Chemokines in allergic lung inflammation

The Chemokine Family

The chemokine family constitutes a group of specialized cytokines, the primary function of which is to regulate the trafficking of leucocytes. All family members are small secreted heparin-binding proteins that can be distinguished from classical chemoattractant molecules (such as bacterial-derived N-formyl peptides, complement fragment peptides C3a and C5a, and lipid molecules such as leukotriene B₄ and platelet-activating factor) on the basis of shared structural similarities. Chemokines have four conserved cysteine residues that form disulphide bonds which are critical for the tertiary structures of the proteins. The chemokine family is organized into four subclasses according to the position of the first two cysteines. The two major subclasses include the CC chemokines, where the cysteines are adjacent, and the CXC chemokines, where the cysteines are separated by one amino acid. Two other subclasses have been identified with, to date, one member in each. The C class has only two cysteines instead of four and has lymphotactin as its member, while the CX3C subclass has three amino acids between the first two cysteines and a mucin stalk at the N-terminal end, and incorporates fractalkine. Chemokines have a short N-terminal domain preceding the first cysteine, a backbone made of β-strands with the connecting loops found between the second and fourth cysteines, and a C-terminal α-helix of 20–30 amino acids.

The chemokine superfamily has rapidly expanded as a result of the availability of large databases of expressed sequence tags (ESTs) and bioinformatics.¹ Indeed, the distinct motifs contained within chemokine structures have made identifying new family members within these EST databases relatively easy. To date, 23 human CC chemokines, 14 human CXC chemokines and one member of both the CX3C and C subclasses have been described. This rapid expansion of the chemokine/receptor family has led to problems when several research groups have described a single chemokine that has become known by multiple names. To counteract this confusion a new nomenclature has established,² where an L (for ‘ligand’) was added (CXCL, CCL, CL and CXXXCL). The chemokines thought to be responsible for attracting the major cell populations involved in the pulmonary response to allergen are shown in Table 1.

Table 1. Specific chemokines that attract the major cell populations involved in the allergic response.

Infiltrating leucocyte	Chemokine
Eosinophils	Eotaxin-1, -2, -3; RANTES, MCP-3, -4;″MIP-1α
Activated T cells	RANTES; MCP-1, -3, -4; SDF-1α; MIP-1α
Th2 cells	MDC; TARC; I-309; Eotaxin
Monocytes	MCP-1, -2, -3, -4; RANTES; MDC;″TARC; MIP-1α

MCP, monocyte chemoattractant protein; MDC, macrophage-derived chemokine; MIP, macrophage inflammatory protein; RANTES, regulated upon activation, normal, T-cell expressed, and secreted; SDF-1, stromal-cell-derived factor 1; TARC, thymus and activation-regulated chemokine; Th, T helper.

Chemokine Receptors

The specific biological effects of chemokines are mediated via interactions with heterotrimeric seven-transmembrane G-protein coupled receptors (GPCRs) expressed predominantly on leucocytes. These chemokine receptors are part of a much larger superfamily of GPCRs that include receptors for hormones, neurotransmitters, paracrine substances, inflammatory mediators, certain proteinases, taste and deodorant molecules, and even photons and calcium ions.³ Chemokine receptors measure ≈350 amino acids in length and consist of a short extracellular N-terminus and an intracellular C-terminus (containing serine and threonine residues that act as phosphorylation sites for receptor regulation). There are seven α-helical transmembrane domains, with three intracellular and three extracellular connecting loops, and a disulphide bond links highly conserved cysteines in extracellular loops 1 and 2. The N-terminus and the third intracellular loop are thought to be essential for specific binding of chemokines. Chemokine receptors couple to heterotrimeric G-proteins through the C-terminus segment and possibly through the third intracellular loop (reviewed in ref. 4).

At present, 18 human chemokine receptors have been identified.^2,5 Chemokine receptors CXCR1 to CXCR5 bind the CXC family of chemokines, whereas the CC family consists of nine receptors (termed CCR1 to CCR9). Specific receptors for lymphotactin (XCR1) and fractalkine (CX₃CR1) have also been identified and cloned. Interestingly, there is a certain amount of promiscuity in the chemokine superfamily with many ligands binding different receptors and vice versa. Some receptors bind only one chemokine, for example CXCR1 binds only interleukin (IL)-8, while others are shared by multiple chemokines; for example CCR1 binds macrophage inflammatory protein-1α (MIP-1α), regulated on activation, normal, T-cell expressed and secreted (RANTES), and monocyte chemotactic protein (MCP)-2 and -3. Another receptor known as the Duffy antigen receptor for chemokines (DARC), expressed on erythrocytes and endothelial cells, is truly promiscuous and has been shown to bind both CXC and CC chemokines. Finally, virally encoded chemokine receptors have been described and are thought to be a mechanism of viral evasion from the immune system.⁶

