Asthma: T-bet – A Master Controller?

Abstract

The transcription factors T-bet and GATA3 are important reciprocal determinants of Th1 and Th2 T helper cell differentiation. Recent evidence suggests that these factors may affect airway immunopathology in asthma.

Asthma remains one of the commonest chronic inflammatory diseases and has a major impact on the life of sufferers. It is associated with allergy mediated by IgE antibodies, eosinophilic airway inflammation and remodelling including collagen deposition beneath the epithelial basement membrane, mucus secretion and increased airway smooth muscle. These changes are thought to contribute to both the episodic airway narrowing that defines the condition and to airway hyperresponsiveness – an increased tendency to narrow the airways to irritant stimuli [1]. T helper cells are divided into subsets: Th1, producing predominantly interferon γ (IFNγ), and Th2 producing interleukin-4 (IL-4) and IL-5 but not IFNγ. IL-4 switches B cells towards IgE synthesis and IL-5 is an eosinophil survival and development factor [1,2]. Evidence is accumulating to support the view that the interplay between these T helper cell subsets is important in asthma. Recent studies have shown that Th2-type cytokines are produced by airway T cells in human asthma, and that allergen-reactive Th2 cells can be isolated from the airway of allergic asthmatics [1].

Th1 and Th2 cells develop from naïve T cells during an immune response, and largely determine the nature of that response through the profile of cytokines they secrete. This process involves a complex pattern of interacting cytokines, transcription factors and signalling pathways [2]. Szabo et al. [3] defined one such transcription factor which they termed T-bet (T box expressed in T cells) which not only induced an IFNγ-producing Th1 phenotype, but also repressed IL-4 and IL-5 production from differentiated Th2 cells. In a recent paper [4], these workers go on to show evidence for decreased numbers of CD4⁺ T cells expressing T-bet in the airway of human asthma patients relative to control subjects. Moreover, deletion of the T-bet gene in mice resulted in airway eosinophilia, Th2 cytokine production, airway hyper-responsiveness and changes of airway remodelling in the absence of allergen sensitisation and inhaled challenge [4]. How do these findings relate to other published work in terms of Th1 and Th2 development and how are they relevant to our understanding of asthma?

Differentiation of Th1 and Th2 Cells

The major determinant of whether an activated naïve CD4⁺ T helper cell develops into a Th1 or Th2 effector/memory cell is the presence of IL-12 or IL-4, respectively. In addition, IFNγ and IL-18 can potentiate Th1 development. The molecular events that programme the Th1 or Th2 phenotype have been extensively studied in vitro and by gene deletion in mice [2]. Immediately after T cell receptor activation, naïve T cells can transcribe messenger RNA for both IFNγ and IL-4 [5], but cytokines or other signals are required to direct Th1 or Th2 polarisation. IL-12 acts through signal transducer and activator of transcription 4 (Stat4) to increase IFNγ production [2]. IFNγ (IFNα in human T cells) maintains expression of the IL-12 β2 receptor chain (IL-12Rβ2), whereas IL-4 activates Stat6 to downregulate IL-12Rβ2 [2]. Deletion of the Stat4 or Stat6 genes results in in vivo deficiency of Th1 or Th2 responses, respectively, but the precise interaction of these STAT proteins with other transcription factors and cytokine gene promoters remains uncertain. Transcription factors interacting with the IL-4 promoter have been shown to be expressed in Th2 cells but not Th1 cells, including c-Maf, NIP45 and JunB, all of which increase IL-4 gene activation, as does NFATc. Gene deletion or overexpression of these factors correspondingly affects IL-4 production [6]. Th2 differentiation involves chromatin remodelling to allow access of these factors to the IL-4 promoter.

GATA3 and T-bet

Recently, two transcription factors have been defined that can control Th1 or Th2 differentiation and override previously programmed cytokine patterns. GATA3 is expressed in Th2 cells but not Th1 cells, and transgenic mice overexpressing GATA3 produced Th2 cytokines even when cultured in Th1-polarising conditions in the absence of IL-4 [6]. Retroviral transfer of GATA3 into Th1 cells could reduce IFNγ production and induce IL-4 and IL-5 [7,8]. Thus not only can GATA3 induce Th2 cytokines, but it can also repress Th1 cytokine production.

T-bet has been suggested to act in a similar fashion for Th1 cells. T-bet was rapidly induced in naïve CD4⁺ T cells cultured in Th1-polarising conditions (in the presence of IL-12 and anti-IL-4 antibodies), and trans-activated a reporter gene driven by the IFNγ promoter. T-bet expression is restricted to Th1 and not Th2 cells [3]. Retroviral gene transfer of T-bet could make Th2 cells activate IFNγ production and repress IL-4 and IL-5. T-bet also induced other Th1 phenotypic changes in developing Th2 cells: expression of the CCR5 chemokine receptor and upregulation of the P-selectin ligand PGSL-1. Furthermore, repression of Th2 cytokines (and IL-2) was not IFNγ dependent as it occurred even in T cells deficient in the IFNγ receptor. How T-bet reduces IL-4 and IL-5 expression remains unclear, as it did not affect expression of a reporter gene driven by the IL-4 promoter [3].

