The development of allergic diseases, such as allergic asthma, depends upon the initiation and maintenance of T-helper cell type-2-skewed allergen-specific immune reactions. Although it is clear that susceptibility to this process is under genetic and environmental control, the fine-tuning and regulation of the type-2 T-helper cell immune response is not yet fully understood. In this second article in the present series, current understanding regarding the involvement of T-cells and antigen-presenting cells is summarised, with emphasis on the interaction between these two types of immune regulatory cells by means of co-stimulatory molecules.
MINI-SERIES “T-CELL CO-STIMULATORY MOLECULES”
Edited by M. Belvisi and K.F. Rabe
Number 2 in this Series
SPECIFIC CD4+ T-CELL SUBSETS CONTRIBUTE TO ALLERGIC AIRWAY DISEASE DEVELOPMENT
CD4+ T-cells play a central role in the initiation and maintenance of the allergic airway reaction. By expression of major histocompatibility complex (MHC) class II molecules and allergen-specific T-cell receptors (TCRs) they link innate and adoptive immune responses. They induce the synthesis of allergen-specific immunoglobulin (Ig)E and recruit and activate effector cells, such as eosinophils, via the secretion of soluble factors. Their pivotal role in controlling the early and late asthmatic reaction was convincingly demonstrated in human patients, as well as in animal models of allergic airway inflammation. Activated T-cells were found in increased numbers in the bronchial tissues of asthmatic subjects 1, 2, and depletion of CD4+ T-cells in a mouse model of allergic airway disease impressively prevented both the development of airway hyperresponsiveness (AHR) and the infiltration of eosinophils into the airways 3.
T-cells were originally subdivided, according to their functional properties, into two polarised subsets 4. Type-1 T-helper cells (Th) predominantly produce the cytokines tumour necrosis factor (TNF)-β, interferon-γ and interleukin (IL)-2 and stimulate a strong cell-mediated immune response, particularly against intracellular pathogens. In contrast, Th2 have the capacity to secret greater amounts of IL-4, -5, -9 and IL-13, and are therefore crucial for allergic immune reactions.
The first report on the specific cytokine profile of T-cells involved in the pathogenesis of human asthma was published in the early 1990s, demonstrating that T-cells in bronchoalveolar lavage fluid (BALF) from asthmatic patients predominantly produce the Th2-type cytokines granulocyte-macrophage colony-stimulating factor and IL-3, -4 and -5 5, 6. The functional importance of these cells for the development of allergic airway inflammation was further persuasively demonstrated by analyses in rodents. Mice lacking pivotal elements required for the induction of Th2 immune reactions, e.g. signal transducer and activator of transcription 6 7, 8, failed to develop symptoms of allergic airway inflammation. Similarly, athymic mice with no functional peripheral T-cells also lack eosinophilic airway infiltration and increased AHR following allergen sensitisation, unless production of the Th2 cytokine IL-5 is restored by systemic administration 9. Although the involvement of Th2 cytokines explains the biological basis of many features of the allergic reaction (e.g. recruitment of eosinophils, activation of mast cells, airway remodelling and IgE production), it has been observed more recently that asthmatic individuals may also exhibit an increased Th1-like response to allergens 10, 11, especially in established disease 12–15. Conversely, acute segmental allergen challenges lead to a reduction in Th1 cytokine-producing T-cell numbers 16. In summary, the dogma of the allergic reaction being purely a Th2-driven disease, especially as an attempt to explain asthma pathology, is questioned; however, the important role of Th2 cytokines, such as IL-4, -5 and -13, has been demonstrated by a large body of evidence.
The important role of the various Th2 cytokines in the development of allergic airway inflammation was further demonstrated by analysis of the pathology of mice overexpressing certain Th2 cytokines, by use of cytokine-blocking agents or genetically deficient mice unable to produce specific cytokines (table 1⇓). In conclusion, these data clearly underline the importance of the Th2 subset in the pathogenesis of allergic (airway) diseases.
