The Baran group at Scripps has a whopper of a total synthesis out in Science. They have a route to the natural product ingenol, which is isolated from a Euphorbia species, a genus that produces a lot of funky diterpenoids. A synthetic ester of the compound as recently been approved to treat actinic keratosis, a precancerous skin condition brought on by exposure to sunlight.
The synthesis is 14 steps long, but that certainly doesn't qualify it for the "whopper" designation that I used. There are far, far longer total syntheses in the literature, but as organic chemists are well aware, a longer synthesis is not a better one. The idea is to make a compound as quickly and elegantly as possible, and for a compound like ingenol, 14 steps is pretty darn quick.
I'll forgo the opportunity for chem-geekery on the details of the synthesis itself (here's a write-up at Chemistry World). it is, of course, a very nice approach to the compound, starting from the readily available natural product (+) 3-carene, which is a major fraction of turpentine. There's a pinacol rearrangement as a key step, and from this post at the Baran group blog, you can see that it was a beast. Most of 2012 seems to have been spent on that one reaction, and that's just what high-level total synthesis is like: you have to be prepared to spend months and months beating on reactions in every tiny, picky variation that you can imagine might help.
Let me speak metaphorically, for those outside the field or who have never had the experience. Total synthesis of a complex natural product is like. . .it's like assembling a huge balloon sculpture, all twists and turns, knots and bulges, only half of the balloons are rubber and half of them are made of blown glass. And you can't just reach in and grab the thing, either, and they don't give you any pliers or glue. What you get is a huge pile of miscellaneous stuff - bamboo poles, cricket bats, spiral-wound copper tubing, balsa-wood dowels, and several barrels of even more mixed-up junk: croquet balls, doughnuts, wadded-up aluminum foil, wobbly Frisbees, and so on.
The balloon sculpture is your molecule. The piles of junk are the available chemical methods you use to assemble it. Gradually, you work out that if you brace this part over here in a cradle of used fence posts, held together with turkey twine, you can poke this part over here into it in a way that makes it stick if you just use that right-angled metal doohicky to hold it from the right while you hit the top end of it with a thrown tennis ball at the right angle. Step by step, this is how you proceed. Some of the steps are pretty obvious, and work more or less the way you pictured them, using things that are on top of one of the junk piles. Others require you to rummage through the whole damn collection, whittling parts down and tying stuff together to assemble some tool that you don't have, maybe something that no one has ever made at all.
What I like most about this new synthesis is that it's done on a real scale. LEO Pharmaceuticals is the company that sells the ingenol gel, and they're interested in seeing if there's something better. That post from Baran's group shows people holding flasks with grams of material in them. Mind you, that's what you need to get all these reactions figured out; I can only imagine how much material they must have burned off trying to get some of these steps optimized. But now that it's worked out, real quantities of analogs can be produced. Everyone who does total synthesis talks about making analogs for testing, but the follow-through is sometimes lacking. This one looks like it'll be more robust. Congratulations to everyone involved - with any luck, you'll never have to do something like this again, unless it's by choice!
Update: here's more from Carmen Drahl at C&E News.