Th2 cells and cytokine networks in allergic inflammation of the lung

The cytokines released from Th2 and Th2-like cells are likely to be central to the pathophysiolgy of asthma and allergy, contributing to aberrant IgE production, eosinophilia and, perhaps, mucosal susceptibility to viral infection. IL-4 has emerged as a central target, not only for B cell IgE production, but also in the commitment of both CD4+ and CD8+ T cells to cells with Th2 effector function capable of secreting IL-5 resultlng in eosinophilic inflammation. In view of the central role of this cytokine and the evidence that glucocorticoids are unable to modify many IL-4 dependent effects, Th2 inhibitors may prove to be novel therapies for the treatment of bronchial asthma.


Introduction
Them is now increasing evidence to suggest that pro-inflammatory cytokines derived from various subsets of T cells play a maior role in the induction of allergic inflammation of the lungs. Additionally, cytokines derived from other cell types such as mast cells and eosinophils may also be of importance in the maintenance/persistence of this disease. This review discusses the interactive cytokine networks that exist in the lungs, which are believed to play an important role in the aetiology of bronchial asthma.
recently, attention has been focused on the role of lymphocytes in the pathophysiology of this disease. In severe acute asthma, an increase in the number of T cells expressing the activation markers HLA-DR, CD25 and VLA-1 has been reported. 2 The realization that CD4 + T cells can be committed to a distinct phenotype that can induce B cells to switch to IgE production and recruit eosinophils to the lungs, mediated via the secretion of cytokines, has led to the hypothesis that lymphocytes play a major role in orchestrating the inflammatory response in the lungs of asthmatic individuals.

Pathology of asthma
Bronchial hyperresponsiveness (BHR) to both specific and nonspecific stimuli is a characteristic feature of bronchial asthma. While the mechanisms underlying this exaggerated responsiveness are still unclear, there is a considerable body of evidence to suggest that mucosal inflammation of the airways is of central importance. Perhaps the most common pathological finding is an increased number of eosinophils and mast cells in the lung mucosa. It is currently believed that damage to the epithelium is mediated by the secretion of eosinophil-derived highly toxic cationic proteins such as major basic protein (MBP).
In addition, ultrastructural abnormalities in the architecture of the lungs of asthmatic individuals have been reported, characterized by subepithelial deposition of collagen types III and V and fibronectin, the amount of which is correlated with the number of myofibroblasts. More CD4+ T cell subsets--the Th2 hypothesis Mature T cells in the periphery can be divided into either CD4 + and CD8 + populations. This subdivision is associated with fundamental differences in their function, in that CD4 + T cells are activated by soluble foreign proteins such as allergens, presented by MHC class II molecules on the cell surface of antigen presenting cells (B cells, macrophages and dendritic cells). In contrast, CD8 + T cells are, in general, activated by intracellular pathogens such as viruses, are MHC class I restricted and are cytotoxic. Murine CD4 + T cells can be further subdivided into two distinct subsets, termed T helper 1 (Thl) and T helper 2 (Th2), on the basis of their restricted cytokine profile and ability to mediate different immune functions. Activated antigennaive resting CD4 + T cell cells secrete mainly IL-2. However, these cells can be primed under the influence of different cytokines (and perhaps dif-(C) 1995 Rapid Science Publishers t J. Coyle and S. Tsuyuki ferent co-stimulatory signals) to either a Thl or that B7-knockout mice had virtually no immune Th2 subset. Thl cells produce TNF-]3 and IFN-7 defects led to the speculation that additional and are involved in delayed hypersensitivity CD28 ligands existed.12This indeed proved to be responses. Th2 cells produce IL-4, IL-5 and IL-10 the case and a second co-stimulatory molecule and provide help to B cells and as such, are termed B7-2 (CD86) has more recently been believed to play a central role in the aetiology of identified, a B7-2 is expressed at extremely low allergic disease. levels on resting B cells, but is rapidly upregulated after cross linking of the B cell antigen receptor or following stimulation with LPS. a4 On Th 11 in hm human B cells, B7-2 expression peaks within 24 h of activation, whereas B7-1 expression peaks While the concept of Th2 subtype of CD4 + T several days later, a5 In addition, a second CD4 + cells was originally defined in murine cell clones, T cell co-receptor, homologous to CD28 has there is now evidence to suggest that a similar been identified, termed CTLA-4 and demoncytokine profile exists in individuals with allergic strated to be a second ligand for B7-1 and B7disease. Studies of T cell clones obtained from 2. a6 The different functional importance of CD28 the blood of patients sensitive to house dust and CTLA-4 is still unclear. However, in contrast mite allergen demonstrated that the majority of to CD28, CTLA-4 expression is increased only these clones produced IL-4 and IL-5 and low after activation, aa and as such may function to levels of IFN-7 and IL-2 after antigen stimulation. 