The Mechanism of Activation

The structural transition from the shut- to the open-channel form of the receptor has been analyzed at 9-Â resolution by comparing the three-dimensional map of the shut form, as described above, with that of the open form, obtained by spraying ACh onto the tubes and then freezing them rapidly within 5 ms of spray impact (31). The rapid freezing combined with minimal delay was needed to trap the activation reaction and minimize the number of receptors that would become desensitized.

Adetailed comparison of the two structures indicated that the binding of ACh initiates two interconnected events in the extracellular domain. One is a local disturbance, involving all five subunits, in the region of the binding sites, and the other an extended conforma-tional change, involving predominantly the two a subunits. which communicates to the transmembrane portion. These experiments give a picture of the receptor in either of the two states (i.e., the shut- and open-channel forms), and thus provide no direct information relating to the possibility that the binding to one site might affect the binding to the other before the channel opens [see Hatton et al. (32)]. However, there is a tight association of a, with the neighboring y subunit (30), which is next to a,; thus some coupling is quite possible.

In the membrane, the exposure to ACh did not bring about any obvious alteration to the outer structure facing the lipids, whereas the M2 helices switched quite dramatically to a new configuration in which the bends, instead of pointing toward the axis of the pore, had rotated (clockwise) over to the side, as shown in Fig. 11.7.

This rearrangement had the effect of opening up the pore in the middle of the membrane, and making it narrowest at the cyto-plasmic membrane surface, where the a-helices now came close enough to associate by side-to-side interactions around the ring. Thus there appear to be two alternative configurations of M2 helices around the pore: one (the shut configuration) stabilized by side-to-

side interactions near the middle of the membrane, and the other (the open configuration) stabilized by side-to-side interactions close to the cytoplasmic membrane surface. These limited sets of interactions, combined with the rigid a-helical folds, might be important in ensuring the precise permeation and fast-gating kinetics that characterize acetylcholine-gated channels.

A tentative alignment of the M2 sequence with the densities in the cytoplasmic leaflet suggests that a line of small polar (serine or threonine) residues would lie almost parallel to the axis of the pore when the channel opens (Fig. 11.7), an orientation that should stabilize the passing ions by providing an environment of high polarizability. The threonine residue at the point of maximum constriction (Torpedo aT244), when substituted by other residues of different volume, has a pronounced effect on ion flow, as if it were at the narrowest part of the open pore (33).The diameter of this most constricted portion of the channel, based on permeability measurements made with small uncharged molecules of different size, is about 10 A (34,35). This value is similar to that indicated by the structural results.

A simple mechanistic picture of the structural transition, derived from these studies, would be as illustrated in Fig. 11.8. First, ACh triggers a localized disturbance in the region of the binding sites. Second, the effect of this disturbance is communicated by axial rotations, involving mainly the a subunits, to the M2 helices in the membrane (see Fig. 11.4). Third, the M2 helices transmit the rotations to the gate-forming side-chains, drawing them away from the central axis; the mode of association near the middle of the membrane is thereby disfavored, and the helices switch to the alternative side-to-side mode of association, creating an open pore.

A more precise description of the extended conformational change, linking to the transmembrane portion, has recently been derived by comparing the 4.6-A structure of the extracellular domain with the crystal structure of the AChBP. It is found that, to a good approximation, there are two alternative extended conformations of the receptor subunits (one characteristic of either a subunit before activation, and the other characteristic of the

Figure 11.7. Transient configuration of M2 helices around the open pore. (a) A barrel of a-helical segments, having a pronounced twist, forms in the cytoplasmic leaflet of the bilayer, constricting the pore maximally at the cytoplasmic membrane surface. The bend in the rods is at the same level as for the closed pore, but instead of pointing inward has rotated over to the side, (b) Schematic representation of the most distant three rods. A tentative alignment of the amino acid sequence with the densities suggests that a line of polar residues (serines and threonine; see Fig. 11.3) should be facing the open pore. [From Un-win (31).]

three non-a subunits) and that the binding of ACh converts the structures of the two a sub-units to the non-a form (30). Evidently, the a subunits are distorted initially by their interactions with neighboring subunits, and the free energy of binding overcomes these distortions, making the whole assembly more symmetrical, analogous to the ligand-bound AChBP.

This transition to the activated conformation of the receptor involves relative movements of the inner and outer parts of the /3-sandwich, which compose the core of the a subunit (see Figs. 11.4-11.6), around the cys-loop disulfide bond, as shown in Fig. 11.9.

Most strikingly, there are 15-16" clockwise rotations of the polypeptide chains on the inner surface of the vestibule next to the mem brane-spanning pore. The M2 segments and also the M2-M3 loops lie directly under these rotating elements. The importance of the M2-M3 loop for gating was first suggested by the group of Schofield as a plausible interpretation of the mechanism by which startle mutations in this area impair the agonist sensitivity of another member of the nicotinic superfamily, the glycine receptor (37). Thus, Lynch et al, (38), on the basis of a scanning alanine mutagenesis study and macroscopic dose-response curves, suggested that both and M2-M3 loops are involved in gating. The single channel work of Lewis et al. (39), based on a preliminary plausible model of the glycine receptor activation mechanism, confirmed that a mutation in M2-M3 (e*K276E) predominantly changes gating. In

Closed Open

Figure 11.8. Simplified model of the channel-opening mechanism suggested by time-resolved electron microscopic experiments. Binding of ACh to both a subunits initiates a concerted disturbance at the level of the binding pockets, which leads to small (clockwise) rotations of the a subunits at the level of the membrane. The rotations destabilize the association of bent a-helices forming the gate and favor the alternative mode of association (Fig. 11.7), in which the pore is wider at the middle of the membrane and most constricted at the cytoplasmic membrane surface. [Adapted from Unwin (36).]

Closed Open

Figure 11.8. Simplified model of the channel-opening mechanism suggested by time-resolved electron microscopic experiments. Binding of ACh to both a subunits initiates a concerted disturbance at the level of the binding pockets, which leads to small (clockwise) rotations of the a subunits at the level of the membrane. The rotations destabilize the association of bent a-helices forming the gate and favor the alternative mode of association (Fig. 11.7), in which the pore is wider at the middle of the membrane and most constricted at the cytoplasmic membrane surface. [Adapted from Unwin (36).]

the nicotinic receptor, too, there is evidence that the M2-M3 region is important in coupling the binding reaction to gating (40). Thus it seems that coupling does not occur directly through Ml, which does not appear to move significantly, but through an interaction of the M2-M3 loop with a part of the extracellular chain associated with the inner sheet, probably the loop between the /31 and j32 strands [see asterisk in Fig. 11.4 and Brejc et al. (25)] and/or the cys loop.

How does binding of ACh bring about the extended conformational change in the a sub-units, converting them into a non-a form? A likely possibility, consistent with the three-dimensional maps and also with the results of biochemical experiments using binding-site reagents (4), is that the C-loop is drawn inward by the bound ACh, bringing it closer to its location in the (non-a) protomer of AChBP. The joined outer sheet could in this way be reorientated and stabilized in the configuration it would have in the absence of subunit interactions, hence favoring a switch toward the relaxed, non-a form of the subunit. Whatever the precise details, the movements that result in the open/shut transition must be fast because it is known that the whole transition from shut to open takes less than 3 ¡jls (how much less is not known) to complete once it has started (41), and the channel often shuts briefly (average 12 jus) and reopens (see below).

Was this article helpful?

0 0

Post a comment