The complement system.
For a biochemical switch to be useful you need
1. A mechanism to detect a change in the environment of a cell
2. An alternative action/pathway to switch to when the change is detected.
The switch on its own is no good if is doesn’t activate some kind of response to the change it detects.
The alternative "something to do" however is no use unless when it is needed you can switch it on. Usually the action to be switched on is a function which is needed only in certain conditions. Thus for it to work you need the switch and the function together. It is a logical "IF-THEN-ELSE" function. This is the central idea of Irreducible Complexity (IC) which Behe raised as a possible candidate for falsification of RMNS as the sole generator of biological complexity and as an indicator of intelligent design.
Switches (according to Behe) with their two alternative responses and some kind of detector system are therefore a good sign of intelligent design.
The complement system (as I understand it) is a system for blowing up cells. Blowing things to pieces is a dangerous job and needs to be handled by careful operators.
For an animal to obtain a system for blowing cells to pieces (bear in mind that an animal is made up of cells) is extraordinary.
To blow up a cell you only need to make a suitably sized hole in it. The rest happens as a matter of course.
Let us start from the position of having a fully functional cell blasting system which is triggered by molecules usually associated with bacteria detected by innate unchanging proteins sticking to carbohydrate molecules only found on the surface of bacteria. These are what I am calling the “dumb immigration officers.” The “dumb immigration officers” don’t do anything very clever they just look for sugar molecules with a foreign appearance and stick to them.
The switch in these systems is a switch between a “snipper” protein with a protective cover and a “snipper” protein that is going around snipping things.
The conversion to an active “snipper” is caused in the “dumb immigration officer” system by a protein complex sticking to certain sugar type molecules only found on bacteria surfaces.
See the molecules involved here
The solution to the evolution of the “intelligence led immigration officer” system is that the same “snipper” switch (presumably by gene duplication and mutation) becomes activated by antibody which has stuck to a foreigner rather than by sticking to bits of the foreigner itself.
Behe asks whether it is reasonable to believe that this development of the cell blasting system together with the “intelligence led immigration officer” system can be explained by a gradual step by step process of incrementally increased function.
The problem with fiddling around with the binding site of one of these snippers is that you are fiddling around with explosives! The only binding sites that will be useful are those which are unique to foreigners….and if you get things wrong you have pressed the suicide button.
Is it reasonable to think of an organism experimenting with a duplicated "snipper" gene until it hits on a binding site that binds antibody stuck to a foreigner? My gut response is that this is not going to work but I suppose it depends on the frequency of useful binding sites as opposed to the lethal ones.