C&SP22: Modeling T cell fate decisions

A. Collaborating Investigators: Penelope Morel, 1 Robin E. C. Lee, 2 James R. Faeder2

B. Institutions: Pitt 1Immunology and 2Computational and Systems Biology Departments

C. Funding Status of Project: Juvenile Diabetes Research Foundation 1-INO-2016-215-A-N (Morel) 06/01/2016-05/31/2017

D. Biomedical Research Problem: The T cell receptor (TCR) is an exquisitely sensitive molecular machine that can translate small differences in peptide/MHC ligands into profoundly different outcomes. Our long-term goal is to understand how T cells perceive signals of differing strengths in order to elucidate how this determines T helper (Th) cell fate. Antigen (Ag) dose affects Th differentiation; and several in vivo models have shown that low Ag dose favors T regulatory (Treg) and Th2 differentiation, whereas high Ag dose induces inflammatory Th1 cells.74-80 On the one hand, Treg cells are critical for the maintenance of self-tolerance and the prevention of autoimmune diseases such as diabetes; on the other, tumors induce Treg cells to suppress immune surveillance. Despite the powerful effect on T cell outcome, how these signals contribute to determining Th cell fate remains poorly understood.

 

 

Fig VIII.6 A. Calibrated time courses for PTEN and FoxO1 following stimulation with high dose antigen . Lines are model predictions with light and dark blue bands indicating 75% and 95% confidence regions. Points are experimental data with error bars showing +/- SEM. B. Single-cell imaging pipeline with automated segmentation and tracking from live-cell time-lapse images. Last panel displays single-cell time courses that can be extracted for hundreds of cell simultaneously. See (Hawse et al, 2015) for details.

E. Methods and Procedures: By coupling mathematical modeling with detailed in vitro activation profiling we have identified several important feedback loops that control the degree of Akt/mTOR activation.81-83 These involve the lipid phosphatase and tensin homolog (PTEN), the transcription factors Foxp3 and FoxO1 and the Ser/Thr kinase Akt.81,82 High dose Ag increases TCR stimulation resulting in PTEN degradation, which drives higher Akt/mTOR signaling and T effector development. Conversely, low TCR stimulation sustains expression of PTEN, which suppresses Akt/mTOR signaling to promote Treg development.81 Based on these and other biochemical mechanisms we have developed a rule-based model in BioNetGen and calibrated it to biochemical data for low and high dose stimulation as shown (Fig. VIII.6A) using Bayesian parameter estimation with the ptempest software, as described in TR&D3 Aim 2.3. This model predicts a sharp threshold for Akt activation with respect to antigen dose that depends on several key variables, most notably PTEN expression level, which exhibit considerable cell-to-cell variation. We are currently conducting experiments in mice that express a fluorescent reporter of TCR activation to determine whether there is a sharp activation threshold as predicted by the model, and in mice heterozygous for PTEN expression to examine how PTEN levels affect this threshold. We will use these data to refine the model and identify additional factors that can affect antigen dose thresholds. In addition, we will use the Amnis ImageStream technology available to the Morel lab and the single-cell live imaging pipeline being developed in the Lee lab (Fig. VIII.6B) to spatially resolve key signaling components, including Akt, PIP3, and NFκB. We will use these data to confirm temporal oscillations observed in preliminary data and to calibrate a coarse-grained spatial model (TR&D3 Aim 1.3).

 

Copyright © 2018 National Center for Multiscale Modeling of Biological Systems. All Rights Reserved.