The ability to accurately quantify antigen-specific responses has become integral to the study of infectious diseases, vaccine responses, autoimmunity and allergy . While a variety of techniques exist to detect T cells that recognize specific epitopes and to analyse the effects of antigen-induced activation on T cell phenotype, many of these methods require sophisticated equipment and significant expertise to perform. An alternative strategy has been the measurement of cytokines produced by antigen-activated T cells, in particular IFNγ, which has been demonstrated to correlate with functional immunity in a range of clinical settings [22, 23].
ELISpot assays have become the standard method to quantify IFNγ production in many laboratories because they are easy to perform, do not require investment in expensive equipment (e.g. flow cytometers) and are sensitive for the detection of even low numbers of antigen-specific cells . However, ELISpot assays have several important limitations: relatively large volumes of blood are required to obtain sufficient PBMCs for analysis; the process of extracting and purifying PBMCs may cause their activation and alter their phenotype [19, 20]; assays cannot be re-run unless stored samples of cells are available.
Due to the high burden of infectious diseases in the developing world, much of the research into T-cell responses to pathogens is focussed on areas where there are limited resources for the extraction and purification of PBMCs and where sample volumes may be restricted by patient age or co-morbidity. In these settings an assay that requires only small volumes of blood and which retains sensitivity despite delays in processing would provide significant benefits. In this study we have described a method to assess antigen-specific T cell responses ex vivo by measuring the up-regulation of MIG and IP10 gene transcription in mononuclear cells, in response to IFNγ released by activated T cells (Figure 1). We analysed responses to both chemokines as a means of internally validating our results and as expected the correlation between MIG and IP10 induction was high. Interestingly, the induction of IP-10 appears to be a more sensitive marker in the presence of low numbers of antigen-specific T cells, possibly reflecting activation of IP-10 gene transcription by cytokines other than IFNγ (Figure 2) .
In contrast with previous studies that have evaluated antigen-specific responses through changes in T-cell cytokines, we have found the monokine assay to have several advantages. The biological amplification of the IFNγ signal in monocytes means that co-stimulatory antibodies are not required ; responses remain stable despite delays in processing (Figure 4), or where samples sizes are small (Figure 5) and sensitivity equivalent to ELISpot is achievable . Furthermore, by enabling the measurements to be made on whole blood, preparation time is reduced and results may be more representative of the response in vivo [28, 29] (Figure 8).
As the assessment of antigen-specific responses in the monokine assay is indirect, one limitation of the assay is its inability to distinguish between a small number of highly activated T cells and a larger number of less activated T cells producing the same amount of IFNγ. Although this necessarily reduces the strength of the relationship between monokine and ELISpot assay results, overall the correlation between these assays is good and significantly, non-responders are clearly identified (Figure 6). Additional benefits of the monkine assay are that by measuring the effects of IFNγ on target cells, rather than the production of IFNγ alone, the monokine assay provides some confirmation of the downstream functionality of the T-cell response and once RNA has been extracted and reverse transcribed, samples can be stored indefinitely.