(D-149) Single-molecule characterization of lynx protein - nicotinic acetylcholine receptor interactions

Nicotinic acetylcholine receptors (nAChRs) are widely expressed in the nervous system. They are involved in brain activities like learning, memory, addiction, and anxiety. As ligand-gated ion channels^[1] comprising five subunits, nAChRs are involved in many neurological and psychiatric diseases. Lynx1 and lynx2, from the Ly-6/neurotoxin protein superfamily, are identified as nAChR modulators^[2]. They are crucial to nAChR activity regulation^[3], preventing excessive excitations^[4]. Recently, we used single-molecule force spectroscopy experiments based on atomic force microscopy (AFM) to quantify lynx-nAChR interactions^[5]. In this work, we used this established assay to investigate the structural and molecular mechanism of lynx-nAChR interactions.

The stably transfected SH-EP1-hα7-nAChR and SH-EP1-hα3β4-nAChR cell lines were cultured in 35 cm² petri dishes to form an adherent monolayer. The recombinant lynx proteins (lynx1 wildtype (WT) and mutants in putative residues essential for receptor binding Y9A, F34A, D55S, and lynx2 WT and Q39H mutant) were expressed in E. Coli cells and purified by affinity and size-exclusion chromatography. The lynx proteins were covalently attached to AFM cantilevers (MLCT-Bio-DC: Bruker Nano), via heterobifunctional PEG linkers. In single-molecule force experiments, the lynx-coated cantilevers were approached and retracted from the cell surface at varying defined speeds. If lynx proteins and nAChRs interact during cantilever-cell contact, the bonded complex(es) will rupture when the cantilever retracts. The signals of unbinding were recorded in the force curves. Several thousands of force-distance curves were recorded for each lynx-nAChR pair. The data was processed using IGOR Pro software to extract the unbinding forces as a function of the loading rate of each unbinding event.

The Bell-Evans model is applied to extract binding kinetics from the single-molecule unbinding data. The plot of the non-linear regressions-fitted most unbinding forces of each lynx-nAChR pair is shown in Figures A and B. The model-estimated dissociated rates of each lynx-nAChR pair (with error bars indicating the standard errors of the fit) are shown in Figure C.

From the data, it is evident that lynx1 proteins generally have a stronger unbinding force with α3β4 nAChRs than with α7 nAChRs, and that the majority of lynx2 proteins have a lower dissociation rate with receptors than lynx1 under the same condition. Lynx1-F34A protein has the lowest binding affinity with nicotinic receptors among all the tested lynx-receptor pairs, whereas the lynx2-Q39H mutant shows the strongest affinity toward the α7 nAChR. In addition, our data show that mutations in both loops 1 (Y9A) and 2 (F34A) of the lynx1 structure do not affect its binding with α3β4 nAChR, but lower its affinity with α7 nAChR significantly, suggesting these two loops determines the affinity of the lynx1- α7 nAChR interaction. Loop 3 might be important in the binding selectivity between receptors, as the D55S mutation has an opposite effect on lynx1 binding for α3β4 and α7 receptors. For lynx2, the mutant we tested was identified from individuals with anxiety. Our data indicate that the Q39H mutant in loop 2 of lynx2 decreases the binding between lynx2 and α3β4 receptor, but enhances its binding with the α7 receptor, highlighting the importance of loop 2 in a potential connection to anxiety.