On the nature of partial agonism in the nicotinic receptor superfamily

Abstract

Partial agonists are ligands that bind to receptors but produce only a small maximum response even at concentrations where all receptors are occupied. In the case of ligand-activated ion channels, it has been supposed since 1957 that partial agonists evoke a small response because they are inefficient at eliciting the change of conformation between shut and open states of the channel. We have investigated partial agonists for two members of the nicotinic superfamily-the muscle nicotinic acetylcholine receptor and the glycine receptor-and find that the open-shut reaction is similar for both full and partial agonists, but the response to partial agonists is limited by an earlier conformation change ('flipping') that takes place while the channel is still shut. This has implications for the interpretation of structural studies, and in the future, for the design of partial agonists for therapeutic use.

Figure 1

Glycine channels show short interruptions when activated by either full or partial agonists. a, Single-channel currents (opening upwards) produced by a full agonist (Glycine) or a partial agonist (Taurine). Open probability increases with agonist concentration (b) reaching a maximum of 0.54 for taurine and 0.96 for glycine (17 - 60 clusters, 3 - 4 patches per point; Hill equation fits; glycine data from ref. 2). c, Short shut times are clearly detectable in channel activations by 1 mM taurine. Points are the digitised record and the continuous line is the best time-course fit obtained during idealization together with the estimated duration of the short gaps. d, Short shut times are the predominant component in shut-time distributions for both agonists. With the usual resolution of 30 μs, a large proportion of such short shuttings are missed, but quite enough are observed to obtain a good estimate of their duration. Dwell-time distributions were fitted here and in Fig. 4d with a mixture of exponentials using ekdist, so they are descriptive and not mechanism dependent.

Figure 2

Global fit of the taurine data with a ‘flip’ mechanism provides a good description of the observations. a, The ‘flip’ activation mechanism fitted by Burzomato et al. (2004) to α1β channel activations by glycine. The success of the fit to the idealised records is judged by its ability to describe the experimental observations displayed in various ways, as in b and c. The results of the fit predict accurately the channel open probability (b) and apparent open- and shut-time distributions at all concentrations as shown by the continuous lines (fit predictions) superimposed onto the data points and the histograms. The dashed lines in the histograms are the distributions predicted if the records were obtained with perfect resolution (i.e. no events were missed; full data display Supplementary Figure 1 and Supplementary Table 1).

Figure 3

Activation of the glycine channel by the full agonist glycine and the partial agonist taurine. a, The three states of the fully-saturated receptor. The resting (shut) channel changes conformation to reach a partially-activated state (‘flipped’), still shut, but with increased affinity for the agonist (blue ellipses). It is from this flipped state that the channel can open (red, right). b, The reaction rates for channel activation by glycine and taurine differ largely in the first step, the flipping. Relative rates are indicated by the size of arrows. The diagram shows also the mean lifetime of each state (μs) and the proportion of time (percent) spent by a bound channel in each of the states at equilibrium. A channel occupied by glycine doesn’t stay long in the resting state, but quickly flips. When taurine is bound, the channel takes a long time to flip, so it spends much more time in the closed resting state. Once the channel has reached the flipped intermediate state, it opens quickly, regardless of which agonist is bound. The probability of opening, rather than unflipping, is over 96% for both agonists, so a burst of many openings results. c, An energy diagram for the three states of the saturated receptor, for full agonist (glycine, orange) and partial agonist (taurine, green). Again, the difference lies largely in the resting-flipped transition. As indicated by arrows, the transition from resting to flipped is downhill for glycine, but uphill for taurine. The overall transition from resting to open is very downhill for glycine, but almost level for taurine, as expected from the maximum response of about 54% open. The calculations used a frequency factor of 10⁷ s⁻¹ and the lines are shifted vertically so they meet at the open state., See also Supplementary movie.

Figure 4

Short interruptions in openings of muscle nicotinic channels occur with both partial and full agonists and cannot be attributed entirely to channel block. a, Single channel currents produced by a full agonist (ACh) and a partial agonist (TMA, openings downwards in the first six rows). b, Short shut times are a major feature of shut-time distributions irrespective of agonist, voltage and concentration (see Supplementary Fig 2 and 4). Some of these brief shuttings could result from fast open-channel block at negative potentials, but not at positive potential, when open channel block is essentially absent. c, Open probability reaches a maximum of 0.94 (± 0.004) for ACh at −100 mV but is lower (0.59± 0.01) for ACh at +80 mV and for TMA at −80 or + 80 mV, 0.78 ± 0.05 or 0.16 ± 0.02, respectively (2 - 4 patches per point; Hill equation fits). These fits give also, for ACh −100 mV: EC₅₀= 5.3± 0.2 μM; n_H= 1.7± 0.1; for ACh +80 mV: EC₅₀= 29± 2 μM; n_H= 1.7 ± 0.1; for TMA −80 mV: EC₅₀= 2.2 ± 0.5 mM; n_H= 1.3 ± 0.4, and for TMA +80 mV: EC₅₀= 6.4 ± 1.9 mM; n_H= 1.2 ± 0.4. d, Because of open-channel block, the amplitude of single-channel currents declines with increasing TMA concentrations at negative potentials (open circles), but not at positive transmembrane potentials (filled circles, 2-3 patches per point).

Figure 5

The ‘flip’ mechanism describes well the activation of acetylcholine receptors by TMA. a, Schematic representation of the ‘flip’ mechanism used here: this has two agonist binding sites. As in Figure 2, the experimental data are displayed as channel open probability values (b) or dwell-time distributions (c, d). The results of fitting the idealised records are used to calculate the predicted P_open - concentration curves and distributions (continuous lines). Both the open probability values and the dwell time distributions are well described by the fitted rate constants for TMA, at both −80 mV (c) and +80 mV (d). The dashed lines in the histograms are the distributions predicted if the records were obtained with perfect resolution (i.e. no events were missed). The full data display for the fits for ACh and TMA is in Supplementary Figures 2, 4, 5 and Supplementary Table 2).