Conformational Switching Dynamics of Syntaxin 1
Activation of syntaxin is key to synaptic membrane fusion, triggering the release of neuotransmitters into the synaptic gap. Although it has been suspected that structural transitions are crucial for the activation, their nature and dynamics were unknown. In a close collaboration with the groups of Reinhard Jahn and Claus Seidel we have monitored these dynamics by single molecule FRET measurements. From these data, distance changes between selected syntaxin residues were derived with sub-millisecond resolution. Molecular modeling revealed a conformational equilibrium between an inactive closed and an active open state, thus explaining the role of regulatory proteins such as munc-18, which, by binding, arrests syntaxin in the closed conformation. The direct observation of a conformational equilibrium also suggested that other regulatory mechanisms such as phosphorylation act by shifting conformational equilibria.
Syntaxin 1 is an essential component of the membrane fusion machinery in all eukaryotes from yeast to humans. It belongs to a family of soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNARE) which have emerged as the leading candidates for mediating membrane fusion. SNARE proteins comprise a superfamily of small membrane proteins distinguished by the SNARE motif, a conserved coiled-coil stretch of 60–70 amino acids. SNARE motifs spontaneously assemble into elongated four-helix bundles.
Three different conformations of syntaxin 1 were studied using single-molecule multiparameter fluorescence detection. First, syntaxin with munc-18 which arrests syntaxin in the closed conformation. Next, free syntaxin was investigated. Finally, syntaxin in the ternary SNARE complex with synaptobrevin and SNAP-25 was examined. In this complex syntaxin 1 is unable to form a closed conformation. For syntaxin-munc-18 the data from FRET analysis agreed with the crystal structure , confirming that the ensemble of syntaxin molecules is dominated by the closed conformation when complexed with munc-18. Data from free syntaxin suggested at least two different conformations. One corresponding to that of syntaxin-munc-18 and one reflecting an open conformation. Thus, only a subpopulation of free syntaxin (ca. 15–30%) is in the closed conformation whereas most molecules adopt a conformation corresponding to an open conformation. Finally, we analyzed syntaxin in the ternary SNARE complex with synaptobrevin and SNAP-25. Again, only one conformation was observed suggesting an open conformation with multiple substates in solution.
Kinetics: Our data show that free syntaxin exists in a mixture of at least two different conformations, one of which is similar to that observed in the complex with munc-18. This finding could reflect just a static conformational heterogeneity, for which each syntaxin molecule is trapped during the observation time within its specific conformation, or, alternatively, structural dynamics, where the individual molecules interconvert in a dynamic equilibrium.
Analysis of these possibilities, revealed that the intensity fluctuations within fluorescence bursts result from individual molecules. Thus free syntaxin switches between an inactive closed and an active open configuration. A relaxation time of 0.8 ms was found, explaining why regulatory proteins are needed to arrest the protein in one conformational state. The presence of dynamic conformational equilibria such as that studied here opens up a fascinating array of questions that can now be experimentally addressed. Whereas it is obvious that interaction with partner proteins are required to arrest syntaxin in one conformational state, it is also possible that the conformational equilibrium is regulated by other mechanisms such as phosphorylation.
Structural Model: By combining FRET results with molecular modelling, we were able to provide a structural model for the conformational transition. For the structural interpretation of the FRET data, we generated a large set of protein conformations that were compatible with these data. FRET data for the closed state agreed with the x-ray structure as shown in Fig. 3. (top part). The green dots denote all possible positions of the Cα-atom of residue 225 (green sphere in Fig. 3) that are compatible with the FRET data. The aim was now to derive an atomistic model for the structure of the open state. Possible positions of residue 225 that are compatible with FRET data of the open state are shown in Fig. 3 (bottom part). No green dots occur in the central region. Thus the position occupied by residue 225 in the closed state is ruled out for the open state. Taking into account all possible positions (data not shown) of the labeled residues [167 (red), 197 (blue), 207 (yellow)], we selected from all FRET-compatible structures the one that required the smallest conformational transition (Fig. 3 bottom part).