4 downregulate the T cell response, a7 Likewise, lymphocytes obtained by bronchoalveo-While the importance of these co-stimulatory lar lavage after allergen provocation of allergic signals for the production of IL-2 is well estabindividuals have increased expression of mRNA lished, the role of these co-stimulatory molecules for IL-3, IL-4 and IL-5. 5 IL-5 mRNA has also been in the production of Th2 pattern of cytokines is identified by in situ hybridization of tissues less clear. However, it has recently been reported obtained by bronchial biopsies of asthmatic sub-that stimulation through CD28 induces a Th2 jects and localized to beneath the epithelial baseresponse in vitro independent of IL-4 producment membrane. 6'7 However, it is important to tion. 18 CD28 mediated stimulation of cloned realize that the Thl/Th2 concept is based on human Th2 clones has also been reported to extremely polarized panels of cytokines which induce responsiveness to IL-4 via the autocrine can be generated in vitro by murine CD4+ T production of IL-lz. 9 In vivo, administration of cells and it is likely that more subtle intermediate CTLA-4Ig can inhibit IL-4 gene expression and phenotypes exist in human disease.
IgE production following infection with Heligmosomoides polygyrus. 2 Further evidence that costimulation modifies the outcome of an immune Antigen specific activation of CD4+ T response is provided by the observation that cells treatment with CTLA-4 Ig prevents IL-4 production following immunization with goat-anti-IgD. 2 Activation of CD4 + T cells require interaction In contrast, mice transgenic for soluble murine between CD3 associated TCR z/[ complex and CTLA-4IG exhibit normal T cell priming and specific antigen presented as peptides by class II cytokine production after immunization with a T bearing cells. However, for complete cellular acti-cell dependent antigen. 22 The reasons for these vation, a second signal is also required. Morediscrepancies are unclear, but may be related to over, antigenic stimulation through the TCR in insufficient levels of CTLA-4Ig in vivo to block the absence of co-stimulation leads to a state of interactions between APC and T cells. unresponsiveness or clonal anergy. The most extensively studied co-stimulatory molecule is CD28, which is constituatively expressed on the Cytokines and allergic inflammation surface of naive CD4 + T cells, the engagement of which is required for IL-2 production and T Regulation of IgE production: The principal cell proliferation. 8 The importance of this feature that distinguishes atopic from non-atopic pathway is illustrated in mice lacking the gene for individuals is their ability to develop IgE anti-CD28 which exhibit impaired lymphokine secre-bodies to foreign proteins. During a CD4-MHC tion. 9 The first ligand for CD28, initially termed class II dependent cognate interaction of T and B B7 (now classified as CD80), was identified on cells, contact-mediated costimulatory signals prothe surface of the B cell and subsequently shown vided by CD4 + T cells such as CD40L and to provide co-stimulatory signals to antigen acti-membrane TNF-0t, initiate B cell activation. vated T cells, m'l However, the demonstration During this process, IL-4 has been demonstrated Th2 cells and cytokines in allergic inflammation in murine 23'24 and human 25'26 systems to instruct recruitment and transgenic mice overexpressing B cells to switch to IgE production. Investigation the IL-5 gene develop peripheral blood, bone of T cell cytokine production indicates that marrow and tissue eosinophilia. 4 Furthermore, enhanced in vitro IL-4 production is correlated administration of anti-IL-5 has been demonstrated with enhanced in vivo IgE production. Whereas to inhibit eosinophilia induced by. nematodes5 T cell membrane-dependent activation signals are or antigen in sensitized animals. The precise regarded as general competence signals to initi-source of IL-5 required for the induction of eosiate B cell activation and induce cytokine responnophil infiltration is at present unclear, although siveness, cytokines such as IL-4 provide a it is likely that it is produced from CD4 + T cells.