Encouraged by these experimental in vitro and in vivo data, trials were performed with the aim of therapeutically targeting Th2 cytokines in human subjects with persistent asthma (Global Initiative for Asthma stage 2 or worse). Single intravenous application of a humanised monoclonal antibody (mAb) directed against IL-5 (SB-240563/mepolizumab) in patients with mild persistent asthma significantly reduced the numbers of eosinophils in the peripheral blood and sputum, but had no significant effect on the late asthmatic response, or on nonspecific AHR to histamine 37. Later, in patients with mild/moderate asthma undergoing a prolonged protocol of repeated antibody treatment, it was shown that, although eosinophils were nearly wiped out in the periphery, signs of eosinophil activation and degranulation were still present in the lung tissues, probably accounting for the lack of efficacy 38. In another trial, in patients with severe persistent asthma, with a different humanised mAb directed against IL-5, SCH55700, numbers of circulating eosinophils were also reduced, but, besides a small increase in baseline forced expiratory volume in one second (FEV1), there was no effect on other clinical indices of disease activity 39. In mild-to-moderate asthmatic subjects, application of a recombinant soluble IL-4 receptor by inhalation during dose-reduction and withdrawal of inhaled corticosteroids prevented a decline in FEV1 or an increase in asthma symptoms 40, 41. Since there was a lack of further evidence of efficacy, this strategy has since been discontinued, as have those with an anti-IL-5 approach.
These rather frustrating outcomes of trials in which the blockade of a single Th2 cytokine was attempted may be explained in various ways: 1) the applied compounds did not sufficiently deplete and antagonise eosinophils and IL-4, respectively; 2) eosinophils might not be totally required or even important in the clinical manifestation of asthma 42; 3) other trigger factors, such as viral infections, exercise or airway pollution, might account for asthma pathology and/or exacerbation; and 4) depletion of a single cytokine is not sufficient, due to the high redundancy of effector function of a wide variety of cytokines 38. It is, therefore, tempting to speculate that elimination of the entire Th2 function or the complete prevention of differentiation towards the Th2 type may overcome the problem of redundant cytokine function (discussed in the next article in the present series 43). Also, different treatment protocols, including early and/or prolonged intervention and the use of antibody cocktails aimed at a variety of cytokines, might result in increased efficacy in asthma therapy.
Furthermore, there are new data showing that strategies targeting T-cells and/or T-cell products that are not linked with a Th1/Th2 bias may be effective in the treatment of asthma. Selective blockade of TNF-α using the TNF receptor-2 Fc fusion protein etanercept, as well as the application of methotrexate, to suppress TNF-α as well as antibody production, led to a reduction in asthma severity 44, 45. In summary, these data are in line with a concept of impaired immunostasis being the main reason for asthma development, with unwanted immune reactions of allergen-specific CD4+ T-cells due to the lack of regulatory T-cell (Tr) functions inducing the allergic cascade, regardless of Th1 or Th2 bias 46.
ALLERGIC AIRWAY INFLAMMATION REGULATION BY SPECIFIC T-CELLS
The term regulatory T-cell (Tr) refers to cells that actively control or suppress the function of other cells, generally in an inhibitory fashion. Tr appear to control the development of autoimmune disease and transplant rejection, and may also play a critical role in controlling the expression of asthma and allergy. Their specific characterisation, as well as their mode of action, is still undergoing intense investigation 47. However, different Tr subtypes with certain attributes and functions may be described.
Naturally occurring Tr are positively selected in the thymus by encountering self antigen. They constitute 5–10% of CD4+ T-cells in the periphery in both mice and humans. These cells constitutively express CD25 (IL-2 receptor), as well as the transcription factor forkhead box protein (Fox) P3 48–50, which acts as a specific factor for the differentiation of the Tr lineage and closely correlates with the suppressor activity of the cells 51.
