The watery interior of a cell is separated from the watery exterior of a cell by a thin double layer of lipids called the plasma membrane. The oily interior of this membrane prevents water and charged molecules from escaping the cell, while allowing hydrophobic (oil-like) molecules through. This system has many significant advantages, but cells frequently need to move charged atoms (ions) across the membrane. This job is primarily performed by two kinds of protiens: channels that create specialized tunnels through the membrane, and transporters that mechanically transfer ions across the membrane. These are distinct activities, usually carried out by different families of proteins. Recent results in the ClC family of membrane proteins, however, have demonstrated that these activities are not as distant from each other as it might seem.Ion channels basically work like tiny pipes that stick through a cell membrane. They have interior pores that are full of water, and usually possess some form of selection mechanism that lets only a particular kind of ion through. Although they can be gated — opened or closed by particular voltage states or molecular signals — they can only move ions with an electrochemical gradient. That is to say, they can only allow their particular ions to move across the membranes in a way favored by both concentration and voltage. By contrast, transporters physically translocate ions without using a watery pore. This allows them to move ions against a concentration gradient as long as they have an energy source, such as another ion's concentration gradient.
The Aside from dealing in ions, these two kinds of activity might seem to have little in common, and in fact most ion channels are not very closely related to ion transporters. However, the ClC family of chloride channels is unusual in that it also includes several members that are transporters. Because all these proteins are presumed to have similar structures in the membrane, there is considerable interest in understanding the key differences that separate ClC channels from ClC transporters. In the case of these proteins, it seems that the line between transporter and channel is easily blurred.
You got transporter in my channel!
The ClC-0 protein from the electric ray Torpedo marmorata has always been classified as a gated channel, but an odd one. Unlike many channels, it has two pores in its active configuration. These pores are closed off or gated by two processes. The fast gating occurs on the millisecond timescale and opens and closes the two pores independently of one another. The slow gating occurs on the timescale of seconds and opens and closes both pores.
The odd thing is that there is a thermodynamic imbalance in this system. All things being equal, we expect that if we track the number of pores that are open over time, we should see the pattern 1 → 0 → 2 → 1 (called J+) with the same frequency that we see the pattern 1 → 2 → 0 → 1 (J-). Instead, researchers have found that J+ is observed more frequently than J-. Originally it was thought that the electrochemical gradient of chloride controlled this asymmetry, but this model didn't quantitatively match the observations. Now Lísal and Maduke have shown that the asymmetry arises from the proton electrochemical gradient, and that this is a molecular vestige of ClC antiporter function (1).
The evidence for this comes from an experiment in which the chloride electrochemical gradient was held constant while the proton electrochemical gradient was changed. If the chloride gradient controls the asymmetry, we should see no differences in the J+/J- ratio during the experiment. Instead, it was observed that switching the direction of the proton gradient changed the behavior of the channel gating from almost exclusively J+ to an even mixture of J+ and J-. Further experiments demonstrated that the J+/J- ratio was proportional to the proton electrochemical gradient, and that it leveled off at 1 when the direction of the gradient was changed to favor movement of protons from inside the cell to outside the cell. This suggests that the gating mechanism is unidirectional.
Because the proton gradient provides energy to produce gating asymmetry, CLC-0 must be a proton transporter as well as a chloride channel. The authors suggest that this reflects a vestigial antiporter activity (similar to several existing members of the family) that has been repurposed as a regulatory mechanism for channel gating. Although channels have very different thermodynamics from transporters, it appears that a single protein can have characteristics of both.
You got channel in my transporter!
But how does an antiporter transform into a channel? Because the structures of these different types of proteins tend to be quite dissimilar, it seems unlikely that the transition between these mechanisms could be accomplished with just a few point mutations. However, sequence homology in a family of proteins usually implies structural homology (with some exceptions), so the ClC channels probably evolved from antiporters in this fashion. This could happen by the destruction of gating mechanisms in the transporter.
In the Cl-/H+ antiporter CLC-ec1, a central binding site for chloride is blocked by two gates. One side is blocked off by Glu148, and the other is putatively formed by the interaction of Tyr445 and Ser107. In principle, removing these gates could give rise to a watery passage through the membrane, and high transport rates similar to those of ion channels. Jayaram et al. therefore performed a series of mutations at these sites and measured the effect on transport rates (2).
From previous work it was known that altering these residues uncoupled chloride transport from proton transport, which is one step towards becoming a channel. Here, Jayaram et al. find that mutating Tyr445 to a smaller side chain does not have a significant effect on chloride transport rates. Mutating Glu148 to smaller residues actually decreases the chloride transport rates. Mutating both residues, however, leads to a 100-fold increase in chloride transport over the wild-type protein. The fastest mutant, moving chloride ions at a rate of more than 35,000 /s, is not quite as fast as a channel. Still, the acceleration is significant.
Jayaram et al. found that increasing the size of the substituted side chain at either site decreased the rate. This is what you would expect for a simple case of larger side chains leading to more constriction of a channel. In order to confirm that there was such a channel, they crystallized one of the mutants and analyzed it to determine whether there was a continuous pathway that water could permeate. Below, you can see a figure I shamelessly stole from their paper, with the WT protein on the left and the E148A/Y445A mutant on the right. The red dots represent a surface made with a 1.4 Å probe designed to mimic water. As you can see, the probe can go all the way through the A/A mutant, but is stopped by the intracellular gate in WT. The double mutant is a tight, but genuine channel.
Because these residues are conserved in both channel and antiporter members of the ClC family, mutations like these are not likely to be the means by which one kind of protein evolves from the other. Nonetheless, they establish that the crystal structure of CLC-ec1 is likely to be a good model for occluded states of gated ClC channels. Moreover, the ease with which a specialized ClC transporter is made into a channel suggests that a progenitor protein could have switched from antiporter to channel in just a few mutations.
The thermodynamics of ion translocation by channels and transporters are quite different, so it was a surprise to discover that the ClC family contained both kinds of activities. These recent papers show that the entanglement is even closer than previously thought. The ClC channel CLC-0 still retains vestigial proton transport activity, and the ClC antiporter CLC-ec1 is only a few mutations away from becoming a channel itself. These findings suggest that substantial changes in operating thermodynamics may result from small evolutionary steps, and point to a shared antiporter past for members of the ClC family.
1. Jiří Lísal, Merritt Maduke (2008). The ClC-0 chloride channel is a 'broken' Cl−/H+ antiporter Nature structural & molecular biology, 15 (8), 805-810 DOI: 10.1038/nsmb.1466
2. H. Jayaram, A. Accardi, F. Wu, C. Williams, C. Miller (2008). Ion permeation through a Cl--selective channel designed from a CLC Cl-/H+ exchanger Proceedings of the National Academy of Sciences, 105 (32), 11194-11199 DOI: 10.1073/pnas.0804503105
Disclaimer: I work on the same floor as all these guys.
No comments:
Post a Comment