLR4 (9q33.1), and TLR9 (3p21.3) have all been shown to be located in regions associated with IBD by genome-wide searches [49]. The TLR4 Asp299Gly polymorphism leads to altered recognition of LPS by the extracellular domain of TLR4, and was associated with CD and UC in a Belgian study [52]. In European populations, the association of the Asp299Gly polymorphism with CD was replicated and an association with colonic disease described [53, 54]. In one German cohort, an association was demonstrated between UC and the TLR4 Thr399Ile polymorphism [55]. However, in another study, the TLR4 polymorphism was found to confer disease susceptibility, while a dominant-negative TLR5 C1174T polymorphism was protective against the development of CD, but not of UC [56]. In a recent study, combined carriership of the alleles TLR9-237C and CD14-260T was increased in a chronic relapsing pouchitis group when compared with findings in UC patients [57]. Several regulatory elements control TLR signaling in extracellular spaces or intracellularly through the reduction of TLR expression; these elements are termed negative regulators of TLRs. The transmembrane protein regulators of TLR signaling are ST2 (IL-1receptor-like1/FIT-1), single immunoglobulin and toll-interleukin-1 receptor (TIR) domain (SIGIRR), and TNF-related apoptosis-inducing ligand receptor 1 (TRAILR) [58]. Of these, only SIGIRR is located in an IBD susceptibility locus on chromosome 11p15.5, though no germline variant studies have been reported to date [49]. In vitro, SIGIRR has been shown to interact with TLR4, IL-1-receptor-associated kinase (IRAK), and TNF-receptor-associated factor 6 (TRAF6) [59]. Its important role in the regulation of intestinal inflammation is illustrated by the development of more severe dextran sodium sulfate (DSS)-induced colitis in SIGIRR-knockout mice as compared with wild-type mice [60]. Intracellular negative regulators of TLR signaling include MyD88s (the short form of MyD88), IRAK-M, suppressor of cytokine signaling 1 (SOCS1), NOD2/CARD15, phosphoinositide 3-kinase (PI3K), Toll-interacting protein (Tollip), and A20 [58]. Macrophages from SOCS1-deficient mice produce increased levels of pro-inflammatory cytokines in response to stimulation with TLR4 and TLR9 ligands [61]. The gastrointestinal phenotype of SOCS1/T cell receptor α (TCRα) double-knockout mice is characterized by earlier development of more severe colitis as compared with TCRα-knockout mice, dependent on IFNγ and IL-4 [62]. In contrast, transgenic mice overexpressing SOCS1 also developed spontaneous colitis with age, and were more susceptible to TNBS-induced colitis, associated with increased expression of IFNγ and TNFα and reduced levels of transforming growth factor (TGF)-β [63]. The SOCS1 gene is located on chromosome 16p13.13 in the IBD1 locus [49]. Tollip is critical for maintaining intestinal epithelium responsiveness to TLR2 ligands [64, 65]. Furthermore, a mutation of lys150glu within this C2 domain causes an inability of Tollip to inhibit LPS-induced NF-κB activation [66]. The Tollip gene lies in the IBD locus on chromosome 11p15.5 [49].

Autophagy genes: roles in innate immunity

Autophagy is an ancient, highly conserved cellular process used by all eukaryotic cells. In its basic form, the autophagy system autodigests intracellular components during starvation conditions to recycle damaged or superfluous organelles and degrade long-living proteins. In the case of intracellular pathogens, the process, which is often termed ‘macroautophagy,’ forms a structure known as an ‘isolation membrane’ that enlarges around the particle to be ingested. The pathogen is further sequestered in a unique double-membrane cytosolic vacuole called an autophagosome that finally fuses with lysosomes for further processing [16, 67]. The association of the ATG16L1 gene with CD led to the study of the autophagy pathway in the context of inflammatory diseases [2]. Reports presented by multiple groups from diverse disciplines have converged to reveal the role of autophagy in innate and adaptive immunity. The first link between autophagy and the innate immune system was shown by the discovery that intracellular pathogens (primarily bacteria and viruses) could be eliminated from cells via the autophagy pathway [68, 69]. One of the signals used to initiate pathogen-induced autophagy seems to be the activation of TLRs [70]. The involvement of autophagy in adaptive immunity has also been established by studies demonstrating that cells expressing MHC class II proteins use the autophagy pathway in processing peptide antigens for presentation to CD4+ T cells. Experiments with mice using targeted deletions of key autophagy genes have shown that lymphocyte homeostasis, as well as T-cell development and central tolerance, are also dependent on this cellular pathway [71, 72]. A recent report showed elevated levels of endotoxin induced by IL-1β production in an experimental model of colitis [73]. In another recent study, mice that expressed reduced levels of ATG16L1 protein displayed defects of Paneth cells in the small intestine [74]. Homozygous deletion of another autophagy gene, Atg5, in the intestinal epithelium of mice also produced abnormal Paneth cells, indicating that these cells are particularly sensitive to autophagy defects [74]. Perhaps most interesting was the finding that CD patients homozygous for the ATG16L risk allele had abnormal Paneth cells in biopsies of uninvolved ileocolic resection samples [75].

