Gastroenterology
Volume 138, Issue 1 , Pages 39-43, January 2010

The Ying and Yang of Bacterial Signaling in Necrotizing Enterocolitis

  • María T. Abreu

      Affiliations

    • Corresponding Author InformationReprint requests Address requests for reprints to: María T. Abreu, MD, Division of Gastroenterology, University of Miami Miller School of Medicine P.O. Box 016960 (D-49) Miami, FL 33101

Division of Gastroenterology, University of Miami Miller School of Medicine, Miami, Florida

published online 23 November 2009.

Article Outline

 

See “Toll-like receptor-4 inhibits enterocyte proliferation via impaired β-catenin signaling in necrotizing enterocolitis,” by Sodhi CP, Shi X–H, Richardson WM, et al, on page 185.

Necrotizing enterocolitis (NEC) is a terrible complication of prematurity. The symptoms can include lethargy, abdominal distention, and bloody stools.1 The condition can be difficult to diagnose and may require exploratory laparotomy to make the diagnosis. The consistent histologic findings of NEC include ischemic necrosis, inflammation, and bacterial overgrowth.2 Treatment includes antibiotics and supportive care, but operative resection of the affected bowel is required when a perforation is present. Because of the epidemic of multiple births secondary to in vitro fertilization, prematurity and its consequences continue to plague neonatal intensive care units. In very low birth weight babies (<1500 g), the risk of proven NEC is approximately 10%.3 In addition to high short-term mortality in infants with NEC, surviving children are delayed in their neurologic development and may be left with short-bowel syndrome.1 Luminal bacteria have been thought to play a causal role in NEC and premature infants fed with formula harbor different intestinal bacteria than full-term, breast-fed infants.4 Because of this, studies have examined using prebiotics or probiotics to prevent NEC.5 Several studies suggest a benefit with probiotic administration.6 Nevertheless, research is needed to prevent NEC and pediatric intensivists worry that probiotics may result in systemic bacteremia.

To study NEC, investigators have turned to animal models to mimic some of the essential disease features. Formula-fed infants seem to carry a higher risk than breast-fed infants and so animal models of NEC generally involve formula feeding of neonatal mice. In addition to formula feeding, neonatal mice are challenged with cold and asphyxia, and sometimes inoculated with a high dose of bacteria orally. Since the identification of Toll-like receptors (TLRs), several investigators have addressed whether TLR recognition of bacteria could play a role in NEC development.

In this issue of Gastroenterology, Dr Hackam and his group7 have explored the mechanisms that lead from innate immune recognition of luminal bacteria to epithelial dysfunction. In particular, the authors explore the role TLR4 plays in NEC. TLR4 is responsible for the recognition of lipopolysaccharide (LPS). Activation of TLR4 by its ligand LPS leads to nuclear factor (NF)-κB activation and proinflammatory cytokine release in many cell types.8 These authors and others have demonstrated that animals with mutant or deficient TLR4 signaling are protected against NEC.9, 10 These results make intuitive sense; germ-free rats or piglets are protected against NEC compared with conventionalized animals.11, 12 To appreciate the results of the current study, it is necessary to review briefly certain aspects of TLR4 regulation in the normal intestine.

A study by Hornef et al has demonstrated that TLR4 signaling is down-regulated after colonization of the intestine with the bacterial flora.13 Specifically, they showed that in isolated intestinal epithelial cells (IECs) from the small intestine, TLR4 and its co-receptor MD-2 are expressed but they are only functional, that is, they are able to activate NF-κB in response to LPS, in fetal IECs in utero. If the newborns are delivered vaginally, the intestinal epithelium down-regulates expression of interleukin-1 receptor-associated kinase 1, a necessary step in TLR4 signaling, and the cells become LPS unresponsive. If, however, the pups are delivered sterilely by Cesarean section, IECs continue to respond to LPS. These data demonstrate that after delivery and colonization of the intestine with bacteria, TLR4 signaling is normally dampened. Presumably, the down-regulation of TLR4 is meant to prevent chronic inflammation in response to commensal bacteria, but this statement is speculative.

