Bacterial-Enterocyte Crosstalk: Cellular Mechanisms in Health and Disease
2003; Lippincott Williams & Wilkins; Volume: 36; Issue: 2 Linguagem: Inglês
10.1097/00005176-200302000-00005
ISSN1536-4801
AutoresHenrik Köhler, Beth A. McCormick, W. Allan Walker,
Tópico(s)Viral gastroenteritis research and epidemiology
ResumoINTRODUCTION The human colon maintains a microbial density approaching 1012 organisms per gram of feces, representing a perfectly balanced ecosystem. The commensal microflora, derived from the Latin "commensalis" meaning "at table together", consists of more than 400 species and lives in perfect harmony with the human intestine (1). Recently the term probiotics was defined by Fuller (2) as a "live microbial food supplement which beneficially affects the host animal by improving its intestinal microbial balance". Commercial probiotics have been used extensively in the prevention and treatment of human conditions from diarrhea of the infant to inflammatory bowel disease (3). However, the mechanisms by which these intestinal bacteria prevent an inflammatory response in reaction to the plethora of foreign antigens available in the intestinal lumen have just begun to come under investigation. In contrast to the myriad of indigenous microorganisms in symbiosis it takes just 10 to 100 single organisms of the pathogen Shigella to destroy this peaceful coexistence and result in bloody mucoid diarrhea associated with fever, nausea, and abdominal cramps. This infectious disease still costs more than half a million people their lives every year and most of them are children under 5 years of age. Even in a developed country such as the U.S.A., bacterial food-borne pathogens may cause an estimated 4.1 million illnesses, 45,000 hospitalizations, and 1,500 deaths every year (4). This continuing health threat has resulted in the development of the Foodborne Diseases Active Surveillance Network (Foodnet) by the Centers for Disease Control and Prevention (CDC). To maintain health and cause disease, which are both active processes, bacteria and the intestinal epithelium must actively interact. This interaction is referred to as "crosstalk" and this research area has become one of the most active in gastroenterology today. Commensal microflora together with enterocytes are hypothesized to act as a gatekeeper to protect the human organism from penetration and disease. On the other hand, bacterial pathogens themselves have developed in their two and a half billion years on earth (5) sophisticated mechanisms, like the type III secretion system, to interact with host cells to promote their own survival. An understanding of the molecular mechanisms by which these microorganisms interact with the intestinal epithelium to promote health or cause disease is expanding rapidly. It is becoming more and more obvious that prokaryotic organisms manipulate the human enterocyte functions in many ways for their own benefit. In contrast commensal microflora and probiotics can provide benefits to the host beyond a local effect in the intestine. These organisms use similar mechanisms to that of the pathogen but these mechanisms are largely undefined because of a lack of sufficient basic research. The interaction of the bacterium with the intestinal epithelial cell (IEC) has, somewhat artificially, been divided into the following categories: i) attachment and invasion, ii) altered epithelial barrier function, and iii) inflammation. All these phenomena are constantly both activated and suppressed at the same time in the human gut by different microorganisms in a highly sophisticated interaction with IEC. In this review we will not comprehensively cover these topics but will rather highlight some of the more recent findings to illustrate the cellular mechanisms that effect microbial-enterocyte crosstalk in health and disease in the human intestine. ATTACHMENT AND INVASION Infectious diarrhea is still the leading cause of death in children in developing countries with the problem being most severe during the first year of life. It accounts for a total mortality of 3.3 million deaths per year (6). The enteropathogens Escherichia coli (E. coli) and Salmonella enterica are closely related and most experts consider Shigella to belong to the species of the genus of Escherichia. The considerable pathogenic diversity superimposed on a relatively conserved genetic backbone is a result of the evolution and dispersion of virulence genes that are spread between the species on transmissible genetic elements. These elements include plasmids, bacteriophages (viruses that infect bacteria), and pathogenicity islands (Table 1). Pathogenicity islands are defined as discrete and relatively large loci containing sets of virulence