Apart from the common genetic variations noted above, other autophagy genes are likely to have a major impact on the response of the innate immune system to intestinal microbiota and susceptibility to IBD [76]. Sequence variants in the autophagy gene IRGM and multiple other replicating loci contribute to CD susceptibility [76]. The IRGM gene is located on chromosome 5q33.1 and encodes a 181 amino acid protein belonging to the p47 immunity-related guanosine triphosphatase family [77]. Experiments with mice have shown that expression of the murine homologue LRG-47 is induced by IFNγ and stimulates macroautophagy to generate large autolysosomal organelles as a mechanism for the elimination of intracellular organisms [78]. Mice deficient in LRG-47 display increased susceptibility to bacterial infections. Human IRGM lacks the IFNγ response element; however, it has been proposed to play a role similar to that of the murine homologue LRG-47 in clearing intracellular pathogens. Singh et al. [79] recently demonstrated the importance of IRGM to autophagy and the clearance of Mycobacterium tuberculosis from human macrophages. Another recent GWA study has revealed that LRR kinase 2 (LRRK2) is a CD susceptible locus [31]. Expression of mutant LRRK2 induced apoptotic cell death in human SH-SY5Y neuroblastoma cells and mouse cortical neurons [80]. In addition, a recent study reported the induction of autophagy by mutant LRRK2, which is of interest given the strong associations between CD and the autophagy genes ATG16L1 and IRGM [31].

Murine genetic models of IBD

Animal models are indispensable for elucidating the mechanisms of IBD pathogenesis. Due to the heterogeneous clinical appearance of human IBD, the numbers of sufficient or deficient gene-targeted mouse strains displaying IBD-like intestinal alterations have been steadily increasing. Most of these models are based on chemical induction, immune cell transfer, or gene targeting, while in others the disease occurs spontaneously without any exogenous manipulation. In several models of chemically induced colitis, DSS is commonly used. It can directly affect the integrity of the mucosal barrier and basal crypts in wild-type mice, as well as in genetically manipulated, T- and B-cell-deficient severe combined immunodeficiency (SCID) or Rag1−/− mice; thus, it is particularly useful for studying the contribution of innate immune mechanisms of colitis [81–83]. Other chemical models, including those created with TNBS and oxazole, are also useful for studying a variety of important aspects of gut inflammation, including cytokine secretion patterns, mechanisms of oral tolerance, cell adhesion, and immunotherapy [82, 84]. Together with these models, genetic models of mice with conditions resembling IBD have enabled the evaluation of the functional role of a gene of interest with its mechanistic pathway. Based on several lines of convincing evidence, it is now considered a fact that, in genetically susceptible hosts, aberrant immune responses and loss of tolerance to environmental factors are major factors contributing to mucosal inflammation. Since defects in innate and/or adaptive immune cells in IBD finally result in non-self-limiting chronic inflammation, we provide a brief overview of selected mouse genetic models of IBD and emphasize their impact on the immune system (see Table 2). Studies with animal models have greatly improved our understanding of the complex field of human IBD, and such studies have unveiled many obscure mechanisms that are, presumably, responsible for disease initiation and progression.

