Dissecting the roles of toxin bivalency, membrane affinity, and stoichiometry in DkTx-activation of TRPV1

Abstract

Recent advances in structural biology show that lipids interact intimately with a wide range of membrane proteins, including ion channels, transporters, and membrane enzymes, implying that protein-lipid interfaces play important roles in protein function. A detailed understanding of the functional roles of these protein-lipid interfaces is required to fully understand the function of both integral and peripheral membrane proteins. A recently obtained structure of the rat TRPV1 channel complexed to its potent agonist, the double knot toxin (DkTx), composed of two inhibitory-cystine-knot (K1 knot and K2 knot), in lipid nanodiscs provides a detailed view of the DkTx−lipid−TRPV1 tripartite complex. This recent structural work provides one of the best available structural snapshots of protein-protein complexes in a native membrane-like milieu. DkTx exhibits an extremely slow wash-off of the toxin after channel activation. This is an intriguing observation given that the TRPV1-DkTx complex structures do not reveal any electrostatic, cation-pi, or pi stacking interactions between the toxin and channel residues, which commonly underpin tight complex formation between proteins. My Ph.D. focuses on elucidating the role of DkTx-membrane interaction in defining the characteristic slow wash-off of DkTx. Detailed electrophysiological studies performed in our laboratory have identified DkTx variants possessing remarkably reduced apparent TRPV1 affinity, all of which map to a toxin-membrane interface identified in the TRPV1-DkTx structure alluded to above. My first approach is to employ tryptophan fluorescence-based membrane partitioning experiments on toxin variants. The results of these experiments reveal a strong correlation between fast wash-off kinetics and mol. fraction partition coefficient forms the basis of the "toxin relay" mechanism. The single knots K1 and K2 of DkTx can individually activate TRPV1, albeit with much poorer potency and considerably faster wash-off rates than wild-type DkTx. Slow wash-off of DkTx has been previously attributed to its bivalency when the single knots act as TRPV1 agonists. However, a recent study from our lab suggests that the toxin wash-off behavior shows a robust negative correlation with its membrane partitioning ability. This poses an intricate problem of bifurcating the role of bivalency or membrane partitioning in DkTx's slow wash-off. I have addressed this query by developing chimeric double-knot toxins by replacing the K1 and K2 knots of DkTx with the Kv channel-targeting membrane partitioning toxin, SGTx, in such a way that they are monovalent with respect to TRPV1 while retaining DkTx-like membrane affinity. These variants washed off slower than single knots, demonstrating the importance of the toxin's membrane affinity in endowing a sustained channel-activating property. In subsequent experiments, I have characterized enzymatically conjugated two double-knots variants, i.e., tetra-knot variants (DkTx)2 and DkTx-(SGTx)2 with hyper-membrane affinities and TRPV1 currents that wash off much slower than wild-type DkTx. These findings show that the membrane affinity of DkTx plays a significant role in its TRPV1 activation properties. I have also ruled out the role of TRPV1-DkTx stoichiometry in the characteristic slow wash-off of DkTx by functionally characterizing the TRPV1 octamutant concatemer. Additionally, I have attempted the N-terminal bioconjugation reaction of DkTx with 2-Pyridinecarboxaldehyde (2-PCA) in a series of experiments where our lab attempted to develop fluorescently labeled DkTx. I have also functionally characterized the cyclic nucleotide gating (CNGC13) channel.