genes that are present within the genome of pathogenic bacteria but are absent from the genome in closely related non-pathogenic strains (7). Many pathogenicity islands encode specialized systems for the delivery of virulence proteins into the host cell. One such system is termed the type III secretion system (TTSS). TTSS are specialized virulence devices that have evolved to modify host cell function through the direct translocation of bacterial proteins into the host cytoplasm (8). The structure has a needle-like appearance in Shigella (9) and Salmonella (10) but is somewhat thicker as the appendage of enteropathogenic Escherichia coli (EPEC) (11). In all cases it bridges between bacterial and host cells.TABLE 1: Selected bacterial factors and their role in the bacterial interaction with the intestinal epitheliumSalmonella The enteric pathogen Salmonella causes a range of diseases from gastroenteritis to enteric (typhoid) fever. Typhoid fever is a protracted systemic illness that results from the exclusively human pathogens S. typhii and S. paratyphi. Without treatment mortality is between 10% and 15%. In contrast, the many nontyphoidal Salmonella strains, such as S. typhimurium and enteritidis, infect a wide range of animal hosts including poultry, cattle, and pigs and usually cause a self-limiting enteritis in humans. Very young, old, or immunocompromised individuals are prone to a systemic disease which causes 500 deaths in the U.S.A. alone every year (4). After entering the GI tract, via contaminated food or water, in a manner similar to other pathogens discussed in this review, Salmonella must penetrate the intestinal mucous layer in the small bowel before encountering and adhering to IEC. Salmonella expresses several fimbriae, which contribute to the adherence process (12). Salmonella typhimurium finally invades its animal host by entering and traversing the epithelial monolayer lining the intestine. This is achieved by initiating an extensive enterocyte-cytoskeletal rearrangement that results in membrane ruffling and bacterial internalization by epithelial cells, which are normally nonphagocytotic (13). Bacterial adhesion to the tips of apical microvilli activates the contact-dependent TTSS through which bacterial products are injected into the host cell (14). Recent research efforts have focused on identifying the responsible virulence proteins and describing host-cell function targeted by these proteins. Members of the Rho family, cell proteins involved in signal transduction, of small Guanine Triphosphatases (GTPases) which includes RhoA, Rac1, and Cdc42, play a central role in regulating the actin cytoskeleton (15). In in vitro studies it was first demonstrated that invasion of Salmonella typhimurium was primarily dependent on Cdc42. Expression of a point mutation on Cdc42 hinders it from binding GTP, which acts in an inhibitory manner preventing bacterial entry. Expression of dominant-negative Rac1 only partially inhibits internalization. In addition, Cdc42 mediates the Salmonella-induced activation of the signaling protein Jun kinase (JNK) (16). However, these studies were conducted in nonpolarized cells, which means that they do not have an apical side and basolateral side. This polarization seems to be a critical cellular characteristic for the intestinal epithelium since the composition of the cellular environment on both sides of the epithelial surfaces, the lumen and subepithelial space, are completely different. In a polarized model epithelium, which probably simulates the in vivo situation more accurately, only Rac1, but not Cdc42, significantly inhibited bacterial entry from the apical side of the host cell (17). It was determined that the modulation of Rho GTPase activity was dependent on two Salmonella effector proteins SopE and Spt (18,19,20). SopE was characterized as a guanine nucleotide exchange factor (GEF) by its ability to stimulate in vitro a nucleotide exchange on Cdc42, Rac1, and RhoA and is required for bacterial entry. A recently described similar bacterial protein, SopE2, possesses comparable activity (21,22). Other bacterial effector proteins, SipA and SipC, are also involved in regulating and timing the invasion process but do not seem to be essential (23,24). Shigella Shigella spp. is a group of gram-negative enteric bacilli that causes acute bacillary dysentery in humans. The syndrome caused by Shigella consists of painful abdominal cramps, nausea and fever, along with blood and mucus in the stool. In its most severe forms, shigellosis is associated with an intense inflammatory reaction that leads to the destruction of the colonic mucosa (25). Shigellosis is mostly a pediatric disease, with greater than 60% of the cases occurring in children