Table 2 Mouse genetic models of IBD
Full size table

Genetics-based therapies for IBD

Since CD and UC are multifactorial diseases, therapeutic approaches based on genetic aspects represent an enormous challenge in terms of gene defects, target cells, and suitable vectors and targeted delivery systems. The underlying molecular mechanisms revealed in tests of genetically engineered animal models suggest that an altered immune response driven by luminal microflora is probably the critical point for both the onset and the chronicity of the pathophysiologic process. Based on these findings, a number of conventional and biological therapeutic approaches have been reviewed by several authors [85, 86]. Recently, we utilized recombinant DNA technologies to prepare and explore the role of homeostatic protein milk fat globule-EGF factor 8 (MFG-E8) in terms of modulating innate-immune responses, which proved to be beneficial in murine experimental colitis [87]. Immunobiological therapies currently in use are mainly based on neutralizing TNF-α by both chimeric and humanized blocking substances, as well as targeting the overexpression of regulatory cytokines to produce therapeutic relevance [88]. Other immunobiological factors shown to be effective in animal models and clinical trials include monoclonal antibodies to leukocyte adhesion molecules (α4, α4β7-integrin), cytokines/cytokine receptors (IL-6, IL-12, IL-18, and IL-2 receptor), recombinant cytokines (IFN-α-2A, IFN-β1α), and antisense or decoy oligodeoxynucleotide (ODN) targeting transcription factors NF-κB/AP-1, which have anti-inflammatory roles in colitis [85, 86, 89]. However, the disadvantages of such therapies include their sometimes undesired side effects due to immune reactions against the non-human parts of the antibodies, as well as higher production costs in comparison with the costs of small molecular compounds. Recently, researchers evaluated gene transfer approaches to increase IL-10 levels in the inflamed gut utilizing several gene therapy vectors, and successfully used them to deliver the regulatory/anti-inflammatory cytokines IL-10 and TGF-β into animal models of human inflammatory and autoimmune diseases [90]. In addition, several studies have demonstrated the prevention of colon inflammation in experimental TNBS colitis in mice after the systemic administration of recombinant Ad5 encoding IL-10 [91]. Interestingly, Lactococcus lactis bacteria genetically modified to secrete murine IL-10 was used to deliver IL-10 to intestinal mucosa. In that study, daily administration via a feeding tube resulted in significant clinical improvement of colitis severity in IL-10−/− and DSS colitis models [92]. As a result of the presence of a large number of stem cells in intestinal crypts and ease of access from the luminal site, the gut is suggested to be an interesting target for therapeutic gene transfer. Unfortunately, gene transfer into the bowel wall is rather demanding, because protective extracellular barriers such as tight junctions, glycocalyx, and mucus are potent safeguards against the entry of extrinsic genetic information. For the purpose of successful gene therapy for IBD, targeting immune cells of gut-associated lymphoid tissue (GALT) is potentially desirable [93, 94]. Since the epithelial barrier prevents the efficient transduction of subepithelial areas after vector administration into the gut lumen, other administration routes or specialized vector systems are required to target enhanced numbers of mononuclear cells in the lamina propria, mesenteric lymph nodes, and Peyer’s patches of the ileum. In vitro and in vivo studies with human and rodent cell lines, as well as animal models using reporter genes, have demonstrated that transduction of intestinal mucosa by local administration of liposomal, retroviral, lentiviral, and adeno-associated (AAV) viral and adenoviral vector systems is feasible [95–98]. In those studies, transduction of colon cells after oral vector delivery was not observed, but could presumably be achieved by rectal application. Owing to the high turnover rate of gut epithelium, the adenovirus-mediated gene expression was, therefore, only transient and declined sharply after 2–3 days. Chen et al. [99] reported successful gene transfer into Peyer’s patches targeting macrophages and epithelial cells by direct injection of recombinant Ad5. It was also shown that the capacity of recombinant Ad5 for targeting non- or semi-permissive cells could be enhanced by developing adenoviral vectors with genetically altered tropism. When evaluated in regard to intestinal gene transfer, the binding of recombinant Ad5 with a modified fiber structure to ubiquitously expressed cellular heparan sulfate receptors increased the transduction of lamina propria mononuclear cell (LPMC) in vitro more than 10-fold [94]. The gut is a self-renewing organ in which most cell types found in intestinal epithelium are constantly shed into the fecal stream and must be replaced by a steady supply of cells generated by rapidly dividing multipotent stem cells. Thus, stem cells are a prime target for gene therapeutic approaches, and the successful genetic modification of intestinal stem cells has outstanding clinical potential for many gastrointestinal diseases, including IBD. Recent research progress in the field of intestinal stem cell biology, particularly the development of in vitro cultivation methods for adult primary intestinal epithelium, may now allow the detailed characterization and experimental manipulation of stem cells [100, 101]. In the future, transplantation of in vitro transfected intestinal stem cells could be an important method for delivering therapeutic genes to the gut.

Future perspectives

Genetic studies have opened a window into the complex biology of IBD, revealing which specific genes related to the innate immune system are involved in IBD pathogenesis (e.g., NOD2, TLR, JAK2, STAT3, and IL-23 signaling) and identifying entirely unexpected paths to disease (e.g., autophagy). However, it is considered that the number of loci now reported to be associated with IBD represents only a small fraction of the genetic risk; thus, additional genetic contributions remain to be discovered. Furthermore, the actual causal variants have been determined in only a minority of loci. The discovery of causal variants will require high-throughput re-sequencing of candidate loci in large cohorts comprised of patients and controls, followed by genotyping to determine the association with disease. The application of expression microarrays cataloging the mRNA expression of thousands of genes may provide additional insight into disease pathophysiology. Furthermore, the application of these approaches to IBD intestinal tissues may result in the sorting of subsets of genes differentially expressed in CD and UC. Understanding the pathways in which genetic factors influence IBD will help to better delineate the biological mechanisms of disease pathogenesis and will eventually lead to effective therapies. It is hoped that the identification of other target genes will also lead to a better understanding of the mechanisms underlying the pathogenesis of IBD and, more importantly, that the identification of such genes will lead to precise diagnoses and the development of new drug therapies to improve the quality of life of affected individuals.