progressive signal inducing proliferation and IL-5 (in addition to ILo3 and GM-CSF) is also switching of B cells to IgE. Inhibition of IL-4 in required for the survival of eosinophils in vivo, either by administration of anti-IL-4 anti-vitro. 37'38 However, in contrast to the requirebodies 24 or by deletion of the IL-4 gene, iv has ment for CD4 + T cell derived IL-5 for the revealed the essential requirement of IL-4 for the recruitment of eosinophils to the lung, 39 the switch to IgE production. However, a recent source of lL-5 required for the survival of eosinostudy has demonstrated that IL-4 independent IgE phils in the lungs is from a non-CD4 + T cell production can occur, although at a greatly source, possibly involving autocrine production 4o reduced level in IL-4 gene targeted mice. 28 from the eosinophil itself. IL-5 however, may However, to what degree such production might not be the only factor involved in recruiting eosibe mediated by already switched B cells remains nophils to the mucosa and/or promoting their to be determined. The situation, however, is survival and activation, as the observation of a more complex in humans as IL-13, the message complete inhibition of eosinophil infiltration into of which has been found in murine and human the airways after antigen challenge in mice CD4 + cell clones of Th0 and Th2 phenotype, as lacking the IL-5 gene (M. Kopf, personal commuwell as CD8 T cell clones, 29 can also induce nication) may also reflect the requirement for immunoglobulin isotype switching to IgE promaturation of eosinophils in the bone marrow.
ductionff The precise importance of IL-4 vs. IL-In this regard, other cytokines including the che-13 as the principal cytokine involved in B cell mokines RANTES and the recently identified isotope switch in allergic disease is at present eotaxin, may also be important in recruiting eosiunclear. However, recent studies have demonnophils to the lung. strated that naive CD4+ T cells (CD45RO--) RANTES, along with IL-8, MCP-1, MCAF and primed through the TCR develop into effector eotaxin belong to the superfamily of structurally cells that secrete IL-5 and IFN-2 and help B cell related low molecular weight cytokines which IgE production via IL-13 and not IL-4. 31 In concontain four cysteines at identical relative positrast, activation of CD4+ memory cells tions with a conserved Cys-Cys-(C-C) motif.
(CD45RO + ) also produce IL-4 and IL-5 and help RANTES was originally identified as an apparently cell IgE through a combination of IL-13 and IL-T cell-specific inducible gene that was expressed by cultured T cell lines. 4-More recently, RANTES IL-4 also plays a central role in the induction has been shown to be produced from activated of the Th2 phenotype as shown by in vitro and platelets. 42 RANTES is eosinophil chemotactic for in vivo experiments using IL-4 gene detected both eosinophils and for 'memory' CD4+ T cells mice. The importance of IL-4 in the cytokine netin vitro and as such may contribute to the eosiworks in the lung are discussed in more detail in nophil recruitment in allergic disease. 4 the next section.
