Flow cytometric analysis allows the simultaneous measurement of multiple cytokines from individual T cells after short-term stimulation with MTB antigens. Comparison of the percentages of T cells secreting all possible combinations of IL-2, IFN-gamma, and TNF-alpha after ESAT-6 or CFP-10 stimulation of PBMC from individuals with LTBI versus active MTB was performed. Higher percentages of CD4+ T cells that are “polyfunctional” triple producers (IL-2+, IFN-gamma+, TNF-alpha+) were found in the PBMC of LTBIs compared with active MTB subjects [17,19], with statistical significance in one study [17]. Also, longitudinal analysis of MTB-specific T cell responses from subjects with active disease (before and after treatment) found that the proportion of IL-2+, IFN-gamma+, TNF-alpha+ triple producing cells was higher, on average, after treatment [17,19]. Measurement of MTB-specific single-positive TNF-α+ CD4 T cells showed the best discriminatory potential for these two disease states in one study [17]. Also, there is a positive correlation of ex vivo proliferative capacity of MTB-specific T cells and the number of TNF-alpha+ IL-2+ IFN-gamma+ cells after MTB antigenic stimulation [19]. Similar to this flow cytometric data, two-color fluorescent Elispot assays for IFN-gamma and IL-2 found more dual producing cells in the LTBIs compared with active TB subjects; furthermore, this signature of IFN-gamma only cells in untreated MTB subjects with active disease shifted to dual IFN-gamma+ IL-2+ T cells in the months after treatment was initiated [20]. Taken together, these data suggest that similar to what is found in other infectious diseases [21,22], high bacterial load is congruent with functional and proliferative exhaustion of antigen-specific T cells, with MTB-specific T cells likely returning to a resting state after treatment and re-acquiring the ability to secrete multiple cytokines and proliferate when antigen is re-encountered.

While analysis of TNF-alpha and IL-2 in addition to IFN-gamma shows great potential to conclusively distinguish LTBI from active MTB subjects, either no differences in the profile of the IL-2+, IFN-gamma+, TNF-alpha+ responses to ESAT-6 or CFP-10 between active and LTBI subjects [23] or that the IL-2+, IFN-gamma+, TNF-alpha+ T cell frequencies were actually higher in the active TB subjects [24] has been reported as well. Use of recombinant proteins in re-stimulation cultures, length and nature of stimulation assays, and differences in patient cohorts and number of events collected in flow cytometric assays could account for why these results differ from other findings. Alternatively, it is possible that T cell functions from other effector lineages must be measured to reliably determine LTBI versus active MTB disease.

IL17 is a cytokine associated with inflammation and anti-viral responses, and is produced by the Th17 lineage of memory T cells. In MTB infection, the frequency of IL17-producing CD4+ T cells after PMA and Ionomycin stimulation was significantly lower in active MTB subjects compared to both LTBI and healthy donors [25]; the frequency of IL17+ T cells after BCG stimulation was also found to be lower in active MTB subjects compared with healthy controls, and IL17 was produced by T cells within PBMC of MTB subjects after ESAT-6/CFP 10 peptide stimulation [26] Also, IL-17 production by CD4+ T cells in response to CFP-10/ESAT-6 peptides was detected from pleural fluid samples from subjects with tuberculous pleurisy (TBP) [27]. This cytokine therefore may prove an important component of anti-MTB immunity and useful for detection of LBTI versus active MTB disease states.

While Th1 cytokines (IFN-gamma, TNF-alpha) and the Th17 cytokine IL-17 are well established as integral components of a successful immune response to infection, these functions can cause significant immuno-pathology to the host. Therefore, T cell effector functions with immunosuppressive roles, such as IL-4, IL-5, IL-10, IL-13, TGF-beta and the direct T cell inhibitory T regulatory cells (Tregs), are often a substantial and integral component of an antigen-specific immune response. TGF-beta and IL-10 are produced in response to PPD from PBMC of infected subjects ex vivo [28]. Cytometric Bead Array assays of culture supernatants after PBMC stimulation in Quantiferon^® Gold In-Tube tests reveal significantly more IL-4 production from the PBMC of active compared with latent TB subjects, with IFN-gamma not significantly different between the two groups [29]. Also, 20–25% of pulmonary TB patients with newly active disease present a diminished TST response [30]; these negative TSTs associate with lower IFN-gamma T cell responses to PPD in ex vivo re-stimulation assays [28,31]. Negative TST results also routinely appear in subjects with miliary TB and other extra pulmonary forms of disease, with reversion to positive responses after MTB treatment [30]. IL-10 and TGF-beta are elevated in active MTB subjects and associated with suppression of MTB-specific immune responses [28]. Also, FOXP3+ CD4 T cells are higher in frequency in pleural fluid compared with peripheral blood [25]; these findings implicate a role of Tregs at the site of MTB replication. The relative Treg frequencies in the PBMC of LTBI and active MTB subjects are not yet clear.