of 1 to 5 years of age, and it is primarily a third-world disease, with approximately 150 million cases every year. Most important is a deadly disease causing about 1 million deaths every year, again primarily in infants and young children (26). The ability of this food-borne pathogen to invade and colonize the colonic epithelium is a key determinant in the establishment of the disease. The invasion capability of Shigella flexneri has been localized to a 31 kilobase (kb) region on its pathogenicity islands (27). This 31kb region codes for many closely linked genes, including the invasion plasmid antigen (Ipa), the membrane expression of invasion plasmid antigens (mxi), and the surface expression of invasion plasmid antigen (spa) genes as well as other, independently expressed genes. One astonishing feature about Shigella infection is that although very few bacteria (10 to 100) can cause dysentery in humans, this microorganism is completely nonpathogenic for experimental animals except certain primates. An additional characteristic of Shigella is that they are unable to invade epithelial cells from the apical surface and therefore most first gain access to the basolateral side before entry (28) (Fig. 1). One mechanism by which Shigella translocates the epithelial layer before entering is the epithelial cell itself is through specialized microfold (M) cells or follicle-associated epithelium (29). There is growing evidence that Shigella can also gain access to the basolateral cell surface by traversing the paracellular space after manipulating tight junctions (30). The mechanism by which Shigella adheres to the host's basolateral cell surface prior to entry is incompletely understood, although the bacterial Ipa complex has been shown to bind to the integrin α5β1 on the host cell (31). Moreover, IpaB may bind directly to the extracellular domain of an enterocyte surface receptor named CD44 (32). Nonetheless, S. flexneri requires contact with the host cell before Ipa proteins are secreted via the Mxi/Spa secretion apparatus (33). In a model of intestinal epithelium it was demonstrated that the lipopolysaccharide (LPS) oligosaccharide side chain, the outermost part of the bacterial membrane, is probably to some extent responsible for the attachment of the bacterium to the basolateral surface of IEC (34). The cellular receptor for this interaction still must be identified. Shigella invasion is also initiated by TTSS-mediated insertion of a pore into the cytoplasmic membrane of the eukaryotic target cell. The bacterial proteins IpaB and IpaC, present in the pore, induce an early polymerization of actin via the carboxy terminal domain of IpaC (35) and most likely mediate the injection of further proteins, such as IpA and IpgD, into the host cell cytoplasm. The process of invasion by which Shigella causes the formation of filipods that are quickly remodeled into lamellipods, thus resulting in a structure that entraps the microorganism (36). Further remodeling leads to the formation of a macropinocytic vacuole that eventually completes the internalization. In a manner similar to the process seen in Salmonella, but at the opposite side of the cell, the Shigella protein IpaC targets the signal transduction pathway of small GTPases of the Rho family but also the protooncogene c-src is recruited (37). IpaA acts through binding of the NH2-terminal domain of the cellular protein vinculin (38), thereby causing actin bundling and formation of a plaque (39). All these events are excellent examples of bacteria manipulating the host cell for its own entry, in this case causing actin nucleation and polymerization at the eukaryotic membrane.FIG. 1.: Pathogenesis of Shigella flexneri. Shigella translocates the epithelial layer before entering the epithelial cell itself through specialized microfold (M) cells epithelium. Shigella can also gain access to the basolateral cell surface by traversing the paracellular space after manipulating tight junctions. To survive in the lamina propria Shigella induces apoptosis in macrophages. Once in contact with the basolateral epithelial surface it induces its phagocytosis by the epithelial cell. Shigella escapes through an intracellular vacuole and multiplies within the cytoplasm. Shigella does not possess flagella and therefore has developed a mechanism to enable intracellular movement. The bacterial protein IcsA assembles host cell actin filaments into a tail that forms at one pole of the bacterium and provides the force that propels the bacterium forward through the cytoplasm and allows horizontal cell-to-cell spread. An infection of epithelial cells with Shigella also results in the secretion of interleukin (IL)-8 which causes recruitment of PMN to the basolateral