Identification of Bifidobacteria as novel activators of AhR pathway in human intestinal epithelial cells 

Get Complete Project Material File(s) Now! »

Development and age-related variation of gut microbiota

Despite the variations due to environmental factors, the microbiota of healthy adults, could be still defined as quite stable. However, during lifetime from newborn to elderly, the microbial density and composition evolve dramatically (Figure 13).
The colonization of the intestinal tract starts very early in the animal life. Recent studies suggested that the bacterial colonization of the newborn already occurs in uterus; indeed the microbiota composition in infant meconium shares features with the bacterial communities identified in placenta and in amniotic fluid, suggesting a prenatally gut colonization (Collado et al. 2016). At birth, the neonatal colonization by microbiota is further influenced by a variety of environmental factors, including the mode of delivery, maternal microbiome and the hygiene of the neonatal environment. Then, the link between the mother and the offspring perpetuates after birth with microbes present in breast milk. Coherently, at the age of 3–4 days, the infant gut microbiota composition is similar to the bacterial community present in colostrum (Collado et al. 2016). Neonatal are rapidly colonized by facultative anaerobic-aerobic bacteria (Staphylococci, Enterococci, Enterobacteria) through the mother and the environment. However, these bacteria, reaching high densities in few days, consume the oxygen in the intestinal tract, with a consequent implantation of strict anaerobic genera such as Bifidobacterium, Bacteroides and Clostridium.
The newly implanted microbial community diversifies during the first months of life and then stabilizes around the age of 3 years, when its composition resembles the microbiota of the adult (Yatsunenko et al. 2012). However, the adult microbiota is likely to maintain the imprinting of the infant colonization, primarily influenced by the mother, method of delivery and breast feeding. Indeed, low level of Bifidobacteria have been observed in babies whose mother gained significant weight during pregnancy as well as in asthmatic patients whose mother have significantly reduced Bifidobacteria species in her breast milk (Grönlund et al. 2007).
The relative stable adult microbiota undergoes a major change in composition as well as in diversity with ageing, characterized by lower Firmicutes/Bacteroidetes ratio in adults over 65 compared to younger adults (Biagi et al. 2010) (Figure 12). However, conflicting results emerged when comparing different cohorts, probably influenced by the reduced bacterial diversity and the different geographical localization (Mueller et al. 2006; Biagi et al. 2010). For example, a small percentage of Firmicutes was observed in Japanese, Finnish and Italian population whilst in Germans the trend was reversed (Mueller et al. 2006). Conversely in an Italian study, young and elderly adults showed a very comparable overall structure of the gut microbiota, whilst centenarians emerged as a separate population, with a significant lower density and a different composition than the adult-like pattern (Biagi et al. 2010). Specific changes in relative proportion of Firmicutes subgroups are observed in extreme old people, with a decrease in the contributing Clostridium cluster XIVa, an increase in Bacilli, and a rearrangement of the Clostridium cluster IV composition. Moreover, the gut microbiota of centenarians was also described as enriched in Proteobacteria that, under some circumstances (e.g. inflammation), could induce pathologies (Biagi et al. 2010).