Functional responses of leucocytes to chemokines

Interaction of chemokines with their counter receptors mediates a series of effects that ultimately result in the directional movement of the leucocyte, namely chemotaxis. This is accomplished by a series of events, starting with a change in shape that occurs within seconds of addition of chemokine to a leucocyte. Polymerization and breakdown of actin leads to formation and retraction of lamellipodia, which function as the limbs of the migrating cell. Stimulation also induces the up-regulation and activation of integrins, which then enable the leucocyte to adhere more firmly to the vascular endothelial cell wall before migrating through to other tissues. Although chemotaxis is the predominant function of chemokines, several other rapid and transient responses are characteristic of the activation of leucocytes by chemokines. These include the rise in intracellular free calcium concentration; the production of microbicidal oxygen radicals and bioactive lipids; and the release of the contents of the cytoplasmic storage granules, such as proteases from neutrophils and monocytes, histamine from basophils and cytotoxic proteins from eosinophils. In addition, evidence suggests that chemokines are involved in the maturation, differentiation, activation and homeostatic trafficking of leucocytes within the immune system. Thus, chemokines have multiple effects on the development and progression of an immune reaction.

Cells involved in airway inflammation

One of the characteristic features of the pulmonary response to allergen in atopic asthmatics is leucocytic infiltration of the airways. The consequence of this recruitment of inflammatory cells is the development of goblet cell hyperplasia and airway hyper-reactivity (AHR) in conjunction with subepithelial remodelling. While eosinophils are the predominant cell type in this infiltrate, significant numbers of lymphocytes and macrophages are also present.

Effector cells

Although eosinophils are the most predominant leucocyte involved in the asthmatic response, it is clear that a number of cells play important effector roles in the evolution of pathophysiology. Mast cells and basophils are also present within the lungs during the early stages after allergen challenge. These effector cells are integral components of the response as their degranulation products can elicit tissue damage. Eosinophils are thought to be responsible for the tissue damage that leads to the disruption of the bronchial epithelium, enhanced AHR and bronchial obstruction. Endobronchial allergen-challenge studies in mild atopics have shown that eosinophils appear in the submucosa as early as 6 hr following allergen challenge. By 24 hr, the majority of these eosinophils have migrated through the bronchial epithelium into the airways where they sit in the bronchiolar epithelial lining fluid and can be collected during bronchoalveolar lavage (BAL). The specific role of chemokines in co-ordinating the recruitment of eosinophils, mast cells and basophils to the lung will be discussed below.

Regulatory cells

Whilst the absolute numbers of T cells in inflammatory infiltrates is low in comparison with eosinophils, the importance of T cells in controlling the allergic reaction is well recognized. T cells are critical mediators of the allergic inflammatory response and as such they, and their secreted products (i.e. IL-4, IL-5 and IL-13), are found in biopsies and BAL fluid from patients. Moreover, in vivo depletion experiments in mice or the use of mice genetically deficient in T cells have shown that a functional CD4 population is critical for the development of allergic inflammation.^7,8 The delivery of functional subsets of T cells to particular tissues or microenvironments is a tightly controlled process involving a complex series of molecules expressed by a variety of cell types. This is especially important for T cells, as effector T cells can be divided into distinct subsets based upon their cytokine profiles and functional properties. T helper 1 (Th1) cells characteristically produce interferon-γ (IFN-γ) and contribute to host defence against pathogens, whereas T helper 2 (Th2) cells produce IL-4 and IL-5 and are associated with allergic reactions involving immunoglobulin E (IgE), eosinophils and basophils.⁹ Th2 cells and the cytokines they secrete are thought to be critically important for the development of injury during allergic reactions such as asthma. Studies with in vivo models of allergic inflammation have shown that blocking Th2 function, or neutralizing key Th2 cytokines, results in abrogation of pulmonary eosinophilia and AHR.^10–12 However, the mechanism by which Th2 subsets are specifically recruited to the lung has been the subject of considerable debate. It is probable that chemokines are intimately involved in this process, which will be discussed in detail below.

Mononuclear phagocytes

Mononuclear phagocytes also play an integral part in the development of allergic inflammatory reactions. They have many functions, both regulatory and effector. Monocytes and macrophages are both rich sources of cytokines and chemokines that can act to magnify or modulate the immune response. In addition they can release mediators, such as histamine or prostaglandins, that contribute to tissue injury. Macrophages may also act as antigen-presenting cells and so can amplify the response. Moreover, there is accumulating evidence that macrophages are involved in tissue repair and contribute to airway remodelling, the long-term consequence of chronic allergic inflammation.