Eotaxin, was originally identified as a factor produced in the BAL of antigen-challenged Development of eosinophilic inflammation: Over guinea-pigs, which, upon subsequent injection, 20 years ago, the development of eosinophilia in induced a selective accumulation of eosinophils nematode infected rodents was demonstrated to in the skin 44 and lungs. 45 Eotaxin has a 53% be lymphocyte dependent. 2 Subsequently, it was homology with MCP-1, 31% with MIP l a and shown that the soluble factor from T cells was 26% with RANTES. mRNA for eotaxin was shown identical to B cell growth factor 2 (BCF II). This to be up-regulated 3 h after allergen provocafactor has been extensively studied and chartion. 46 The role of eotaxin and other C-C cheacterized and is now termed interleukin-5 (IL-5). mokines in the airways of asthmatic individuals IL-5 has been shown to promote the growth and remains to be determined. differentiation of eosinophils in culture and, in contrast to IL-3 and GM-CSF, acts as a terminal CD8+ Th2 like cells---a mechanism for viral eosinophil differentiation factor. In vivo, adminis- ,, IL-2) after in vitro stimulation. CD8 + T cells suggest that steroids may not be able to inhibit mediate lysis of viral infected cells and inhibition some of the effects and the production of IL-4, of viral replication through the production of and may therefore be intrinsically unable to sup-IFN-. Moreover, it has been suggested that press the central underlying immunological pro-CD8 + T cells down-regulate the CD4 + T cell cesses that drive eosinophilic inflammation in the driven reslonse dependent on the production mucosa.
of IFN-'. 47;'48 However, recent observations from More recently, it has been shown that pred-Erard and colleagues have shown that in the nisolone therapy fails to down-regulate mRNA for presence of IL-4, activated CD8+ cells can IL-4 and IL-5 in steroid resistant asthmatics. 55 switch their function to produce Th2 cyto-While the precise mechanisms by which steroid kines. 49 Additionally, these cells produce less resistance occurs is unclear, it has recently been IFN-, and lose their cytotoxic potential. This demonstrated that IL-4 can down-regulate the suggests that in allergic individuals, CD8 + T cell glucocorticoid receptor affinity in vitro. 56 These function may be re-directed if activated by a speobservations suggest that inhibition of IL-4 may, cific antigen in the presence of IL-4. In this in addition to providing an alternative therapy to context, it has recently been demonstrated that steroids, prove to be a useful adjunct therapy in MHC class I restricted viral antigen-specific acti-the treatment of asthma by possibly restoring vation of CD8 + T cells in the presence of IL-4, normal steroid receptor function.
leads to a phenotype that produces IL-5 and reduced amounts of IFN-t. 5 Moreover, in vivo, the induction of an IL-4-dependent Th2 pheno-Cgtokin rgulion type can switch viral-specific CD8+ T cells to Commitment produce IL-5 and induce eosinophil recruitment into the lungs. 5 These observations suggest that Activated antigen-naive resting CD4 + T cell IL-4 is important not only in regulation of cells from non-allergic donors secrete mainly IL-CD4 + T cell commitment, but has dramatic 2, with a very low production of IL-4 and IFN-,. effects on CD8 + T cell function. Thus, this IL-4 These cells must be 'primed' or 'committed' to mediated switch of CD8 + T cells to a Th2 like either a Thl (IFNq, producing) or Th2 (IL-4 phenotype, may not only exacerbate asthma producing) phenotype. It is now well established severity by secreting IL-5, but the reduction in in vitro that IL-4 is essential for the commitment IFN-, secretion may impair the normal host of naive CD4 + T cells to the Th2 phenotype response, leading to delayed viral clearance from after either stimulation through the CD3 the lung. complex 5v'58 or by a specific antigen. 59 Likewise, studies performed in mice lacking the IL-4 gene IL-4 and steroid resistant asthma: Glucocortico-have demonstrated the essential role of this steroids are the single most effective class of cytokine in the development of Th2 immune drug able to control symptoms and suppress response, as ex vivo stimulation of CD4+ T overt airway inflammation in asthma, and are cells from these mice causes the secretion of increasingly widely used as first line therapy. It greatly reduced amounts of IL-5 and IL-10, and has recently been reported that treatment of fail to mount an IgE response. Likewise, these allergic individuals with systemic prednisolone mice have a marked attenuation of IL-5 depeninhibits the expression of mRNA for IL-4 and IL-5 dent lung eosinophil recruitment following aem-mRNA, whilst increasing mRNA for IFN-, sug-allergen provocation. The initial source of IL-4 gesting that one of the therapeutic effects of this required for Th2 commitment is at present drug is to inhibit Th2 cell activation. 