The precise function of FoxP3, which until recently was thought to be expressed only by natural CD25+ Tr, is not known, but absence of FoxP3 in humans results in immunodysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome 52, 53, also know as X-linked autoimmunity–allergic dysregulation syndrome 54. Patients with IPEX syndrome share many phenotypic features with scurfy mice, which have a natural mutation in the gene encoding FoxP3, and lack natural CD25+ Tr. Scurfy mice develop diarrhoea, malabsorption, autoimmune haemolytic anaemia, thrombocytopenia, leukocytosis, lymphadenopathy, hepatosplenomegaly, hypogonadism and dry skin, which are all corrected by adoptive transfer of natural CD25+ Tr into these mice 55. In addition, forced expression of the FoxP3 gene, which encodes a transcription repressor, converts naive murine T-cells into Tr that phenotypically and functionally resemble natural CD25+ Tr.
Although the absence of FOXP3 expression in patients with IPEX syndrome and scurfy mice eliminates development of natural CD25+ Tr, whether IPEX syndrome patients and scurfy mice also lack adaptive Tr that differentiate in the periphery from CD25- T-cells is not yet clear. However, since patients with IPEX syndrome develop eczema, food allergy, elevated IgE levels and peripheral eosinophilia, associated with elevated Th2 cytokine production 54, the present authors suggest that IPEX syndrome patients lack both natural CD25+ Tr and antigen-specific adaptive Tr, which may be particularly effective in regulating Th2 responses to allergens.
In mice, natural CD25+ Tr have been shown to limit allergen-induced airway inflammation but not the development of airway hyperreactivity 56. In these experimental settings, pre-activation of natural Tr by nonspecific signals appears to be required in order to obtain strong inhibition of allergen-induced airway inflammation 57.
In children who have outgrown cow’s milk sensitivity, an increased frequency of circulating natural Tr is associated with decreased in vitro proliferative responses to the specific allergen bovine β-lactoglobulin 58. Natural CD25+ Tr from nonallergic, but not from allergic, donors suppress proliferation and IL-5 secretion by allergen-stimulated CD4+CD25- effector T-cells, indicating that natural Tr may also suppress allergic immune responses to inhaled allergens, e.g. cat allergen or grass-pollen 59. Interestingly, in allergic donors, the inhibitory activity of CD25+ Tr on the production of Th2 cytokines was most suppressed when pollen counts were highest during the specific season. Accordingly, natural Tr from both allergic and nonallergic individuals suppressed T-cell proliferation and Th2 cytokine production in response to birch allergen most potently outside, but not during the birch pollen season 60.
Although it is clear that natural Tr are positively selected in the thymus after encounter with self antigens, it is likely that allergen-specific adaptive Tr develop in the periphery after encounters with exogenous allergens (e.g. from food or plants). Adaptive Tr do not have clearly defined markers that would make them easily identified or isolated; thus their further characterisation is difficult to accomplish.
Several studies have analysed the significance of co-stimulatory pathways in the generation and expansion of Tr. First, it was demonstrated that rat Tr respond less well than CD25- T-cells to conventional co-stimulation, but are readily expanded in vitro by superagonistic CD28-specific mAbs, which are potent mitogens for all T-cells without the need for TCR engagement 61. In vivo, functional Tr are preferentially expanded over other T-cell subsets by CD28 stimulation. These data suggest that CD28-driven activation of Tr may be effective in the treatment of inflammatory and autoimmune diseases, but side-effects should be carefully taken into account 62.
In addition, co-stimulation of CD4+ T-cells with anti-CD52 mAb leads to the development of Tr that suppress the polyclonal responses of CD4+ T-cells 63. These Tr had the potential to suppress graft-versus-host disease-like pathology in severe combined immunodeficiency syndrome mice injected with human peripheral blood mononuclear cells. It was suggested that anti-CD52-induced Tr, which can be expanded and which confer antigen specificity, have the potential to be a desirable tool in cellular immunotherapy.
