A Feed-Forward Loop Involving Hyaluronic Acid and Toll-Like Receptor-4 as a Treatment for Colitis?
Article Outline
See “Regulation of colonic epithelial repair in mice by Toll-like receptors and hyaluronic acid,” by Zheng L, Riehl TE, and Stenson WF, on page 2041.
Toll-like receptors (TLRs) continue to fascinate as possible triggers of inflammatory diseases. Ten TLRs are expressed in humans and their main function is to sense microbial products to mount an appropriate host defense response.1 TLR-4, the receptor for lipopolysaccharide (LPS) from gram-negative bacteria, has been a particular focus for many researchers and a role for TLR-4 in the pathogenesis of diseases such as septic shock, atherosclerosis, asthma/allergy, and arthritis have been reviewed.2 In the gut, however, things seem to be somewhat different. Mice deficient in TLR-4, or its signaling adapter MyD88, have been shown to be more susceptible to colitis, particularly in the well-known dextran sodium sulfate (DSS) model.3, 4 The basis for this is not known, but the assumption has been that TLR-4 senses LPS derived from commensal bacteria, which in the context of the gut leads to protection from injury. Induction of cyclo-oxygenase (COX)-2 and production of prostaglandin E2 (PGE2) have been suggested as a possible mechanism, because PGE2 protects the epithelium from damage.5 Studies in the current issue of Gastroenterology by Zheng et al now provide new insights.6 They find that DSS is sensed by macrophages in the gut—possibly via TLR-4—and induces production of hyaluronic acid (HA) fragments. These in turn, again acting via TLR-4, induce COX-2 and protect the gut from the inflammatory effects of DSS. The HA fragments can be administered therapeutically in the model, providing us with the intriguing possibility that modulation of TLR-4 by HA fragments might have potential as a novel way to treat colitis.
HA is a complex glycosaminoglycan that is assembled by HA synthases (HAS). The high molecular weight form is a well-characterized component of the extracellular matrix, where its role is in tissue hydration, acting as a key “shock absorber” in joints.7 However, fragments of HA are generated during tissue injury and inflammation and these fragments have been shown to be proinflammatory.8 This effect is thought to occur via HA binding to TLR-2 and/or TLR-4. Higher molecular weight fragments of HA have been shown to be anti-inflammatory in lung inflammation and this effect is thought to be via immobilization of TLR-4, which somehow prevents signaling.9 Because TLR-4 has been shown to be protective in models of colitis, Zheng et al sought to determine whether HA might be protective in DSS colitis. Such an effect could be via HA inducing COX-2 in macrophages in a TLR-4–dependent manner in the colon.
First, Zheng et al demonstrated that DSS can induce an increase in HA production in the lamina propria of mice in an MyD88-independent manner. This effect was quite dramatic, with a progressive increase in HA occurring extracellularly between the crypts. There was also an increase in plasma HA. The effect on HA was probably via the induction of HAS-2 and HAS-3 in macrophages; there was a 12-fold increase in mRNA for these isoforms in the distal colon, which was again MyD88-dependent. DSS also induced HAS-2 and HAS-3 in peritoneal macrophages in vitro in an MyD88-dependent manner. These isoforms of HAS have been shown to synthesize HA fragments in inflammation. The role of MyD88 suggested TLR involvement and it was shown that DSS is endocytosed by peritoneal macrophages in an MyD88-dependent manner in vitro. The authors inferred that the effect of DSS was occurring via TLR-4 and it was shown that DSS did not increase COX-2 in TLR-4–deficient macrophages in vitro. This result provides the surprising conclusion that DSS might act directly via TLR-4, although this was not shown conclusively. If this is correct, the DSS model would not solely rely on LPS being released from commensals, which heretofore has been thought to be a mechanism for the proinflammatory effects of DSS. HA itself was next tested and it was shown that fragments with a molecular weight of 7500 daltons were able to induce tumor necrosis factor production in peritoneal macrophages.
Zheng et al then carried out in vivo experiments with HA fragments. Intraperitoneal injection of HA induced an increase in plasma HA, which was sustained in wild-type but not MyD88-deficient mice. Intraperitoneal HA also induced HAS-2 and HAS-3 in the distal colon. HA also induced the expression of tumor necrosis factor, macrophage inflammatory protein-2, and COX-2 in macrophages. This effect was shown to depend on MyD88. HA increased the expression of COX-2 in cells localized to the base of the crypts. These cells were identified as macrophages. Furthermore, HA administered intraperitoneally had an anti-inflammatory effect in the DSS model. This effect was also shown to depend on MyD88 and TLR-4. HA was able to rescue the colitis induced by DSS in wild-type mice, but not in TLR-4–deficient mice. Similarly, HA did not have a protective effect in COX-2–deficient mice. HA administration is therefore able to treat established colitis via a mechanism involving TLR-4 and COX-2.