Mechanisms for Sensing Microbial Signals by IECs

The maintenance of barrier and immunoregulatory functions, is guaranteed by the ability of intestinal epithelial cells (IECs) to act as a frontline sensor of microbes and to integrate commensal bacteria-derived signals into antimicrobial and immunoregulatory responses.
IECs express pattern-recognition receptors (PRR) that enable them to act as a dynamic sensor of the microbial environment and as active participants of the mucosal immune response. The PRRs sense evolutionarily conserved microbe-associated molecular patterns (MAMPs) of microorganisms (e.g. flagellin, peptidoglycan, lipoteicholic acid) that trigger sequential activation of intracellular signalling pathways, leading to the induction of cytokines and chemokines to modulate the early host resistance to infection. Members of PRR include Toll-like receptors (TLRs) and Nucleotide binding Oligomerization Domain (NOD)-like receptors (Figure 15). Toll-like receptors (TLRs) (Figure 15) are PRRs expressed by various cells in the gastrointestinal tract, including intestinal epithelial cells (IEC) and resident immune cells in the lamina propria. TLR signalling is involved in either maintaining intestinal homeostasis or the induction of an inflammatory response. TLRs recognize a wide range of microbial fragments and therefore sense antigens derived from both the microbiota and from invading pathogens. TLR2, dimerizing with TLR1 or TLR6, recognizes bacterial cell wall lipoproteins. Lipopolysaccharide (LPS) produced by Gram-negative bacteria is recognized by TLR4, whereas flagellin is recognized by TLR5. In addition, bacterial DNA is recognized by TLR9. Coherently with their role, TLR2, 4, and 5 are generally expressed at the cell membrane, whereas TLR9 is expressed intracellularly. However, in IEC, TLR9 has been reported to be also expressed at the cell membrane (Lee et al. 2006). Under homeostatic conditions, IEC show low expression of TLR2 and TLR4 and therefore they are unresponsive to their TLR stimuli (Otte et al. 2004). Additionally, a bacterial polysaccharide (PSA) produced by B. fragilis activates TLR2 directly on Foxp3+ regulatory T cells to induce mucosal tolerance (Round et al. 2011). However, under inflammatory conditions, epithelial TLR expression is increased, which contributes to both inflammation as well as immune tolerance (Otte et al. 2004). The TLRs signalling, dependent on two adaptor molecules MyD88 (all TLRs except TLR3) and TRIF (TLR3 and TLR4), induces the production of antimicrobial peptides (AMP) belonging to the C-type lectin family (e.g. Reg3β and Reg3ϒ), in response to bacterial signals (Kawasaki & Kawai 2014). Thus, the fine regulated cross-talk between the host and the microbiota, is also mediated by the TLR signalling. Indeed, MyD88-/- mice are associated with both a shift in bacterial diversity and a greater proportion of segmented filamentous bacteria (SFB) in small intestine (Larsson et al. 2012) that could have potential pathogenic roles. It has recently been reported that the expression and secretion of soluble protein of the lectin family (galectin-9) is supported by the TLR9 activation, inducing tolerogenic dendritic cells (DC) along with the development of the Treg cells (de Kivit et al. 2013). Furthermore, a recent study demonstrated that Clostridium butyricum increased iTreg generation via a TLR2-dependent induction of TGF1 by DCs (Kashiwagi et al. 2015). This TLR2-dependent generation of tolerogenic DC favours regulatory T-cells induction which is similar to what was reported for Bacteroides fragilis (Round et al. 2011) suggesting a more general mechanism used by several commensal bacteria.

Intestinal Microbiota and the Immune System beyond the Gut

While the link between microbiota and the local mucosal immune response has been largely described, more limited is the knowledge on the impact of commensal bacteria on peripheral responses. The development of antibiotics and the improvement of hygiene have led to a significant reduction in infections but also to an increased susceptibility to autoimmune and allergic diseases (Russell et al. 2012; Bach 2002). For the human population, antibiotics are seen as a major modifiers of the beneficial human-microbiota crosstalk, together with alterations caused by other exogenous factor, such as urbanization, global travel and dietary changes (Dethlefsen et al. 2007). In experimental models, antibiotic administration modifies microbiota structure and is linked to an increase susceptibility, for example, to allergic airway inflammation and food allergies (Bashir et al. 2004; Noverr et al. 2005). The long-term consequences of microbial perturbation through the intensive use of antibiotics are difficult to discern, although chronic conditions, such as asthma, have been associated with childhood antibiotic use and altered microbiota. More direct evidences have been provided for the role of microbiota on peripheral responses and in particular the innate immune cell development. Germ-free (GF) mice display reduced proportions and differentiation potential of myeloid cell progenitor populations of both yolk sac and bone marrow origin. The defect in myelopoiesis resulted in less resistance and more severe pathogen burden following to Listeria monocytogenes infection, rescued by the re-colonization of the mice gut with a complex microbial community (Khosravi et al. 2014). These findings reveal that gut microbiota directs innate immune cell development via promoting hematopoiesis. Conversely, the gut microbiota can also alter autoimmune conditions. GF mice develop significantly less severe diseases in models of experimental autoimmune encephalomyelitis (EAE) (Ochoa-Reparaz et al. 2010). On the other hand, colonization with segmented filamentous bacteria (SFB), promotes autoimmune arthritis through the induction of antigen specific TH17 cells, which in turn promote auto-antibody production via B cell expansion in germinal centres (Wu et al. 2010). Additionally, commensal microbiota is also describe to be involved in inflammatory diseases, such as inflammatory bowel diseases (IBD), and in metabolic diseases (e.g. obesity, diabetes) (detailed in section 1.4.5).