Chemokines involved in the allergic inflammatory response

The chemokines are potent chemoattractants and evidence suggests that they play a critical role in directing inflammatory cell recruitment during pulmonary allergic inflammation. Immunohistochemical analyses of chemokine expression in lung biopsies from allergic asthmatics and in lungs taken from mice during models of allergic airway disease have determined that a range of chemokines are expressed (Table 2). Moreover, these experiments have yielded important facets of chemokine function. In models of disease it is possible to make a detailed examination of expression during a time-course study after allergen challenge. These studies have revealed important features of chemokine expression that may be critical to our understanding of their action in vivo, i.e. their temporal and spatial distribution. It has been reported that the pattern of chemokine expression changes during the progression of disease.¹³ This inevitably dictates the temporal recruitment of different leucocyte populations. It is presumed that chemokines generate chemotactic gradients in vivo. The location, range and distribution of these chemotactic gradients depend on the spatial location of the cellular source for the specific chemokine. The mapping of the chemotactic gradients and their pattern of expression are essential in determining and ultimately understanding the specific role of a given chemokine in the inflammatory disease process. As yet, the majority of this work has been carried out in the mouse but some studies are analysing human biopsies collected at multiple time-points following segmental allergen challenge in humans.¹⁴

Table 2. Cellular sources of chemokines and their receptors involved in asthma.

Chemokine	Cellular source in the lung	Cognate receptor	Receptor expression in asthma
Eotaxin (CCL11)	Epithelial cell, alveolar macrophage,″endothelial cell, smooth muscle cell,″fibroblast, eosinophil, lymphocyte	CCR3	Eosinophils, basophils, mast cells
Eotaxin-2 (CCL24)	Unknown	CCR3	Eosinophils, basophils, mast cells
RANTES (CCL5)	Epithelial cell, smooth muscle, eosinophil	CCR1, CCR3, CCR5	Eosinophils, monocytes, activated″T cells
MCP-1 (CCL2)	Epithelial cell, fibroblast,″alveolar macrophage	CCR2	Monocytes, macrophages, basophils,″activated T cells
MCP-3 (CCL7)	Epithelial cell	CCR1, CCR2, CCR3	Eosinophils, basophils, monocytes,macrophages, activated T cells
MCP-4 (CCL13)	Epithelial cell	CCR2, CCR3	Eosinophils, basophils, monocytes,″macrophages, activated T cells
MDC (CCL22)	Smooth muscle cell, alveolar″macrophage, bronchiolar epithelium	CCR4	Monocytes, Th2 cells
TARC (CCL17)	Bronchiolar epithelial cells	CCR4	Monocytes, Th2 cells
MIP-1α (CCL3)	Epithelial cell	CCR1, CCR5	Eosinophils, monocytes, activated″T cells
IL-8 (CXCL8)	Alveolar macrophage	CXCR1, CXCR2	Neutrophils

IL-8, interleukin-8; MCP, monocyte chemoattractant protein; MDC, macrophage-derived chemokine; MIP, macrophage inflammatory protein; RANTES, regulated upon activation, normal, T-cell expressed, and secreted; TARC, thymus and activation-regulated chemokine.

In order to highlight some of the important functions of chemokines in mediating the allergic response, the actions of three particular CC chemokines – eotaxin, MCP-1 and macrophage-derived chemokine (MDC) – will be discussed. These chemokines are vital for the development of particular facets of the pathophysiology associated with asthma, and discussion of their functions in detail will exhibit some important features of chemokine function.

Eotaxin–eosinophil recruitment

The eotaxins are a group of three chemokines united by the fact that they all mediate recruitment of eosinophils via interaction with the receptor CCR3. Eotaxin-1 (known as eotaxin) was originally identified in the BAL fluid of allergen-sensitized and challenged guinea-pigs.¹⁵ Thereafter, the mouse and human genes were cloned, with the human gene exhibiting 58% identity with the guinea-pig and mouse genes.^16–18 In vitro experiments have shown eotaxin to be a potent and selective eosinophil chemoattractant. Eotaxin expression is increased in bronchial biopsies of mild asthmatics and in BAL cells obtained 6 hr after segmental allergen challenge.^14,19,20 Moreover, soluble eotaxin has been measured in the serum of asthmatics, and eotaxin levels correlated with disease severity, especially during acute asthma.²¹ The kinetics of eotaxin production in the BAL fluid of asthmatics is similar to that of other key, pro-asthmatic chemokines such as RANTES, MIP-1α and MCP-1. Eotaxin levels peak at 4 hr following allergen challenge and decrease by 24 hr. This contrasts with the secretion of IL-5, which increases gradually in BAL fluid after allergen challenge and peaks at 24 hr after challenge, suggesting that while eotaxin initiates eosinophil recruitment, IL-5 maintains lung eosinophilia.²² In situ hybridization and immunohistochemical staining have demonstrated that the main cell types expressing eotaxin are epithelial cells,^23,24 but macrophages, endothelial cells, smooth muscle cells, fibroblasts, T cells and mast cells have also shown positive expression.