surface of the epithelium. The epithelial secretion of PEEC is the likely but not fully established factor responsible for the transepithelial migration of PMN.E. coli In developing countries, enteropathogenic E. coli (EPEC) is one of the most common pathogens. In Brazil, for example, EPEC can be isolated from stools of over 40% of infants with acute diarrhea and was associated with a mortality of 7% (40). The pathogenesis of EPEC is in some way unique for enteric bacterial pathogens since it is essentially noninvasive and produces no toxins. The attachment of EPEC to the epithelial cell, described as localized adherence, results in a so-called attaching and effacing lesion (A/E) (41,42). EPEC also uses its TTSS to deliver bacterial effector proteins like EspA and EspB into the host cell to alter the cytoskeleton (11,43). However, the most fascinating aspect of EPEC pathogenesis is that it inserts, through the type III secretion system, its own receptor into the host cell. Rather than searching for a receptor it provides its own receptor and uses it when needed (Fig. 2). Thus, EPEC is able to insert the Tir receptor into the host cell membrane where it serves as the receptor for the bacterial protein intimin after it is phosphorylated on tyrosine by the host cell (44,45). In addition, intimin also possesses binding sites for β1-integrin (46).FIG. 2.: EPEC pathogenesis. The attachment of EPEC to the epithelial cell, described as localized adherence, results in a so-called attaching and effacing lesion (A/E). The initial step of adherence of EPEC to the epithelial cell is mediated by the bundle forming pilus (BFP). Subsequently EPEC insert through the type III secretion system the Tir receptor into the host cell membrane where it serves as the receptor for the bacterial protein intimin. Finally, a cytoskeletal rearrangement occurs after Tir-intemin binding, resulting in the formation of a pedestal like structure. Partially adapted from (7).COMMENSAL MICROFLORA AND PROBIOTICS CAN INHIBIT PATHOGEN WITH THE ENTEROCYTE INTERACTION The simplest mechanism by which the commensal microflora and also probiotics function in the prevention of adhesion of a pathogen is by competing for the binding site on the epithelium, a phenomenon called competitive exclusion. Lactobacillus spp. directly inhibit the attachment of Salmonella, E. coli, and other foodborne pathogens to human intestinal cell lines (3,47,48,49) through either a competition of attachment sites or in some instances due to secretion of bactericidal factors. The polyamine derivate piperidine, which presumably is produced by intestinal microflora, has been shown to inhibit the binding and internalization of Salmonella and Shigella to intestinal epithelial cells in vitro (50). Whether this is a relevant in vivo mechanism still must be determined. A fascinating example of how nonpathogenic bacteria inhibit pathogen adhesion using the host cell as an effector, and not competing with binding sites with the pathogen, was demonstrated recently for Lactobacillus plantarum 299v. This probiotic bacterium increased the transcription and excretion of the mucins MUC2 and MUC3 in goblet cells and thereby inhibited the adherence of EPEC to the intestinal surface (51). Both pathogens and commensal bacteria adhere to specific glycoconjugates on the microvilli in the intestine; recently it was demonstrated that they also influence the glycosylation pattern themselves. The intestinal microflora can at least partially stimulate specific glycosylation via stimulation the enzymes α2,3/6-sialyltranferase and α1,2-fucosyltransferase on the surface of intestinal epithelial cells. It was demonstrated that germfree mice show an immature glycosylation pattern longer than animals conventionally colonized by normal microflora (52). Since many pathogenic bacteria use intestinal brushborder glycoconjugates as their target host cell receptor, these findings might help explain the regional specificity and age-related susceptibility to certain enteropathogens. EPITHELIAL BARRIER Besides regulating the uptake of fluid, electrolytes, and nutrients the intestinal epithelium has the task of establishing a barrier to the penetration of pathogens. In addition to the possibility of traversing the cell after invasion, the most vulnerable point of penetration is the paracellular space between epithelial cells. The human gut has developed a sophisticated network of tight junction proteins which seal this route to prevent microbial translocation but to allow the transport of electrolytes, etc. The development of intestinal barrier function is determined by the assembly of tight junction (TJ) and adherens junction proteins (53,54) (Fig. 3). These macromolecular complexes of