Role of Cytokine Combinations on CD4+ T Cell Differentiation, Partial Polarization, and Plasticity: Continuous Network Modeling Approach

Abstract

Purpose: We put forward a theoretical and dynamical approach for the semi-quantitative analysis of CD4+ T cell differentiation, the process by which cells with different functions are derived from activated CD4⁺ T naïve lymphocytes in the presence of particular cytokine microenvironments. We explore the system-level mechanisms that underlie CD4+ T plasticity-the conversion of polarized cells to phenotypes different from those originally induced. Methods: In this paper, we extend a previous study based on a Boolean network to a continuous framework. The network includes transcription factors, signaling pathways, as well as autocrine and exogenous cytokines, with interaction rules derived using fuzzy logic. Results: This approach allows us to assess the effect of relative differences in the concentrations and combinations of exogenous and endogenous cytokines, as well as of the expression levels of diverse transcription factors. We found either abrupt or gradual differentiation patterns between observed phenotypes depending on critical concentrations of single or multiple environmental cytokines. Plastic changes induced by environmental cytokines were observed in conditions of partial phenotype polarization in the T helper 1 to T helper 2 transition. On the other hand, the T helper 17 to induced regulatory T-cells transition was highly dependent on cytokine concentrations, with TGFβ playing a prime role. Conclusion: The present approach is useful to further understand the system-level mechanisms underlying observed patterns of CD4⁺ T differentiation and response to changing immunological challenges.

Keywords: CD4+ T cells; ODE; cytokines; heterogeneity; micro-environment; plasticity; regulatory network.

FIGURE 1

Methodology. Using available experimental data we constructed (A) the regulatory network, (B) Boolean functions (Martinez-Sanchez et al., 2015), and (C) ordinary differential equations (current article). (D) We then determined the resulting steady state for different concentrations and exogenous cytokines.

FIGURE 2

CD4+ T cell transcriptional-signaling regulatory network. The regulatory network was constructed using available experimental data. (A) The network includes transcription factors (rectangles), autocrine cytokines and their signaling pathways (ellipses) and exogenous cytokines (diamonds). Interactions leading to activation are represented by black arrows, while those leading to inhibition with red dots. (B) Sample attractors of the system.

FIGURE 3

CD4+ T cell fate as a function of the concentration of single exogenous cytokines: IL12, IFNG, IL2, IL4, IL6, Il21, TGFB, IL10, and IL27. From an initial state TH0, a CD4+ T cell may acquire diverse phenotypes on an abrupt or gradual transition, depending on critical concentrations of environmental cytokines. The plot shows the difference between the values of the initial Th0 state and the final steady state at different concentrations of exogenous cytokines. We observe that the presence of either IL12 or IFNg is sufficient for Th1 polarization, as well as IL4, is sufficient for TH2 polarization. On the other hand, IL2 alone does not lead to an effector phenotype. Similarly, the presence of either IL6 or IL21 alone is sufficient for Tfh induction, as is the case of TGFB and IL10, leading to Th3 and Tr1, respectively. IL27 alone does not lead to any fate transition in this model.

FIGURE 4

T-CD4 cell fate as a function of exogenous cytokine concentrations define diverse phenotype-associated environments. From the Th0 initial state, a CD4+ T cell evolves to different phenotypes, depending on critical concentrations of environmental cytokines as shown in Table 1: Th1 (IFNG and IL12), Th2 (IL4, Il2), Th17 (Il21, TGFB), Treg (IL2, TGFB), Tfh (IL21), Th9 (IL4, TGFB), Tr1 (IL10, IL27), and Th3 (TGFB). The plot shows the difference between the values of the initial Th0 state and the final steady state at different concentrations of exogenous cytokines. The transition may be abrupt or gradual and, interestingly, may involve an intermediate state, as in the cases Th0 - > Tfh - > Th17 (C), and Th0 - > Th3 - > Th9 (F).

FIGURE 5

Phenotype space diagrams for Th1 and Th2 polarization and plasticity as a function of the relative concentration of environmental IFNg and IL4, and expression of transcription factors. (A) Diagram for cell differentiation assuming an initial Th0 state. As the external concentration of IFNGe increases, the system develops an abrupt transition from Th0 to Th1. Similarly, an increase in external IL4e drives an abrupt transition from TH0 to Th2. For moderate concentrations of IFNGe and IL4e (< 0.8), we observe two wide zones of Th1 or Th2 prevalence with a sharp boundary, meaning that small variation of cytokines at these zones may change cell polarization. A transition zone with no defined polarization appears at higher concentrations of these cytokines (white and gray areas). (B,C) Plasticity diagrams assuming full Th1 (B) or Th2 (C) polarized states (i.e., induced by INFg = 1 and IL-4 = 1 in diagram A, respectively). No phenotypic transitions are observed under variable concentrations of environmental IL4e and autocrine IFNGe. (D) Plasticity diagram of Th1 cells assuming an environmental concentration of IL4e = 1. Cells require the production of initial high levels of autocrine IFNG and expression of TBET to maintain a Th1 phenotype. If the initial expression levels decrease, especially in the case of autocrine IFNG, it will transit into a Th2 cell. (E) Plasticity diagram of Th2 cells assuming an environmental concentration of IFNGe = 1. The cell requires the production of high levels of autocrine IL4 and expression of GATA3 to maintain a Th2 phenotype. If the initial expression levels decrease it will transit into a Th1 cell. At high expression levels of initial GATA3 and low initial IL4, there exists a transition zone where the cell cannot be classified.

FIGURE 6

Three-dimensional phenotype space diagrams for Th17 and iTreg polarization and plasticity as a function of the relative concentrations of IL2, IL21, and TGFB in the microenvironment. In the differentiation diagram (A) we observe alternative phenotypic regions defined by relative concentrations of environmental cytokines. The regions may be either separated by a sharp boundary or by a more gradual transition zone (labeled in white). The plasticity diagram (B) indicates a polarized behavior for Th17 versus Tfh phenotype determined by a high or low concentration of external TGFB. A richer behavior ensues when the initial state is Treg, as shown in the plasticity diagram (C). We observe a similar structure as that depicted in A, except that the Th0 zone is absent.