Eotaxin-2 was identified by random sequencing of ESTs in an activated monocyte cDNA library.²⁵ Although eotaxin and eotaxin-2 are functionally similar, they share only 39% identity at the amino acid level and differ almost completely at the N-terminal region. Intradermal injection of eotaxin-2 into rhesus monkeys elicits eosinophil recruitment at the injection site.²⁶ Increased mRNA for eotaxin-2 has been exhibited in skin biopsies isolated during the allergen-induced late-phase cutaneous response,²⁷ and in bronchial biopsies derived from non-atopic and atopic asthmatics.¹⁴ Eotaxin-2 protein has not yet been demonstrated in the airways of asthmatic patients. However, a functional role for eotaxin-2 in allergic inflammation has been postulated in studies showing that glucocorticoids down-regulate eotaxin-2 mRNA expression in nasal polyps.²⁸ Moreover, allergen challenge increases eotaxin-2 mRNA expression in mouse models.²⁹

Eotaxin-3 has only 36% and 32% identity with eotaxin and eotaxin-2, respectively, yet shows considerable functional identity.^30,31 Eotaxin-3 is chemotactic for eosinophils and basophils, but is 10-fold less potent than the other two eotaxins. Eotaxin-3 mRNA expression is induced in endothelial cells by IL-4 and IL-3. It seems that eotaxin-3 is only expressed in humans, as it has not been identified in the mouse.²⁹ A recent study has shown that eotaxin-1 and eotaxin-2 expression was increased in asthmatics compared to controls, but that only eotaxin-3 expression was significantly increased following allergen challenge.³² The authors suggest that eotaxin-3, rather than eotaxin or eotaxin-2, may account for the ongoing eosinophil recruitment to asthmatic airways in the later stage (24 hr) following allergen challenge. Although injection of eotaxin-3 into cynomolgus monkeys induces eosinophil recruitment at the injection site,³¹ the functional role of this chemokine in the allergic pulmonary response remains unproven.

Functional evidence of a role for eotaxin in allergic asthma has come from a number of studies using animal models. Neutralizing anti-eotaxin-1 antibodies have been found to decrease lung and lavage eosinophilia in mice.^17,33,34 In mice, this decrease in eosinophilia is accompanied with abrogation of airway hyper-reactivity.³³ Somewhat surprisingly, mice rendered genetically deficient in eotaxin-1 show only a partially resolved allergic response.^35,36 This may be as a result of compensation by other eosinophilic chemokines (such as eotaxin-2 or RANTES) or perhaps highlights the inherent differences in using knockout mice as opposed to neutralizing antibodies. As yet there have been no experiments determining the effect of neutralizing eotaxin-2 during in vivo models of airway disease.

One property of eotaxin-1 highlights an important facet of chemokine biology – that of synergy with other inflammatory mediators. At an inflammatory site, multiple mediators are present with overlapping functions. It is tempting to speculate whether this seeming redundancy serves to amplify an inflammatory response or is part of a complicated process to limit an inflammatory reaction by feedback inhibition. Experiments investigating the particular relationship between the eosinophil chemoattractant eotaxin-1 and the cytokine IL-5 have revealed important information regarding the recruitment of eosinophils to an allergic site.

The specific role of IL-5 and eotaxin in the regulation of eosinophil trafficking has initiated investigation into the mechanisms of co-operation between these two molecules. Under basal conditions, eosinophils normally reside within the bone marrow and tissues. In response to specific stimuli (inflammatory or parasitic), increased numbers of eosinophils migrate to the site of inflammation. Investigation of eosinophil trafficking in animal models has revealed that IL-5 and eotaxin co-operate to perform fundamental roles under basal conditions and during allergy to regulate recruitment of eosinophils to specific sites.^37–39 Eotaxin is constitutively expressed in a number of tissues but allergen challenge in sensitized animals leads to an early increase in production with subsequent recruitment of eosinophils.^17,40 This eosinophilia is thought to occur in response to the secretion of eotaxin from pulmonary endothelial and epithelial cells, which has been observed in a variety of models and in asthmatic patients.^7,19,41,42 Furthermore, eosinophil migration through the tissues may be potentiated by other chemokines interacting with CCR3, such as RANTES, MIP-1α and MCP-3.

Recruitment of eosinophils into tissues is limited by the low circulating numbers of these leucocytes. A rise in circulating numbers occurs following allergen challenge and this is significantly inhibited by neutralizing antibodies to IL-5 in mice.^43–45 In this context it has been suggested that IL-5 and eotaxin function in a co-ordinated manner, with IL-5 acting to mobilize eosinophils from the bone marrow and eotaxin directing their local recruitment into tissues.

