Another chemical element that you don't see much in pharmaceuticals is silicon. Hey, it's right under carbon in the periodic table, and forms four tetrahedral bonds just like carbon does, so why not, eh?

Now, if you're like me, you grew up reading old science fiction stories that posited silicon-based life forms. That seemed pretty plausible to me when I was a kid, and rather a long shot as I got older, but learning chemistry for real made me realize just how unlikely that is. For one thing, silicon-silicon bonds get progressively weaker as you try to make longer and longer chains, as opposed to carbon chains, where there's no real effect. Silicon's more unstable to oxidation than carbon is, too. If you open up a tank of methane, it'll just hiss all over the room. But if you open up a tank of silane, you'd better have the fire department on the line already.

And silicon doesn't form double bonds very well at all, not with itself or carbon (which means, practically speaking, no alkenes and no aromatic rings) or even with oxygen (which means no analogs of amides, for one big thing.) It gives you a new appreciation for carbon, it does.

Your nose can tell that there's something off about the element. It isn't fooled by its position in the periodic table. Many organosilanes have a distinctive, hard-to-describe smell, a sort of flat, spicy, camphor-like reek, and this smell persists over a fairly wide range of structures that normally would be enough to mask it.

But sulfur smells like Satan's socks, and it's vital. There's no problem with working some single-bonded silicon into your molecules, at least on paper. Reasonable organosilanes are stable to normal sorts of things, and there's no particulary toxicity associated with the element. When I was doing my post-doc in Germany, I even saw ads for silicon-containing supplements, which claimed that it was vital for health. That's pushing it, to say the least, but at least it's not vital for sickness.

There sure aren't many examples, though. I'm virtually certain that no human drug has ever been marketed with a silicon atom in it. DuPont actually took a fungicide to market with one, but pharmaceutical chemists look a bit askance at what the crop science folks can get away with. (Where's the challenge, we keep thinking a bit unfairly, in dosing something that doesn't have a gut or a liver?)

There was a cholinesterase inhibitor in development a few years ago with a silicon, and recently there have been some reports of organosilane-based protease inhibitors. A few other such one-offs show up in the literature. From the scattered reports, you can tell that folks have every so often worked up the nerve to take one into the clinic, but nothing's made it all the way through. That keeps many teams from making a big effort, frankly. Who wants to be the first to find out that there's a problem with, say, liver enzymes after ten years of dosing? Most companies would rather let someone else turn over that card.

I've made a silicon analog or two myself over the years, and reaction from my colleagues and supervisors has been, well, mixed. Some fans of the weird cheered the compounds on when they saw them, while other people rolled their eyes almost audibly. None of the compounds were active enough to force any issues, though.

But one small English company is trying to break the silaceous ice, targeting silicon compounds for pharmaceutical use specifically because they believe they've been underexplored. Good look to Amedis of Cambridge, I say. Perhaps they can make the element respectable.