In contrast with the down-regulation of TLR4 described, studies of TLRs in NEC both in newborn mice and humans indicate that NEC is associated with increased expression of TLR4 in the intestinal mucosa compared with healthy infants.9, 10, 14, 15 The Hackam laboratory has shown that TLR4–mutant C3H/HeJ mice are protected from NEC in a model in which 10- to 14-day-old mice are fed formula multiple times a day and undergo hypoxia twice a day for 4 days.10 In previous studies, they found that LPS and hypoxia increased TLR4 expression on IECs. TLR4 activation resulted in decreased IEC proliferation. Likewise, other studies have demonstrated that in mother-fed mice, TLR4 mRNA expression decreases whereas TLR4 expression increases in formula-fed animals exposed to cold asphyxia stress.9 These results suggest that TLR4 signaling is harmful in NEC. Formula feeding and other stressors cause an increase in TLR4 expression. At least part of the way in which TLR4 signaling makes mice susceptible to NEC is through inhibition of epithelial proliferation and restitution.

In the current paper, Sodhi et al7 delve into the mechanism for the decrease in proliferation of IECs in NEC. The authors test the hypothesis that TLR4 alters β-catenin signaling. β-Catenin is intimately linked with IEC proliferation.16 In the proliferative state, Wnt binds to the Frizzled receptor, thereby leading to inhibition of glycogen-synthase kinase 3β (GSK3β) through its phosphorylation.17 The absence of functional GSK3β permits nuclear translocation of β-catenin and activation of genes that promote proliferation (Figure 1). In the absence of Wnt, GSK3β phosphorylates β-catenin leading to its ubiquitination and destruction, thereby preventing its availability in the nucleus. The authors show through detailed biochemical studies and transfection studies that TLR4 activation leads to a decrease in nuclear β-catenin. Transfecting constitutively active β-catenin bypasses the effect of LPS and restores proliferation in LPS-treated cells in culture (IEC-6 cells). They provide data to support a model in which the effect of LPS, is mediated by Akt. We are shown that TLR4 signaling blocks Akt phosphorylation that allows GSK3β to remain active (unphosphorylated) and to phosphorylate β-catenin leading to its destruction (Figure 1, left panel). Importantly, inhibition of GSK3β with lithium chloride circumvented the negative effect of LPS on β-catenin and restored proliferation of IEC-6 cells.

  • View full-size image.
  • Figure 1. 

    Intestinal epithelial cells in the proliferative state integrate a variety of growth signals. Wnts are proteins that interact with the receptor frizzled (Fzd) and its co-receptor LRP. Once activated, disheveled (Dsh) phosphorylates glycogen synthase kinase 3 β leading to its decreased function. By decreasing its function, less β-catenin is phosphorylated. When β-catenin is phosphorylated, it is targeted for ubiquitination and proteosomal degradation. In the presence of Wnt, less β-catenin is degraded and is then available to enter the nucleus where it binds the Tcf transcription factor leading to expression of genes involved in proliferation including cyclin D1. On the left is a model of what occurs in the NEC-susceptible intestine. Whereas normal infant intestinal epithelial cells express low levels of TLR4 (or have decreased downstream signaling), prematurity, formula feeding, and hypoxia result in increased TLR4 expression through unknown mechanisms. TLR4 inhibits Akt phosphorylation. Akt phosphorylates and inactivates GSK3β. By inhibiting Akt phosphorylation, GSK3β remains in its active state and phosphorylates β-catenin leading to its degradation. Hackam and colleagues show that expression of active β-catenin can bypass the inhibition of proliferation caused by LPS in the NEC intestine.

To prove in vivo relevance, Hackam et al studied formula-fed versus mother-fed mice and found an increase in GSK3β protein and a decrease in β-catenin in the ileum of formula-fed mice. Similar findings were observed in surgical specimens from NEC babies. Not many studies have been able to look at fresh tissue to corroborate the findings from animal models and in vitro studies.

Although the Hackam study demonstrates a novel pathway from TLR4 signaling to altered proliferation, it is still not perfectly clear that the only way by which TLR4 sensitizes the gut to development of NEC is through decreased intestinal epithelial proliferation. Other events such as proinflammatory cytokine expression, reactive nitrogen expression, and impaired cell migration probably also conspire in promoting intestinal damage.9, 10 In the Hackam study, adenoviral delivery of dominant-negative TLR4 restores proliferation, but no comment is made about whether this alters the severity of the NEC. We also do not know how decreased epithelial proliferation culminates in the most dramatic feature of human NEC—ischemic necrosis of the bowel.2