MCP-1 polarization of effector T cells

The allergic response is a prototypical Th2 reaction, whereby polarized lymphocytes drive the immune response along a particular path via the secretion of specialized cytokines. In addition to the initiation and maintenance of leucocyte accumulation, CC chemokine members may have the capacity to drive the inflammatory reaction by augmenting or directionally differentiating T lymphocytes towards a Th1- or a Th2-type response. For example, MCP-1 has been shown to drive undifferentiated in vitro T-cell populations towards an IL-4-producing Th2 type of cell, while MIP-1α appears to promote the development of a Th1 type response by enhancing IFN-γ secretion and decreasing IL-4 production.⁴⁶ Further evidence of the role of MCP-1 in controlling Th2 differentiation comes from a study which shows that mice deficient in MCP-1 are unable to mount Th2 responses.⁴⁷ Lymph node cells from immunized MCP-1^−/− mice synthesize extremely low levels of IL-4, IL-5 and IL-10, but normal levels of IL-2 and IFN-γ. Consequently these mice are not able to accomplish the normal immunoglobulin subclass switch that is characteristic of Th2 responses. This influence on Th2 polarization will probably have far-reaching effects on disease pathogenesis.

Experiments using a mouse model of allergic airway disease have shown that MCP-1 has an important role in the early stages of the response to allergen. Neutralization of MCP-1 during the sensitization and challenge phases of an allergen model ultimately affects recruitment of lymphocytes, monocytes and eosinophils, as well as development of AHR.³³ In contrast, administration of antibody during either the sensitization or the challenge phase had no affect on eosinophil recruitment.³³ Blocking experiments in this and other models have shown that neutralization of MCP-1 results in decreased recruitment of monocytes, but also reduces the levels of Th2 cytokines in the lavage and in serum IgE.³³ Moreover, levels of inflammatory mediators such as histamines and prostaglandins are reduced in the BAL.^33,48 These wide-ranging effects indicate that MCP-1 is acting as more than just a monocyte chemoattractant. It is probable that MCP-1 has an important role in driving the asthmatic response, perhaps as early as during the primary sensitization phases.

MDC: migration of antigen-specific effector T cells

Induction of allergic airway disease is dependent not only on the generation of polarized effector T cells but also on the selective recruitment of these antigen-specific Th2 cells to the lung tissue. The recent observation that chemokine receptor expression appears to be tightly regulated on effector T cells puts forward an attractive mechanism for this recruitment. Examination of polarized effector T cells in vitro has shown that Th cells not only express a restricted panel of receptors for chemokines but that they migrate differentially in response to the chemokines that bind to these receptors.^49,50 In the context of the allergic pulmonary response, eotaxin and thymus and activation-regulated chemokine (TARC)/MDC are among the chemokines that seem to attract selectively Th2, but not Th1, cells. Eotaxin binds CCR3 with high affinity and fidelity.⁵¹ MDC interacts specifically with CCR4⁵² and has been shown to attract Th2 cells in preference to Th1 cells.^50,53–55 CCR8 is also selectively expressed on Th2 cells,⁵⁶ particularly after activation of the T-cell receptor.⁵⁷ The attraction of these Th2 cells by selected chemokines may represent a mechanism by which an allergen-driven reaction escalates with the production of IL-4 and IL-5, both of which are necessary for the differentiation and activation of eosinophils.

The fact that chemokine receptors are differentially expressed on T-cell subsets and that this differential expression results in selective migration of Th1 and Th2 cells to particular chemokines has led to the hypothesis that chemokines are responsible for the selective recruitment of antigen-specific Th2 cells to the lung following allergen challenge. In support of this hypothesis, a recent study has found differential chemokine receptor expression on T cells in bronchial biopsies taken from patients following segmental allergen challenge.⁵⁸ A significantly greater number of CCR4⁺ T cells were documented in biopsies after allergen challenge as compared to prechallenge or non-allergic patients. Moreover, the ligands for CCR4, MDC and TARC were found to be up-regulated on airway epithelium in atopic asthmatics following allergen challenge. CCR4 was found to co-localize with the Th2 cytokine IL-4, strongly suggesting that Th2 lymphocytes are recruited to the airways via interaction of MDC/TARC with CCR4.⁵⁸

Evidence for functional roles of chemokines in attracting effector Th2 cells has been investigated in vivo using mouse models of allergic airway disease. Neutralization of CCR4 ligands with antibodies specific for MDC or TARC has been shown to reduce airway hyper-reactivity as well as lung eosinophilia.^55,59 Interestingly, blockage of TARC in vivo resulted in a decrease in the levels of Th2 cytokines within the BAL.⁵⁹ Although blockage of MDC resulted in a decrease of eosinophils within the lung interstitium, eosinophils in the airway lumen were unaffected, suggesting that MDC plays a key role in the retention of eosinophils in the lung tissue.⁵⁵ In contrast to these studies using neutralizing antibodies to CCR4 ligands, the CCR4 knockout mouse shows no defect in the development of allergic airway disease.⁶⁰ This difference may reflect intrinsic differences in the experimental model systems used or mechanistic differences in blocking ligands versus receptors.