5 Steroids uncertain. As discussed below, mast cells and may also be effective by inducing apoptosis of basophils can produce IL-4 following crosseosinophils in the inflamed mucosa, by suppreslinking of IgE bound to the FceR1. 6 However, it sing IL-3, GM-CSF and IL-5 synthesis. 52 is unlikely that mast cells would function as the However, steroids are unable to suppress IL-4 primary source of IL-4 as in murine systems IgE induced up-regulation of the endothelial adheproduction is strictly IL-4 dependent. Moreover, sion ligand for eosinophils, and VCAM-153 and adoptive transfer of CD4 + T cells alone from can enhance IL-4-driven IgE production in normal mice to IL-4 gene deleted mice results in normal human lymphocytes. Furthermore, in a the development of an IgE response following Th2-1ike T cell line, IL-4 selectively inhibits dexaimmunization, suggesting that for the induction methasone-induced apoptosis. In this context, it of the B cell isotype switch to IgE, IL-4 is has recently been demonstrated that in vitro derived solely from CD4+ T cells. 6 More steroids are unable to inhibit IL-4 secretion from recently, a small subpopulation of T cells murine Th2 cells. 54 Together, these observations derived from the spleen and bearing the markers 242 Mediators of Inflammation Vol 4 1995 Th2 cells and cytokines in allergic inflammation CD4+/NKI.1 + have been shown to rapidly between IL-4 and IFN-7 production will deterproduce IL-4 mRNA upon activation, indepenmine whether a Thl or a Th2 phenotype is prodent of IL-4 itself. 62 These observations suggest duced. This cross regulation has been elegantly that possibly these cells can provide the first demonstrated in models of parasite infection source of IL-4 at the outset of the immune where neutralizing antibodies against IL-4 or IFNresponse which subsequently primes naive 7 have been used to resolve or promote infec-CD4 + T cells to the Th2 phenotype, tion by biasing the immune response in favour IL-10 is a major cytokine produced by several of Thl or Th2 immunity. different cell types. Mouse IL-10 is secreted from IL-12 is also believed to play a central role in Th2, but not Thl cells. 6 In addition, mouse the regulation of the immune response. IL-12, CD5+ B cells produce IL-10 and it has been like TNF-cz, IL-1 and IL-6, is secreted principally suggested that autocrine production of IL-10 by from macrophages in response to bacterial or these cells acts as a growth factor. 64 Murine mac-viral pathogens.  has been reported to facilrophages are also a source of IL-10. In contrast, itate the development of the Thl phenotype, effithe production of IL-10 by human cells is not ciently prime cells for IFN-7 production and restricted to Th2 cells, and approximately one inhibit the commitment of IL-4 producing cells, v third of CD8 + T cell clones, as well as Th0 and Similarly in vivo, IL-12 shifts the immune Thl cells, produce IL-10 upon restimulation. 65 In response from an IL-4 producing to an IFN-7 comparison to other cytokines, IL-10 is syntheproducing phenotype and provides protection sized at a late stage after activation, with maximal against Leishmania major infection where Th2 expression of mRNA 24 h and protein production responses are lethal. 71 More recently, it has been 48h later. 65 IL-10 inhibits production of IFN-7 demonstrated that IL-12 will inhibit IL-4, IL-5, IL-6 and the proliferation of Thl, but not Th2 cells. 63 and IL-13 mRNA and potentiate expression of Moreover, inhibition was observed only in the mRNA for IFN-7, IL-2 and IL-10 after infection presence of macrophages and not B cells, sugwith Schistosoma mansoni. 72 Studies using neugesting an indirect mode of action via the tralizing antibodies to IFN-7 demonstrated that antigen-presenting cell itself. 66 This inhibition the effects of IL-12 on expression of IL-2, IL-4 appears to be unrelated to antigen processing, and IL-13 were mediated through IFN-7 producas IL-10 also inhibited cytokine production from tion. However, the suppressive effects of IL-12 on Thl cells induced by superantigens. 