A Mechanistic Model to Estimate Avidity Effects in Multivalent SH2 Binding

Introduction: Binding avidity can occur in bivalent or multivalent interactions when the overall affinity is larger than the sum of the individual interactions. It is a well-documented phenomenon in antibody binding, whether this refers to homobivalent ligands such as IgE, IgG, and IgD that possess two identical epitopes, or heterobivalent ligands with dual specificity to multiple targets (Vaquelin, Br. J. Pharmacol, 2013). Cell signaling interactions can also use multivalent interactions to drastically increase sensitivity and specificity in network connections. For example, recruitment of the tyrosine kinase ZAP-70, a key regulator of T cell receptor activation, requires both phosphotyrosine sites in the immunoreceptor tyrosine-based activation motif (ITAM) sequences in the CD3𝜁 chain of the T-cell receptor. Not only are the two SH2 domains of ZAP-70 recruited with 100x greater affinity than either SH2 domain individually, but deletion of one of ZAP-70’s SH2 domains completely prevents localization of the protein to the T-cell receptor (Isakov, JEM, 1998; Sloan-Lancaster, JCB, 1998). ZAP-70 is one of ten bivalent SH2-domain containing proteins in the human proteome that may have the capacity for bivalent avidity interactions with complementary doubly-phosphorylated partners. Others include PI3K regulatory proteins, SHP2, SYK, PLCg1/2, and RASA1, each of which plays a vital role in growth and survival pathways that drive many diseases such as cancer. By exploring the conditions under which these tandem SH2 proteins can exhibit a strong avidity effect, novel regulatory roles can be identified, and there is potential for elucidation of new targets to halt unchecked tumor growth.

Materials and

Methods: An ODE model developed by Dr. Casim Sarkar to track binding configurations in multivalent ligand-receptor interactions was adapted and utilized to investigate the ability of tandem SH2 domain proteins to bind in a bivalent manner to phosphotyrosine (pTyr) sites under different conditions (Errington, PNAS, 2019). These conditions include the length and rigidity of the linker that separates each component as well as the individual binding affinity of each SH2/pTyr pair. By tracking these binding events through a simulated association/dissociation, an effective binding affinity of the binary interaction can be estimated, thus allowing for the predicted avidity effect to be calculated. In this way, sets of conditions that maximize the avidity effect can be identified, and this provides insight into which pTyr pairs in the body are likely binding partners of each tandem SH2 protein. To validate this computational model, bio-layer interferometry (BLI) experiments were conducted to quantify the binding affinity of different bivalent and monovalent interactions. One major interaction tested is between the SHP2 tandem SH2 domain and sites Y627/Y659 in Gab1. Gab1 was phosphorylated through co-expression with a tyrosine kinase using a synthetic toolkit we previously developed. Following purification, it was biotinylated to allow for binding to a streptavidin-coated BLI probe. Monovalent affinities were calculated by mutating one pTyr site to a phenylalanine, and the double mutant was used as a negative control (Figure 1).

Results, Conclusions, and Discussions: Simulations showed that each family of tandem SH2 domains has a unique linker connecting their SH2 domains that determines the set of pTyr sites in which they can experience a strong avidity effect. Distances between pTyr sites that more closely matched the length of the SH2 linker were the best optimized, with increased rigidity within the linker requiring a closer match to achieve similar results. Therefore, proteins with longer, helical-rich linkers such as ZAP-70 and PIK3R1 have a more narrow range of pTyr pairs they experience avidity with, whereas proteins with short, flexible linkers like SHP2 and PLCg1 experience a broader range of potential strong interactions (Figure 2A). Additionally, maximal avidity was achieved when each SH2/pTyr pair experiences a moderate monovalent affinity for one another (Figure 2B). Stronger avidity was also shown to correlate with longer binding interactions, suggesting that multivalent interactions not only help shape the preferred interactions within signaling pathways, but also drive the temporal dynamics of these systems (Figure 2C). Work is ongoing to incorporate crystal structures in order to improve the estimation of distance pTyr sites in fully folded proteins. Experimental validation was also performed on multiple tandem interactions. The SHP2-Gab1 interaction was utilized to demonstrate a clear avidity effect. Beyond this, interactions between PLCg1 and nine pTyr sites on the EGFR intracellular tail were explored through BLI. Each SH2 domain of PLCg1 is capable of binding to 4-5 pTyr sites on EGFR, making it difficult to predict how the two proteins will interact. To elucidate the preferred binding partners, simulations were conducted to calculate the avidity effect of each pair and rank-order them accordingly. Four of these pairs (Y1016/1092, Y1016/1138, Y1092/1197, Y1138/1197) were isolated, cloned, and phosphorylated to validate the predicted rankings using BLI. The ability to use this model to predict the binding interactions of these tandem SH2 domains can help elucidate novel signaling dynamics within disease pathways, providing insight into the most sensitive regions to target treatment interventions and potentially identifying mechanisms of resistance to current treatments.

Acknowledgements (Optional):