The role of eosinophils in inflammatory bowel disease

Eosinophils are proinflammatory leucocytes that constitute a small percentage of circulating blood cells. In the healthy state, most of these cells reside in the gastrointestinal tract within the lamina propria of the stomach and intestine. They differentiate in the bone marrow from progenitor cells under the influence of interleukin (IL)-3, IL-5, and granulocyte-macrophage colony stimulating factor. IL-5 also stimulates their release into the peripheral circulation.¹ They then migrate to the gastrointestinal tract in response to eotaxin, a chemokine that is constitutively expressed throughout the gastrointestinal tract. This chemokine binds to the CCR-3 receptor on eosinophils and is required for their homing to the gastrointestinal tract.^2,3 However, constitutive expression of eotaxin is not sufficient for tissue eosinophil trafficking because some gastrointestinal segments (such as the tongue and oesophagus) express eotaxin but are normally devoid of eosinophils.¹ So this may explain why the help of other cytokines is needed to complete the homing of eosinophils to the intestines. One such cytokine is IL-5, which increases the circulating pool of eosinophils and primes eosinophils to have enhanced responses to eotaxin.¹

Eosinophils secrete toxic inflammatory mediators that are stored in preformed vesicles and also synthesised de novo following cellular activation. The major proteins secreted by eosinophils are eosinophilic cationic protein, major basic protein, eosinophil protein X, eosinophil derived neuroendotoxin, and eosinophil peroxidase. These cause damage to tissues, insert pores into membranes of target cells, and increase smooth muscle reactivity by generating toxic oxygen radicals.⁴

Eosinophils and the role they play in inflammatory diseases of the gastrointestinal tract have become a point of interest in recent literature. Inflammatory bowel disease (IBD) includes two major chronic diarrhoeal illnesses, ulcerative colitis (UC) and Crohn’s disease (CD). The inflammatory process in these illnesses involves many inflammatory cells, such as lymphocytes, macrophages, mast cells, neutrophils, and eosinophils.⁵ The two most important roles that eosinophils play in IBD appear to be as proinflammatory and promotility agents thus producing effects such as diarrhoea, inflammation, tissue destruction, formation of fibrosis and strictures and, as recently suggested, even repair.⁶

IBD probably starts by an unknown antigenic stimulus likely coupled with genetic predisposition that leads to increased production in the intestinal tract of chemoattractants to a variety of inflammatory cells, including eosinophils.⁷ The pathogenesis of IBD may also include an aberrant response of the intestinal mucosa to components of the normal flora through cross reaction with self antigens, and which are not appropriately downregulated.⁸ When recruited to intestinal tissue, eosinophils partly contribute to the inflammatory process through release of various toxic proteins and cytokines that drive the inflammatory process in concert with other inflammatory and immune cells. One of the more important of these immune cells include the T lymphocytes which were also found to express the CCR-3 eotaxin receptor which is what draws these cells to colocalise with eosinophils during the inflammatory reaction.⁹ Eosinophils participating in the inflammatory phase of IBD then remain activated to possibly finally contribute to the repair process, as shown by Lampinen and colleagues⁶ in this issue of Gut (see page 1714).

Eotaxin is a potent chemotactic agent for eosinophils that is inherently expressed in intestinal tissues. This protein has been found to be overexpressed in IBD and more so in active than in inactive stages.^10,11 In addition, levels in quiescent CD were found to be higher than in quiescent UC.¹⁰ The finding of elevated eotaxin was found to be associated with larger numbers of eosinophils in intestinal biopsies of patients with active IBD.¹² Also, the number of degranulated eosinophils was higher in these patients compared with controls. Lampinen et al have shown that the numbers of activated eosinophils were higher in patients with active and inactive UC compared with normal controls.⁶

In addition to its role in the differentiation and release of eosinophils from bone marrow, IL-5 is also a chemotactic agent, although less potent, for eosinophils which possess specific receptors for this cytokine. Eosinophils also secrete IL-5 which in turn stimulates their own proliferation and differentiation which contributes to a further increase in their numbers.^4,13,14 Lampinen and colleagues showed that IL-5 was increased in rectal perfusion fluids in UC patients and contributed to eosinophil recruitment to the intestinal mucosa in these patients. They also showed that the inflammatory effects could be reduced by antibodies to IL-5.¹⁵ IL-5 has also been shown to stimulate smooth muscle hypercontractility in the intestine, which in cases of worm infestations helps propel and expel the parasite.¹⁶ This phenomenon may play a role in increased motility in UC.

Eosinophils are one of the many inflammatory cells involved in the pathogenesis of IBD. The inflammatory response also involves, but is not limited to, neutrophils, mast cells, and T lymphocytes, mainly Th2 in UC and Th1 in CD.¹⁷ Th2 lymphocytes express CCR-3 receptors that are also present on eosinophils and they colocalise with eosinophils in inflamed tissues.³ Th2 type lymphocytes secrete several types of cytokines, some which serve to induce adhesion molecules in the microvasculature that are required for eosinophil diapedesis and for the priming and prolonged survival of eosinophils.¹⁸ Therefore, although increased eotaxin expression in tissues is involved in the early part of inflammatory and allergic responses, the response needs to be maintained by antigen specific Th2 cells that generate IL-4 and IL-5 which serve as growth and stimulation factors for eosinophils.^3,9 Neutrophils also participate in the active inflammatory stage, mostly through liberating reactive oxygen species.¹⁹

Many authors have demonstrated the increase in numbers of mast cells in IBD.^20–22 Also, larger numbers of degranulated mast cells were found in active areas of IBD.²³ Among the many cytokines that are released from mast cells is IL-5. Lorentz et al found that there was increased secretion of IL-5 from mast cells in IBD. Also, in their study, increased IL-5 was associated with increased tissue eosinophilia in IBD.²⁴ Tryptase is another chemotactic agent for eosinophils that is secreted from mast cells.²⁵ It has also been shown to induce proliferation of smooth muscle and fibroblasts in the lung.^26,27

Finally, which inflammatory cells contribute to the repair process in IBD? Lampinen and colleagues⁶ showed in their article that the number of activated eosinophils was actually higher in the quiescent phase of UC than in the active phase.⁶ Thus it seems possible that they in some way contribute to the repair process. Previous studies that have examined the role of eosinophils in the repair process have linked eosinophils to activation of fibroblasts to explain the phenomenon of fibrosis and stricture formation in CD.^28,29 For example, Xu et al examined the roles of mast cells and eosinophils in CD and found that they may affect fibrosis by directly influencing intestinal fibroblast properties.²⁹ Given the difference in fibrogenic response between UC and CD, it would be interesting to explore if there is a variation in the degree or nature of eosinophil activation between UC and CD to explain the variability of fibrosis between the two processes. On the other hand, there may be other unknown factors that account for the difference in response of fibroblasts in UC compared with CD.

Identification of many components in the inflammatory process in IBD has uncovered many potential areas for therapeutic intervention. Treatment options in IBD now include, in addition to the conventional therapies of anti-inflammatory agents, the choice of other agents such as immune system modulators, cytapharesis, and biological therapies.³⁰ Potential biological therapies may involve blocking the effect of eosinophils whether through the recruitment phase or the effector phase, by antagonising the effect of the mediators that they release. However, given the diverse roles that eosinophils are apparently playing in the pathophysiology of IBD (proinflammatory versus repair) it would be necessary to accurately identify the mechanism for each process to be able to reach a balance of blocking the inflammatory effects without interfering with the repair mechanism. Although these venues have been partially explored experimentally, it remains to be seen if they will find practical use in the treatment of IBD.