Another mouse model of allergic airway disease has been used to address the role of chemokines in the specific migration of allergen-specific Th2 cells.⁶¹ This model was based on the adoptive transfer of polarized effector Th cells. Tracking of the transferred allergen-specific Th2 cells after repeated antigen challenge established that both the CCR3/eotaxin and the CCR4/MDC axes contribute to the recruitment of Th2 cells to the lung, demonstrating the in vivo relevance of the expression of these receptors on Th2 cells.⁶¹ CCR3/eotaxin was found to be important in the earlier stages of the response, whereas repeated antigen stimulation resulted in the predominant use of the CCR4/MDC pathway. This finding emphasizes the relevance of previous in vitro results and demonstrates for the first time in vivo that CCR3 and CCR4 are more than markers of Th2 cells but have a critical pathophysiological significance in the development of allergic airway disease (AAD) (as determined by their impact on AHR and eosinophilia).⁶¹

CCR8 (the receptor for I-309/CCL1) is also selectively expressed on Th2 cells, indicating that this receptor may also be important in recruitment of Th2 cells to the lung. Indeed, a recent study using mice genetically deficient in CCR8 demonstrates that CCR8 is important in the migration of antigen-specific T cells to the lung and in the ensuing development of AHR.⁶² Furthermore, a significant increase in the number of CCR8-bearing T cells in atopic asthmatics following allergen exposure was recently documented,⁵⁸ implicating I-309 and CCR8 in the specific recruitment of Th2 cells in vivo. Although there were fewer CCR8⁺ than CCR4⁺ T cells detected, the number of CCR8⁺ T cells infiltrating the airway mucosa was found to correlate with the degree of airflow limitation during the late-phase reaction. Interestingly, the authors of the study could not detect expression of the CCR8 ligand I-309 in either airway mucosa or in airway epithelial cells, leading them to speculate that either TARC acts as an in vivo ligand for CCR8 or that there is another, as yet unidentified, ligand for CCR8 in the airway mucosa.

Do chemokines play a functional role in asthma?

Multiple chemokines and receptors are expressed during the allergic reaction, and while the extent of investigation has been limited in humans to expression analysis and direct intradermal injection of chemokines in vivo, functional analysis of chemokines and their receptors has been conducted in animal models of asthma. Mouse models of allergic airway disease have been invaluable in attempts to dissect the roles that individual cells and molecules play in the development of different pathophysiologies associated with asthma. These models are particularly suited for the study of chemokine function in vivo for a number of reasons. Primarily, multiple leucocyte subpopulations are recruited to the lung in a sequential manner. Moreover, some of these populations are clearly regulatory in nature (e.g. Th2 cells), while others have an effector role (e.g. eosinophils) and the immunological/physiological functions of one or the other can be distinguished with different end-point assays. Importantly, the recruitment of leucocyte subsets to the lung can be monitored at different, distinct anatomical sites – perivascular, interstitial and in the airway lumen. Finally, the anatomical location of these leucocyte subsets can be correlated with physiological end-points that are essential for disease: AHR and mucus hypersecretion. These models have been used to define the roles that chemokine/receptor interactions play in vivo to mediate the recruitment of specific leucocyte subtypes to the lung over time following allergen challenge. A summary of the findings of these studies is given in Table 3. Studies with neutralizing antibodies, genetically deficient mice and small-molecule antagonists have revealed critical roles for eotaxin, RANTES, MCP-1, MCP-3, MCP-5, MDC, MIP-1α and stromal-cell-derived factor 1 (SDF-1α) in eosinophilia and airway hyper-reactivity. These studies have involved the blockage of a single chemokine at intervals in a number of different models of allergic airway disease. While there are limitations to these studies, they clearly show that chemokines play an important role in the disease process. Interestingly, they have shown that chemokines not only contribute to tissue inflammation by recruiting and activating leucocytes, but they also mediate degranulation and cause mediator release from effector cells such as basophils, mast cells, neutrophils and eosinophils.

Table 3. Functional analysis of chemokines and their receptors in vivo.

