The Case for Absolute Ligand Discrimination: Modeling Information Processing and Decision by Immune T Cells

Abstract

Some cells have to take decision based on the quality of surroundings ligands, almost irrespective of their quantity, a problem we name "absolute discrimination". An example of absolute discrimination is recognition of not-self by immune T Cells. We show how the problem of absolute discrimination can be solved by a process called "adaptive sorting". We review several implementations of adaptive sorting, as well as its generic properties such as antagonism. We show how kinetic proofreading with negative feedback implement an approximate version of adaptive sorting in the immune context. Finally, we revisit the decision problem at the cell population level, showing how phenotypic variability and feedbacks between population and single cells are crucial for proper decision.

Balancing specificity, sensitivity, and speed of ligand discrimination by zero-order ultraspecificity

“…The multistep kinetic proofreading scheme only has specificity, and therefore other mechanisms that can balance all of the four characteristics should be considered. Altan-Bonet proposed a kinetic proofreading model with feedback regulation to balance both specificity and sensitivity [15,21], then François and colleagues performed in silico evolution to obtain an "adaptive sorting" mechanism for specificity, sensitivity to target, and insensitivity to non-target [4,17]. These studies indicate that all or some of the four discrimination properties are generated by nonlinear responses of biochemical reaction systems.…”

“…Some of the previous works have not explicitly modeled the receptor dephosphorylation reaction [4,[9][10][11]. However, in general, phosphorylated molecules are dephosphorylated by phosphatase enzymes.…”

“…To balance specificity and sensitivity, several modifications have been proposed, e.g., by extending the multistep scheme [13] or by incorporating feedback loops [14][15][16]. In addition, Lalanne et al applied in silico evolution to obtain an "adaptive sorting" model [4,17] that balances the specificity, sensitivity to target, and insensitivity to nontarget. The adaptive sorting model is a modified version of a multistep kinetic proofreading model with a feedback regulation motif, which is similar to the bacterial twocomponent system for initiating an adaptive response to an external signal [18].…”

Attack and Defense in Cellular Decision-Making: Lessons from Machine Learning

“…To model cellular decision-making, we use the general class of "adaptive sorting" or "adaptive proofreading" models, which account for many aspects of immune recognition [14,33] and can be shown to capture all relevant features of such cellular decision-making close to a decision threshold [34]. An example of such a model is displayed in Fig.…”

“…For instance, T cells have to discriminate between foreign and self-ligands, which is challenging since foreign ligands might not be very different biochemically from self-ligands [12,13]. Decision-making in an immune context is equally prone to detrimental perturbations in a phenomenon called ligand antagonism [14]. Antagonism appears to be a general feature of cellular decision-makers: It has been observed in T cells [15], mast cells [16], and other recognition processes like olfactory sensing [17,18].…”