65 IL-10 is a mRNA for IL-5 and IL-6 were relatively refractory potent suppresser of macrophage activation and to IFN-T 72 raising the possibility that IL-12 may inhibits the production of IL-lcz, IL-l]3, IL-6, IL-8, be a more important regulator of IL-5 production TNF-z and GM-CSF from monocytes at both than IFN-7. In human T cells, IL-12 can switch the protein and transcriptional level. IL-10, like the in vitro recall response of allergen-specific T IL-4, up-regulates class II MHC expression on cells from atopic donors, from a Th2 to a Thl small resting B cells and maintains their viability phenotype, suggesting that IL-12 can not only in vitro, although in contrast to IL-4, IL-10 commit cells to a Thl phenotype, but also has failed to increase expression of CD23. Thus, IL-the potential to reverse established immune 10 appears to down-regulate the immune response. response by inhibiting the production of pro-IFN-cz has a wide range of immunomodulatory inflammatory cytokines. In this context, it has activity and has been shown to suppress IgE pro- 74 -75 recently been demonstrated that IL-10 will duction both in vivo and in vitro. However, a inhibit the accumulation of eosinophils in the direct effect on B cells is unlikely as IFN-cz fails lung after antigen challenge, associated with a to inhibit B cell switch in vitro unless IFN-7 is supression of TNF-cz in the bronchoalveolar present. 7. The hypothesis that IFN-cz mediates its lavage fluid. 67 suppressive effect via IFN-7 production is sup-IFN-7 has been shown to inhibit proliferation ported by the finding that IFN-cz induced of Th2 cells and to inhibit IL-4 induced Th2 pheinhibition of anti-IgD induced IgE and IgG1 pronotype commitment. 8 Furthermore, IFN-7 has duction is associated with an increase in splenic been shown to prime cells for further IFN-7 pro-mRNA for IFN-T. 77 This hypothesis has recently duction. In vivo, IFN-7 has been shown to inhibit been supported using human CD4 + T cells, the recruitment of eosinophils to the lungs of showing IFN-cz increases the frequency of IFN-7 sensitized mice. 69 Moreover, anti-IFN-7 poten-producing CD4+ T cells and antagonizes the tiated the antigen-induced eosinophil accumulasuppressive effects of IL-4 on IFN-7 production. 7 tion, suggesting that endogenous IFN-7 secreted As a consequence, both IL-12 IFN-7 and IFN-cz following aerosol antigen provocation acts to may favour the induction and maintenance of inhibit mucosal Th2 immune responses. .9 It has Thl cells and counteract Th2 driven allergic reactherefore been proposed that the balance tions (Fig. 1).
Cytokines from a non-CD4+ T cell source of human eosinophils with calcium ionophore also leads to the production of IL-8. 85 However, Mast cells and basophils: Mast cell metaplasia in while IL-8 generation by monocytes or neuthe bronchial mucosa is a pathological feature of trophils can be induced by IL-I[ and TNF-a, chronic asthma. Mast cells have been shown to these cytokines failed to induce IL-8 secretion be a rich source of cytokines and as such may from eosinophils. Eosinophils can also secrete not only contribute to the acute responses TGF-8 and TGF-187 and as such may account following antigen challenge, but may also play an for the eosinophil-derived stimulatory capacity important role in the persistence of airway for fibroblast proliferation. 88 This is particularly inflammation. Murine non-B/non-T cells (either important in bronchial asthma, where changes in mast cells or basophils) have been reported to the architecture of the lung may contribute to secrete IL-4 and IL-6, 79 as well as TNF-a and the irreversibility of this disease. Likewise, human GM-CSF upon cross linking of the Fce recepeosinophils synthesize and secrete IL-6 9 which is tor. 79 Likewise, human mast cells also secrete able to facilitate IL-4 dependent IgE production, IL4 upon IgE specific activation. More recently, and synergizes with IL-3 and GM-CSF for the it has been shown that murine mast cells can maturation of multipotential granulocyte progenialso synthesize and secrete IL-13 upon IgE tors. Interestingly, IL-6 can also promote the dependent activation and induce IgCe transecretion of IgA in mucosal tissue, which may scripts. 