Continuous Modeling of T CD4 Lymphocyte Activation and Function

Abstract

T CD4+ cells are central to the adaptive immune response against pathogens. Their activation is induced by the engagement of the T-cell receptor by antigens, and of co-stimulatory receptors by molecules also expressed on antigen presenting cells. Then, a complex network of intracellular events reinforce, diversify and regulate the initial signals, including dynamic metabolic processes that strongly influence both the activation state and the differentiation to effector cell phenotypes. The regulation of cell metabolism is controlled by the nutrient sensor adenosine monophosphate-activated protein kinase (AMPK), which drives the balance between oxidative phosphorylation (OXPHOS) and glycolysis. Herein, we put forward a 51-node continuous mathematical model that describes the temporal evolution of the early events of activation, integrating a circuit of metabolic regulation into the main routes of signaling. The model simulates the induction of anergy due to defective co-stimulation, the CTLA-4 checkpoint blockade, and the differentiation to effector phenotypes induced by external cytokines. It also describes the adjustment of the OXPHOS-glycolysis equilibrium by the action of AMPK as the effector function of the T cell develops. The development of a transient phase of increased OXPHOS before induction of a sustained glycolytic phase during differentiation to the Th1, Th2 and Th17 phenotypes is shown. In contrast, during Treg differentiation, glycolysis is subsequently reduced as cell metabolism is predominantly polarized towards OXPHOS. These observations are in agreement with experimental data suggesting that OXPHOS produces an ATP reservoir before glycolysis boosts the production of metabolites needed for protein synthesis, cell function, and growth.

Keywords: CTLA-4; T CD4 cells; T cell receptor; lymphocyte activation; mTOR; mathematical model; metabolism; regulatory network.

Figure 1

Conceptualization of the time evolution of the OXPHOS-glycolysis shift along the bioenergetic and biosynthetic profile of T CD4 cells under antigenic stimulation. Adapted from (25).

Figure 2

Metabolism module including basic elements involved in glycolysis and OXPHOS regulation, as described in section 2.1. AMPK is a central energy sensor of the AMP/ATP ratio and displays a negative feedback loop with MTORC1. This loop defines a switch driving either OXPHOS or glycolytic activity. Continuous and dotted lines represent activator and inhibitory pathways, respectively.

Figure 3

A 51-node network of the early biochemical interactions induced by TCR activation, co-stimulation, and phenotype-inducing cytokines. Inputs of the network are the MHC-antigen, the co-stimulatory molecules CD80/86, and external cytokines (blue polygons). The set of interacting rules defining the dynamical system for Boolean modeling is shown in Supplementary Material 1 . Boolean rules were translated to continuous functions as shown in Material and Methods, and the complete set of ordinary differential equations is provided in Supplementary Material 2 . Green and red rectangles represent stimulatory and inhibitory nodes of the activation core, respectively. Pink and yellow rectangles represent phenotype-inducing transcription factors and cytokines produced endogenously, respectively. White rectangles represent the metabolism module including AMPK as a central regulatory element, as shown in figure 2. Construction of the network can be consulted in (40). LAT: LAT-Gads-SLP-76 complex; IL2G: IL-2 gene; L4e, IFNγe, IL10e, TGFβe and IL21e: exogenous cytokines.

Figure 4

Dynamics of transcription factors and metabolism under optimal engagement conditions of TCR with the MHC-antigen complex, and CD28 with CD80/86, during a stimulation time τ _{MHC / A} = τ _{CD 8086} = 15 units. Left panel: Sustained T-cell activation associates to low-level activity of CTLA-4 (with high decay rated d_{CTLA 4} = 5). (A) TCR and CD28 are fully activated as far as antigenic stimulation persists, leading to (B) sustained expression of the transcription factors AP-1, NFAT, and NFκB, and (C) sustained expression of mTORC1 and IL-2. (D) Metabolic profile: After initial activation, the AMP/ATP ratio decreases, leading to a reduction of the activity of the nutrient sensor AMPK and a temporary increment of OXPHOS. With the course of time, OXPHOS decays and glycolytic activity increases up to a steady level, in parallel with the AMP/ATP ratio. Right panel: Checkpoint blockade associated to high-level activity of CTLA-4 (with low decay rate d_{CTLA 4} = 0.5). (E) Initially, TCR and CD28 are fully activated. With the course of time, the expression level of dimerized CTLA-4 increases, displacing CD28 from co-stimulatory molecules, with the concomitant induction of anergy. This is manifested as (F) transitory expression and down-regulation of the activation transcription factors AP-1, NFAT, NFκB, and (G) transitory expression and down-regulation of mTORC1 and IL-2. (H) Metabolic profile: At the beginning, the metabolic activity shows an identical pattern as in the case of sustained activation; eventually, the inhibitory action of CTLA-4 induces the decay of glycolysis.