Studies have addressed the role of TLR4 in inflammatory bowel disease. We and others have shown that the absence of TLR4 leads to more severe epithelial injury after damage with dextran sodium sulfate.18, 19 In the absence of TLR4, epithelial proliferation is decreased. Mechanistically, TLR4 signaling in IECs induces expression of cyclo-oxygenase (COX)-2, prostaglandin E2, and endothelial growth factor receptor ligands, which lead to increased proliferation.20 At face value, these results are the opposite of Dr Hackam's in that, in inflammatory bowel disease models, TLR4 seems to promote epithelial proliferation. But anyone who works with animal models knows that the devil is in the details. The small intestine and colon may behave very differently with respect to TLR regulation. In addition, all of the colitis models are performed in adult mice (typically >8 weeks of age). Indeed, the Hackam study looked at the effect of systemic LPS on IEC proliferation in adult mouse ileum and colon and did not see a decrease in proliferation in these cells, supporting the notion that there is both a developmental and site-specific effect of TLR4 signaling in the lower gastrointestinal tract. The small bowel normally has very little bacterial content compared with the colon. It is still not clear what is the cause and the effect of such dramatic differences in bacterial density between the small bowel and colon. Are there factors in the small intestine that make it inhospitable for bacteria, such as Paneth cells and secretion of antimicrobial peptides, TLR-mediated expression of defensins by IECs, and small bowel motility sweeping away bacteria? And is the colon a more hospitable place for bacteria?—secretion of mucins that allow a biolayer of bacteria to thrive, decreased motility, and no Paneth cells.

Another of the issues that is not yet clear is whether TLR4 is unique among TLRs in its role in IEC proliferation or whether the phenomenon occurs with other TLRs as well. When conventionally housed mice are compared with germ-free mice, expression of TLR2, TLR3, TLR4, and TLR5 is increased, suggesting that colonization may induce expression of TLRs.21 Results with TLR9 expression are mixed; one study showed an increase in expression and another a decrease in expression with colonization.21, 22 One model that could be proposed based on a composite of studies is that TLR4 is expressed constitutively until activated by LPS in the newborn intestine and then its intracellular signaling cascade is turned off, for example, decreased expression of IRAK1. It stands to reason that colonization with bacteria will alter expression of TLRs. The question is whether TLR signaling is enhanced by bacterial colonization or silenced. Another level of complexity is how prematurity and artificial nutrients (formula) may alter regulation of TLRs. At least for TLR4, the weight of the evidence is that TLR4 signaling is normally decreased in the intestinal epithelium after birth. The work of the Hackam group shows us that when TLR4 signaling is not down-regulated, the newborn intestine develops inflammation.

Like all good studies, more questions emerge. Does macrophage expression of TLR4 contribute to the inflammatory process as well? Data from Stappenbeck have revealed that MyD88/COX-2-expressing stromal cells provide a stimulus to intestinal crypt stem cells to proliferate.23 Could other cell types therefore also be playing a role in the decreased proliferation observed with NEC? The results of the current study not only offer a mechanistic understanding of how TLR4 leads to decreased proliferation through the Wnt pathway, but also substantiate further using TLR4 antagonists to prevent NEC.15 Localized enteric approaches to blocking TLR4 signaling may be very beneficial in premature infants and may decrease the risk for systemic immunosuppression from TLR4 antagonists.