proteins form contiguous rings at the apices of epithelial cells. Claudins (55) and occludin (56) have been identified as tight-junction specific integral membrane proteins. Juxtaposed to the tight junction membrane the membrane-associated guanylate kinase-like homologues (MAGUKs) zonula occludens (ZO)-1 and ZO-2 are located. They are reported to play a general role in creating and maintaining specialized membrane domains by cross-linking multiple integral membrane proteins at the cytoplasmic surface of plasma membranes (57,58). The adherens junction proteins (like E-cadherin in epithelial cells) are located adjacent to the tight junctions. They are also connected to underlying cytoskeletal components and are important in maintaining the integrity of other intercellular junctions (59,60). Tight junction and adherens junction complexes will for simplicity be referred to here as the tight junction (TJ). It is feasible that pathogens aim to destroy the integrity of the epithelial barrier to gain easy access to the gut interstitium, which can further give access to the blood stream and thereby allow systemic spreading of the organism.FIG. 3.: Interaction of a pathogen with the epithelial barrier. Tight junctions are macromolecular complexes of proteins. Claudin and occludin have been identified as tight-junction specific integral membrane proteins. ZO-1 and ZO-2 are play a general role in creating and maintaining specialized membrane domains by cross-linking multiple integral membrane proteins at the cytoplasmic surface of plasma membranes. The adherens junction proteins E-cadherin is located adjacent to the tight junctions, and also connects to the underlying cytoskeletal components (e.g., Cingulin, 7H6, ZA-1TJ, symplexin, α-actinin, vinculin, radixin) maintaining the epithelial integrity. A limited number of pathogens have the ability to translocate across the tight junctional seal into the paracellular space between IEC after manipulating the tight junction complex. Partially adapted from (57).Pathogens can impair the epithelial barrier The list of pathogens identified which affect the barrier is surprisingly short, and even shorter is the list of mechanisms shown to be responsible (61). The most notable mechanism, although indirect, to modify TJ function is the biochemical modification of the actin binding protein Rho, which has been associated with Clostridium difficile toxins A and B (62) and E. coli cytotoxic necrotizing factor 1 (63). The only pathogen-derived factor identified so far affecting claudins is a cytotoxic enterotoxin (CPE) produced by the bacterium Clostridium perfringens that binds to claudin-3 and claudin-4. A carboxy (COOH) terminal fragment of the toxin binds claudin but is non-cytotoxic. This results in an epithelial cell line with the removal of claudin-4 from the cell surface, whereas other claudins remain at the junction. Also, a drop in transepithelial electrical resistance (TER) is observed after monolayers are treated with CPE (64). Claudins share a similar membrane topology with occludin. Occludin spans the membrane four times with cytoplasmic NH2 and COOH terminals and forms two extracellular loops (Fig. 3). Disruption of occludin is known to result in a drop in TER. Therefore, occludin seems to be another important protein for the maintenance of barrier function, although it needs claudin to be fully effective. First E. coli and Yersinia have been shown to disrupt the epithelial barrier by dephosphorylating occludin (65,66,67). The mechanisms by which EPEC affects tight junctions are well studied and particularly interesting since this pathogen is primarily noninvasive (see previous discussion). One mechanism is the phosphorylation of the 20kDa light chain of myosin (MLC20) in the eukaryotic cell initiated by EPEC via an increase in intracellular calcium and the formation of calcium-calmodulin complexes leading to an increased TJ permeability (68). The mechanism by which occludin is dephosphorylated is found to be orchestrated with the help of the TTSS and one secreted protein, which probably is EspF (66). Just recently Chen et al. (69) reported that PKC signaling regulates ZO-1 translocation and increased the paracellular flux in T84 colonocytes exposed to clostridium difficile Toxin A. It was shown previously by Jepson et al. (70) that infection of an epithelial monolayer with S. typhimurium led to apical pole clustering of F-actin, E-cadherin, and ZO-1. In addition recent studies suggest that Shigella flexneri can also regulate a variety of proteins involved in the epithelial barrier (30). S. flexneri rapidly decreases the TER of T84 cell monolayers after apical and basolateral contact, indicating the physical disruption of tight