Chemokine/receptor	Method	In vivo effect	Reference
Eotaxin (CCL11)	Neutralizing antibodies	Decrease in eosinophil recruitment, AHR	33
	Neutralizing antibodies	Reduced Th2 recruitment, AHR	61
	Gene knockout	Partial reduction in eosinophil recruitment	35
	Gene knockout	No effect	36
MCP-1 (CCL2)	Neutralizing antibodies	Reduction in AHR, lavage and tissue eosinophilia″and inflammatory mediator release	33,48,78
MCP-3 (CCL7)	Neutralizing antibodies	Reduced lavage eosinophilia	79
MCP-5 (CCL12)	Neutralizing antibodies	Reduction of tissue eosinophil recruitment and AHR	33,80
RANTES (CCL5)	Neutralizing antibodies	Reduced eosinophilia	78
	Receptor antagonist	Reduction in tissue and lavage eosinophilia and AHR	33
MIP-1α (CCL3)	Neutralizing antibodies	Reduced eosinophilia	78
		Partial reduction in eosinophilia and AHR	33
MDC (CCL22)	Neutralizing antibodies	Reduction of tissue eosinophil recruitment and AHR	55
		Reduced Th2 recruitment	61
TARC (CCL17)	Neutralizing antibodies	Reduction of tissue eosinophil recruitment, AHR and″decreased Th2 cytokine production	59
SDF-1α (CXCL12)	Neutralizing antibodies	Reduction of tissue eosinophil recruitment and AHR	81
CCR1	Gene knockout	Decreased Th2 cytokines and airway remodelling	82
CCR2	Gene knockout	Reduced airway hyper-reactivity and BAL histamine	48
	Gene knockout	No effect	83
	Gene knockout	Increased inflammation, AHR and remodelling	84
CCR4	Gene knockout	No effect	60
CCR8	Gene knockout	Decreased eosinophilia and serum IL-5	62
CCR6	Gene knockout	Decreased eosinophilia, AHR and lung IL-5	85
IL-8R	Gene knockout	Increased B-cell recruitment, serum IgE and decreased AHR	86
CXCR4	Neutralizing antibodies	Reduction of tissue eosinophil recruitment and AHR	81

This table summarizes results from various investigators after abrogating chemokine ligand or receptor function by genetically deficient mice, neutralizing antibodies or receptor antagonists. The effect column denotes the resulting change in phenotype after intervention. For further details of each experiment please see individual references.

AHR, airway hyper-responsiveness; BAL, bronchoalveolar lavage; IgE, immunoglobulin E; IL-5, interleukin-5; IL-8R, interleukin-8 receptor; MCP, monocyte chemoattractant protein; MDC, macrophage-derived chemokine; MIP, macrophage inflammatory protein; RANTES, regulated upon activation, normal, T-cell expressed, and secreted; SDF-1, stromal-cell-derived factor 1; TARC, thymus and activation-regulated chemokine; Th2, T helper 2.

Redundancy or co-ordinated effect?

In vitro and in vivo characterization of chemokine functions suggests that there is redundancy within the family, as many chemokines have overlapping actions and promiscuous receptor usage. Furthermore, expression studies in humans and animal models have determined that multiple chemokines are expressed in the lung in asthma. This might imply that some of the chemokines detected by expression analysis are redundant during the allergic response. However, detailed study of animal models of airway inflammation has shown that the production of these chemokines in vivo is organized and occurs in a co-ordinated manner. Examination of murine models of pulmonary allergic inflammation have shown that leucocyte recruitment and development of AHR involves the action both of chemokines that are primarily eosinophilic (e.g. eotaxin, RANTES, MCP-5 and MIP-1α) and non-eosinophilic (e.g. MCP-1, MDC).^33,55 Taken together, these studies argue against redundancy as chemokines seem to exert a critical role at different stages and on different pathways of the development of allergen-induced lung eosinophilia. Although the exact contribution of individual chemokines varies according to the particular model used, it is clear that chemokines function in a tightly controlled manner, with particular chemokines operating at key stages of the response.

Prospects for therapeutic intervention

The fact that chemokines play a role in multiple aspects of the pathological response to allergen makes them attractive as novel therapeutic targets. Modulation of cell recruitment during an inflammatory reaction is very likely to ameliorate the associated physiological symptoms. There is compelling evidence for this to be the case as there is a large body of evidence showing that blockage of chemokines suppresses inflammatory diseases in a variety of animal models. Moreover, the success of small-molecule inhibitors of GPCRs in the treatment of various other diseases has led the pharmaceutical and biotechnology industries to investigate the production of small-molecule inhibitors of chemokine receptors. Based on this strategy, several methods of antagonizing chemokine receptors are being investigated: small-molecule antagonists; modified chemokines; neutralizing antibodies; and viral antagonists.

