Single molecule FRET reveals pore size and opening mechanism of a mechano-sensitive ion channel

Abstract

The mechanosensitive channel of large conductance, which serves as a model system for mechanosensitive channels, has previously been crystallized in the closed form, but not in the open form. Ensemble measurements and electrophysiological sieving experiments show that the open-diameter of the channel pore is >25 Å, but the exact size and whether the conformational change follows a helix-tilt or barrel-stave model are unclear. Here we report measurements of the distance changes on liposome-reconstituted MscL transmembrane α-helices, using a ‘virtual sorting’ single-molecule fluorescence energy transfer. We observed directly that the channel opens via the helix-tilt model and the open pore reaches 2.8 nm in diameter. In addition, based on the measurements, we developed a molecular dynamics model of the channel structure in the open state which confirms our direct observations.

Abstract

Bacterial cells are full of fluid, and they will burst if they are not able to respond to a build up of pressure. Fortunately, the membrane of a bacterial cell contains channels that can detect the increased mechanical stress on the cell membrane and then open to relieve the pressure.

In many bacterial cells, the last defence against the cell exploding is called the mechanosensitive channel of large conductance (MscL). This is made of five proteins, each of which consists of TM1 and TM2 helixes, which are responsible for opening and closing the channel. Two models have been proposed to explain how the channels are opened. In the barrel-stave model, the TM1 helix moves, while the TM2 helix remains stationary. This results in an open pore that is lined with TM1 and TM2 helixes in the same way that wooden staves line a barrel. In the helix-tilt model, both helixes tilt towards the membrane to open the channel.

Wang et al. have now used a technique called single-molecule fluorescence resonance energy transfer (FRET) to explore the structure of the open channel in E. coli in order to determine which model is correct. In this technique an individual channel is labeled with two different fluorescent molecules. By illuminating the channel with light of a wavelength that excites the first fluorescent molecule, and measuring the strength of the fluorescence from the second molecule, it is possible to work out the distance between the two molecules. From this, the structure of the channel and how it opens and closes can be explored.

Previous attempts to measure the diameters of open channels using fluorescence techniques have suffered from issues caused by the use of large numbers of fluorescent molecules. This has made it necessary to use computational modeling to extract the required data. By looking at a series of individual proteins, Wang et al. overcame these problems and found that the diameter of the fully open pore is 2.8 nm. The result provides strong support for the helix-tilt model.