Living cells have molecular machines for free energy conversion, for example, sliding machines in muscle and other cells,¯agellar motors in bacteria, and various ion pumps in cell membranes. They are constructed from protein molecules and work in the nm (nanometer), pN (piconewton) and ms (millisecond) ranges, without inertia. In 1980s, a question was raised of whether the input±output or in¯ux±ef¯ux coupling in these molecular machines is tight or loose, and an idea of loose coupling was proposed. Recently, the long-distance multistep sliding of a single myosin head on an actin ®lament, coupled with the hydrolysis of one ATP molecule, was observed by Yanagida's group using highly developed techniques of optical microscopy and micromanipulation. This gave direct evidence for the loose coupling between the chemical reaction and the mechanical event in the sliding machine. In this review, I will brie¯y describe a historical overview of the input±output problem in the molecular machines of living cells.
Input±output in molecular machinesConsider a series of solid gears. The ratio of the angle of rotation of a gear in the input to the angle of rotation of another gear in the output is constant and independent of the speed of rotation or the torque for rotation. In such a way, are the input and output or in¯ux and ef¯ux in molecular machines of living cells tightly coupled? Does one chemical reaction in the input always generate a unit mechanical movement in the output?The molecular machines are very small, so that the input free energy is not much larger than the energy of thermal noise. Nevertheless, the machine is neither cooled nor isolated, but works at room temperature in water. The structural units of the machine, the protein molecules, are not rigid, but have thermal¯uctuation. The energy exchange among these units and their environment may be positively involved in the process of free energy conversion. The relation between input and output in the machine would not be straightforward. Intermediate processes in the machine would make the in¯ux±ef¯ux coupling loose. It is very likely that living cells invented the loose coupling mechanism in order to give an ef®cient conversion of small free energy in thermal noise.To understand the working principle of molecular machines, it is important to know, ®rst of all, their function. We have to de®ne their input and output, or in¯ux and ef¯ux, and carry out quantitative investigations on the relation between them under various conditions.