Ligand-induced activation of AhR signaling pathway

The AhR activation is generally mediated by the binding of diverse classes of molecules, to the ligand binding pocket in the LBD of the receptor. Although some authors reported the activation of AhR signalling in the absence of detectable exogenous ligands, the evidences for a ligand-independent activation seemed discordant as some chemicals, firstly reported to activate AhR through a ligand-independent mechanism, were lately identified as weak AhR ligands (e.g. omeprazole (Gerbal-Chaloin et al. 2006)). In this regards and in light of the aim of this manuscript, the herein discussion will focus only on the AhR signalling induced by ligand binding.
A huge number of molecules originated from the environment, diet as well as endogenous compounds, have been described as ligand of AhR (Barouki et al. 2007; Murray & Perdew 2017), although the physiological role in some cases remains partly unclear. The best characterised high affinity ligands for AhR include a variety of environmental contaminants such as halogenated aromatic hydrocarbons (HAHs), among which dioxin, and polycyclic aromatic hydrocarbons (PAHs). However, other molecules have been recently identified as AhR ligand and activators of AhR signalling. Consequently, AhR is not a mere dioxin-receptor but it is now recognised as more promiscuous receptor.
The activation of AhR depends upon two main characteristics of the ligand: the affinity, which is the property of attraction between the ligand and the receptor, and the intrinsic effect, referring to the receptor occupancy. A potent agonist is thus defined as a compound with both strong affinity and high efficacy. Conversely, a molecule could bind to the receptor with low efficacy, resulting in no activation of the receptor, while having a good affinity for it. This is the case of antagonist molecules, which have good or high affinity for the receptor and so compete with the agonist for the binding.
Another important characteristic of ligands is the potency, defined as the concentration of the ligand required to produce a detectable effect, crucial to describe the inducible effects of the specific compound and, in our contest, fundamental for evaluating the physiological impact. The variety of ligand activating AhR include not only strong agonists and antagonist but also partial agonists, which show both agonistic and antagonistic properties.

AhR-Dependent Genes and Negative Feed-Back Regulation of the AhR Signalling

The cytochrome P450 1A1 (CYP1A1), a drug metabolizing enzyme, was the first described as dependent on AhR activation signalling. However, later on, AhR was found to regulate the expression of a variety of other drug metabolizing enzymes, including cytochrome P450 1A2 (CYP1A2), cytochrome P450 1B1 (CYP1B1), glutathione-S transferase (GST) A1, NAD(P)H:quinone oxidoreductase 1 (NQO1), UDP-glucuronosyltransferase (UGT) 1A, aldehyde dehydrogenase(ALDH)3A1. Nonetheless, the list of genes influenced by AhR activation continues to grow. Indeed, recent works suggested the impact of AhR on PPARγ- and estrogen receptor (ER)- related gene expression, on the gene-encoding aromatase cytochrome P450 19A1 (CYP19A1) and on enhanced expression of genes encoding epidermal growth factor(EGF)-like proteins (Haarmann-Stemmann et al. 2009; Kerkvliet 2009; Puga et al. 2009; Beischlag & Perdew 2005).
Interestingly, AhR also regulates the expression of its repressor AhRR to ensure a negative regulator loop. Indeed multiple mechanisms evolved to suppress sustained AhR activity, implying that the prolonged receptor signalling is physiologically deleterious. Among these regulatory mechanisms, the AhR-dependent expression of AhRR as well as CYP1A1 constitute a negative regulatory feedback loop through the metabolic depletion of exogenous and endogenous ligands.