8 Additionally, mast cells produce IL-3, subsequently 'arm' eosinophils to secrete their which together with IL-5, facilitate expansion of granular contents following activation. the eosinophil lineage and enhance eosinophil survival. IL-3 can also prime mast cells for cytokine production ( Table 1). Metachromatic cells Th2 inhibitors--novel asthmatic drugs? may therefore have an important role as rapid sources of cytokines triggered by IgE-dependent Important theoretical questions surround the cell activation. Indeed, a positive feedback loop rational and therapeutic value of Th2 inhibitors may operate involving IgE, mast cells and Th2 as they arise. A central question relates to the cytokines derived from CD4 + T cells. Thus, role of IL-4 in the longevity and maintenance of antigen exposure would lead to the production Th2 immune responses. While inhibition of IL-4 of IL-4 from CD4 + T cells which would then would prevent the switch B cells to IgE producexpress CD40L upon activation and facilitate the tion, it would also inhibit the differentiation of production of IgE, which in turn would 'arm' naive T cells to the Th2 phenotype. However, mast cells with the specific antibody. Activation therapeutically this would be an unachievable of mast cells and basophils upon re-exposure goal. As allergic individuals retain their sensitivity to the antigen would then lead to IL-4 producto allergens over many decades, it might be suption and amplification of the Th2 response posed that IgE-switched B cells and IL-4 com- (Fig. 2). mitted Th2 cells retain their phenotypes for prolonged periods. If this is the case, continuous Eosinophils: While there is a considerable body exposure to environmental antigens would be of evidence to suggest that eosinophils are final expected to reactivate these immune mechanisms effector cells in the pathogenesis of allergic in an IL-4 independent manner. If allergic disease and bronchial asthma, mediated largely 81 82 through the secretion of cationic proteins, these cells also have the capacity to synthesize CD4+ T cells, provided with appropriate co-stimulatory signals, secrete IL-4 and express CD4OL, and as a consequence instruct B cells to produce IgE. Likewise, Th2 cells produce IL-5 to recruit eosinophils into the lung. Interestingly, activated eosinophils may also express CD40L to help B cell IgE production. IgE can then bind to the FCRI on mast cells, which, when activated, provide a further source of IL-4 to maintain and amplify the immune response.
immune 'memory' (persisting responsiveness to recall antigens) is IL-4 independent, inhibitors of Th2 cytokine production rather than commitment will prove to be of greater therapeutic value. However, in some murine systems of ongoing IgE synthesis, there is evidence that IL-4 is required for the maintenance of the Th2 phenotype. 9 However, the situation may be more complex in humans, where IL-13 dependent ongoing IgE synthesis may be more important.
However, inhibition of IL-4 may offer advantages in steroid resistant asthma by preventing/reversing impaired steroid receptor function 55 and in viral mediated exacerbations of asthma, where IL-4 may be of central importance in switching cytot0xic CD8 + T cells to a Th2 like phenotype. 49'50 Likewise, selective inhibition of IL-5 offers the possibility of selectively suppressing eosinophil infiltration into the airways. However, whether other eosinophil factors such as eotaxin can replace IL-5 in the lung of allergic individuals remains to be determined. Finally, one important point to determine is whether agents that suppress the production of Th2 cytokines from T cells are also effective in suppressing the secretion of cytokines from, for example, mast cells and eosinophils, which may be important sources of cytokines in established chronic asthma.

Conclusions
The cytokines released from Th2 and Th2-1ike cells are likely to be central to the pathophysiology of asthma and allergy contributing to aberrant IgE production, eosinophilia and, perhaps, mucosal susceptibility to viral infection. IL-4 has emerged as a central target, not only for B cell IgE production, but also in the commitment of both CD4 + and CD8 + T cells to cells with Th2 effector function capable of secreting IL-5 resulting in eosinophilic inflammation. In view of the central role of this cytokine and the evidence that glucocorticoids are unable to modify many IL-4 dependent effects, Th2 inhibitors may prove to be novel therapies for the treatment of bronchial asthma.