junctions. Furthermore, S. flexneri has the capacity to translocate across the tight junctional seal into the paracellular space between IEC. This observation strongly raises the possibility that Shigella can penetrate the intestinal epithelium directly as well as via a M-cell dependent invasion route thereby gaining access to the basolateral side of the epithelium. The S. flexneri serotype 2a targets claudin-1 from the apical side of an epithelial cell monolayers. Furthermore Shigella flexneri serotype 2a and 5, but not the non-invasive Escherichia coli strain F-18, share the ability to regulate expression of ZO-1, ZO-2, E-cadherin, and to dephosphorylate occludin. The precise factors inducing this manipulation still must be identified. An indirect regulation appears likely, probably triggered by the effect of the pathogen on the cytoskeleton, which is closely linked to the TJ complex. The in vivo relevance of these in vitro systems which detail effects of pathogens on the intestinal epithelial barrier is still to be determined. Commensals can enhance the mucosal barrier So far only a few bacteria are known to enhance the epithelial barrier. Bacteroides thetaiotaomicron is a prominent genetically manipulatable member of the normal mouse microflora (1). It has been shown to fortify the epithelial barrier in mice by inducing the expression of decay-accelarating factor that inhibits cytotoxic damage from microbial activation of secreted complement products (71). It could be demonstrated that Lactobacillus plantarum 299v can prevent the E. coli-induced increase in intestinal permeabilty in a rat model (72). In addition Madsen et al. (73) have shown that a commercial mixture of probiotic bacteria (VSL#3) containing Bifidobacterium longum, B. infantis, B. breve, Lactobacillus acidophilus, L. casei, L. delbruecki, L. plantarum, and Streptococcus salivarius can enhance the epithelial barrier in IL-10 knockout mice which serves as a model for inflammatory bowel disease. However, although the epithelial barrier integrity was determined accurately using mannitol flux and other physiologic parameters, individual tight junction proteins were not investigated. It remains a challenge to dissect this effect at the cellular and subcellular level. INFLAMMATION Proinflammatory mechanisms The pathophysiology of enteritis caused by bacterial pathogens is characterized by movement of electrolytes and water as well as polymorphonuclear leukocytes (PMN or neutrophils) into the intestinal mucosa and lumen from the underlying vasculature. These inflammatory cells can subsequently be detected in the feces. Therefore this section will primarily deal with the inflammatory response leading to transepithelial PMN migration (Fig. 4). Epithelial inflammation following infection with Salmonella typhimurium has been studied extensively. In vitro experiments have shown that interaction between S. typhimurium and polarized intestinal epithelial cells can induce the epithelial cells to release potent chemoattractants that direct the transmigration of PMN (74–76). For example, IL-8 is secreted from epithelial cells basolaterally and functions to attract PMN from the intravascular space through the lamina propria into the infected mucosa (74). A pathogen-elicited-epithelial-chemoattractant (PEEC) is secreted apically by epithelial cells and directs PMN migration across an epithelial monolayer through tight junctions (75). The continuous and polarized secretion of IL-8 and PEEC by epithelial cells is hypothesized to be responsible for directing PMN movement into the intestinal lumen in response to non-typhoidal Salmonella infections in humans. Activation of the IL-8 promoter during Salmonella invasion of cultured epithelial cells is primarily dependent upon the transcription factor nuclear factor (NF)-κB. In addition, recent evidence indicates that S. typhimurium flagellin is transported across a polarized model intestinal epithelium by a transcellular route and activates NF-κB by binding to the Toll-like receptor 5 (TLR5), one of a family of pattern recognition receptors involved in the innate immune response (77,78), located at the basolateral side of the epithelial cell.FIG. 4.: Orchestration of PMN transmigration by Salmonella typhimurium. S. typhimurium flagellin is transported across polarized model intestinal epithelium by a transcellular route and activates NF-κB by binding to the Toll-like receptor 5 (TLR5) at the basolateral side of the epithelial cell. The activation of NF-κB results in the production and secretion of interleukin (IL)-8 from epithelial cells basolaterally and functions to attract PMN from the intravascular space through the lamina propria. The secreted S. typhimurium protein SipA activates though an unknown