This immunosuppressive arm of the immune response may arise to dampen the inflammation and reduce immuno-pathology during onset of active disease, or alternatively, MTB may have evolved to induce type 2 responses to inhibit autophagy in macrophages, as IFN-gamma promotes while IL-4 and IL-13 directly inhibit this process [32]. Regardless of the cause, the apparent emergence of these Type 2 cytokines render them attractive candidates for analysis when seeking to define bio-signatures of LTBI and active MTB disease states for immunodiagnostics.

Elispots, like flow cytometry-based Intracellular Cytokine Staining (ICS), can directly determine the frequency of immune cells responding to a particular antigen, a core competency for immune diagnostics. In fact, while the putative naïve T cell repertoire is comprised of a vast number (≥10¹²) of TCR specificities, for any single infective agent, the frequency of circulating antigen-specific memory cells is quite low, typically in the range of 1:10,000–1,000,000. While capable of single-cell resolution, detection of this rare cell pool can be a challenge for flow cytometry, whose lower limit of sensitivity for identifying functional T cell frequencies is reported to be 0.02% [33]. By contrast, in routine 96-well Elispot assays, limits of 0.00025% (1 spot forming unit per 400,000 input PBMC) are commonly observed [34]; this equates to a near 100-fold greater detection sensitivity compared with flow cytometry. In fact, the Elispot assay has demonstrated signal linearity for cultures in the range of 1 × 10⁶ PBMC [35]; thus, the only true constraint on this technique appears to be the number of cells available for testing, size can also be increased to accommodate larger cell numbers. Also, Elispots are currently one of the only techniques permitting quantification of secretory activity at the single cell level. Since detection by ICS occurs prior to cytokine release, there is the potential for misleading results due to post-translational modulation of protein expression [36]. Further, while both ICS and Elispots measure cytokine secretion independent of kinetics within the culture period, a significant fact given the unsynchronized nature of the responding T cell pool, the duration of the ICS assay is limited by the toxicity of protein transport inhibitors. Therefore, detection of some cytokines may not be possible when ICS is used.

While ELISAs and bead-based measurement of cytokines from culture supernatants offer some advantages [37], there are limitations that preclude their acceptance as a T cell diagnostic platform for MTB disease states. For example, bulk supernatant assays lack the sensitivity for extremely low frequency signal detection. Spot size gating of Elispots using quantitative image analysis platforms permits a clear distinction between the diagnostically relevant large T cell-derived spots and smaller non-specific bystander spots. Since bulk assays cannot provide information on a ‘per cell’ basis, the contribution of Ag-specific responses may be diluted by background from the innate immune system, resulting in overall signal flattening; this issue is most relevant in PBMC cultures where the Ag-specific T cell pool is vastly outnumbered. The sensitivity threshold for cytokine measurements in culture supernatants is further diminished in these assays by analyte dilution in the surrounding milieu, absorption by bystander cells, and enzymatic degradation.

Elispots not only provide superior quantitative performance, they are also amenable to multiplex analyses carried out simultaneously (single well), in parallel wells, or in multi-color assays. Given the relatively small sample size required per test, multiple cytokine Elispots can be performed in parallel. Well established dual and even triple-color Elispots, both enzymatic and fluorescent, allow multiparameter resolution at the single cell level and are currently used in many research settings, including clinical diagnostics as well as therapeutic design [38,39,40]. The ever-expanding availability of discrete fluorochromes, when combined with multi-laser instrumentation and fully automated sample acquisition and data analysis, provide the framework for unsurpassed multi-functional analysis of antigen-specific T cell frequencies in the near future.

Beyond performance criteria, the Elispot assay possesses a number of other characteristics favorable for widespread adoption in developing countries where MTB is most devastating. Compared with flow cytometry, Elispots require one-tenth less cells per test, a fact that is crucial under conditions where samples are precious (remote testing) and/or limiting (pediatric test subjects). Smaller sample sizes also mean less reagents used per test. Also, Elispot data acquisition/analysis is far easier to perform, less time consuming, and dramatically lower in cost than flow cytometry. In fact, data from a 96-well Elispot plate can be acquired and analyzed by an image-based platform, such as the ImmunoSpot, in the same amount of time it takes a skilled flow cytometer operator to collect one sample. Also, while flow cytometry struggles with a lack of user-independent gating capabilities, and therefore suffers from inter-user and inter-lab subjectivity [41], automated platforms promote the standardization of Elispot data analysis and greater reproducibility across sites. These features make the Elispot assay the ideal choice for high-throughput testing applications as could be applied in large-scale subject profiling.