Joined: April 2007
One of Behe's favorite examples of an "intelligently designed" system are molecular motors, and "the" bacteria flagellum in particular. He claims that not only is it irreducibly complex, but it even resembles motors that humans build. His first claim has been easily refuted. His second claim has as well, but it is nevertheless the interesting bit here.
At first glance, the comparison has merit. It certainly looks similar in some ways to a human motor. But does it actually operate the same?
It turns out it probably does not. There are plenty of examples of where the design analogy serves not useful purpose (biology, physics, chemistry, astronomy, etc). However, here is a case where the design analogy seems to have significantly impeded scientific understanding of a subject. Only by going to back to the fundamentals, physics in this case, was the answer actually discovered. I apologize if this has already been covered, but I could find no mention of it.
This stems primarily from a paper by Alexander Vologodskii in Physics of Life Reviews last year, Energy Transformation in Biological Molecular Motors. The basic premise of the paper is simple. Molecular motors cannot operate like man-made motors because they don't have inertia. If you look at a car engine, for instance, a chemical reaction (gasoline exploding) pushes a piston down. The explosion and following expansion of gasses (the power stroke) is fairly brief. After that ceases there is no longer any force pushing on the piston. It relies on inertia to carry it through the rest of the cycle and ultimately start the cycle again.
This is fine on human length scales. However, on molecular scales this starts to become a problem. When you scale motor downs the inertia decreases more rapidly than the size of the motor due to rapidly increasing viscosity of water. By the time you get down to molecular size inertia is pretty much gone. As soon as you stop applying force the motor pretty much instantly stops (compared to the size of the motor itself). In order for the motor to operate in that manner a continuous force would have to be exerted, and that is not the case (ATP by its nature is a one-shot deal) nor would it be efficient to do it this way if it could be. Long-range interactions also do not work, so applying some sort of alternating push and pull like you see in man-made electrical motors does not work. This means that none of the principles used in human motors work. And attempts to look at the systems in question from that perspective have ultimately failed. This has not stopped people from calling it a "power stroke", a clear analogy to human machinery despite the fact that it is nothing like human machinery and can't really even be called a "stroke"
So then, how do molecular motors work when none of the principles we are familiar with from our machines work? As you probably guessed, they work by a method that is completely different than human machines. In the end, molecular motors are made of a relatively small number of molecules, mostly proteins. And one property that proteins have is that they can change their shape. This is known from enzymes, for instance, which can undergo allosteric changes when they bind to other molecules. Molecular motors take advantage of this. Of course human machines can change their shape to an extent. Piezoelectrics come to mind. But this works considerably differently, and no one would consider using piezoelectrics in the sort of roles molecular motors often fill (such as moving objects along a cable or rotorary motors). In essence, instead of the movement of the motor being caused by a chemical reaction, the movement of the motor is a chemical reaction. The binding of a ligand causes the equilibrium conformation of the protein to change. This causes the random thermal changes in the protein to tend towards a different state than they previously did, causing a net change in protein shape. Basically it acts like a rectifier for random molecular motion. Unbinding causes the equilibrium situation to change back, causing the protein's random shape changes to tend towards a different conformational state.
Imagine you had a cylinder full of balls and you wanted to move them to an adjoining cylinder connected by a long tube. The cylinders both have movable floors. Now the logical way to do this would be to raise the floor of the side with the balls and tilt it so the balls all roll out of the cylinder and along the tube into the other cylinder. But what if your balls cannot roll? And what if they are bouncing around the cylinder and never stopped? Then by raising the floor of first cylinder and lowering the floor of the other you can cause the probability that a ball will be in the original cylinder very small and the probability of it being in the other very large. However, moving the cylinders takes energy, and that is where the ATP or ion gradient or whatever power source is being used comes in. In essence you are dealing with a variant of "Maxwell's Demon", where the ligand binding or unbinding acts as the demon that traps a single fast-moving particle on one side of a partition or the other. Only it is a hungry demon and it uses far more energy than is gained from trapping the particles. The problem with these analogies is that it is not obvious how you would get such a system to do useful work, and certainly not how you could get them to do useful work in an efficient manner. Nevertheless details studies of the equilibrium energy states of single molecular motors show they do indeed change in a manner that would be expected from such a system. Further, analysis of enzymes show their conformational changes operate in much the same manner.
Now this article does not specifically mention bacterial flagella, but considering the length scales in question "the" bacterial flagellum cannot possibly operate similarly to a human motor, and most likely operate in a similar manner to this. Ultimately, instead of some specially-designed machinery operating in a similar manner to human machinery we have yet another a ubiquitous feature of proteins jerry-rigged to fit a new and not intuitively obvious role. So much for the argument from design.