Back to Article Outline

References 

  1. Henry MC, Moss RL. Neonatal necrotizing enterocolitis. Semin Pediatr Surg. 2008;17:98–109
  2. Ballance WA, Dahms BB, Shenker N, et al. Pathology of neonatal necrotizing enterocolitis: a ten-year experience. J Pediatr. 1990;117:S6–S13
  3. Uauy RD, Fanaroff AA, Korones SB, et al. Necrotizing enterocolitis in very low birth weight infants: biodemographic and clinical correlates (National Institute of Child Health and Human Development Neonatal Research Network). J Pediatr. 1991;119:630–638
  4. Dai D, Walker WA. Protective nutrients and bacterial colonization in the immature human gut. Adv Pediatr. 1999;46:353–382
  5. Caplan MS. Probiotic and prebiotic supplementation for the prevention of neonatal necrotizing enterocolitis. J Perinatol. 2009;29(Suppl 2):S2–S6
  6. Bin-Nun A, Bromiker R, Wilschanski M, et al. Oral probiotics prevent necrotizing enterocolitis in very low birth weight neonates. J Pediatr. 2005;147:192–196
  7. Sodhi CP, She XH, Richardson WM, et al. Toll-like-receptor-4 inhibits enterocyte proliferation via impaired β-catenin signaling in necrotizing enterocolitis. Gastroenterology. 2009;138:185–196
  8. Medzhitov R, Preston-Hurlburt P, Janeway CA. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature. 1997;388:394–397
  9. Jilling T, Simon D, Lu J, et al. The roles of bacteria and TLR4 in rat and murine models of necrotizing enterocolitis. J Immunol. 2006;177:3273–3282
  10. Leaphart CL, Cavallo J, Gribar SC, et al. A critical role for TLR4 in the pathogenesis of necrotizing enterocolitis by modulating intestinal injury and repair. J Immunol. 2007;179:4808–4820
  11. Musemeche CA, Kosloske AM, Bartow SA, et al. Comparative effects of ischemia, bacteria, and substrate on the pathogenesis of intestinal necrosis. J Pediatr Surg. 1986;21:536–538
  12. Sangild PT, Siggers RH, Schmidt M, et al. Diet- and colonization-dependent intestinal dysfunction predisposes to necrotizing enterocolitis in preterm pigs. Gastroenterology. 2006;130:1776–1792
  13. Lotz M, Gutle D, Walther S, et al. Postnatal acquisition of endotoxin tolerance in intestinal epithelial cells. J Exp Med. 2006;203:973–984
  14. Gribar SC, Sodhi CP, Richardson WM, et al. Reciprocal expression and signaling of TLR4 and TLR9 in the pathogenesis and treatment of necrotizing enterocolitis. J Immunol. 2009;182:636–646
  15. Chan KL, Wong KF, Luk JM. Role of LPS/CD14/TLR4-mediated inflammation in necrotizing enterocolitis: pathogenesis and therapeutic implications. World J Gastroenterol. 2009;15:4745–4752
  16. Grigoryan T, Wend P, Klaus A, et al. Deciphering the function of canonical Wnt signals in development and disease: conditional loss- and gain-of-function mutations of beta-catenin in mice. Genes Dev. 2008;22:2308–2341
  17. Clevers H. Wnt/beta-catenin signaling in development and disease. Cell. 2006;127:469–480
  18. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recognition of commensal microflora by Toll-like receptors is required for intestinal homeostasis. Cell. 2004;118:229–241
  19. Fukata M, Michelsen KS, Eri R, et al. Toll-like receptor-4 is required for intestinal response to epithelial injury and limiting bacterial translocation in a murine model of acute colitis. Am J Physiol Gastrointest Liver Physiol. 2005;288:G1055–G1065
  20. Fukata M, Chen A, Klepper A, et al. Cox-2 Is Regulated by Toll-Like Receptor-4 (TLR4) Signaling: Role in Proliferation and Apoptosis in the Intestine. Gastroenterology. 2006;131:862–877
  21. Lundin A, Bok CM, Aronsson L, et al. Gut flora, Toll-like receptors and nuclear receptors: a tripartite communication that tunes innate immunity in large intestine. Cell Microbiol. 2008;10:1093–1103
  22. Ewaschuk JB, Backer JL, Churchill TA, et al. Surface expression of Toll-like receptor 9 is upregulated on intestinal epithelial cells in response to pathogenic bacterial DNA. Infect Immun. 2007;75:2572–2579
  23. Brown SL, Riehl TE, Walker MR, et al. Myd88-dependent positioning of Ptgs2-expressing stromal cells maintains colonic epithelial proliferation during injury. J Clin Invest. 2007;117:258–269

 Conflicts of interest The author discloses no conflicts.

PII: S0016-5085(09)02048-4

doi:10.1053/j.gastro.2009.11.031

Refers to article:

  • Additional Online Content AvailableEditorial Accompanies ArticleVideo Abstract Toll-Like Receptor-4 Inhibits Enterocyte Proliferation via Impaired β-Catenin Signaling in Necrotizing Enterocolitis , 27 September 2009

    Chhinder P. Sodhi, Xia–Hua Shi, Ward M. Richardson, Zachary S. Grant, Richard A. Shapiro, Thomas Prindle, Maria Branca, Anthony Russo, Steven C. Gribar, Congrong Ma, David J. Hackam
    Gastroenterology January 2010 (Vol. 138, Issue 1, Pages 185-196)

Gastroenterology
Volume 138, Issue 1 , Pages 39-43, January 2010

Article Tools

* Image Usage & Resolution