Descending past mere irritants and past disfiguring killers, we arrive at the bottom of the pit. These are compounds that are to humans what a spray-can of insecticide is to flies.
I mean that literally. Back in the 1930s, a group at IG Farben in Germany was searching for new classes of compounds to kill insect pests. After trying out several classes of organofluorines, Gerhard Schrader's lab made a phosphoramide fluorine derivative in 1935. That was a pretty potent compound, and a whole new area of research opened up.
Two days before Christmas of 1936, Schrader used the fluorine intermediate to make the first compound of the type we know as nerve gas. This was what's now called Tabun or GA, and it was the one of the most potent insecticides the lab had produced. After the Christmas holidays, he and his lab assistant were continuing their work when they noticed that they were getting short of breath, and that their vision was dimming. They evacuated the lab, which was a good call - a few minutes more would surely have killed both of them.
Those side effects are among the earliest signs of nerve gas poisoning, no matter what the agent. That's because they all work by the same mechanism. Some of the compounds are easier to make than others, some are more potent (so you don't have to make as much,) and some are more stable (so you can keep them in storage until you feel the need to commit mass murder.) But all of them do the same thing: irreversibly inhibit esterase enzymes.
A little-known fact is how broad-spectrum that activity is, and how little that matters. There are probably dozens of enzymes that a nerve agent shuts down in vivo, and this wholesale disruption would probably cause death in hours or days. Those pathways never get the chance, though, because the enzyme that counts is acetylcholinesterase.
Here's why: a number of different compounds are used in the nervous signaling for neuron-to-neuron crosstalk, but real workhorse is a small one called acetylcholine. It's made in the neuron, stored in vesicles up near the cell surface, and released to float across the synapse. Once it makes it across, it binds to one or more members of two families of proteins (muscarinic receptors and nicotinic receptors.) That docking sets off further signaling inside the receiving neuron. Fellow medicinal chemists and biologists know this as a prime example of a G-protein coupled receptor mechanism; it's a theme that shows up in many other signaling pathways.
The thing is, a signal across the synapse isn't a continuous current. It's a pulse across a gap, and when the signal has been received, the synapse has to be cleared. That's the job of the acetylcholinesterase enzyme. It's extremely efficient at breaking down acetylcholine, insuring that the signaling pathway doesn't stay switched on.
And nerve agents are extremely efficient at deactivating the enzyme. One molecule of nerve gas, if it makes it to the enzyme, will shut it down. When you consider that each enzyme molecule would otherwise turn over thousands upon thousands of acetylcholines - well, things get out of hand very quickly. The acetylcholine piles up in the synapse, causing all the receptors on the receiving neuron to get switched to an unnatural full-on overload. The entire nervous system goes down within minutes (at best) under these conditions - no interpretable signals to the muscles can get through at all. The limbs, the heart, the lungs all shut down or spasm uncontrollably.
Schrader and his assistant felt what they did because those organs were the first to be affected by the Tabun vapors, which were absorbed by the moist tissues of the eyes and taken up through their lungs. Their intercostal muscles were being partially inactivated (shortness of breath,) and the blast of acetylcholine signaling switched on the M1 muscarinic receptors in their pupillary muscles, causing them to contract. [Note added later - the shortness of breath was more likely due to bronchial effects, or the beginning of effects on the diaphragm muscle. The effects on the eyes are complex, probably involving both m1 and m3 receptors.] Further exposure starts to affect other muscle groups, and you get a mixture of muscarinic and nicotinergic symptoms. The only people who can tell you how this feels live in Japan (thanks to Aum Shinrikyo) and in northern Iraq (thanks to Saddam Hussein.) I should warn you, the New Yorker article that link goes to is very difficult to take. It's vital reading for an understanding of chemical warfare in Iraq, but it'll give you bad dreams.
Another thing that isn't widely known is that cholinesterase inhibition actually has positive medical uses. It's one of the few therapies now available for Alzheimer's disease, for example. The idea, which is admittedly a crude one, is to crank up the volume of the brains' acetylcholine signaling to compensate for the damage of AD. It works, a little, for a while. Of course, the sorts of drugs you use for this therapy need to be a bit less. . .efficacious than nerve gas. Ideally, they're weak, reversible inhibitors of the enzyme (as opposed to butt-kicking irreversible ones,) and they should tend to concentrate in the brain while getting cleared from the rest of the body.
The dose makes the poison, indeed. We'll return to Gerhard Schrader in the next article, after he learned to treat his compounds with more respect.
[Post edited after inital version - cleaned up some pharmacology and added links.]