T-CELLS SHOW SPECIFIC DIFFERENTIATION AND MIGRATION IN ALLERGIC AIRWAY INFLAMMATION
The differentiation of T-cell function depends upon the recognition of a specific peptide presented by antigen-presenting cells (APCs) in combination with additional co-stimulatory signals, as described in the first article in the present series 64. In recent years, a variety of studies have broadened knowledge of the cellular interaction leading to T-cell differentiation in allergic airway inflammation.
The allergic immune response is initiated by pulmonary dendritic cells (DCs), derived from blood monocytes with low granulocyte-differentiation antigen (Gr)-1 and high CX3C chemokine receptor 1 densities 65. They are located underneath the basal membrane of the airway epithelium, where they encounter and incorporate allergens and migrate to the draining thoracic lymph nodes (fig. 1⇓) 66. As soon as the DCs carrying the allergen peptides arrive in the T-cell area of the local lymph node, they stimulate specific naive or memory T-cells by increasing expression of co-stimulatory and MHC class II molecules. Under conditions in which allergen exposure is accompanied by an inflammatory stimulus, differentiation towards functional effector T-cells, Tr or memory T-cells is induced 67. Activated effector T-cells leave the lymph node and enter the pulmonary tissues, the site of further antigen deposition, where they are retained in a nonproliferative state, recognising antigens presented by local DCs 68. They further enhance the recruitment and maturation of other immune cells involved in the allergic cascade, such as eosinophils and mast cells, via the secretion of cytokines and chemokines.
IMPACT OF INNATE IMMUNITY ON T-CELL DIFFERENTIATION AND DEVELOPMENT OF THE ASTHMA PHENOTYPE
The innate immune system links environmental factors with the specific responses of adaptive immunity. A family of evolutionarily highly conserved receptors, known as Toll-like receptors (TLRs), are expressed by APCs and recognise pathogen-associated molecular patterns. This interaction leads to secretion of different cytokines by the APCs, including IL-12, particularly promoting the development of Th1 69. The epidemiological observation that the modified exposure to microbial particles in westernised areas is associated with a reduced prevalence of atopic diseases, the so-called “hygiene hypothesis”, may thus be explained by the reduced production of IL-12 shifting the immune responses to a predominantly Th2 type. This hypothesis was supported by recent findings that genetic variations in the TLR2, TLR4, TLR6 and TLR10 genes were associated with atopy in children 70–74.
Similar to TLRs, CD14 is a pattern recognition receptor for certain microbial molecules and expressed by APCs. Associations of different polymorphisms within the CD14 gene and cardinal features of atopic diseases, such as IgE levels 75, eczema 76, asthma severity 77 and bronchial hyperresponsiveness 78, were identified. In contrast, other studies did not replicate these associations 72, 79, possibly due to genetic variations in different study populations or differences in the kinetics of the microbial exposure 80.
CO-STIMULATORY SIGNALS REGULATE ALLERGIC AIRWAY INFLAMMATION
The development of allergen-induced airway inflammation and AHR is regulated by fine-tuning of several co-stimulatory factors. The first co-stimulatory signal for activation of naive T-cells is delivered by the interaction of CD28 with its ligands expressed on DCs, CD80 (B7.1) and CD86 (B7.2) (fig. 1⇑) 81. This interaction induces the expression of anti-apoptotic molecules of the B-cell leukaemia/lymphoma gene product (Bcl) family, permitting the clonal expansion of the activated T-cell 82 and inducing the production of cytokines IL-2, -4 and -5. It is currently a matter of debate as to whether the differential ligation of CD28 with either CD80 or CD86, respectively, favours differentiation into Th1- or Th2-type T-cells 83, 84. Certainly, the cytokine milieu at the time of antigen presentation is crucial to the type of T-cell differentiation; increased production of IL-12 by DCs favours the development of Th1-mediated cellular immune responses, whereas the preferential synthesis of IL-4 and -6 by DCs that are localised at mucosal surfaces (DC2 cells) may trigger polarisation towards the Th2 phenotype 85, 86.