AhRR Expression and Regulation of AhR Signalling

AHRR was isolated and identified in 1999 from a mouse intestinal cDNA library and firstly described as an “AHR-like” protein, hypothesized to compete for binding to ARNT and work as a negative regulator of the receptor activity (Kawajri & Fujii-Kuriyama 2007; Baba et al. 2001; Mimura et al. 1999). The AhRR was then better characterized as a highly evolutionarily conserved bHLH/PAS protein in vertebrates, closely related to the AHR (Hahn et al. 2009; Mimura et al. 1999; Haarmann-Stemmann et al. 2007) and localized in chromosome 13C2 in mouse, 1p11 in rat and 5p15.3 in human gene (Baba et al. 2001). Orthologous proteins have been described in several mammalian species, including mice, humans, rats, amphibians, and many types of bony fish (Tsuchiya et al. 2003; Korkalainen et al. 2001).
In humans, as well as in mouse and rats, AhRR expression has been detected in numerous tissues and cell lines revealing a tissue- and cell-type specific expression (Tsuchiya et al. 2003; Korkalainen et al. 2001). Notably, in adult human, AHRR was detected in liver, breast, colon, kidney, lung, bladder, uterus, ovary, adrenal gland, with a prominent high expression in testis. Furthermore, AhRR mRNA expression was also evidenced in the lung, kidney, spleen, and thymus of the human fetus, but not in brain, liver, heart, and muscle (Tsuchiya et al. 2003). This observed embryo-fetal expression, suggests a possible developmental role for the AHRR, although in one study, AHRR-/- mice are fertile and the offspring appear to develop normally (Hahn et al. 2009; Hosoya et al. 2008).
Studies on AhRR structure revealed high sequence similarity with the AhR in the N-terminal region in which are residing both NLS and NES, homologous to AHR (Kawajiri & Fujii-Kuriyama 2007). However, aside the N-terminal region, the repressor significantly diverges from AHR towards the C-terminus (Kewley et al. 2004). Owing to its structural similarity to the AhR in the N-terminal half, the AhRR can also dimerize with ARNT and bind to XREs. However, since the C-terminal part of the AhRR protein lacks the transactivation domain (TAD) (Baba et al. 2001; Mimura et al. 1999), the DNA binding is proposed to recruits co-repressors and to function as a transcription repressor of AhR activation (Mimura et al. 1999). Additionally, AhRR is also lacking the PAS-B motif and LBD, making the protein unable to bind ligands (Baba et al. 2001; Mimura et al. 1999) (Figure 29).

