Justifying total synthesis

Ernest Rutherford famously said that all science is either physics or stamp collecting. Over the years, plenty of people have put forward a similar point about total synthesis – that other parts of organic chemistry are just there to support the grand enterprise of replicating nature’s molecular structures. Working out mechanisms and inventing new reactions is useful, of course – but only because they can help to make natural products. You know, real chemistry.

If you go back to the days of R B Woodward, whose status as the major deity of synthetic chemistry is still unchallenged, I’d say that attitude was probably justified. Often the only way to be sure of a compound’s structure was to synthesise it, and a good number of new reactions came out of the work done along the way. The most difficult problems (and the most innovative and interesting solutions) were all to be found making natural products from scratch. But is that still the case? 

I should disclose that my own PhD work was in the field. My views changed a bit while I was still doing the work, but they’ve evolved most while I’ve been in industry, and I can’t help but think that the world has moved on. If you look at the classic reasons for doing total synthesis, I just don’t see how they hold up these days. 

Advances in instrumentation have largely wiped out the proof-of-structure argument, for one. I know that there are still cases where natural products have been wrongly assigned, but how often has this happened in recent years for molecules of any importance? I’ll grant this reason a bit of weight, but not enough to carry the entire field. And as for discovering new reactions, it’s my distinct impression that most new ones are now found by people who are trying to find them, rather than people who are trying to make a specific molecule. We have so many synthetic techniques now, compared to the Woodward days, that most molecules can be banged out one way or another if you’re willing to throw enough post-docs at them. Discovering new reactions just slows things down. 

There’s no doubt that total synthesis is difficult. Natural products can be ridiculously complicated, and I mean that literally: I’ve seen chemists start laughing as they see a new structure. But it’s a nervous sort of laughter, because they’re thinking of the people who will try to make the compound from the ground up, and what they’re going to have to go through. Horrible stories are everywhere in the field’s history, with people spending years of their lives working on huge synthetic sequences that have to be junked because of an insurmountable problem at (say) step forty-five, or going through fiendish contortions to make what turns out to be a wrongly-assigned structure. My own problems weren’t in that league, but I still have stories to tell that I haven’t yet learned to laugh at. 

The sheer difficulty of the enterprise is traditionally what made pharmaceutical companies hire people who had worked in total synthesis. It gives you an excellent all-around fluency in synthetic methods, and (even more importantly), it teaches you to overcome adversity and to solve problems. Drug discovery has plenty of adversity for everyone, and it’s better to select for chemists who already know what it’s like to have things fail over and over. 

So total synthesis does have its uses. But R B Woodward never said anything about providing the drug industry with good hires, and it’s doubtful if he would have considered that enough of a reason by itself. But is there another way to prepare chemists for the rigours of inventing new medicines?

None of this stops total synthesis being an interesting field. It still consumes a lot of energy and a lot of funding, and some of the people doing it have just as much attitude as ever. It’s just a lot harder to justify than it once was. 

Derek Lowe is a medicinal chemist experienced in preclinical drug discovery