Small-molecule antagonists

This theoretical strategy has been put into practice with some success in the development of small-molecule antagonists for CCR5 and CXCR4 in view of the critical role that these receptors play in human immunodeficiency virus (HIV) infection.^63–65 Targeting of chemokine receptors to suppress cell recruitment to the lung is an area of intense investigation for novel asthma therapy. Receptors expressed on eosinophils are of particular interest as they represent the most prominent infiltrating leucocyte. CCR3 is an obvious choice as it is expressed not only on eosinophils, but also on Th2 cells, both of which are critical in the development of the asthmatic response.^18,51,53 A range of chemokines signal via CCR3, mainly the eotaxins, but also RANTES, MCP-4 and MCP-3, all of which have been documented to be up-regulated in asthma.⁶⁶ Signalling through CCR1 and CCR3 has been blocked successfully in vitro using a single compound (UCB35625);⁶⁷ however, as yet no in vivo data is available. Another area of investigation is the suppression of Th2 responses, as this population of effector cells is thought to be responsible for the initial development and subsequent escalation of the allergic response. Receptors on Th2 cells include CCR3, CCR4 and CCR8. To date there are no compounds that selectively target chemokine receptors, but these receptors are currently the subject of intense investigation.

Modified chemokines

The discovery and development of pharmaceutical antagonists, however, is a long and expensive process and consequently other strategies are being pursued. Modified chemokines and N-terminal peptides can be engineered to allow them to retain binding specificity and affinity to a receptor while blocking intracellular signalling and therefore function. A modified version of RANTES, whereby an additional methionine residue was added to the N-terminus,⁶⁸ decreased both cellular inflammation and AHR in an in vivo mouse model of allergic pulmonary disease.³³ Similarly, addition of an aminooxypentane residue at the N-terminus of RANTES (AOP-RANTES) has been shown to inhibit HIV-1 infectivity in macrophages and lymphocytes.⁶³ Another CCR3 antagonist, termed CKβ7, has been generated by an N-terminal alanine–methionine swap of MIP-4.⁶⁹ Whereas Met–RANTES inhibits eosinophil effector function through antagonizing CCR1 and CCR3, CKβ7 specifically antagonizes CCR3. CKβ7 is a more potent CCR3 antagonist than Met–RANTES and prevents signalling through CCR3 at concentrations of 1 nm. However, the success of modified chemokines or N-terminal peptides as antagonists depends mostly on their capacity to fully occupy the chemokine receptor/s at nanomolar concentrations, competing with the natural ligand(s) binding and thus blocking signalling. One of the advantages of using a modified ligand is that most of the receptors used by that ligand can be blocked, or partially blocked, by a single antagonist.^70,71

Therapeutic antibodies

Generation of specific monoclonal antibodies (mAbs) against chemokines or their receptors represents another strategy for modifying chemokine function. A range of in vivo studies with mouse models of allergic disease have demonstrated the benefits of blocking chemokines³³ (reviewed in ref. 2); however, the range of chemokine functions, particularly those in lymphocyte homeostasis, suggests that targeting receptors may be a more effective therapeutic prospect. A neutralizing mAb against CCR3 blocks chemotaxis and calcium flux induced by all CCR3 ligands in human eosinophils in vitro.⁷² Moreover, a neutralizing monoclonal anti-CCR3 antibody was shown to inhibit eosinophil recruitment to the skin in an in vivo guinea-pig model.⁷³ However, the effect of this antibody on allergic pulmonary inflammation has yet to be determined.

Viral antagonists

Another potential source of chemokine antagonists comes from the observation that many viruses use chemokine antagonists to subvert immune responses.⁶ Chemokine homologues such as vMIP-II were probably pirated by viruses for broad antagonistic activity. vMIP-II is encoded by the Kaposi's sarcoma herpes virus (HHV8).^74,75 This viral chemokine antagonizes many of the Th1-associated receptors such as CCR1, CCR2 and CCR5, but stimulates Th2-associated receptors such as CCR3 or CCR8.⁷⁶ Other viruses use membrane-expressed chemokine receptor homologues, such as US28, a protein encoded by cytomegalovirus, to soak up chemokines to suppress host responses.⁷⁷ A noteworthy feature of most viral chemokines or chemokine-binding proteins is their broad chemokine or receptor-binding capabilities, which suggests that viruses need to circumvent chemokine redundancy for effective immune subversion.

Conclusion

As our characterization and understanding of the chemokine family of ligands and receptors has grown, so has our appreciation of its complexity and pleiotropy. From a first classification as a group of cytokines specializing in leucocyte recruitment, the role of the chemokine family has expanded to include maturation, differentiation, homing, activation and homeostatic trafficking of leucocytes within the immune system and in response to inflammation. One of the particular features of the chemokine family that makes these molecules particularly attractive as therapeutic targets is their specificity. Unlike the pleiotropic effects of cytokines, chemokines target specific leucocyte subtypes. Thus, an agent designed to target a particular chemokine/receptor interaction would be expected to have a physiologically limited effect, with a reduced set of side-effects. The challenge in the future is to design specific and selective chemokine receptor antagonists. We are optimistic that a greater understanding of the spatial patterns of chemokine expression and function during allergic disease will enable us to design novel therapeutic strategies to limit the pathophysiological side-effects of allergen exposure.