CD154 expression has been demonstrated to identify an antigen-specific T cell population following short-term stimulation with pathogen-derived antigens [1, 2]. The method is technically straightforward, is not HLA-restricted and is compatible with intracellular staining for cytokine expression, making it an attractive alternative to tetramer staining in some situations. We recently presented the first detailed description of CD154+ T cell responses following birch allergen stimulation in allergic and non-allergic subjects . In this study, we explore the frequency and phenotype of T cells responding to grass pollen – a true seasonal allergen – and cat allergen, a perennial allergen. The findings are confirmatory of previous reports, and include: a higher frequency of T cells responding to cat dander compared to grass pollen allergen, a higher frequency of late-differentiated CD27- T cells responding to grass pollen compared to cat dander allergen; a mixed T cell responses to allergens, but with a higher Th2: Th1 ratio in sensitised subjects. We also demonstrate a close correlation between the frequency of Th2 cells responding to cat dander allergen and the cat allergen-specific IgE level and show that the assay system cannot differentiate between asymptomatic sensitisation and allergic disease.
The allergen-specificity of T cell responses in our experiments have not been confirmed, which is a weakness of the study. The CD154+ population in our experiments is far greater than the allergen-specific cell frequency reported using other methods , and most likely represents an enriched population. The vast majority of our positive findings relate to the CD154/cytokine double-positive population, the frequency of which is more in keeping with previous literature. Demonstrating allergen specificity in this assay system is not straightforward: intracellular staining for CD154 precludes live cell sorting; specificity for pathogen-reactive T cells has been demonstrated using a modified method that utilises extracellular CD154 staining , however, the frequency of responding cells following allergen stimulation is small, and cloning experiments – with attendant limitations – would be required. A previous study with CD154/tetramer co-staining after prolonged cell culture followed by re-stimulation questioned the validity of CD154 as a marker for allergen-specific T cells (ref). However, these conditions are known to invalidate the assay, which is only reliable after short-term stimulation .
Although the allergen-specificity of the T cell responses in this assay system have not been formally demonstrated, our previous paper highlights important supportive information: the frequency of Th2 cells responding to birch allergen correlates with the levels of birch-specific IgE, providing a direct link between a cellular parameter that purports to be antigen-specific and a serological parameter that is definitely allergen-specific; the Th2: Th1 ratio after PBMC stimulation with a control allergen to which an atopic subject was not sensitised was low; the frequency of T cell responding to birch allergen increased during the birch season . Confirmation of a strong correlation between Th2 cells responding to cat dander and levels of cat-dander IgE in this paper provides further support, in addition to confirmation of phenotypic data established using MHC class II tetramers. In keeping with our previous studies of responses to birch allergen  and other studies [7, 8, 11], we found that responses to cat and grass allergens were very mixed, with activated Th2, Th1 and Tr1-like lymphocytes identified in varying proportions in allergic and non-allergic participants. Although the frequency of Th2 cells was significantly higher amongst allergic participants compared to non-allergic controls, there was considerable overlap, and the frequency of Th1 cells did not differ significantly. The overlap of Th2 responses to cat allergen was particularly marked; notably, a modified Th2 response to cat allergen in health has been reported by Platts-Mills et al. characterised by strong IL-4 and IL-10 production, albeit in highly-exposed subjects.
Using correlation analysis, we demonstrate a novel relationship between the frequencies of T cell subsets in grass and cat allergen tolerance, maintaining a low Th2: Th1 ratio and an appropriate frequency of IL-10+ T cells. This relationship was dysregulated in allergic patients, in whom the correlation between T cell subsets was weaker or lost and the Th2: Th1 ratio maintained at higher levels. This mathematical relationship between the frequency of IL4 expression and IFNγ/IL10 responses in health further supports what Akdis and colleagues described as a “fine balance between Th1, Th2 and Tr1-like responses” .
We also confirm important differences between T cell responses to cat allergen and grass pollen allergens. Studies with MHC class II tetramers suggest a higher frequency of circulating T cells responding to perennial compared to seasonal allergens : on the basis of the CD154 response, our data support this view, although the finding must be interpreted with caution as comparison is difficult when stimulating with different allergen extracts, even when the experiments are optimised. It has been hypothesised that responses to perennial allergens are directed at a broader range of epitopes compared to seasonal allergens, resulting in a larger composite signal [5, 13, 14].
Secondly, we demonstrate that the frequency of CD27- effector memory T cells within the Th2 compartment was significantly higher in grass-allergic participants, supporting the view that T cell responses to seasonal allergens comprise a higher proportion of late-differentiated T cells . In contrast, Th2 cells in cat-allergic individuals were consistent with an early-differentiated phenotype. A number of groups have demonstrated a predominantly early-differentiated phenotype in T cells responding to cat allergen using MHC tetramer experiments [4–8]. A recent study by Wambre and colleagues also demonstrated early- and late-differentiated T cell phenotypes for perennial (house dust mite) and seasonal (birch) allergens . It has been hypothesised that chronic allergen exposure could maintain a persistent central memory profile for perennial allergens. Wambre and colleagues investigated the differentiation stages of Aln g 1-specific T cells before and after specific immunotherapy for alder pollen allergy . Pathogenic responses, defined as Th2 cytokine production from CRTh2+ cells, were specifically associated with the late-differentiated CD27- T cells, dominating in allergic individuals but absent in healthy control subjects. Specific immunotherapy induced depletion of this CD27- Th2 cell population in favour of a predominantly CD27+ “protective” T cell response, therefore indicating the expression of CD27 can define distinct subsets of allergen-specific T cells. Similarly, Bonvalet et al. illustrated a reduction in CD27- Th2 cells following grass pollen-specific immunotherapy, although this did not correlate with clinical efficacy . Certainly, central memory T cells have been associated with persistent long-lived responses to antigens and in persistent viral infections. Literature investigating virus-specific lymphocytes hypothesised the nature and timing of antigen stimulation may influence the generation of memory T cells , in which chronic viral infection, such as cytomegalovirus, is associated with a central memory T cell phenotype . In addition, CD27 expression has been utilised to discriminate active and latent tuberculosis infection, in which the loss of CD27 expression marks active infection, compared to latent, chronic infection exhibiting high CD27 expression .
Asymptomatic sensitisation – where IgE results are positive in the absence of a clinical history of symptoms upon exposure – is an important issue in clinic. An assay capable of discriminating the two would be highly desirable, particularly in the settings of food, drug and venom allergy. Unfortunately, the Th2:Th1 ratio and other T cell-based markers in this assay were not able to discriminate subjects with asymptomatic sensitisation and true clinical allergy to grass, birch and cat allergens. It is possible that more refined assays with a greater number of markers could have improved predictive value, or it may be that T cell responses are too upstream of early phase symptoms to be useful.