Figure 5

Values of the decay rate of CTLA-4, d_{CTLA 4}, as a function of the stimulation time, τ _stim = t_{CD 8086} = t_MHC/A , associated to states of no activation (yellow), sustained activation (blue), or regulated activation (pink). For τ _stim < 7 units no activation arises; sustained activation ensues for 7 ≤ τ _stim ≤ 10, and regulated activation for τ _stim > 10. The first stage implies that a minimum stimulation time is necessary to boost activation; the second one, that CTLA-4 requires a minimal time of activation to be induced and perform inhibition; the third stage reveals that regulated activation only arises at later stimulation times and when the CTLA-4 decay rate is low, that is, d_{CTLA 4} < 1. Therefore, CTLA-4 should be induced for a time long enough to overcome a threshold level and its activity should be maintained to perform inhibition. The oscillatory behavior of the regulation threshold is associated to the inhibitory action of CTLA-4 which depends in turn on TCR and CD28 activation. This induces a stimulation-inhibition cycle which is downstream-propagated throughout the network. Consequently, the threshold d_{CT LA4} leading to regulation is determined by the expression level attained by activation inducers at a given phase of this cycle.

Figure 6

Dynamics of transcription factors and metabolism after incomplete stimulation during an activation time τ _{MHC / A} = τ _{CD 8086} = 15 units. Left panel: Anergy is induced by strong TCR stimulation and weak CD28 co-stimulation (A _{MHC / A} = 1 and A _{CD 8086} = 0.5). (A) TCR and CD28 are transiently activated, decaying both at time τ _{MHC / A} = τ _{CD 8086}. At this time the anergy factor NDRG1 is temporarily expressed. (B) NFAT, NFκB, and AP-1 are down-regulated by the inhibitory action of NDRG1. (C) Similarly, mTORC1 and IL2 are only transiently expressed. (D) The metabolic profile is similar to that associated to regulated activation induced by CTLA-4 ( Figure 4H ). Right panel: Anergy is induced by weak TCR stimulation and strong CD28 co-stimulation (A _{MHC / A} = 0.5 and A _{CD 8086} = 1. (E) TCR and CD28 show full and low -level activation, respectively, both decaying at time τ _{MHC / A} = τ _{CD 8086}; however, the anergy factor NDRG1 remains unexpressed. (F). NFAT, NFκB, and AP-1 are unexpressed. (G) mTORC1 is fully activated, but decays along the antigenic stimulation. (H) The metabolic profile is similar to that associated to regulated activation induced by CTLA-4 ( Figure 4H ).

Figure 7

Dynamics of master transcription factors and interleukin production by effector cells under optimal antigenic recognition and low-level activity of CTLA-4: (A) Th1 profile: Sustained expression of T-bet and INF-gamma production, (B) Th2 profile: Sustained expression of GATA3 and IL-4 production, (C) Th17 profile: Sustained expression of RORγT, as well as Il-17 and IL-21 production. (D) The Th1, Th2, and Th17 phenotypes display an initial transient OXPHOS phase and then a stable glycolytic metabolism. (E) Treg profile: Sustained expression of Foxp3 and dimeric CTLA-4 with joint production of IL-10 and TGF-β. (F) Treg displays an initial transient OXPHOS phase followed by glycolytic metabolism; however, with the course of time this is replaced by a metabolism based on OXPHOS.