The immunesystem is an important part of the body since, it guards the organism against infections and cancer. For this purpose, highly specialized cell types, e.g. T cells and B cells have evolved as part of the immunesystem. T cells play an important role during a variety of immune system responses to pathogens and cancer. They are able to kill infected and abnormal cells or aid in activating different other cell types as part of an immune response. One of these different subtypes of T cells are T helper cells. They mature from naive T cells during an immune response after antigen stimulation.

Different lineages of T helper cells secrete different sets of characteristic cytokines helping to orchestrate the immune response and aiding in activating other cell types of innate and adaptive immunity. The development of T helper cells into different lineages, depends on the cytokine environment that induces different master regulators (cf. figure 1).

T helper cells differentiate along different lineages, e.g. type I (Th1) or type II (Th2) helper cells. While Th2 cells activate B cells to secrete antibodies, Th1 cells mediate the cellular immuneresponse by activating macrophages, that act against intracellular pathogens. Th1 cells are especially involved in Type IV hypersensitivity reactions.

The master regulator for Th1 differentiation is thought to beT-bet. This transcription factor is also responsible for inducing the secretion of the major Th1 cytokine interferon γ (IFN-γ) and theinterleukin 12 receptor (IL-12Rβ2) β2 chain (IL-12Rβ2). The induction of T-bet itself depends on T-Cell receptor signals and the presence of the cytokines interleukin 12 (IL-12) and IFN-γ (cf. figure 1).

The work presented in this paper [2], combines experimental work with mathematical modelling to further understand the regulatory network that underlies T-bet expression. The authors found experimental evidence that T-bet is expressed in two waves. The first wave is IFN-γ dependent. The second one relies on IL-12 and is independent of IFN-γ. The second wave also coincides with STAT4 binding to the T-bet enhancer. Moreover, the two waves seem to be coordinated by the TCR signal. During the first wave, the IFN-γ signal acts synergistically with the TCR signal whereas, the expression of the IL-12Rβ2 chain is repressed. The IL-12 mediated phase begins with the end of the TCR signal, thereby releasing the IL-12Rβ2 repression that renders the cells responsive to the IL-12 signal.

Figure 1: The different T helper cell lineages, their master regulators and characteristic cytokines. Figure taken from [1].

Figure 2: Previous Model (One Loop Model) of T-bet regulation based on the current literature. Figure taken from [2].

Figure 3: Two Loop Model of T-bet regulation. Figure taken from [2].

To properly explain the regulations underlying the experimental observations, the authors analyse two mathematical models. The first model, termed One-Loop Model, is based on the current literature knowledge (cf. figure 2). The attempt to fit this model simultaneously to the experimental T-bet, IL-12Rβ2 and IFN-γ expression profiles fails.

In contrast, the Two-Loop model proposed by the authors (cf. figure 3) explains the regulatory mechanisms that lead to the observed expression profiles. The Two-Loop model contains two new features, IL-12 dependent T-bet expression and antigen-dependent IL-12Rβ2 repression. The authors show, that a model including both additional regulatory circuits performs better than models that only include one of them.

The Two-Loop Model was validated by comparing experimental results with the simulation results for these experiments. Blocking either IL-12 or IFN-γ signaling gave equivalent results in simulations and experiments as well (cf. figure 4). The absence of IFN-γ, results in the complete loss of the first T-bet wave in the model as well as experiments. IL-12 absence strongly reduced the second T-bet peak. The upregulation of the IL-12Rβ2 chain (cf. figure 4A/B) during the second IL-12 dependent T-bet wave, supports the existence of the T-bet IL-12β2 feedback. Furthermore, IFN-γ seems to accelerate this feedback, which can be seen from the delaying effect in the IFN-γ deficient experiments/simulations (cf. figure 4, blue curves). Both loops act independently of each other and sequentially in time.

The paper provides further experimental evidence, that the proposed Two-Loop Model contains the necessary core regulations for primary Th1 activation. Repression of Stat4 and Gata3 transcription do not seem to play a major role under the used conditions. In addition, the second wave of T-bet expression seems to imprint Th1 cells for later IFN-γ production in recall responses. The frequency of IFN-γ producing cells shows T-bet dosage dependence pointing towards a certain time window where T-bet expression is decisive.

Figure 4: Experimental Validation of the Two-Loop Model (Ref. 2). Expression kinetics of T-bet (A), IL-12Rβ2 (B), IFN-γ (C). Black curves are the expression kinetics under Th1 inducing conditions. IFN-γ or IL-12 signaling was blocked to obtain the blue and red curves respectively. Figure taken from [2].

The presented study is a good example for the integration of experimental studies with mathematical modelling. The authors use experimental data to propose a mathematical model explaining their observations. They are able to show that, their new model performs much better than the previous explanations available in the current literature. The mathematical model helps to understand the core regulatory features that underlie the two waves of T-bet expression during T-cell priming and thus underlines their regulatory hypothesis. It also allowed to predict effects that occured if certain parts of the included regulatory pathways were knocked down. This study clearly shows the benefit of modelling for hypothesis testing in experimental studies.