Alkaline Phosphatase: Keeping the Peace at the Gut Epithelial Surface

Vertebrate intestinal surfaces are in constant contact with a vast consortium of commensal bacteria. To preserve mutually beneficial host-microbial relationships, gut epithelia have evolved strategies to limit the proinflammatory potential of resident gut microbes. In this issue of Cell Host & Microbe, Bates and colleagues report that intestinal alkaline phosphatase, whose expression is induced during establishment of the microbiota, dephosphorylates lipopolysaccharide and promotes mucosal tolerance to commensal bacteria in zebrafish. Vertebrate intestinal surfaces are in constant contact with a vast consortium of commensal bacteria. To preserve mutually beneficial host-microbial relationships, gut epithelia have evolved strategies to limit the proinflammatory potential of resident gut microbes. In this issue of Cell Host & Microbe, Bates and colleagues report that intestinal alkaline phosphatase, whose expression is induced during establishment of the microbiota, dephosphorylates lipopolysaccharide and promotes mucosal tolerance to commensal bacteria in zebrafish. Most vertebrates harbor complex communities of coevolved bacteria in their intestines. In humans, this consortium is comprised of thousands of distinct species and as many as 10 to 100 trillion organisms (Turnbaugh et al., 2007Turnbaugh P.J. Ley R.E. Hamady M. Fraser-Liggett C.M. Knight R. Gordon J.I. Nature. 2007; 449: 804-810Crossref PubMed Scopus (3001) Google Scholar). These bacteria make important contributions to the health of their hosts by increasing the efficiency of digestion and synthesizing essential vitamins. In addition, indigenous microbial communities help to protect the host by occupying niches that might otherwise harbor pathogenic organisms. Despite the benefits conferred by their prokaryotic partners, humans and other vertebrates must cope with several serious problems posed by a close relationship with dense microbial populations. One issue is how to manage the threat of microbial invasion from large bacterial communities. This has been solved in part through the evolution of mechanisms, such as secretion of antimicrobial proteins and immunoglobulins, which compartmentalize microbes on the luminal side of the gut epithelial barrier. However, another problem involves the vast amount of lipopolysaccharide (LPS) that is produced by the luminal microflora. LPS is a major constituent of Gram-negative bacterial cell walls, which is recognized by the innate immune system, eliciting inflammatory responses that are designed to clear bacterial infections. The major receptor for LPS in mammals is Toll-like receptor 4 (TLR4), which triggers signaling cascades through MyD88, an adaptor molecule that functions downstream of multiple TLRs. TLR activation leads to proinflammatory cytokine expression (e.g., tumor necrosis factor), neutrophil recruitment to infected sites, and systemic responses such as fever. At high concentrations, LPS becomes toxic by overstimulating TLR4 signaling, leading to an excessive inflammatory response that results in adverse reactions such as septic shock. Since a large fraction of the intestine's indigenous bacteria are Gram-negative, the load of LPS in the gut is enormous. However, in healthy humans, indicators of intestinal inflammation such as neutrophil presence are minimal (Neish, 2002Neish A.S. Microbes Infect. 2002; 4: 309-317Crossref PubMed Scopus (134) Google Scholar). So, given the presence of 100 trillion bacteria and their associated LPS, why isn't the gut continuously inflamed? In this issue of Cell Host & Microbe, Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar make significant progress toward resolving this paradox by uncovering a mechanism whereby LPS is detoxified at the gut epithelial surface. They have utilized a zebrafish model, which has key advantages as a system for dissecting host-bacterial associations. As in mice, zebrafish have a complex intestinal microbiota (Rawls et al., 2004Rawls J.F. Samuel B.S. Gordon J.I. Proc. Natl. Acad. Sci. USA. 2004; 101: 4596-4601Crossref PubMed Scopus (615) Google Scholar) and can be reared germ-free, allowing manipulation of the intestine's bacterial colonizers for the study of host-bacterial relationships. However, in contrast to mice, zebrafish gene expression is easily manipulated through morpholino antisense technology, making it possible to routinely elucidate the contributions of specific host genes to maintaining host-microbial associations. Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar have further demonstrated that zebrafish, like mammals, respond to LPS through a MyD88-dependent mechanism that induces recruitment of neutrophils into the intestine and elicits toxic responses to high LPS concentrations. To resolve the question of how the vertebrate intestine copes with high LPS loads, Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar focused on alkaline phosphatases (APs), a class of enzymes that are prominent in the gut. Although alkaline phosphatases have been proposed to play a role in the breakdown of dietary lipids, they can also dephosphorylate the lipid moiety of LPS, resulting in reduced LPS toxicity in mammals (Koyama et al., 2002Koyama I. Matsunaga T. Harada T. Hokari S. Komoda T. Clin. Biochem. 