The role of CD28 in the pathogenesis of human asthma was studied using T-cells obtained from BALF and bronchial biopsy specimens from asthmatic patients 87–90. Interestingly, the proliferation of human BALF-derived T-cells from patients with atopic asthma after in vitro stimulation with allergen was inhibited by the addition of anti-CD86, but not of anti-CD80 mAb 90, underscoring the critical role of CD86 in allergen-induced T-cell (Th2) responses.
The activation of T-cells through CD28 is negatively regulated by the inducible molecule cytotoxic T-lymphocyte antigen (CTLA)-4. Accordingly, inhibition of CD28 co-stimulation by means of CTLA-4–Ig suppressed the production of IL-5, -13 and -16 and RANTES (regulated on activation, normal T-cell expressed and secreted) upon allergen-specific in vitro stimulation by T-cells isolated from bronchial biopsy specimens from asthmatic subjects 88, 89. A more detailed analysis of these experiments revealed that allergen-mediated Th2-type cytokine production in bronchial biopsy specimens was dependent upon signalling via both CD80 and CD86 87.
In human bronchial asthma, an association between certain single nucleotide polymorphisms (SNPs) within the CTLA-4 gene and increased total serum IgE levels (+49A/G 91), bronchial hyperresponsiveness (BHR) (-1147C/T and +49A/G 92) and asthma severity (-318C/T 92) was demonstrated. In two other cohorts, no association of the +49A/G and -318C/T polymorphisms and the development of the asthma phenotype was found 93, 94.
The critical role of CD28 signalling in the development of allergic airway diseases was further confirmed in mice deficient in this co-stimulatory pathway. Mice deficient in CD86 developed less airway inflammation in response to allergen sensitisation and airway challenges than their normal littermates 95. In the same line, CD28-deficient or CD80/CD86 double knockout mice showed virtually no signs of allergic airway inflammation (e.g. production of IgE, eosinophilic airway inflammation and development of AHR) following allergen sensitisation and airway challenges (table 2⇓) 95, 97. Mice deficient in CTLA-4 developed a lethal lymphoproliferative disease, which reflects the importance of this molecule in the restriction of T-cell-mediated immune responses 109.
After initial T-cell activation, effector T-cell function seems to be less dependent on the engagement of CD28 110, 111. Other co-stimulatory molecules, which are only expressed upon T-cell activation, take over and seem to play the dominant role in the control of the ongoing immune reaction.
A prominent member of this second line of T-cell control is inducible co-stimulatory antigen (ICOS). Initial studies on its functional properties demonstrated that co-stimulation via ICOS mainly resulted in high expression of IL-10 and -4 and minor production of other Th1 and Th2 cytokines during the initial priming and effector T-cell responses 112–114. It was further demonstrated that the production of certain cytokines correlated with the expression density of ICOS on the cell surface 115. Intermediate ICOS expression was associated with high production of Th2 cytokines, whereas high levels of ICOS predominantly translated into high IL-10 production. In line with these findings, it was shown that mature pulmonary DCs in the bronchial lymph nodes of mice exposed to respiratory allergen induced the development of Tr, in a process that required T-cell co-stimulation via the ICOS−ICOS ligand (ICOS-L) pathway. The Tr produced IL-10 and showed potent inhibitory activity; when adoptively transferred into sensitised mice, Tr blocked the development of AHR. Both the development and the inhibitory function of regulatory cells were dependent upon the presence of IL-10 and on ICOS−ICOS-L interactions. These studies demonstrate that Tr and the ICOS−ICOS-L signalling pathway are critically involved in respiratory tolerance and downregulating pulmonary inflammation in asthma 116, 117.