receptor a protein kinase C (PKC)-dependent signal transduction pathway the apical secretion of the pathogen-elicited-epithelial-chemoattractant (PEEC). PEEC directs PMN migration across the epithelial monolayer through tight junctions. The continuous and polarized secretion of IL-8 and PEEC by epithelial cells is responsible for directing PMN movement into the intestinal lumen in response to non-typhoidal Salmonella infections in humans.By contrast, the final and perhaps rate-limiting step of PMN migration to the intestinal lumen is thought to be governed by the epithelial secretion of PEEC (75). It was recently reported that S. typhimurium activates a protein kinase C (PKC)-dependent signal transduction pathway (independent of NF-κB) that orchestrates transepithelial PMN movement. PKC has long been identified as the cellular receptor for the lipid second messenger diacylglycerol (DAG), and is therefore a crucial element in signal transduction pathways. At present, we know that S. typhimurium initiates an ADP ribosylation factor (ARF) 6 and phospholipase D (PLD) dependent signaling cascade that, in turn, directs activation of PKC, PEEC, and subsequent PMN movement (79). Of interest, bacterial invasion was not necessary to elicit PMN migration. The application of the bacterial secreted protein SipA to the cell surface is sufficient to initiate the inflammatory response (80,81). Although an effector molecule, like SipA in Salmonella, has not been identified in EPEC, a similar mechanism leading to the migration of PMN induced by the noninvasive EPEC seems likely (82). It is noteworthy that a novel bacterial flagellin has been identified in enteroaggregative E. coli which stimulates the secretion of IL-8 by the host target intestinal epithelial cell (83). The release of IL-8 by the epithelium in response to EPEC has been attributed to the activation of NF-κB, in a manner similar to many other stimuli. EPEC has also been shown to activate this pathway by stimulating the phosphorylation and degradation of IκBα (43), the inhibitory molecule that in the noninflammed state keeps NF-κB in the cytoplasm, preventing it from moving into the cell nucleus to initiate transcription of IL-8 etc. Additional involvement of upstream signaling pathways, such as of the mitogen-induced protein (MAP) remain to be studied. The response of intestinal epithelial cells to bacterial LPS is of particular interest. This response must be a highly conserved mechanism across species since an induction of immune defense mechanism against the microbial flora resulting in chronic infection must be avoided. In the human system, the TLR4/MD2/CD14 complex (a complex of transmembranous and surface receptors closely linked to intracellular signal transduction pathways) has been demonstrated to serve as a receptor for LPS (84). Despite the expression of Toll-like receptors (TLR) on intestinal epithelial cells they remain largely unresponsive to external LPS (85,86). Most recently it was demonstrated that the continuous exposure of intestinal epithelial cells to LPS leads to a tolerance. Interestingly the LPS was internalized by the epithelial cell and co-localized with TLR4 receptor at the Golgi apparatus (87), suggesting a possible immuno-regulatory function of TLR4. There is mounting evidence to suggest that mammalian cells also have an intracellular receptor that detects LPS in the cytoplasm of bacterial-infected cells (86) and activates nuclear factor-κB (NF-κB). Indeed, recent studies suggest that this response is mediated by a cytosolic, plant disease resistance-like protein termed CARD4/NOD1 (88). In particular, CARD4/NOD1 has been found to mediate NF-κB and JNK activation by invasive S. flexneri. Also, the extracellular regulated kinase (ERK) is used by S. flexneri to elicit neutrophil transmigration across a model intestinal epithelium. This activation is linked to the length of the polysaccharide part of the bacterial LPS (34), which further is involved in cell-to-cell spread of the bacterium (89,90). Shigella does not possess flagella and therefore has developed a fascinating mechanism to enable itself to spread horizontally within the epithelium. The bacterial protein IcsA assembles host cell actin filaments into a tail that forms at one pole of the bacterium (91) and provides the force that propels the bacterium forward through the cytoplasm (92). Since icsA mutants administered intragastrically to macaque monkeys cause only mild clinical symptoms of dysentery (93,94) it can be concluded that the ability of Shigella to disseminate intra- and intercellularly is crucial for the clinical inflammatory manifestations of the disease. It was recently shown (95) that the icsA mutant is invasive in an epithelial model