2002; 35: 455-461Crossref PubMed Scopus (120) Google Scholar). When Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar colonized germ-free zebrafish with a normal microbiota or exposed them to purified LPS, they induced expression of an intestine-specific isoform of alkaline phosphatase (IAP) that associates specifically with the apical brush border of gut epithelial cells (Bates et al., 2006Bates J.M. Mittge E. Kuhlman J. Baden K.N. Cheesman S.E. Guillemin K. Dev. Biol. 2006; 297: 374-386Crossref PubMed Scopus (374) Google Scholar). Morpholino knockdown of MyD88 revealed that IAP is induced through a MyD88-dependent mechanism, suggesting that its expression is governed by TLRs. What is the functional significance of microflora-induced IAP expression? To answer this question, Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar first demonstrated that dephosphorylated LPS is less toxic to zebrafish. Morpholino knockdown experiments revealed that IAP suppresses LPS mediated toxicity to zebrafish, as would be predicted on the basis of its LPS dephosphorylating activity. Moreover, in the absence of IAP, gut microflora elicited abnormally high levels of neutrophil recruitment to the small intestine, indicating that IAP plays a key role in suppressing proinflammatory responses to commensal microbes. These findings suggest that IAP acts as a microflora-controlled LPS detoxifying mechanism that functions at the epithelial surface. In this way, IAP is part of a negative feedback loop that senses and detoxifies LPS in order to maintain homeostasis (Figure 1). As microflora-driven expression of IAP requires MyD88, this indicates that MyD88-dependent TLR signaling not only activates proinflammatory pathways, but also triggers anti-inflammatory responses that maintain homeostasis with the normal flora. The discovery that IAP modifies the proinflammatory potential of LPS is reminiscent of a mammalian LPS detoxification mechanism that involves the enzyme acyloxyacyl hydrolase (AOAH). AOAH is expressed systemically, and cleaves acyl chains from the lipid A portion of LPS such that it is no longer recognized by TLR4, thereby preventing prolonged systemic inflammatory reactions following Gram-negative infections (Lu et al., 2005Lu M. Zhang M. Takashima A. Weiss J. Apicella M.A. Li X.H. Yuan D. Munford R.S. Nat. Immunol. 2005; 6: 989-994Crossref PubMed Scopus (54) Google Scholar, Shao et al., 2007Shao B. Lu M. Katz S.C. Varley A.W. Hardwick J. Rogers T.E. Ojogun N. Rockey D.C. Dematteo R.P. Munford R.S. J. Biol. Chem. 2007; 282: 13726-13735Crossref PubMed Scopus (64) Google Scholar). The localization of IAP to the apical surface of epithelial cells suggests a mechanism whereby LPS from the luminal microflora is detoxified while LPS from invading pathogens activates appropriate proinflammatory responses. As a result of its apical location, IAP likely modifies luminal LPS specifically at the epithelial surface while leaving intact LPS that is encountered in subepithelial tissues. Thus, in the event of bacterial invasion across the epithelial barrier, the unmodified subepithelial LPS could activate the inflammatory responses required to clear bacterial infection. A crucial remaining question concerns the cellular localization of the TLRs that drive the LPS responses uncovered by this study. One plausible model would place the relevant TLRs on the surface of epithelial cells, where they could be accessed by LPS from the luminal microflora. However, this model must account for the fact that LPS can drive distinct anti- and proinflammatory responses. Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar have shown that under homeostatic conditions, microflora-derived LPS elicits anti-inflammatory expression of IAP. In contrast, in the absence of IAP or in the presence of excess LPS, inflammatory responses, including neutrophil recruitment, are triggered. One way to explain these findings is to propose that intact LPS must exceed a threshold concentration in order to activate inflammatory pathways. In this model, the choice of whether to mount an anti-inflammatory or proinflammatory response would thus be governed both by the affinity of LPS binding to its receptor(s) and the rate at which IAP dephosphorylates LPS. Although it remains to be determined whether IAP performs a similar function at the mammalian intestinal surface, the activity of this enzyme increases during postembryonic establishment of the microbiota in both fish (Bates et al., 2006Bates J.M. Mittge E. Kuhlman J. Baden K.N. Cheesman S.E. Guillemin K. Dev. Biol. 2006; 297: 374-386Crossref PubMed Scopus (374) Google Scholar) and mammals (Henning, 1985Henning S.J. Annu. Rev. Physiol. 1985; 47: 231-245Crossref PubMed Google Scholar). This suggests that IAP may contribute to maintaining mutualistic bacterial-host relationships in both groups. If so, the findings from the Bates et al., 2007Bates J.M. Akerlund J. Mittge E. Guillemin K. Cell Host Microbe. 2007; 2 (this issue): 371-382Abstract Full Text Full Text PDF PubMed Scopus (458) Google Scholar study could lead to new insight into human disorders, such as inflammatory bowel disease, which are characterized by dysregulated inflammatory responses to intestinal microbes. Intestinal Alkaline Phosphatase Detoxifies Lipopolysaccharide and Prevents Inflammation in Zebrafish in Response to the Gut MicrobiotaBates et al.Cell Host & MicrobeDecember 13, 2007In BriefVertebrates harbor abundant lipopolysaccharide (LPS) in their gut microbiota. Alkaline phosphatases can dephosphorylate and detoxify the endotoxin component of LPS. Here, we show that expression of the zebrafish intestinal alkaline phosphatase (Iap), localized to the intestinal lumen brush border, is induced during establishment of the gut microbiota. Iap-deficient zebrafish are hypersensitive to LPS toxicity and exhibit the excessive intestinal neutrophil influx characteristic of wild-type zebrafish exposed to LPS. Full-Text PDF Open Archive