Although it was suggested that T-bet expression may be stochastic [9], it has recently been shown that T-bet is upregulated in lymphoid and myeloid cells by IFNγ in a Stat1-dependent manner within 6 hours, and this early T-bet activation is Stat4 independent [10]. However, Grogan et al. [5] showed that at 72 hours T-bet upregulation in Th1 was not seen in Stat4-deficient mice, though this could be an indirect effect via IFNγ [6]. GATA3 expression can occur in the absence of Stat6, but Stat6-dependent pathways upregulate expression [7]. Taken together these data suggest that initial Th1 differentiation is dependent on IFNγ-induced T-bet expression, whilst GATA3 will direct Th2 differentiation. Full expression of a polarised immune response may require expansion, selective survival and stabilisation of cytokine mRNA of Th1 cells via IL-12-induced Stat4, with amplification of IFNγ production by IL-18, or Th2 cells via IL-4-induced Stat6 [2]. A key part of Th1 development is retention of IL-12Rβ2 – and hence IL-12 responsiveness – whilst Th2 cells lose IL-12Rβ2. T-bet was found to induce IL-12Rβ2 expression [9], so that part of the phenotype resulting from T-bet deficiency may result from a loss of ability to retain IL-12 responsiveness. These control elements are summarised in Figure 1. Whether the same control mechanisms apply in human cells remains to be fully confirmed.

Figure 1. T-bet in Th1 and Th2 CD4⁺ T cell differentiation.

Naïve CD4⁺ T cells upregulate T-bet in response to IFNγ in a Stat1-dependent manner leading to upregulation of IFNγ production and repression of Th2 cytokines, together with expression of IL-12Rβ2, hence IL-12 responsiveness. Increased IFNγ production and Th1 commitment occurs through IL-12 acting via Stat4 (which is not required for but may augment T-bet expression) and by IL-18. IL-4 downregulates IL-12Rβ2 and GATA3 increases Th2 cytokine expression (which is further enhanced by c-Maf, NIP45, JunB and NFATc) and represses IFNγ production. Th2 cytokines are involved in B-cell switching to IgE production (IL-4 and IL-13), eosinophil infiltration (IL-5 and IL-9), and airway remodelling and mucus secretion (IL-13): all of which may contribute to airway hyperresponsiveness and other features of asthma. The absence or relative deficiency of T-bet favours spontaneous airway features of asthma, in part through favouring Th2 development.

GATA3 and T-bet in the Airway

How then do these Th1 and Th2 control factors act in the asthmatic airway? Finotto et al. [4] found relatively fewer T-bet-expressing CD4⁺ T cells in the airways of asthmatics than control groups, which fits with previous findings that T cells in asthmatic subjects are predominantly of a Th2 phenotype (which do not express T-bet) [4]. These authors went on to show that T-bet deficiency alone in mice could induce airway changes similar to those seen after allergen challenge of sensitised rodents.

Animals heterozygous for T-bet deficiency (T-bet^+/−) did not show any increase in IL-4, tumour necrosis factor α or transforming growth factor β in the airways compared with wild-type controls [4]. The heterozygotes did, however, show airway hyperresponsiveness, increased expression of IL-5 in the airway, and airway remodelling without challenge, similar to that seen in the homozygous deficient animals. In culture CD4⁺ T cells from the heterozygous mice were not different from wild type in terms of Th2 cytokine production in neutral conditions [11].

One interpretation of these finding in T-bet^−/− mice would be that Th2 cytokines may be basally repressed by T-bet, and that absence of T-bet allows their expression. Transgenic overexpression of GATA3 in T cells led to cells with characteristics of a Th2 phenotype, so the balance of T-bet and GATA3 may determine the default T cell phenotype [12]. Another possibility is that environmental antigens drive Th2 development in the absence of T-bet, because they cannot induce IFNγ (and IL-12Rβ2) and thus Th1 development is deficient. It will be of interest to know if T-bet^−/− T cells are IL-12 responsive. Thus the T-bet^−/− findings suggest that Th1 cells and IFNγ production may be a major regulator of basal Th2 responses. It has also been postulated that regulatory cells producing IL-10 attenuate Th2 responses to allergen. However, recent data suggest that these IL-10-producing regulatory cells express neither T-bet nor GATA3 [13], so these cells should still be active in the absence of T-bet.

It would be of some interest to know whether the Th2-mediated pathology in T-bet^−/− mice is restricted to the airway or whether there is ‘allergic’ inflammation at other sites. Sensitisation and inhaled ovalbumin challenge failed to induce a further increase in airway hyperresponsiveness in either T-bet^−/− or T-bet^+/− mice (in this model inhaled allergen will induce airway hyperresponsiveness in sensitised wild-type mice) [4]. It would also be of interest to know if specific IgE or ovalbumin-specific T cells were induced in the lungs after challenge. Since T-bet is expressed in non-lymphoid cells [10], it will be important to look for any direct effects in other cell types such as airway smooth muscle or epithelium.

Although overproduction of Th2 cytokines in the airway, like T-bet deficiency, can induce an asthma phenotype, it is of note that these changes can also be seen after sensitisation and inhalation challenge in mice deficient in IL-4, IL-5, both IL-4 and IL-5, or IL-9 [14,15]. Thus targeting a single (or several) Th2 cytokine in asthma may not be effective, and indeed recent human trials of monoclonal antibodies blocking IL-5 were disappointing although soluble IL-4 receptor had some effect [16,17]. Although human airway Th2 cells from asthmatics are still responsive to IL-12 in vitro [18], exogenous IL-12 had no effect on asthma symptoms or lung function despite reducing blood and sputum eosinophilia [19]. In animal models most workers report that transfer of Th1 cells into the airways can exacerbate Th2-mediated lung pathology, and Th1 cells can themselves cause disease [20], so switching on T-bet in asthma might not be beneficial.

The confirmation of T-bet and GATA3 as Th1 and Th2 control factors in vitro and in vivo is of great interest. Whether this will allow the therapeutic manipulation of the immune response that has been a goal ever since the description of the Th1/Th2 dichotomy is uncertain, but these are tantalising targets.