system, but did not cause transepithelial neutrophil transmigration. In a manner similar to Salmonella infection of epithelial cells, infection with Shigella results in the secretion of IL-8 which results in recruitment of PMNs to the basolateral surface of the epithelium (26). The epithelial secretion of PEEC, resulting in transepithelial PMN migration, is likely but has not been fully established. The fact that the inflammation is not necessarily fighting the pathogen but in contrast can further promote spread of infection has been well documented. In the case of Shigella (96) transepithelial PMN migration promotes the access of the bacterium to the basolateral cell surface, from where it exclusively invades the epithelium. In the case of Shiga toxin in other bacteria like enterohemorrhagic E. coli (EHEC) the inflammation enhances the toxins penetration to the subepithelial space (97) where in vivo it now has access to the intravascular space and is capable of causing hemolytic uremic syndrome (HUS). Anti-inflammatory mechanisms Nonpathogenic bacteria may have the ability to directly influence the inflammatory response elicited by pathogens by downregulating specific signaling pathways. One goal would obviously be to use non-pathogenic microorganisms to treat inflammatory bowel disease. A non-pathogenic strain of E. coli has been found to exhibit efficacy similar to that of mesalazine in patients with ulcerative colitis (98). Results using a probiotic cocktail in pouchitis have been impressive (99). Here, however, we would like to highlight the few known cellular mechanisms which might be responsible for some of the effects seen in vivo. Most recently, it has been shown for the first time that nonpathogenic bacteria may directly influence the intestinal epithelial cells to limit immune activation by inhibiting the ubiquination and degradation of IκB (100,101). As a result, the nuclear penetration of NF-κB becomes arrested, leading to a significant reduction in the amount of IL-8 that is secreted from the intestinal epithelial cells (Fig. 5). Cadaverine, a polyamine, product of an pathway encoded by an antivirulence gene, found in most nonpathogenic E. coli but not present in the pathogen Shigella flexneri, expresses anti-inflammatory activity in vivo and in vitro some of which might be due to the inhibition of Shigella enterotoxin (95,102). Furthermore the metabolic product of cadaverine piperidine, presumably produced by the intestinal microflora, was demonstrated to inhibit the activation of PKC in response to infection with Salmonella typhimurium thereby preventing the migration of neutrophils across a model intestinal epithelium (50). The lipoteichoic acid from the probiotic strains Lactobacillus johnsonii and acidophilus have been shown to downregulate the inflammatory response in epithelial cells to lipopolysaccharide and enteric bacteria. In particular, the expression of the key proinflammatory cytokine interleukin-8 was inhibited by the probiotic products (103). Many other effects of nonpathogenic bacteria altering inflammation in the intestine wait to be explored.FIG. 5.: Nonpathogenic bacteria interfere with NF-κB activation in intestinal epithelial cells. The transcription factor NF-κB is activated by binding of bacterial products to the epithelial cell. Commensal bacteria have devised schemes to interfere with NF-κB activation and hence expression of genes involved in the inflammatory response. Nonpathogenic bacterial (demonstrated for nonpathogenic Salmonell a, reference 100) block ubiquination and degradation of IκB, an inhibitor that binds to and sequesters NF-κB in the cytoplasm. When IκB is degraded, NF-κB is released and moves to the nucleus where it switches on target genes involved in intestinal inflammation. Adapted from (101).SUMMARY Bacterial pathogens have developed mechanisms to manipulate the intestinal epithelial cell for their own purposes (Table 2). They frequently inject proteins into the host cells membrane or cytoplasm stimulating the cytoskeleton to initiate their phagocytosis. They can further alter the epithelial barrier (104) and elicit an inflammatory response by stimulating the epithelial release of proinflammatory factors. Only a few mechanisms by which commensal microflora counteracts these steps and vice versa, as shown most recently by Fagarsan et al. (105), are known, but the area should provide a most interesting future research effort. Understanding bacterial-epithelial interaction in the intestine might ultimately provide new strategies not only for the treatment of infections but also inflammatory diseases.TABLE 2: Selected protective mechanisms of commensal microflora and probiotics at the intestinal epithelial level
Referência(s)