Taken together, the study provides us with a model (Figure 1). DSS acting via TLR-4 in macrophages induces the production of HA fragments, which feed back on the macrophage to promote the TLR-4–dependent induction of COX-2. This leads to the production of PGE2, which protects the gut from inflammation via a cytoprotective effect on epithelial cells. As the authors also speculate, there may be a further “feed-forward” effect whereby exogenous HA induces endogenous HA synthesis, propagating the anti-inflammatory effect.
Figure 1.
DSS induces HA fragments that, via TLR-4, are anti-inflammatory in colitis. DSS administration is a well-utilized model of colitis. Zheng et al now provide complex insights into the effect of DSS in the gut.6 They demonstrate that DSS can act via TLR-4 in macrophages to induce HAS leading to the production of HA fragments (blue lines). These fragments in turn, also acting via TLR-4, induce COX-2 leading to PGE2 production, which has a cytoprotective effect and is anti-inflammatory in the epithelium (red lines). These fragments can also induce further fragment production in a feed-forward effect, propagating the anti-inflammatory response. Administering HA fragments are anti-inflammatory in the DSS model, possibly indicating a new therapeutic approach for colitis.
A number of questions arise from this study. First, why are HA fragments anti-inflammatory in the gut in this model, and yet in other contexts, including in lung, they are proinflammatory? HA can clearly induce inflammatory mediators in peritoneal macrophages and yet the net effect in vivo here is to be anti-inflammatory. Might there be other mechanisms to explain the effect of administering HA in the DSS model? The observation that the effect is TLR-4–dependent could be due to the severity of the inflammation in the TLR-4–deficient mice, which might be more difficult to block. Second, what are we to make of the data suggesting that DSS might act via TLR-4? Despite this, a proinflammatory outcome dominates, presumably because of other effects of DSS, whereas when HA fragments are acting via TLR-4 in the model, they are supposedly anti-inflammatory. Precisely how DSS might act directly via TLR-4 is not clear, but in the context of this model it might differ from HA fragments. It is possible that the HA fragments are also acting via TLR-2, which might have an anti-inflammatory effect via interleukin-10 production. Another question is whether HA would be anti-inflammatory in other models of colitis. If this were the case, it might suggest a general anti-inflammatory effect for HA in the gut. Whether similar mechanisms exist in humans in relation to TLR-4 having a protective effect in the gut have yet to be clearly demonstrated.
In terms of possible clinical utility, the most exciting data concern the anti-inflammatory properties of HA in colitis. This warrants further investigation in the context of human disease, where HA administration or enhancement of endogenous HA production might have potential as a novel therapeutic approach.
References
- . The IL-1 receptor/Toll-like receptor superfamily: 10 years of progress. Immunol Rev. 2008;226:10–18
- . Therapeutic targeting of Toll-like receptors as a treatment for immune and inflammatory diseases and cancer. Pharm Rev. 2009;61:177–197
- Recognition of commensal microflora by TLRs is required for intestinal homeostasis. Cell. 2004;118:229–241
- . Toll-like receptors and intestinal epithelial repair. Curr Opin Gastroenterol. 2008;24:103–107
- MyD88-dependent positioning of Ptgs2-expressing stromal cells maintains colonic epithelial proliferation during injury. J Clin Invest. 1997;117:58–269
- . Regulation of colonic epithelial repair in mice by Toll-like receptors and hyaluronic acid. Gastroenterology. 2009;137:2041–2051
- Hyaluronan fragments: an information-rich system. Euro J Cell Biol. 2006;85:699–715
- Hyaluronan fragments induce nitric oxide synthase in murine macrophages through a NF-kappaB-dependent mechanism. J Biol Chem. 1997;272:8013–8019
- Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nat Med. 2005;11:1173–1179
Conflicts of interest The author discloses no conflicts.
PII: S0016-5085(09)01837-X
doi:10.1053/j.gastro.2009.10.015
© 2009 AGA Institute. Published by Elsevier Inc. All rights reserved.
Refers to article:
-
Regulation of Colonic Epithelial Repair in Mice by Toll-Like Receptors and Hyaluronic Acid
, 03 September 2009