Role of AhR on intestinal epithelial barrier function

The proliferation and maintenance of the intestinal epithelium is the front line physical barrier for the protection toward environmental contaminants as well as potential harmful microbes. AhR-/- mice showed an impaired proliferation of colonic crypt stem cells with a consequent defect in cell renewal and integrity of the epithelium (Stockinger et al. 2014). Additionally, an enhanced apoptosis of epithelial cells was observed when AhR was depleted in mice (Chinen et al. 2015). The role of AhR on epithelial integrity was further described to be mediated by the ligand-dependent activation of the receptor. Indeed, AhR agonist FICZ in in vivo models ameliorates intestinal obstruction- and DSS- driven intestinal permeability by rescuing the expression of epithelial tight junction (TJ) proteins zonula occludens-1 (ZO-1), Occludin and Claudin-1 (Yu et al. 2018; Han et al. 2016). This ligand-dependent effect of AHR on epithelial permeability was confirmed by in vitro experiments in which Caco-2 cells evidenced significant change in TJ distribution following to FICZ exposure (Yu et al. 2018). In this context, a recent study showed that loss of Notch1 signalling in Caco-2 cells, counteracted the development of TJs induced by FICZ (Liu et al. 2018); consequently, a mechanism mediated by the up-regulation of Notch1 was proposed for the AhR-mediated protection against intestinal damages. Similarly, bacterial derived indoles has been shown to increases epithelial TJ integrity and reduce intestinal inflammation (Bansal et al. 2010), although it is still not clear if this effect is mediated by AhR.
Despite its significance at epithelial level, it should be point out that AhR is expressed in a variety of other cells, thus the effects on the epithelium could be mediated by signalling through other cell types. For example, the AhR-dependent production of IL-22 by immune cells (ILCs and TH17), have a major effect on IEC proliferation and production of antimicrobial peptides (Kiss et al. 2011; Qiu et al. 2012). In turn, IEC may modulate the effect induced by AhR on other cell types. Indeed, constitutive CYP1A1 activity in IECs, but not in adaptive immune cells, restricts the availability of AHR ligands to cells in the intestinal lamina propria, resulting in loss of AHR ligand-dependent ILC3 cells (Schiering et al. 2017). In turn, the reduction in ILC3 affect the level of IL-22 in the colon with a consequent increased sensitivity toward Citrobacter rodentium infections (Schiering et al. 2017). Conversely, constitutive expression restricted to adaptive immune cells showed normal number of ILC3 in steady state and a survival rate to Citrobacter rodentium infection similar to wild type mice (Schiering et al. 2017). Consequently, this suggest a major role of IECs in molecular clearance, fundamental for the availability of AhR ligands to other cell types and a consequent control of the AhR signalling (Schiering et al. 2017).
Aside from the epithelial proliferation, the barrier function in intestine is also guaranteed by the production of mucus. In a lung epithelial cell line, AhR activation by TCDD was shown to induced the expression of MUC5AC (Wong et al. 2010). Considering that MUC5AC is also expressed in human intestine (Guyonnet Duperat et al. 1995), it is conceivable that similarly to what observed in the lung, AhR could be involved in the modulation of mucus production in the intestine. However, to my knowledge not such studies have been reported.

Table of contents :

Chapter 1. Introduction 1 The gastrointestinal tract 
1.1.1. Stomach
1.1.2. Intestines 3 The immune system in the GI tract
The Human Gut Microbiota
1.3.1. Composition of the gut microbiota
1.3.2. Development and age-related variation of gut microbiota
1.3.3. The study of gut microbiota 18 The Role of Gut Microbiota in the Host-Bacteria Cross-Talk
1.4.1. The Barrier Homeostasis in Intestine
1.4.2. Mechanisms for Sensing Microbial Signals by IECs
1.4.3. Intestinal Microbiota and the Immune System beyond the Gut
1.4.4. Microbial-derived Metabolites in Host-Bacterial Cross-Talk
1.4.5. Dysbiosis in Human Pathology
Aryl Hydrocarbon Receptor: Description, Characterization and Physiological Role
1.5.1. Expression and Localization
1.5.2. Structure and Functional Domains
1.5.3. AHR pathway
1.5.4. Physiological role of AhR
1.5.5. Cross-talk with other signalling pathways
Chapter 2. Rationale and Objectives 
Chapter 3. Results 
Paper I: Identification of the novel role of butyrate as AhR ligand in human intestinal epithelial cells. 
Paper II: Identification of Bifidobacteria as novel activators of AhR pathway in human intestinal epithelial cells 
Chapter 4. Discussion 
Chapter 5. Supplementary 
Choline-TMA catabolism by gut microbiota
5.1.1. Comments on Objectives and Methods
5.1.2. Experimental set-up
5.1.3. Comments on the Results and Discussion
Bibliography 

GET THE COMPLETE PROJECT