Other authors have shown that adoptively transferred ICOS+ cells were capable of inducing airway inflammation upon subsequent allergen airway challenges, highlighting the essential role of ICOS in the development of allergic airway inflammation 115.
One study showed an association of SNPs within the ICOS gene (-693A/A and -1413A/A) with increased total IgE and allergic sensitisation to airborne allergens in a Hutterite population. Furthermore, the -1413 SNP is located at a nuclear factor-κB-binding site and peripheral blood mononuclear cells homozygous for the A allele (-1413A/A) produce significantly greater amounts of the cytokines IL-4, IL-5 and IL-13 and TNF-α. This points towards a role of ICOS in allergic diseases in humans 118.
Mice lacking ICOS or ICOS-L expression are strongly impaired in T-cell IL-4 production, resulting in impaired production of antigen-specific class-switched antibodies, such as IgG1, IgG2a, IgA and, especially, IgE (table 2⇑) 100, 101, 103. Interestingly, ICOS-deficient mice still develop profound IL-5-mediated eosinophilic airway inflammation upon allergen sensitisation and subsequent airway challenges, indicating that ICOS may not primarily be involved in global Th2 differentiation, but is rather required for the expression of IL-4 in the effector phase 100, 119.
Another inducible co-stimulatory molecule involved in the pathogenesis of the allergic airway reaction is CD134 (OX40). This molecule is mainly expressed by activated Th2. OX40 is required for the cytoplasmic expression of the transcription factors Bcl-xL and Bcl-2, which deliver anti-apoptotic signals for pre-activated T-cells 120. These signals are essential for the accumulation of T-cells during the primary immune response, as well as during the subsequent formation of a memory cell pool. In mice lacking OX40 signalling, strong impairment of allergic airway inflammation, with significantly reduced development of AHR, Th2 cytokine production and IgE synthesis, was observed (table 2⇑) 105, 107.
Co-stimulation is also an effective means of downregulating T-cell functions. Signalling through the negative co-stimulatory molecule, programmed cell death (PD)-1, leads to arrest in cell cycle phase G0–G1, thus limiting an ongoing immune reaction 121. The importance of PD-1-mediated T-cell suppression for peripheral tolerance was supported by the observation that PD-1-deficient mice developed autoimmune diseases, e.g. cardiomyopathy 122, autoimmune glomerulonephritis and arthritis 123.
In a murine model of airway inflammation, allergen sensitisation and airway challenge lead to increased expression of PD-1 ligand 1 (PD-L1) on pulmonary DCs, macrophages and B-cells isolated from lung tissues 124. Despite this abundant expression of PD-L1, this molecule seems not to be involved in the control of allergic airway inflammation, since blockade by mAb did not alter the allergic phenotype 124. In contrast, another PD-1 ligand, PD-L2, was expressed only in very low amounts by lymphocytes of unchallenged mice, with only moderate increases after allergen challenge. However, blockade of this specific ligand significantly inhibited the development of AHR, lung eosinophilia and production of Th2 cytokines 124. This gives PD-L2 a so-far unique role in the negative regulation of allergen-induced airway responses.
A complex network of enhancing and inhibitory co-stimulatory signals regulates T-cell differentiation and effector functions. Owing to their central role in allergen-mediated airway inflammation, T-cell co-stimulatory molecules constitute promising targets for novel therapeutic interventions. The state of the art regarding experimental data on the use of these signals in modulating allergen-induced T-cell responses, and the outlook for future approaches targeting co-stimulation for intervention in allergic airway diseases is discussed in more detail in the third and final part of the present series 43.
Previous articles in this series: No. 1: Beier KC, Kallinich T, Hamelmann E. Master switches of T-cell activation and differentiation. Eur Respir J 2007; 29: 804–812.
- Received July 18, 2006.
- Accepted January 6, 2007.
- © ERS Journals Ltd